The discovery of novel reactivities with carbon–carbon double bonds (CC) is not only useful for chemical synthesis and theoretical studies, but is also important for applications in biology, because CC bonds are abundant in nature. As we all know, CC bonds are fundamental structures of alkenes, where the CC π bond is localized above and below the C–C σ bond wherein π electrons are relatively far away from the nuclei and are loosely bound, so that they can be easily attacked to construct a new bond.1,2 Indeed, addition reactions are one of the common reactions of CC bonds, such as reactions with HX (X = Cl, Br, I, OH, SH, RS, etc.) as described in Scheme 1a.3 Additionally, CC bonds can also be cleaved by oxidation as in the typical ozonolysis of alkenes (Scheme 1b).4 However, because of their unique structure, reductive cleavage of alkenes (Csp–Csp) has rarely been reported so far.
Scheme 1
(a) Typical addition reactions in alkenes. (b) Oxidative cleavage of alkenes by ozone. (c) Schematic representation of H2S-mediated reductive cleavage of CC bonds and four phenothiazine ethylidene malononitrile derivatives, with single- or dual-color turn-on fluorescence responses to H2S.
In particular, the cleavage of CC bonds in styrenes has been activated by a hard Lewis acid and ethanethiol.5,6 In addition, Shi and coworkers reported reductive cleavage of Csp–Csp bonds using rhodium based catalysis.7 In 2014, Bogdanov et al. reported that CC bonds in 1,10-disubstituted isoindigos could be reductively cleaved by aqueous hydrazine hydrate.8 In this work, we discovered an interesting H2S-mediated reductive cleavage of CC bonds under mild conditions.As is well known, H2S with sulfur at the lowest valence is an excellent reductant and nucleophile, found predominantly as HS– at physiological pH and therefore displaying higher nucleophilicity compared with many other thiols in cells. Consequently, addition reactions have been a commonly used strategy for developing fluorescent H2S probes in recent years. Highlighting the popularity and influence of this reaction mechanism, many fluorescent probes were well designed with rapid and specific responses to H2S by disrupting the conjugated π-system of CC bonds with addition reactions.9–13 In addition, aryl nitro groups could be cleaved by thiolysis, triggering the fluorescence turn-on for the detection of H2S or HS.14–17 In the current work, we report our finding of H2S-mediated reductive cleavage of CC bonds under mild conditions (Scheme 1c), anticipated to be a new strategy for devising fluorescent H2S probes. Concomitant with this has been the development of a dual-color fluorescent H2S probe, which was successfully applied to monitor endogenous H2S in vivo.
Results and discussion
Design, synthesis and characterizations
Phenothiazine (PTZ), having a non-planar butterfly conformation, provides strong fluorescence, and has been used as an electron donor in photoelectric materials with a variety of applications.18–22 We found that phenothiazine ethylidene malononitrile derivatives, PTZ-P1, PTZ-P2, PTZ-P3 and PTZ-P4, could effectively react with H2S via reductive cleavage of CC bonds to yield a new fluorescent compound PTZCNSF.Firstly, we use PTZ-P1 as an example to discuss this novel reaction. PTZ is a strong electron donor whereas the dicyano group is a strong electron acceptor. The strong intramolecular charge transfer (ICT) in PTZ-P1 gave rise to fluorescence quenching. Upon reacting with H2S, PTZ-P1 exhibited turn-on fluorescence with high selectivity and sensitivity.To dissect the reaction mechanism, PTZ-P1 was used to react with NaHS and the green fluorescent product, PTZCNSF, was successfully isolated. The NMR results of PTZCNSF (Fig. S25 and S26†) demonstrated the disappearance of the pyridine moiety and single crystal X-ray analysis was further used to confirm the structure. Single crystals of PTZ-P1 and PTZCNSF for X-ray diffraction were obtained by slow evaporation of dichloromethane solutions, and ORTEP drawings are depicted in Fig. 1. During the transformation from PTZ-P1 to PTZCNSF, C13, C14, C16 and C17 were connected by S2 from H2S via an intramolecular cyclization reaction to yield PTZCNSF. It is surprising that the C17–C18 double bond in PTZ-P1 was reductively cleaved and the pyridine moiety was cut off from the main structure, which was consistent with the NMR and MS analysis (Fig. 2a and b).
Fig. 1
ORTEP diagrams of PTZ-P1 (a) and PTZCNSF (b) with ellipsoids adjusted to 50% probability. Solvent molecules were deleted for clarity.
Fig. 2
MS data of PTZ-P1, PTZCNSF, and the red emitting products of PTZ-P4 after the reaction with H2S.
Three more derivatives, PTZ-P2, PTZ-P3, and PTZ-P4, were then designed and synthesized. For PTZ-P3, BODIPY was introduced for its good photostability, narrow emission, and high fluorescence quantum yield. In PTZ-P4, a red emitting fluorophore (BOBPY) was chosen to conjugate with phenothiazine ethylidene malononitrile. BOBPY and derivatives were first developed by Jiao and coworkers,23 and they are a kind of N2O-type benzopyrromethene boron complex, wherein axial positions are substituted by corresponding boronic acids. We chose BOBPY here because of its excellent stability and high fluorescence quantum yield in different media. We hypothesized that both green and red emitting fluorophores will be released after reductive cleavage, and these could be used as dual-color fluorescent probes for sensing H2S.Detailed synthetic procedures are elucidated in the ESI.† Products and intermediates were fully characterized using 1H NMR, 13C NMR, HRMS and MALDI-TOF spectra (Fig. S8–S24, S27–S29†). All four derivatives could effectively react with H2S via reductive cleavage of CC bonds to produce PTZCNSF. However, pure single samples of the other part were difficult to isolate, because of the high reactivity of the –CH2 moiety after reductive cleavage. Fortunately, the MALDI-TOF data of PTZ-P4 revealed that the –CH2 moiety could form R–CH3 monomers and dimers (Fig. 2c) under these reductive conditions. Although the exact reaction pathway is still under investigation, MS, NMR, and single-crystal X-ray analysis undoubtedly confirmed this H2S-mediated reductive cleavage of CC bonds.
Absorption and fluorescence response of probes to H2S
With these probes in hand, their responses to H2S were studied. Initially, the absorption spectra of PTZ-P1 upon addition of H2S were assessed in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400) using aqueous NaHS as the H2S source. As depicted in Fig. S5a†, PTZ-P1 showed a main absorption peak at 330 nm along with a weak-intensity ICT band at around 500 nm. The absorption intensity decreased upon the gradual addition of H2S (0–800 μM). However, PTZ-P1 showed a turn-on fluorescence response at 488 nm with a 25-fold enhancement (Fig. S5b†). The absorption and fluorescence of PTZ-P2 responding to H2S were similar to those of PTZ-P1 (Fig. S6†). However, the increase of the fluorescence intensity was lower (only a 5-fold increase), which may be attributed to the stronger electron withdrawing strength of the dipyridyl moiety compared with the pyridyl group.Subsequently, spectra titration experiments were performed to further investigate the response of PTZ-P3 towards H2S. The characteristic absorption peaks at 330 nm decreased upon gradual addition of H2S (0–700 μM), whereas the peaks at 525 nm did not change much (Fig. 3a). The decrease of the absorption peak at 330 nm revealed the reductive cleavage of PTZ-P3 induced by H2S, and was consistent with those of PTZ-P1 and PTZ-P2. As shown in Fig. 3b, very weak emission of PTZ-P3 (10 μM) was displayed in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400) without H2S because of the strong ICT, but the emissions at 480 nm and 540 nm were both remarkably increased upon gradual addition of H2S (0–700 μM). The enhanced emission at 480 nm was caused by PTZCNSF after reductive cleavage, which was consistent with those of PTZ-P1 and PTZ-P2. However, heightening emission at 540 nm was related to the BODIPY moiety being released from PTZ-P3.
Fig. 3
(a and b) Absorption and fluorescence spectra of PTZ-P3 (10 μM) in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400) with the addition of H2S (0–700 μM), λex = 330 nm. Inset in (b): photoimages of PTZ-P3 with and without H2S under daylight (left) and an ultraviolet lamp (365 nm; right). (c) The fluorescence spectra of PTZ-P3 (10 μM) in the presence of H2S (0.5 mM), Na2S2 (0.2 mM), cysteine persulfides and polysulfides (0.2 mM) and other analytes (0.5 mM) in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400), excitation: 500 nm. (d) Photoimages of PTZ-P3 with H2S (0.5 mM) and other analytes under daylight (above) and an ultraviolet lamp (365 nm, below). (e and f) Absorption and fluorescence spectra of PTZ-P4 (10 μM) in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400) with the addition of H2S (0–600 μM), λex = 580 nm. Inset in (f): photoimages of PTZ-P4 with and without H2S under daylight (left) and an ultraviolet lamp (365 nm; right). (g) The fluorescence spectra of PTZ-P4 (10 μM) in the presence of H2S (0.5 mM), Na2S2 (0.2 mM), cysteine persulfides and polysulfides (0.2 mM) and other analytes (0.5 mM) in PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400), excitation: 580 nm. (h) Photoimages of PTZ-P4 with H2S (0.5 mM) and other analytes under daylight (above) and an ultraviolet lamp (365 nm, below).
In the presence of H2S, the ethylenic bond was reductively cleaved, so that optical properties of PTZ-P4 were observed both from PTZCNSF and BOBPY. Bearing this point in mind, the responses of absorption and fluorescence of PTZ-P4 to H2S were studied by gradual addition of NaHS solution into a PBS buffer (pH = 7.4)/DMSO (1/2, 2% v/v PEG 400) solution containing 10 μM probe. As illustrated in Fig. 3e, the absorption peaks of PTZ-P4 at 384 nm were exhibited as decreasing, while the absorption peaks at 628 nm displayed almost no change after gradually adding H2S (0–600 μM). As expected, the fluorescence spectra were consistent with PTZ-P1 in the green region, but the red fluorescence was so strong that the weak green fluorescence was covered with the insignificant ratiometric change. As shown in Fig. 3f, the emission intensity at 638 nm (λex = 580 nm) increased about 30-fold for PTZ-P4 upon addition of 600 μM H2S.
Selectivity of PTZ-P3 and PTZ-P4 to H2S
The selectivity of the PTZ-P3 and PTZ-P4 towards H2S was further identified. PTZ-P3 and PTZ-P4 (10 μM) were both treated respectively with various biologically relevant analytes in PBS buffer (pH = 7.4)/DMSO (1 : 2, 2% v/v PEG 400) for 10 min. As shown in Fig. 3c, d, g and h, the turn-on fluorescent responses of PTZ-P3 and PTZ-P4 are highly selective for H2S versus biologically relevant thiols, reactive oxygen species (ROS) such as H2O2 and ClO–, ions including K+, Na+, Ca2+, HSO3–, SO32–, SO42– and S2O32–, and so on. Cysteine, cystine, glutathione, Na2S2, and cysteine persulfides and polysulfides24 induced very little increase of fluorescence intensity. Therefore, both PTZ-P3 and PTZ-P4 showed high selectivity for H2S.
Imaging exogenous H2S in living cells
After developing these probes based on the novel reductive cleavage of CC bonds, we explored the applications of PTZ-P4 in monitoring H2S under physiological conditions. The cytotoxicity of PTZ-P4 was evaluated in HeLa cells using a MTT assay.As described in Fig. S7†, PTZ-P4 exhibited relatively low toxicity towards HeLa cells with good viability. Putting PTZ-P4 into practice, exogenous H2S was detected in living HeLa cells. Firstly, HeLa cells were exposed to 20 μM PTZ-P4 for 5 h, then NaHS solution was used as the exogenous H2S source at 100 μM, incubating with cells for another 3 h. Compared with cells treated with only PTZ-P4, there was obvious green and red fluorescence in cells treated with both H2S and PTZ-P4 (Fig. 4).
Fig. 4
Confocal images of exogenous H2S in HeLa cells without (top) and with (bottom) 100 μM H2S in the presence of 20 μM PTZ-P4. Scale bar: 20 μm.
Imaging endogenous H2S in living C. elegans
Endogenous H2S mainly originates from sulfur-containing amino acids metabolized by at least three enzymes: cystathionine β-synthase (CBS), cystathionine γ-lyase (CSE) and 3-mercaptopyruvate sulfur transferase (3-MST).25C. elegans is an excellent in vivo model system for monitoring physiological H2S with clear molecular mechanisms of H2S action.26,27 Specifically, it has been shown that the germline-deficient glp-1 (e2144) mutants displayed increasing production of endogenous H2S, while the deletion mutation in cth-1, the gene encoding the H2S synthesizing enzyme cystathionine γ-lyase, resulted in down-regulated H2S levels.27–30 To further understand the features of PTZ-P4, in vivo imaging was employed to visualize endogenous H2S in C. elegans. Concentrations of PTZ-P4 and time periods used for feeding were titrated to ensure enough PTZ-P4 absorption and metabolism without significant biotoxicity (data not shown). In living C. elegans, PTZ-P4 should be absorbed, distributed, metabolized and excreted, and so 200 μM PTZ-P4 was used to make sure that there was enough PTZ-P4 in C. elegans for in vivo imaging. As a representative example shown in Fig. 5a, both green and red fluorescence were observed in wild-type N2 worms after being fed with 200 μM PTZ-P4 for 48 h. Additionally, H2S was mainly accumulated in intestinal cells, predominantly in the cytoplasm, and in apical membranes as well (Fig. 5a and data not shown).
Fig. 5
(a) Confocal images captured the H2S distribution in wild type N2 with 200 μM PTZ-P4; the scale bar was 25 μm. (b) Fluorescence images of endogenous H2S in wild type N2, germline-deficient glp-1 (e2144) mutants and cth-1 (ok3319) mutants incubated with 200 μM PTZ-P4. Normal, elevated and reduced endogenous H2S levels are shown from left to right; the scale bar was 200 μm. (c) Fluorescence intensities were measured along a line crossing the anterior of the intestine.
Consistent with changes of H2S levels in different strains, the red fluorescence intensity of glp-1 (e2144) mutants was obviously stronger than that in wild-type N2 worms, while cth-1 (ok3319) mutants exhibited notably weaker red fluorescence compared with wild-type N2 worms (Fig. 5b and c). This result suggested that red fluorescence signals could be clearly captured with significant changes and successfully reflected different H2S levels under different physiological conditions. Of note, differences in green fluorescence among different strains were also detected, albeit to a less appreciable level (Fig. 5c), due to the weak absorbance of PTZCNSF at 405 nm (the excitation wavelength of the fluorescence microscope). Altogether, these results provided compelling evidence that PTZ-P4 is highly sensitive and selective with dual fluorescent colors to detect endogenous H2S in living systems and can be used as a potential probe for in vivo imaging of endogenous H2S.A significant bottleneck in detecting H2S is the effective imaging of endogenous H2S in vivo, a problem that restrains the biological applications. In the current study, the dual-color fluorescent probe PTZ-P4 showed high selectivity for H2S versus cysteine or glutathione. It also demonstrated a good response concurrently with the application in different strains of C. elegans, showing distinct differences in brightness for wild type N2, glp-1 (e2144) and cth-1 (ok3319) mutants.
Conclusions
In summary, we have discovered H2S-mediated reductive cleavage of CC bonds under mild conditions, which was successful as a new strategy to design fluorescent probes for effective detecting of H2S. Of these probes, PTZ-P4 displayed dual-color turn-on fluorescence upon in vivo sensing of endogenous H2S in different strains of C. elegans, showing distinct differences in brightness for wild type N2, glp-1 (e2144) and cth-1 (ok3319) mutants. As far as we know, this is the first report that H2S can reductively cleave CC bonds under mild conditions. As CC bonds are actively involved in various organic reactions, the discovery of this reductive cleavage of CC is anticipated to be valuable not only for the development of CC bonds in synthetic chemistry and theoretical studies, but also for the design of new fluorescent H2S probes for bioimaging and sensing applications.
Conflicts of interest
The authors declare no competing financial interest.Click here for additional data file.Click here for additional data file.
Authors: Ismail Ismail; Zhuoyue Chen; Lu Sun; Xiuru Ji; Haishun Ye; Xueying Kang; Haojie Huang; Haibin Song; Sarah G Bolton; Zhen Xi; Michael D Pluth; Long Yi Journal: Chem Sci Date: 2020-07-09 Impact factor: 9.825
Scheme 1
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5