Design and synthesis of a small molecular NIR-II chemiluminescence probe for in vivo-activated H2S imaging

Edited by Marco Garavelli, Università di Bologna, Bologna, Italy; received March 24, 2022; accepted December 28, 2022 by Editorial Board Member Shaul Mukamel
February 14, 2023
120 (8) e2205186120

Significance

Chemiluminescence is a peculiar phenomenon that the chemical energy could convert into luminescence, which has been widely explored for molecular imaging. To address the limitations of chemiluminescence in the visible window, such as low resolution and penetration, an H2S-activated unimolecular probe CD-950 with chemiluminescence in the NIR-II region was developed. By virtue of high efficiency of resonance energy transfer between chemiluminescence donor and fluorophore acceptor, CD-950 showed bright signal at 950 nm specifically triggered by H2S. H2S-activated NIR-II chemiluminescence performed greatly increased resolution and higher signal-to-noise ratio than that of NIR-II fluorescence in vivo. The superiority of NIR-II chemiluminescence signifies its importance as a chemiluminescence-based probe for biological imaging.

Abstract

Chemiluminescence (CL) with the elimination of excitation light and minimal autofluorescence interference has been wieldy applied in biosensing and bioimaging. However, the traditional emission of CL probes was mainly in the range of 400 to 650 nm, leading to undesired resolution and penetration in a biological object. Therefore, it was urgent to develop CL molecules in the near-infrared window [NIR, including NIR-I (650 to 900 nm) and near-infrared-II (900 to 1,700 nm)], coupled with unique advantages of long-time imaging, sensitive response, and high resolution at depths of millimeters. However, no NIR-II CL unimolecular probe has been reported until now. Herein, we developed an H2S-activated NIR-II CL probe [chemiluminiscence donor 950, (CD-950)] by covalently connecting two Schaap’s dioxetane donors with high chemical energy to a NIR-II fluorophore acceptor candidate via intramolecular CL resonance energy transfer strategy, thereby achieving high efficiency of 95%. CD-950 exhibited superior capacity including long-duration imaging (~60 min), deeper tissue penetration (~10 mm), and specific H2S response under physiological conditions. More importantly, CD-950 showed detection capability for metformin-induced hepatotoxicity with 2.5-fold higher signal-to-background ratios than that of NIR-II fluorescence mode. The unimolecular NIR-II CL probe holds great potential for the evaluation of drug-induced side effects by tracking its metabolites in vivo, further facilitating the rational design of novel NIR-II CL-based detection platforms.
Activity-based optical sensing has emerged as an indispensable technique for disease diagnosis and biological analysis (13). Chemiluminescence (CL) assays rely on emissive excited state formed from a chemical reaction instead of the external light excitation, which attenuates the autofluorescence and scattering caused by the external light excitation (4). Therefore, CL exhibited outstanding sensitivity, micrometer resolution at depths of millimeters, and high signal-to-background ratios (SBR) (57). However, current developed CL probes usually emit light in the visible (400 to 650 nm) region, which limits their further applications in vivo because of low resolution and penetration (813). On the other hand, optical imaging in the near-infrared (NIR) region, which was usually defined as NIR-I (650 to 900 nm) and NIR-II regions (900 to 1,700 nm), shows reduced photon absorption and scattering, high spatial resolution, and deep penetration (1417). Although some NIR CL nanoprobes have been reported (1824), most of them are based on the intermolecular chemiluminescence resonance energy transfer (CRET) effect between CL donor and fluorophore acceptor. CRET refers to nonradiative energy transfer from a CL donor to a luminophore acceptor candidate (Fig. 1A) (2529). As compared with the Förster resonance energy transfer (FRET) (30), CRET without the input of excitation light presents relative higher resolution and deeper penetration. For example, Zhang et al. reported a NIR-II CL nanoprobe by assembling chemical donor and two alternative dye acceptors into nanoparticle, and then employed successive CRET and FRET to realize the conversion of chemical energy to NIR-II CL emission (31). Due to multistep operation and low intermolecular energy conversion efficiency, such modification suffers from energy loss and extra energy relay processes, leading to diminished NIR-II CL intensity. In addition, the probe embedded in nanoparticle may also have difficulties in proximity to targeted molecules. To overcome these defects, it is desired to develop unimolecular NIR CL probe by covalently connecting high-energy CL donors and candidate NIR fluorophore acceptors.
Fig. 1.
Design and illustration of CD-950 for NIR-II CL in vivo imaging and hepatotoxicity detection. (A) The illustration of the principle for CRET. (Left of the dotted line: A common CL moiety emits green luminescence. Right of the dotted line: By conjugating a common CL moiety to the NIR fluorophore, the green luminescence of common CL moiety is diminished and the excited fluorophore illuminates red luminescence.) (B) Chemical structure of the Schaap's dioxetane core and its emission mechanism. (C) A series of previous reported chemical structures of the CL probes based on Schaap’s dioxetane core with different emission wavelengths, and the reported CD-950 in this work. (D) Schematic illustration of CD-950 triggered by H2S in situ to achieve NIR-II CL imaging for hepatotoxicity detection.
To design the unimolecular NIR CL probe, it is crucial to select a CL donor that could exhibit bright CL emission under physiological conditions. Metastable Schaap’s dioxetanes have gained increased interest in bioimaging because of their innate ability towards analyte-specific responses by customized designing the phenol-caging groups (3237). Removal of the caging group from a phenol results in an unstable phenolate−dioxetane, which disassembles via a chemically initiated electron exchange luminescence (CIEEL) mechanism to generate chemiexcited benzoate ester and consequently yielding CL signals (38, 39) (Fig. 1B). Recently, activatable CL probes based on Schaap’s dioxetane structures (Fig. 1C) showed three orders of magnitude increase in CL intensity when the electron-withdrawing group was bonded to an optimal probe (40). Furthermore, some CL probes based on the Schaap’s dioxetane skeleton yielding NIR-I emission have been employed through extended conjugation or fluorophore attachment (CRET) (4147). However, unimolecular CL probe based on Schaap’s dioxetane producing NIR-II emission has not been reported to date. Similarly, the selection of a suitable NIR-II organic fluorophore acceptor is also vital. To achieve effective energy transfer, a large spectral overlap between acceptor absorption spectrum and donor emission spectrum is necessary (25). At the same time, the fluorophore candidate should exhibit excellent stability and high fluorescence (FL) quantum yield in an aqueous solution. As a common NIR-II organic fluorophore (Fig. 1D), donor–acceptor–donor (D-A-D) based on benzo[1,2-c:4,5-c’]bis([1,2,5]thiadiazole) (BBTD) core with an adjustable structure has been employed in numerous biological applications due to its superior stability and imaging performance under physiological conditions (4852). Based on these characteristics, BBTD-based derivatives are promising as acceptors to assist the CRET process.
Metformin is a prescribed medicine for type-2 diabetes, cognitive disorders, cardiovascular diseases etc (53). However, long-term or overdose intake of metformin has side effects, such as heartburn and hepatotoxicity, which usually relates to the overexpression of H2S in liver (54). Despite efforts on NIR-I and NIR-II fluorescence (NIR-II FL) probes for H2S detection, H2S-activated NIR-II CL imaging probe has barely been reported (55). Herein, we reported an H2S-activated NIR-II CL unimolecular probe (CD-950) for highly specific in situ imaging of metformin-induced hepatotoxicity (Fig. 1D). We also showed the feasibility of H2S triggered NIR-II CL luminescence through excellent intramolecular CRET with efficiency of 95.0% between two Schaap’s dioxetane CL moieties (C-700, λem, max = 700 nm, Fig. 2) and one NIR-II D-A-D fluorophore (D-970, λabs, max = 740 nm, Fig. 2). In addition, CD-950 was selectively activated by H2S over other reactive species for a long duration (~60 min) and deeper tissue penetration (~10 mm), allowing in vivo imaging of subcutaneous tumors and metformin-induced liver injury. By taking advantage of high-contrast resolution under NIR-II CL imaging, the probe CD-950 achieved a 2.5-fold higher SBR sensing than NIR-II FL in the metformin-induced hepatotoxicity detection, highlighting the potential for assessment of medication side effects through tracking drug-induced metabolites.
Fig. 2.
The synthesis procedure of CD-950.

Results

Design and Synthesis of CD-950.

The synthetic procedure for CD-950 was depicted in Fig. 2. The outlined information was divided into three main sections, including synthesis of C-700, D-970, and CD-950. Compound 1 was synthesized by a modified procedure reported in the studies (40, 45). Firstly, bromine in the benzene was replaced by an aldehyde group, which was then reduced to a benzyl alcohol to obtain compound 3. By following the synthetic routes, including iodination reaction (4), Heck coupling reaction (5), oxidation (6), and hydrolysis of ester, the corresponding product 7 was afforded. The compound 7 was reacted with a dicyanomethylchromone via Knoevenagel Condensation to yield compound 8, which was regarded as the skeleton of NIR-I CL. Next, an H2S-cleavable 2,4-dinitrobromobenzene group (54) was introduced at the phenol position of compound 8 to form the precursor of C-700, 9. By the oxidation of singlet oxygen, the precursor 9 underwent [2+2] cycloaddition to give CL donor C-700. In the second part, we synthesized the fluorophore (D-970). Similarly, starting from 4-bromotriphenylamine, a nitration product 11 was prepared firstly, followed by Miyaura borylation to obtain compound 12. Subsequently, compound 13 was prepared via a Suzuki Coupling reaction between compound 12 and BBTD. Through the reduction of nitro and cyclization reaction, we finally obtained a blue solid D-970. With C-700 and D-970 in hand, a simple amidation reaction was performed to obtain the target NIR-II CL probe CD-950. 1H NMR, 13C NMR, and ESI-MS characterization confirmed the successful preparation of CD-950 molecule (from SI Appendix, Figs. S18–S40).

Response of CD-950 toward H2S.

We first evaluated the optical properties of C-700 and D-970, respectively. C-700 and D-970 showed maximum emission and absorption peaks at 700 nm and 740 nm, respectively (Fig. 3A). Due to high spectral overlap between C-700 and D-970, two units showed a potentially efficient CRET process. As expected, Na2S addition showed a maximum NIR-II CL signal enhancement at 950 nm by 32.9-fold higher than that of CD-950 without Na2S treatment (Fig. 3B), indicating an intramolecular CRET process. The emission wavelength of CD-950 was blue-shifted about 20 nm relative to that of D-970 (Fig. 3B). These electron-donating amino groups strengthened the intramolecular charge transfer effect in D-970 and decreased its bandgap, resulting in a maximum emission peak at 970 nm. In CD-950, these amino groups changed to amido bonds and reduced their electron-donating ability, leading to a blue shift in the emission spectrum of the CD-950. In addition to an obvious NIR-II CL signal, the working solution had a significant color change in the presence of H2S. In Fig. 3C, C-700, Nano-970 (nanomicelles encapsulating C-700 and D-970, SI Appendix, Fig. S4), and CD-950 showed immediate and obvious color change after Na2S addition. At the same time, distinct NIR-II CL signals were observed in Nano-970 and CD-950, respectively. Although C-700 also showed a noticeable color change, signals were not detected in the NIR-II window. NIR-II CL signal of CD-950 exhibited standard CL kinetic profile, which quickly increased up to maximum and then decreased to a negligible level (Fig. 3D). NIR-II CL half-life of CD-950 was 23.5 min. In addition, the reaction kinetic studies revealed the activation occurred within 10 s after the addition of Na2S solution.
Fig. 3.
Characterization of CD-950 in vitro. (A) Absorbance (black) spectrum of D-970 and emission (red) spectrum of C-700. (B) NIR-II CL spectrum of CD-950 in the absence (black) or presence (red) of Na2S, NIR-II FL spectrum of D-970 (purple). (C) These pictures of left and middle were gained by camera in the absence (black) or presence (red) of Na2S, the right one was gained by NIR-II in vivo imaging instrument. (D) CL kinetic profiles of CD-950 in PBS in the presence of Na2S. (E) NIR-II CL signal of CD-950 and NIR-II CL signal of Nano-970 in 96-well plate through stacked slices of chicken ham at varied depth. (F) The SBR value of CD-950 and Nano-970 in each depth of chicken ham. (G) The NIR-II CL intensity and NIR-II FL intensity of CD-950 treated with different concentration [Na2S] (0 to 200 μM). (H) The linear relationship between CD-950 and Na2S concentration in the range of 0 to 200 μM. (I) The NIR-II CL signal enhancement of CD-950 at 1,000 nm after addition of RONS (20 μM), metal ions (100 μM), GSH (100 μM), L-cys (100 μM) in PBS.
Tissue penetration depth is important for in vivo bioimaging. Hence, NIR-II CL signals of CD-950 and Nano-970 were recorded by a 1,000-nm long-pass filter through chicken ham as mimic tissues (Fig. 3E). NIR-II CL signals of activated CD-950 were detectable up to 10 mm, while the detection threshold for Nano-970 was 6 mm. The difference in tissue penetration depth was attributed to the higher NIR-II CL intensity of CD-950 at 1,000 nm. When the thickness of chicken ham was up to 2 mm, NIR-II CL intensity of CD-950 exhibited a maximum SBR of 8.9, about ~4.7 times higher than that of Nano-970 (Fig. 3F). These results clearly indicated that covalent conjugation contributed to higher luminescence intensity of unimolecular probe CD-950 due to efficient intramolecular CRET process.
To verify the probe’s specific response to H2S, the fluctuating signals of CD-950 were recorded after reacting with different concentrations of H2S. The NIR-II CL emission intensity of CD-950 increased with increasing Na2S concentration while the NIR-II FL signal almost remained unchanged (Fig. 3G). Meanwhile, NIR-II CL intensity of CD-950 at 1,000 nm showed a wide linear correlation with Na2S concentration ranging from 0 to 200 μM (Fig. 3H). We further tested the selectivity of CD-950 for H2S against other biologically relevant reactive sulfur, oxygen, and nitrogen species and metal ions, such as Ca2+, Fe2+, and Mg2+. As shown in Fig. 3I, CD-950 showed negligible NIR-II CL response toward interfering species. Additionally, the response to H2S was minimally perturbed by the presence of physiological levels of GSH and L-cysteine. These results suggested that CD-950 exhibited optimal selectivity and response to H2S. Therefore, CD-950 could be employed as a reliable and precise NIR-II CL probe for sensitive detection of H2S.

Response Mechanism and Theoretical Calculation.

Given the excellent optical properties of CD-950, a proposed activation mechanism was shown in Fig. 4 (the separate response mechanism of C-700 was showed in SI Appendix, Fig. S5). The dioxetane-fluorophore conjugate (I) decomposed to generate chemiexcited precursor (II) along with the departure of 2,4-dinitrothiophenol. The high activity of precursor (II) was spontaneously triggered in a CIEEL mechanism to access the chemiexcited benzoate, which transfers its chemiexcited energy to the NIR-II fluorophore, resulting in excitation of the fluorophore (III). Then, the excited intermediate decays to its ground state (benzoate-fluorophore conjugate ester IV) accompanying by NIR-II photon emission.
Fig. 4.
Proposed H2S-activated pathway of CD-950 for generating NIR-II CL emission.
To further elucidate the energy resonance transfer process of CD-950, relevant calculations based on the density functional theory and time-dependent density functional theory were performed at Cam-B3LYP/6-311G* level (5658). Firstly, the calculated emission spectrum of C-700-R (C-700-R is the product that the C-700 after response to H2S) and excitation spectrum of D-970 exhibited significant overlap, which was in agreement with the experimental results (SI Appendix, Fig. S6). Thus, both experimental and theoretical results verified the potential possibility of CRET in CD-950 and Nano-970 system. Next, it clearly showed that a smaller excitation energy of Compound [IV (1.65 eV)] than that of C-700-R (2.37 eV) (SI Appendix, Table S1 and Figs. S7–S9). By comparing the oscillator strength (f value) of the S1 → S0 transition, the f value of compound (IV) (f = 0.77) is higher than that of Nano-970-R (Nano-970-R is the product of the Nano-970 after response to H2S, f = 0.43), which might explain the difference of their luminous intensities (SI Appendix, Table S1 and Figs. S7–S9). The optimized excited state of the compound (IV) exhibited a centroid distance of 1.97 nm between the paired D-970 and C-700-R (SI Appendix, Fig. S10). In combination with the bright NIR-II CL signal of CD-950, this distance may be suitable for achieving CRET (59, 60). By analogy to the FRET energy conversion formula, a 95.0% CRET efficiency was obtained (SI Appendix, Scheme 1) (61). Thus, CD-950 showed a clear superiority in CL imaging in NIR-II window.

In Vitro NIR-II CL Imaging of H2S.

NIR-II CL imaging of CD-950, activated by H2S in cells, was then investigated. The potential cytotoxicity in MC38-luc cells, MCF-7 cells, and normal LO2 cells was evaluated by a CKK-8 assay. These cells showed good cytocompatibility after incubation for 24 h with different concentrations of CD-950 (up to100 μM) (SI Appendix, Fig. S11), suggesting the excellent biosafety of CD-950. After treating these cells with the probe for 30 min, NIR-II CL imaging signals were detected in MC38-luc, while no NIR-II CL signals in MCF-7 and LO2 cells were recorded (Fig. 5A). Because the NIR-II FL imaging signals of CD-950 were always turn-on under 808 nm laser irradiation and the intensity depended on the concentration of CD-950, the NIR-II FL intensity in three experimental groups was almost the same (Fig. 5 A and B). As a control (SI Appendix, Fig. S12), NIR-II CL signals in CD-950-treated MC38-luc cells reduced substantially when pretreated with amino-oxyacetic acid (AOAA, an inhibitor of endogenous H2S) and ZnCl2 (a scavenger of endogenous H2S). These results were attributed to the fact that MC38-luc expressed high levels of H2S while MCF-7 and LO2 barely expressed H2S, confirming the feasibility of CD-950 for sensitive detection of H2S in living cells.
Fig. 5.
H2S activated imaging of CD-950 in cells. NIR-II CL and NIR-II FL imaging (A) and corresponding intensity (B) of MC38-luc, MCF-7, and LO2 cells after incubation with CD-950 (50 μM) for 30 min. (C) NIR-II CL and NIR-II FL imaging of LO2 cells preincubated with Na2S (0 to 100 μM) for 10 min, followed by incubation with CD-950 (80 μM) for 30 min. (D) NIR-II CL and NIR-II FL intensities in C. (E) The linear relationship between NIR-II CL intensity and Na2S concentration in the range of 0 to 100 μM.
To further evaluate the detection reliability of CD-950 in response to varying H2S concentrations in cells, we firstly added different concentrations of Na2S solution into LO2 cells for 10 min and then incubated with CD-950 for 30 min. As shown in Fig. 5C, the corresponding NIR-II CL intensity gradually increased with increasing Na2S concentration, while NIR-II FL intensity remained unchanged. Mean intensity values of NIR-II CL and NIR-II FL were shown in Fig. 5D. The quantitative analysis of NIR-II CL intensities showed a linear relationship with Na2S concentration (Fig. 5E). These results further revealed the potential of CD-950 for accurate evaluation of H2S in cells.

In Vivo NIR-II CL Imaging of H2S in Tumor.

Benefitting from the excellent in vitro NIR-II CL performance of CD-950, in vivo imaging of H2S on MC38-luc tumor-bearing mice was then investigated. Firstly, the biocompatibility of CD-950 was examined. After intravenous injection of CD-950 (200 μM, 200 μL) for 24 h, hematoxylin and eosin (H&E) staining results revealed that no noticeable inflammation or lesions in the main organs (heart, liver, spleen, lung, and kidney) were observed, further verifying good biocompatibility of CD-950 (SI Appendix, Fig. S13). When the tumor size reached up to ~100 mm3, 100 μM CD-950 in 40 μL solution (containing 85% saline, 10% dimethylsulfoxide (DMSO), 5% Tween-80) was injected into tumor-bearing mice and NIR-II CL and NIR-II FL images of the mice were acquired at different time points. As described in Fig. 6 A and B, obvious NIR-II CL signals at the tumor site after 1 min injection of CD-950 were observed. NIR-II CL signals increased gradually, reached the plateau at 3 min, and then reduced to the baseline after 20 min. At the same time, NIR-II FL signal maintained a stable intensity range in the tumor region after injection. As described in Fig. 6C, the SBR ratio of NIR-II CL intensity was 3.1-fold than that of NIR-II FL intensity at 3 min after injection, further indicating a higher SBR of NIR-II CL imaging than NIR-II FL imaging.
Fig. 6.
NIR-II CL and NIR-II FL imaging of subcutaneous tumors treated with CD-950. (A) MC38-luc tumor-bearing mice imaging taken at different time points after subcutaneous injection of CD-950. (B) NIR-II CL and NIR-II FL intensities of MC38-luc tumor-bearing mice at different time points in Fig. 6A. (C) The corresponding SBR value of NIR-II CL (3 min) and NIR-II FL (3 min) in Fig. 6A. (D) NIR-II CL imaging of saline-pretreated tumor-bearing mice, L-Cys-pretreated tumor-bearing mice, Na2S-pretreated tumor-bearing mice, AOAA-pretreated tumor-bearing mice, and ZnCl2-pretreated tumor-bearing mice. (E) The corresponding NIR-II CL intensity in Fig. 6D.
Next, to precisely evaluate the feasibility of probe CD-950 in response to the dynamic variation of H2S in the tumor region, we pretreated tumor-bearing mice with saline, L-Cys, Na2S, AOAA, and ZnCl2, respectively (Fig. 6D). Firstly, when L-Cys solution was injected into tumors to stimulate H2S levels, NIR-II CL signal was significantly enhanced compared to the control group. To further verify the imaging ability toward exogenous H2S, we then injected Na2S into the tumor site prior to in vivo imaging. As expected, a stronger NIR-II CL emission was observed after pretreatment with Na2S. When AOAA was pretreated for 6 h, NIR-II CL signal significantly reduced. Similarly, intertumoral injection of ZnCl2 to scavenge H2S resulted in no obvious luminous signal in the tumor area (Fig. 6E). All these results further reflected the accurate response of CD-950 to the fluctuation of H2S in the tumor.

NIR-II CL Imaging of H2S in Metformin-Induced Liver Injury In Vivo.

To further validate the advantages of in vivo NIR-II CL imaging, a metformin-induced hepatotoxicity model was established (Fig. 7A). Firstly, CD-950 was administered via intravenous injection and the blood half-life (12.2 min) of CD-950 indicated that the probe rapidly metabolized from the blood circulation (SI Appendix, Fig. S14). When intravenous injection of CD-950 to saline-treated mice, in vivo and ex vivo of NIR-II FL imaging showed that the liver exhibited bright NIR-II FL signal under 808 nm laser irradiation by a 1000-nm long-pass filter with 50 ms exposure (Fig. 7 B and C and SI Appendix, Fig. S15). However, no significant NIR-II CL emission was observed in the liver of healthy mice. Meanwhile, we tested the NIR-II CL imaging signal of CD-950 in the mice pretreated with different doses of metformin. When mice were pretreated with 2 mg of metformin for 7 d, a turn-on NIR-II CL imaging signal was readily observed at 2 min after injecting CD-950, which was 2.2 times higher than saline-treated mice (Fig. 7 B and D). Due to the fact that a low dose of metformin only induced a low concentration of H2S, no obvious NIR-II CL signal was detected at 4 min of postinjection (Fig. 7 B and D). Next, a noticeable and persistent NIR-II CL signal was observed when the amount of metformin increased to 4 mg and 6 mg, respectively. Maximal NIR-II CL intensity in hepatotoxicity mice was 3.9 times higher than control mice and the longitudinal time of NIR-II CL imaging was up to 15 min (Fig. 7 B and D). Similarly, a clear ex vivo NIR-II CL imaging signal was detected in the liver when the organs were resected from metformin-treated (6 mg) mice at 2 min of postinjection (SI Appendix, Fig. S16). The maximal SBR in NIR-II CL imaging of the target site was up to 4.2, which is 2.5-fold higher than that of NIR-II FL imaging (1.7), as displayed in Fig. 7E. According to the intensity profiles of NIR-II CL and NIR-II FL treated with 6 mg metformin, the liver size in the horizontal direction was determined to be 15.5 mm and 28.4 mm, respectively. The obvious difference in liver size from NIR-II CL and NIR-II FL imaging suggested that NIR-II CL imaging accurately located the margin of the liver due to higher sensitivity and resolution than NIR-II FL imaging in the same animal model.
Fig. 7.
NIR-II CL and FL imaging of mice with varying degrees of hepatotoxicity induced by metformin. (A) Schematic illustration of the probe’s detection of hepatotoxicity induced by metformin in mice model. (B) NIR-II FL and CL imaging after i.v. injection of CD-950 with different treatment groups. (C) The dynamic NIR-II FL signal intensity as a function of time after i.v. injection of CD-950 under 808 nm laser irradiation. (D) The dynamic NIR-II CL signal intensity as a function of time after i.v. injection of CD-950. (E) SBR profiles along the liver in Fig. 7B (Red dotted line indicated by the white arrow represents NIR-II FL and green dotted line represents NIR-II CL). (F) ALT and AST values in the blood serum of mice with different treatment groups. (G) Representative H&E staining of liver tissue. Red cycles of ROI 1 to 3 indicate the enlarged areas in the intact tissue slice. Black arrows: swollen hepatocytes, blue arrow: lipid cavity, red cycle: hepatic necrosis.
To further confirm the degree of liver damage, we assessed the activities of two important hepatic indicators, aspartate aminotransferase (AST) and alanine aminotransferase (ALT), and results were presented in Fig. 7F. AST and ALT levels in metformin-treated mice were markedly elevated as the metformin dose increased from 2 to 6 mg, which was further confirmed by different histopathological damages. As shown in Fig. 7G and SI Appendix, Fig. S17, H&E staining results also demonstrated the corresponding characteristic of liver damage such as swollen hepatocytes, lipid cavity, and hepatic necrosis when treated with different dosages of metformin. Based on these results, CD-950 could be applied to visualize the upregulation of hepatic H2S and evaluate the severity of the metformin-induced liver injury.

Discussion

We first constructed an H2S-activated unimolecular NIR-II CL probe (CD-950) with the maximum emission wavelength at 950 nm. Based on the rational design of CL donor (C-700) and fluorophore acceptor (D-970), an efficient CRET process was achieved under physiological conditions. Unimolecular probe CD-950 achieved a bright NIR-II CL and also exhibited additional salient features of deeper tissue penetration (~10 mm), long duration (~60 min), and specific and fast H2S response. In addition, the dynamic fluctuations of H2S in cancer cells and living tissues were easily detected by CD-950 probe. More importantly, NIR-II CL imaging showed superior sensitivity and resolution compared to NIR-II FL imaging in the mice model with metformin-induced hepatotoxicity of mice model, indicating its significance in the prediction of drug-induced side effects. We believe that our designed strategy will further motivate the development of the NIR-II CL sensing platform for biological imaging and clinical translation applications.

Materials and Methods

All reactions were carried out at room temperature unless stated otherwise. Chemicals and solvents were either A.R. grade or purified by standard techniques. Thin layer chromatography: silica gel plates Merck 60 F254: compounds were visualized by irradiation with UV light. Column chromatography (FC): silica gel Merck 60 (particle size 0.040 to 0.063 mm), eluent given in parentheses. Reverse-phase high pressure liquid chromatography (RP-HPLC): C18 5u, 250 × 4.6 mm. Preparative RP-HPLC: C18 5u, 250 × 21 mm. NMR spectra were recorded using Bruker Advance operated at 400 MHz.

In Vitro NIR-II CL and FL Imaging.

NIR-II CL spectrum of CD-950 (50 μM) was acquired in the presence of Na2S (20 μM) aqueous solution containing 85% saline, 10% DMSO, 5% Tween-80 at 37 °C. NIR-II FL spectrum of CD-950 was obtained under 808 nm laser irradiation with exposure time of 50 ms. For tissue penetration study, 100 μM of CD-950 or Nano-970 was placed in a black 96-well plate, which was overlaid by chicken breast tissues with the desired thickness of 2, 4, 6, 8, and 10 mm, respectively. To study the H2S triggered optical behaviours of CD-950, the concentration of CD-950 was elevated up to 100 μM. For cancer cell imaging experiment, MC38-luc, MCF-7, and LO2 cells in a 96-well plate were divided into three groups and precultured at 37 °C for 24 h in an incubator containing 5% CO2. Then, 50 μM CD-950 in 20 μL solution (containing 85% saline, 10% DMSO, 5% Tween-80) was incubated for another 30 min. For the control experiment, MC38-luc cells were preincubated with PBS, ZnCl2 (300 μM), and AOAA (50 μM) for 6 h, respectively. Then, 50 μM CD-950 in 20 μL solution was incubated for another 30 min. For Na2S concentration-dependent experiment, LO2 cells were preincubated in Na2S (0 to 100 μM) solutions for 10 min. Then, 80 μM CD-950 in 40 μL solution (containing 85% saline, 10% DMSO, 5% Tween-80) was incubated for another 30 min. After incubation, each well was washed twice with PBS, and imaging was performed under a NIR-II CL (no additional excitation light source, the exposure time was 900 ms) and NIR-II FL (under an 808 nm laser, the exposure time was 50 ms) imaging setup.

In Vivo NIR-II CL Imaging of H2S in Subcutaneous Tumor.

For in vivo NIR-II CL and NIR-II FL imaging in the subcutaneous tumor: 100 μM CD-950 in 40 μL solution (containing 85% saline, 10% DMSO, 5% Tween-80) was used as an intratumor injection to mice when the tumor volume reached ~100 mm3. For the saline-pretreated group, the mice were intratumorally injected with saline (50 µL), followed by subcutaneous injection of CD-950, 60 min later. For L-Cys-pretreated group, the mice were intratumorally injected with L-Cys (1 mM, 50 μL), followed by subcutaneous injection of CD-950, 2h later. For Na2S-pretreated group, the mice were intratumorally injected with Na2S (1 mM, 50 μL), followed by subcutaneous injection of CD-950, 20 min later. For AOAA-pretreated group, the mice were intratumorally injected with AOAA (0.5 mM, 50 µL), followed by subcutaneous injection of CD-950 after 12 h. For ZnCl2-pretreated group, the mice were intratumorally injected with ZnCl2 (1 mM, 50 μL), followed by subcutaneous injection of CD-950 after 60 min. After injecting the probe, continuous imaging was carried out and images were collected at different time points. NIR-II CL image was acquired for 900 ms with a 1,000-nm long-pass filter. NIR-II FL image under 808 nm laser irradiation was acquired for 50 ms with a 1,000-nm long-pass filter.

In Vivo NIR-II CL Imaging of H2S in Metformin-Induced Liver Injury.

Via intravenous injection 200 μM CD-950 in 200 μL solution (containing 85% saline, 10% DMSO, 5% Tween-80) was administered to the mice with various degrees of liver injury. The whole imaging process lasted for 60 min, we took photos and collected data at 2, 4, 8, 15, and 60 time points, respectively. NIR-II CL and NIR-II FL images were acquired as described previously.

Data, Materials, and Software Availability

All the data (including synthetic procedures, spectral data, theoretical calculation, H&E staining results, and in vitro and in vivo data) supporting the findings of this study are available within the article and SI Appendix.

Acknowledgments

This research work was supported by the National Natural Science Foundation of China (No. U21A20377,21874024) and the Natural Science Foundation of Fujian Province (No. 2020J02012).

Author contributions

Z.C., L.S., and J.S. designed research; Z.C., Y.W., J.L., R.W., Q.L., C.W., L.L., and J.S. performed research; J.L. and L.L. contributed new reagents/analytic tools; L.S., R.W., and Q.L. analyzed data; and Z.C. and J.S. wrote the paper.

Competing interests

The authors declare no competing interest.

Supporting Information

Appendix 01 (PDF)

References

1
S. Gnaim, D. Shabat, Activity-based optical sensing enabled by self-immolative scaffolds: Monitoring of release events by fluorescence or chemiluminescence output. Acc. Chem. Res. 52, 2806–2817 (2019).
2
J. Huang, K. Pu, Activatable molecular probes for second near-infrared fluorescence, chemiluminescence, and photoacoustic imaging. Angew. Chem. Int. Ed. 59, 11717–11731 (2020).
3
X. Chen et al., Recent progress in the development of fluorescent, luminescent and colorimetric probes for detection of reactive oxygen and nitrogen species. Chem. Soc. Rev. 45, 2976–3016 (2016).
4
Y. Yang, F. Zhang, Activatable chemiluminescent molecular probes for bioimaging and biosensing. Anal. Sens. 1, 75–89 (2021).
5
Y. Jiang, K. Pu, Molecular probes for autofluorescence-free optical imaging. Chem. Rev. 121, 13086–13131 (2021).
6
J. Huang et al., Molecular optical imaging probes for early diagnosis of drug-induced acute kidney injury. Nat. Mater. 18, 1133–1143 (2019).
7
U. Haris, H. N. Kagalwala, Y. L. Kim, A. R. Lippert, Seeking illumination: The path to chemiluminescent 1,2-Dioxetanes for quantitative measurements and in vivo imaging. Acc. Chem. Res. 54, 2844–2857 (2021).
8
J. Huang et al., Chemiluminescent probes with long-lasting high brightness for in vivo maging of neutrophils. Angew. Chem. Int. Ed. 59, e202203235 (2022).
9
P. Li et al., A new polymer nanoprobe based on chemiluminescence resonance energy transfer for ultrasensitive imaging of intrinsic superoxide anion in mice. J. Am. Chem. Soc. 138, 2893–2896 (2016).
10
Y. Yu et al., Electrochemiluminescence detection of reduced and oxidized glutathione ratio by quantum dot-layered double hydroxide film. Analyst. 141, 3305–3312 (2016).
11
C.-L. Shen et al., Trigonal nitrogen activates high-brightness chemiluminescent carbon nanodots. ACS Mater. Lett. 3, 826–837 (2021).
12
J. Yang et al., Turn-on chemiluminescence probes and dual-amplification of signal for detection of amyloid beta species in vivo. Nat. Commun. 11, 4052 (2020).
13
M. A. Tzani et al., Direct and indirect chemiluminescence: Reactions, mechanisms and challenges. Molecules. 26, 7664–7693 (2021).
14
H. D. Li, Activity-based NIR fluorescent probes based on the versatile hemicyanine scaffold: Design strategy, biomedical applications, and outlook. Chem. Soc. Rev. 51, 1795–1835 (2022).
15
M. W. Yang et al., Chemiluminescence for bioimaging and therapeutics: Recent advances and challenges. Chem. Soc. Rev. 49, 6800–6815 (2020).
16
G. S. Hong, A. L. Antaris, H. J. Dai, Near-infrared fluorophores for biomedical imaging. Nat. Biomed. Eng. 1, 0010 (2017).
17
Y. Sun et al., Rhomboidal Pt(ii) metallacycle-based NIR-II theranostic nanoprobe for tumor diagnosis and image-guided therapy. Proc. Natl. Acad. Sci. U.S.A. 116, 1968–1973 (2019).
18
H. C. Shen et al., Rational design of NIR-II AIEgens with ultrahigh quantum yields for photo-and chemiluminescence imaging. J. Am. Chem. Soc. 144, 15391–15402 (2022).
19
C.-K. Lim et al., Chemiluminescence-generating nanoreactor formulation for near-infrared imaging of hydrogen peroxide and glucose level in vivo. Adv. Funct. Mater. 20, 2644–2648 (2010).
20
C. Lu et al., Light-free generation of singlet oxygen through manganese-thiophene nanosystems for pH-responsive chemiluminescence imaging and tumor therapy. Chem. 6, 2314–2334 (2020).
21
G. Qian, J. P. Gao, Z. Y. Wang, Near-infrared chemiluminescence tunable from 900 nm to 1700 nm from narrow-bandgap compounds and polymers. Chem. Commun. 48, 6426–6428 (2012).
22
M. Li, X. Huang, J. Ren, Multicolor chemiluminescent resonance energy-transfer system for in vivo high-contrast and targeted imaging. Anal. Chem. 93, 3042–3051 (2021).
23
D. G. Y. J. Jeon, Chemiluminescence resonance energy transfer–based nanoparticles for quantum yield–enhanced cancer phototheranostics. Sci. Adv. 6, eaaz8400 (2020).
24
L. Wu et al., H2S-activatable near-infrared afterglow luminescent probes for sensitive molecular imaging in vivo. Nat. Commun. 11, 446 (2020).
25
H. N. Kagalwala, A. R. Lippert, Energy transfer chemiluminescent spiroadamantane 1,2-dioxetane probes for bioanalyte detection and imaging. Angew. Chem. Int. Ed. 61, e202210057 (2022).
26
X. Huang, L. Li, H. Qian, C. Dong, J. Ren, A resonance energy transfer between chemiluminescent donors and luminescent quantum-dots as acceptors. Angew. Chem. Int. Ed. 45, 5140–5143 (2006).
27
Y. J. Han, J. Jose, E. Mei, K. Burgess, Chemiluminescent energy-transfer cassettes based on fluorescein and nile red. Angew. Chem. Int. Ed. 46, 1684–1687 (2007).
28
D. F. R. E. H. White, Intramolecular energy transfer in chemiluminescence. J. Am. Chem. Soc. 89, 3944–3945 (1967).
29
D. Lee et al., In vivo imaging of hydrogen peroxide with chemiluminescent nanoparticles. Nat. Mater. 6, 765–769 (2007).
30
I. L. Medintz, J. H. Konnert, H. Mattoussi, A fluorescence resonance energy transfer-derived structure of a quantum dot-protein bioconjugate nanoassembly. Proc. Natl. Acad. Sci. U.S.A. 101, 9612–9617 (2004).
31
Y. Yang et al., NIR-II chemiluminescence molecular sensor for in vivo high-contrast inflammation imaging. Angew. Chem. Int. Ed. 59, 18380–18385 (2020).
32
C. Peukert et al., Enzyme-activated chemiluminescent siderophore-dioxetane probes enable the selective and highly sensitive detection of bacterial pathogens. Angew. Chem. Int. Ed. 61, e202201423 (2022).
33
W. W. An et al., A chemiluminescent probe for HNO quantification and real-time monitoring in living cells. Angew. Chem. Int. Ed. 58, 1361–1365 (2019).
34
J. Cao et al., Chemiluminescent probes for imaging H2S in living animals. Chem. Sci. 6, 1979–1985 (2015).
35
S. Son et al., Chemiluminescent probe for the in vitro and in vivo imaging of cancers over-expressing NQO1. Angew. Chem. Int. Ed. 58, 1739–1743 (2019).
36
L. S. Ryan, A. R. Lippert, Ultrasensitive chemiluminescent detection of Cathepsin B: Insights into the new frontier of chemiluminescent imaging. Angew. Chem. Int. Ed. 57, 622–624 (2018).
37
R. An, S. Wei, Z. Huang, F. Liu, D. Ye, An activatable chemiluminescent probe for sensitive detection of γ-Glutamyl transpeptidase activity in vivo. Anal. Chem. 91, 13639–13646 (2019).
38
F. Augusto et al., Efficiency of electron transfer initiated chemiluminescence. Photochem. Photobiol. 89, 1299–1317 (2013).
39
W. Adam et al., Solvent-cage effect (viscosity dependence) as a diagnostic probe for the mechanism of the intramolecular hcemically initiated electron-exchange luminescence (CIEEL) triggered from a spiroadamantyl-substituted dioxetane. J. Am. Chem. Soc. 121, 958–961 (1999).
40
O. Green et al., Opening a gateway for chemiluminescence cell imaging: Distinctive methodology for design of bright chemiluminescent dioxetane probes. ACS Cent. Sci. 3, 349–358 (2017).
41
S. Gnaim et al., Modular access to diverse chemiluminescent dioxetane-luminophores through convergent synthesis. Angew. Chem. Int. Ed. 61, e202202187 (2022).
42
H. N. Kagalwala et al., Chemiluminescent 1,2-dioxetane iridium complexes for near-infrared oxygen sensin. Angew. Chem. Int. Ed. 61, e202115704 (2022).
43
J. S. Huang, Y. Jiang, J. C. Li, J. G. Huang, K. Y. Pu, Molecular chemiluminescent probes with a very long near-infrared emission wavelength for in vivo imaging. Angew. Chem. Int. Ed. 60, 3999–4003 (2021).
44
J. S. Huang, J. G. Huang, P. H. Cheng, Y. Y. Jiang, K. Y. Pu, Near-infrared chemiluminescent reporters for in vivo imaging of reactive oxygen and nitrogen species in kidneys. Adv. Funct. Mater. 30, 2003628 (2020).
45
K. J. Bruemmer, O. Green, T. A. Su, D. Shabat, C. J. Chang, Chemiluminescent probes for activity-based sensing of formaldehyde released from folate degradation in living mice. Angew. Chem. Int. Ed. 57, 7508–7512 (2018).
46
O. Green et al., Near-infrared dioxetane luminophores with direct chemiluminescence emission mode. J. Am. Chem. Soc. 139, 13243–13248 (2017).
47
N. Hananya, A. Eldar Boock, C. R. Bauer, R. Satchi-Fainaro, D. Shabat, Remarkable enhancement of chemiluminescent signal by dioxetane-fluorophore conjugates: Turn-on chemiluminescence probes with color modulation for sensing and imaging. J. Am. Chem. Soc. 138, 13438–13446 (2016).
48
L. Zhang et al., Improving quantum yield of a NIR-II dye by phenylazo group. Adv. Healthc. Mater. 9, e1901470 (2020).
49
A. L. Antaris et al., A small-molecule dye for NIR-II imaging. Nat. Mater. 15, 235–242 (2016).
50
Q. Yang, H. Ma, Y. Liang, H. Dai, Rational design of high brightness NIR-II organic dyes with s-d-a-d-s structure. Acc. Mater. Res. 2, 170–183 (2021).
51
Y. Fang et al., Design, synthesis, and application of a small molecular NIR-II fluorophore with maximal emission beyond 1200 nm. J. Am. Chem. Soc. 142, 15271–15275 (2020).
52
A. L. Antaris et al., A high quantum yield molecule-protein complex fluorophore for near-infrared II imaging. Nat. Commun. 8, 15269 (2017).
53
R. A. Miller et al., Biguanides suppress hepatic glucagon signalling by decreasing production of cyclic amp. Nature. 494, 256–260 (2013).
54
C. J. Cone, A. M. Bachyrycz, G. H. Murata, Hepatotoxicity associated with metformin therapy in treatment of type 2 diabetes mellitus with nonalcoholic fatty liver disease. Ann. Pharmacother. 44, 1655–1659 (2010).
55
D. A. Jose et al., Sensing and bioimaging of the gaseous signaling molecule hydrogen sulfide by near-infrared fluorescent probes. ACS Sens. 5, 3365–3391 (2020).
56
A. D. Becke, Density-functional exchange-energy approximation with correct asymptotic behavior. Phys. Rev. A 38, 3098–3100 (1988).
57
C. Lee, W. Yang, R. G. Parr, Development of the Colle-Salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 37, 785–789 (1988).
58
C. Adamo, D. Jacquemin, The calculations of excited-state properties with Time-dependent Density Functional Theory. Chem. Soc. Rev. 42, 845–856 (2013).
59
J. Lou, X. Tang, H. Zhang, W. Guan, C. Lu, Chemiluminescence resonance energy transfer efficiency and donor-acceptor distance: From qualitative to quantitative. Angew. Chem. Int. Ed. 60, 13029–13034 (2021).
60
P. Wu, L. Brand, Resonance energy transfer: Methods and applications. Anal. Biochem. 218, 1–13 (1994).
61
Y. X. Yuan et al., Fluorescent TPE macrocycle relayed light-harvesting system for bright customized-color circularly polarized luminescence. J. Am. Chem. Soc. 144, 5389–5399 (2022).

Information & Authors

Information

Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 120 | No. 8
February 21, 2023
PubMed: 36787363

Classifications

Data, Materials, and Software Availability

All the data (including synthetic procedures, spectral data, theoretical calculation, H&E staining results, and in vitro and in vivo data) supporting the findings of this study are available within the article and SI Appendix.

Submission history

Received: March 24, 2022
Accepted: December 28, 2022
Published online: February 14, 2023
Published in issue: February 21, 2023

Keywords

  1. chemiluminescence
  2. second near-infrared window
  3. bioimaging
  4. H2S
  5. metformin-induced hepatotoxicity

Acknowledgments

This research work was supported by the National Natural Science Foundation of China (No. U21A20377,21874024) and the Natural Science Foundation of Fujian Province (No. 2020J02012).
Author Contributions
Z.C., L.S., and J.S. designed research; Z.C., Y.W., J.L., R.W., Q.L., C.W., L.L., and J.S. performed research; J.L. and L.L. contributed new reagents/analytic tools; L.S., R.W., and Q.L. analyzed data; and Z.C. and J.S. wrote the paper.
Competing Interests
The authors declare no competing interest.

Notes

This article is a PNAS Direct Submission. M.G. is a guest editor invited by the Editorial Board.

Authors

Affiliations

Zhongxiang Chen
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Ying Wu
State key Laboratory of Chemical Resource Engineering, College of Chemistry, Beijing University of Chemical Technology, Beijing 10010, China
Jianyong Liu
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Rongrong Wu
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Qian Li
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Chenlu Wang
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
Luntao Liu
Ministry of Education (MOE) Key Laboratory for Analytical Science of Food Safety and Biology Institution, College of Chemistry, Fuzhou University, Fuzhou 350108, P. R. China
State key Laboratory of Chemical Resource Engineering, College of Chemistry, Beijing University of Chemical Technology, Beijing 10010, China

Notes

1
To whom correspondence may be addressed. Email: [email protected].

Metrics & Citations

Metrics

Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.


Citation statements




Altmetrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

View Options

View options

PDF format

Download this article as a PDF file

DOWNLOAD PDF

Get Access

Login options

Check if you have access through your login credentials or your institution to get full access on this article.

Personal login Institutional Login

Recommend to a librarian

Recommend PNAS to a Librarian

Purchase options

Purchase this article to get full access to it.

Single Article Purchase

Design and synthesis of a small molecular NIR-II chemiluminescence probe for in vivo-activated H2S imaging
Proceedings of the National Academy of Sciences
  • Vol. 120
  • No. 8

Media

Figures

Tables

Other

Share

Share

Share article link

Share on social media