Malononitrile as the 'double-edged sword' of passivation-activation regulating two ICT to highly sensitive and accurate ratiometric fluorescent detection for hypochlorous acid in biological system.

Hypochlorous acid (HOCl) as one of the most important reactive oxygen species in the organism, its role is more and more recognized. In fact, in recent years, various HOCl fluorescent probes have been developed unprecedentively based on various mechanisms. However, because most of the mechanisms are based on the oxidation characteristics of HOCl, the excellent detection performance of probes depends on the activation ability of some functional groups to reaction sites. The C[bond, double bond]C bond in the probe is often oxidized by HOCl to realize HOCl detection. However, due to the break of conjugated structure, the probe often present as a quenchable or turning on fluorescence emission. In this work, malononitrile was introduced as the "double-edged sword" of passivation-activation when in HOCl fluorescent probe was designed. Passivation-activation regulated two ICT (Intermolecular Charge Transfer, ICT) processes to ratiometric fluorescent detection for HOCl. Highly sensitive and accurate detection realized efficient application in biological imaging.

Introduction

Reactive oxygen species, keeping a certain concentration range under normal physiological conditions, had been proved to play an important role in maintaining cell morphology and maintaining basic cell function [1]. It was generally believed that HOCl can be obtained in organisms by hydrogen peroxide and chloride ions catalyzed by myeloperoxidase (MPO) [2]. As one of the representative substances in reactive oxygen species, HOCl played an irreplaceable role in daily life and vivo [3]. On the one hand, HOCl was used as bleach, disinfectant, deodorant, etc [[4], [5], [6], [7]]. Especially, HOCl played a crucial role as the main disinfectant on the novel coronavirus (novel coronavirus (2019-nCoV)) incident. On the other hand, the excessive HOCl could cause various diseases such as cardiovascular disease [8], arthritis [9], arteriosclerosis [10] and cancer [11,12]. Therefore, it was of profound significance to realize accurate detection of HOCl in vivo [[13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]].

Anisaldehyde, as an important intermediate in organic synthesis, was widely used in medicine, food and other fields [[33], [34], [35], [36], [37]]. To as we know, anisaldehyde could be as a spice in condiments in China, which fully demonstrated anisaldehyde owns particularly biocompatible and lower toxic.

Due to its attractive properties by controlling the emission wavelength of the probe molecules through the donor-acceptor system, dyes based on dicyanoisophorone (DCO) have received widespread attention [36,37]. Besides, the CC bond of (DCO) was often used as the reaction site of HOCl, which could be oxidized to aldehydes or ketones by HOCl under mild conditions [[38], [39], [40], [41]]. Due to the change of molecular structure, the molecular realized the ratiometric fluorescence detection to HOCl [42]. Therefore, it was a wise choice to design a ratiometric HOCl fluorescent probe combining anisaldehyde and DCO.

Considering the above, we synthesized a novel HOCl fluorescent probe () based on anisaldehyde (donor) and DCO (acceptor) fluorescent dye (Scheme 1 ). Owing to the presence of electron-withdrawing CN group and electron-donating group —OCH3, the ICT-1 of probe was triggered to show obvious fluorescent signals. However, the electron-withdrawing CN group of probe attacked by HOCl, the ICT-1 process from —OCH3 moiety to CN group was disturbed. Simultaneously, the ICT-2 process was triggered from —OCH3 moiety to electron-withdrawing ketone group, which the fluorescence spectra displayed ratiometric manners to HOCl in solution.

Scheme 1

The synthesis route of the probe .

At the same time, the spectral test results were consistent with the above analysis, probe showed good selectivity and low detection limits. Most importantly, the probe had low toxicity and high specificity for HOCl, which could achieve in vivo and in vitro monitoring of HOCl.

Material and methods

The preliminary of experiments

The details of the material, setting of the spectra and preparation of the imaging experiment were available in the supporting information.

The preparation and characterization of AI

The synthetic route of was showed in Scheme 1. was synthetized according to the literature report [[43], [44], [45]]. was characterised by NMR and HR-MS: 1H NMR (600 MHz, DMSO-d):δ 7.67 (d, J =7.6 Hz, 2 H), 7.28 (s, 2 H), 6.99 (d, J =7.6 Hz, 2 H), 6.84 (s, 1 H), 3.80 (s, 3 H), 2.61 (s, 2 H), 2.54 (s, 2 H), 1.02 (s, 6 H) (Fig. S1); 13C NMR (151 MHz, DMSO-d 6): δ 170.83, 161.05, 156.96, 138.19, 130.94, 130.10, 129.11, 127.69, 123.16, 122.27, 114.92, 114.53, 113.71, 75.65, 55.59, 42.76, 38.62, 31.90, 27.91, 27.57(Fig. S2); HR-MS m/z: calcd for 304.15756; found: m/z: 304.15681(Fig. S3).

Results and discussion

Theoretical calculation

In our original design, the electron-withdrawing CN group of probe was attacked by HOCl, and then weaker electron-withdrawing ketone group was formed. Due to the decrease of the ability of electron absorption, the wavelength of absorption and emission would be blue shifted. Therefore, we selected the Gaussian 09 program to predict the optical properties of . By the density functional theory (DFT) and B3LYP/6-31G(d) method, the optimized structure and energy levels of and were obtained (Fig. 1 ). The LUMO (Lowest unoccupied molecular orbital) and HOMO (Highest occupied molecular orbital) of and were uniformly distributed throughout the molecule. Compared with , the HOMO and LUMO of increased, and the gap enlarged, which was caused by the decreased electron-absorbing ability from -CN to carbonyl. Based on these results, we initially believed that could be considered as an efficient ratiometric HOCl fluorescent probe.

Fig. 1

Theoretical calculation of HOMO / LUMO energy gaps of and .

UV–vis and fluorescence spectra of AI titrated with HOCl

In order to study the optical response of the probe to HOCl, the detection methods of UV–vis and fluorescence spectra were carried out (Fig. 2 a). Firstly we analyzed the UV–vis spectra of probe upon increasing the concentration of HOCl in DMSO/PBS (1/1, v/v, pH = 7.4). As Fig. 2a showed, the absorption peak of 417 nm gradually disappeared, meanwhile the absorption peak of 345 nm gradually increased and an equal absorption point appeared at 360 nm with increasing of HOCl, which meant that reacted with HOCl to generate a new substance (). Subsequently, fluorescence property of was also studied in DMSO/PBS (1/1, v/v, pH = 7.4). As Fig. 2b showed, displayed maximum emission peak appeared at 570 nm under excitation at 370 nm in the system. When HOCl was added gradually, the emission peak at 570 nm disappeared, while the emission peak at 495 nm strengthened gradually, showing a good response to HOCl. The reasons for this phenomenon could be attributed to the ICT-1 process form —OCH3 to DCO group was interrupted and that the ICT-2 process from —OCH3 to ketone group was formed. Based on the above results, could be used as a ratiometric probe for identifying and detecting HOCl.

Fig. 2

Absorption spectra of (10 μM) with increasing HOCl (0–60 μM) (a) and (b) fluorescence spectra of (10 μM) with increasing HOCl (0–60 μM) (λex =370 nm) in solution (DMSO:PBS = 1:1, pH = 7.4).

The selectivity of AI

In order to verify the selectivity of , some representative analytes, such as (H2O2, NO2 −, ONOO−, T-BuOO−, 1.0 mM), other amino acids including (Cys, Hcy, GSH, L-Glu, L-Phe, L-Gly, L-Asp, L-Met, L-Ala, L-Thr, L-Try, 1.0 mM), and common interfering molecules (F−, Cl−, HPO4 2-, H2PO4 −, HCO3 −, CH3COO−, CO3 2-, SCN−, SO3 2-, S2O3 2-, S2- 1.0 mM) were added into the system of DMSO: PBS(1:1, v/v, pH = 7.4) containing respectively [46]. As shown in Fig. 3 a, the fluorescence intensity at 495 nm showed negligible changes. But, the fluorescence intensity showed significantly changed responding with HOCl (100 μM), which fully confirmed that had good selectivity for HOCl. In addition, under the coexistence of other analytes, still showed an effective response to HOCl (100 μM), indicating that the presence of other analytes did not affect the detection of HOCl by (Fig. 3b).

Fig. 3

(a) Fluorescence intensity of (10.0 μM) after response to 1 eq HOCl and 100 eq other analytes each at 495 nm (λex =370 nm). (b) The competitive test response of under the presence of various analytes.

The detection limit and pH dependent of the AI for HOCl

In addition, we also studied the limits of , and observed a good linear relationship I495 in the range of 0−50 μM. The LOD of for HOCl (3σ/slope) [[25], [26], [27], [28], [29], [30], [31], [32], [33], [34]] was calculated to be 0.84 μM (Fig. 4 A).

Fig. 4

(a) the fluorescence intensity at 495 nm was obtained under different pH conditions in the presence and absence of HOCl. (b) The linearity between Iprobe+HOCl / Iprobe with 10 μM probe and 0−50 μM HOCl in mixture system (λex =370 nm).

We studied the change of fluorescence intensity (I495) in different pH ranges (2.0–12.0) to further explore the optical properties of (Fig. 4B). After a series of tests, we found that the probe showed a good response to HOCl under pH 5–7, indicating that the probe could be used in the physiological environment.

Proposed mechanism

The original intention of our design was that the probe with combining with anisaldehyde and DCO owned a large conjugated system, owing to the presence of electron-withdrawing CN group and electron-donating group —OCH3, the ICT-1 of probe was triggered to show obvious fluorescent signals. Then the electron-withdrawing CN group of probe attacked by HOCl, the ICT-1 process from —OCH3 moiety to CN group was disturbed. Meanwhile, the ICT-2 process was triggered from —OCH3 moiety to electron-withdrawing ketone group. Therefore, the fluorescence spectrum showed ratiometric change after responding to HOCl. In order to confirm the theoretical calculation and spectra experimental results, we adopted HR-MS and 1H-NMR to study the reaction mechanism between and HOCl (Fig S4). The m/z of the oxide with reacting with HOCl was 256.14633 by calculated, the data that we obtained was 257.1384([M+H]+), which preliminarily confirmed our conjecture. The 1H NMR data of in DMSO-d6 showed that characteristic peaks in the spectrum did not increase or decrease, but some chemical shift occurred (Fig. 5 ), which was ascribed to the stronger electron-withdrawing CN group had left and the weaker electron-withdrawing ketone group had formed. As shown in Scheme 2 , the possible recognition mechanism between and HOCl was proposed.

Fig. 5

1H NMR titration spetra of in DMSO-d upon addition of excess HOCl.

Scheme 2

The reaction mechanism of to HOCl.

Imaging in HeLa cells

To test the application of biological experiments, we first carried out cytotoxicity experiments according to literature methods [[46], [47], [48], [49]]. The results showed that the cells survival rate was more than 85 % (CCK-8 method), indicating that the probe had the future to be applied into cells (Fig. S5). The cell imaging of was obtained by fluorescence confocal microscope. Reference was made to above spectral experiments, we selected yellow channels (λ em = 550–590 nm) and blue channels (λ em = 470–530 nm) for observation. was incubated to the HeLa cells and cultured at 37℃for 20 min. Fluorescence signal was observed in the yellow channel. After adding HOCl to the above medium, the culture continued at the same temperature for 20 min, it could be observed that the fluorescence signal was enhanced of the blue channel and the yellow channel was weakened (as shown Fig.6 b, 6c), which showed that could detect exogenous HOCl at the cell level.

Fig. 6

(a) 10 μM was incubated with HeLa cells for 20 min; (b) then 20 μM HOCl was incubated for 20 min; (c) or then 50.0 μM HOCl was pre-treated with HeLa cells for 20 min; (d) 10 μM LPS was incubated for 20 min after (a): Scale bar: 20 μm. λex =405 nm, λem1 = 470−530 nm; λem2 = 540−590 nm.

According to previous reports, cells produce endogenous HOCl under the stimulation of LPS (a lipopolysaccharide that stimulates cells to produce HOCl) [50]. Therefore, LPS and HeLa cells were incubated for 12 h and then adding for incubation for 20 min, as shown in the Fig. 6d, obvious fluorescence signal could be observed in the yellow channel and the blue channel overtly. Comprehensive consideration with cell imaging experiments, could be used as a ratiometric probe to detected endogenous and exogenous HOCl.

Imaging of mice

Based on all the research results of the cells experiments, we further tested the practicality of in vivo imaging. Due to the background fluorescence of the mice, we selected the fluorescence signal of 500–590 nm to detect HOCl in the mice (Fig. 7 ). was injected into the subcutaneous of the mouse, and strong fluorescence signal was observed. When HOCl was injected into the same area, the fluorescence signal gradually enhanced with the time delay. The above results indicated that probe could identify and detect exogenous HOCl in mice.

Fig. 7

In vivo images of nude mice (A) Fluorescence imaging of the control group; (B) 20.0 μM was injected; (C–D) 50.0 μM HOCl was injected for 10 and 20 min; λex =405 nm, λem = 470−530 nm.

Conclusions

In summary, we synthesized a novel ratiometric fluorescent probe () based on anisaldehyde (donor) and DCO (acceptor) fluorescent dye to detected HOCl. A series of experimental results showed that probe had better fluorescence response, highly selectivity and lower detection limit to HOCl. In addition, could be used to detect endogenous and exogenous HOCl in HeLa cells and also successfully applied into mice imaging. Based on the above experimental results, we believed provides a powerful tool for better understanding the contributions of HOCl in physiological and pathological processes.

CRediT authorship contribution statement

Yan Shi: Investigation, Writing - original draft. Fangjun Huo: Funding acquisition. Caixia Yin: Conceptualization.

Declaration of Competing Interest

The authors report no declarations of interest.