Synthesis of TA1
As shown in Scheme 1, the structure of target molecule TA1 comprises three parts: 1) H2Sn-reactive triggering part, 2) Rhodol-based two-photon fluorophore bearing mitochondria targeting TPP, and 3) anti-inflammatory drug, indomethacin. The TA1 was synthesized by the following procedure. First, H2Sn-reactive site 2 was prepared by s-acetylation of thiosalicylic acid using benzoyl chloride. The triggering unit, 6, was synthesized via tert-butyldimethyl silyl (TBDMS) protection at both sides of aliphatic hydroxyl group of 2,6-bis(hydroxymethyl)-p-cresol, followed by 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) coupling reaction with 2. Then, the mono-deprotection of the TBDMS group in 6 was conducted in the presence of amberlyst-15 to yield 7. Compound 7 was then conjugated with indomethacin through EDC coupling to produce 8. Finally, the other TBDMS group of 8 was further deprotected using amberlyst-15 and subsequently reacted with 4-nitrophenyl chloroformate for activation. The resulting intermediate was reacted with TPP-conjugated two-photon Rhodol fluorophore 4 (Rhodol-TPP) which was synthesized by EDC coupling of Rhodol and 3, successfully yielding the desired product, TA1 (Scheme 1). The detailed procedures of the synthesis and structurally characterized evidence for the compounds are provided in supporting information.
Characterization of TA1
To determine whether TA1 could react with H2Sn to give an off–on fluorescence change, the photophysical change of TA1 (10 μM) was investigated using UV–Vis absorption and fluorescence spectroscopy under simulated physiological conditions (10 mM PBS buffer, pH 7.4), in the presence and absence of Na2S3 (100 μM), H2Sn donor. UV–Vis absorption intensity of TA1 (10 μM) markedly enhanced at 512 nm upon addition of Na2S3 (100 μM, 10 mM PBS buffer, 0.2% DMSO, 100 μM CTAB) (Fig. 1a). Time-dependent fluorescence intensity changes of TA1 (10 μM) were observed, and they indicated a marginal fast (30 min) and gradual increase of its fluorescence intensity at 542 nm, upon treatment with Na2S3 (Fig. 1b, c). These results confirm that the TA1 reacts with H2Sn and induces cleavage of H2Sn-reactive triggering part to release the corresponding fluorophore, where the fluorescence off–on change is attributed to the ring-opening of Rhodol fluorophore upon self-immolation reaction (Fig. 2a). The fluorescence responses of TA1 (10 μM) to H2Sn were also evaluated to investigate the sensitivity to different concentrations of Na2S3 (0 to 10 μM). As a function of Na2S3 concentration, gradually increasing fluorescence intensity of TA1 (λex = 512 nm) was observed, with a center at 542 nm (Fig. 1d). A linear correlation of various concentrations of Na2S3 with fluorescence intensities at 542 nm was also observed (Fig. 1e). Taken together, we conclude that TA1 can effectively respond to H2Sn with reliable sensitivity under physiological conditions.
For two-photon confocal-microscopic imaging of the probe, TPA cross sections of TA1 and TA1 + H2Sn were initially investigated with rhodamine 6G as the reference molecule. Upon addition of Na2S3 (100 μM), TA1 exhibited 66 GM with maximum TPA cross-section value at 800 nm under physiological conditions (PBS buffer, 0.5% DMSO, 100 μM CTAB), which validates that TA1 can be sufficiently sensitized by two-photon absorption (Fig. 1f).
To determine the selectivity of TA1 toward H2Sn over other biological species, fluorescence experiments with a series of other biologically relevant species were also performed. The TA1 showed a high selectivity for Na2S3 over amino acids and other nucleophilic sulfur species such as glutathione (GSH), cysteine (Cys), homocysteine (Hcy), S2O32−, HSO3−, and Na2S (Fig. 1g). Moreover, TA1 was also found to be inert to other reductive species such as ascorbic acid. Upon treatment of a mixed solution of Na2S (200 μM) and ClO− (50 μM) to generate H2Sn in situ, TA1 showed a strongly enhanced fluorescence at 542 nm as well. These results conclude that TA1 can selectively respond to H2Sn in the biological media containing various potential interferences.
Proposed Mechanism of H2Sn-responsive Activation of TA1
To verify that the proposed self-immolation cleavage mechanism of the theranostics system shown in Fig. 2a is reasonably operated, HPLC analysis of TA1 in the presence of Na2S3 was undertaken. As shown in Fig. 2b, the retention time for TA1, Rhodol-TPP, and indomethacin was 6.6, 5.5, and 7.4 min, respectively. Time-course experiment upon reaction with H2Sn gave a Rhodol-TPP peak, cleaved from TA1, which gradually increased. Indomethacin release began after 1 h, indicating that TA1 can release both Rhodol-TPP and indomethacin, simultaneously, upon reaction with H2Sn (Fig. 2c). Moreover, the ESI–MS spectrum of TA1 in the presence of Na2S3 showed two peaks of m/z 380.12 and 759.30, corresponding to indomethacin and Rhodol-TPP, respectively (Fig. S13). This result supports the mode of action that TA1 simultaneously releases both indomethacin and Rhodol-TPP in the presence of H2Sn.
Selective Activation of TA1 in vitro
The abovementioned results suggest that TA1 selectively reacts with H2Sn and could be suitable for precise drug delivery as a potential theranostic agent. The TA1 was then applied to mouse macrophage cell line, RAW264.7, as a bioassay model. First, from the LDH cytotoxicity assay, we found the low cytotoxicity of TA1 in RAW264.7 cells, at various concentrations after 24 h-incubation (Fig. S14), thus suggesting that it could be further applied to an anti-inflammatory therapeutic system. Subsequently, confocal-microscopy images of the live RAW264.7 cells having endogenous H2Sn were obtained in the presence of TA1 (10 μM) at 37 °C. The group treated with TA1 exhibited brighter fluorescence as it responded to H2Sn compared to that exhibited by the control group (Fig. 3a, b and S15). The group further treated with exogenous Na2S2 (5 μM) displayed stronger fluorescence (Fig. 3c and S15). To further examine the responsiveness to endogenously produced H2Sn by perturbing the pool, RAW264.7 cells were pre-incubated with LPS (1 μg mL−1, 16 h), which can induce an inflammatory environment to trigger the overexpression of CSE mRNA, and thus promote the production of endogenous H2Sn. Upon subsequent treatment of TA1 (10 μM, 2 h), the cells displayed a remarkable increase in fluorescence intensity (Fig. 3d and S15).
On the contrary, the pretreatment of DL-propargylglycine (PAG, 1 mM; CSE inhibitor) significantly attenuated the fluorescence intensity of TA1, thus confirming that CSE contributed to the endogenous generation of H2Sn (Fig. 3e and S15). Moreover, mitochondrial localization of TA1 was demonstrated via co-localization assays with Mito-tracker Red (Fig. 3b-e and S16), thereby proving the accessibility of TA1 to mitochondrial H2Sn. In addition, two-photon fluorescence microscopy images were collected from RAW264.7 cells for investigating the responsiveness of TA1 toward H2Sn upon excitation at 800 nm (Fig. 3f-j and S15). These results are consistent with those from the one-photon fluorescence microscopy experiment. Collectively, the results thus indicate that TA1 can react with endogenous cellular H2Sn in living cells, which can be directly visualized by fluorescence off–on changes with both one-photon and two-photon fluorescence microscopy.
Anti-inhibitory Effects of TA1 in vivo
Following the results that TA1 selectively responds to endogenous and exogenous H2Sn in living cells and shows diagnostic abilities, the therapeutic effect of TA1 against inflammation was subsequently investigated by various biological tests. First, the western blot analysis of inflammation-induced RAW264.7 cells treated by LPS was implemented to explore COX-2 levels. The group treated with LPS exhibited high expression of COX-2 levels and another group treated with N-Acetyl cysteine (NAC, 1 mM for 12 h), a quencher of LPS-mediated inflammation, displayed decreased COX-2 levels (Fig. 4a). However, cells incubated with further treatment of TA1 exhibited a significant reduction of COX-2 expression compared to that in the control group. This result signifies that TA1 selectively releases indomethacin (IMC) upon reaction with H2Sn, existing in the inflammatory environment, to reduce COX-2 levels. We also observed that PGE2 production increased in LPS-induced inflammatory response, whereas the levels were markedly reduced in TA1 treated RAW264.7 cells (Fig. 4b). Moreover, an inflammation-induced mouse model was established to confirm the in vivo theranostic potential of TA1. Ahead of investigation, LPS was intraperitoneally injected in mice to cause hepatotoxicity and inflammation. Upon the injection of TA1 into the LPS-induced acute liver injury (ALI) mouse model, the fluorescence expressions were observed to verify whether TA1 is triggered at the inflammatory site. As seen in Fig. 4c and S17, the marked fluorescence enhancement of in vivo and ex vivo imaging was observed in the liver because TA1 released both Rhodol-TPP and IMC upon reaction with H2Sn at the inflammatory site. Besides, the blood serum of the ALI mouse model was isolated to examine PGE2 level, which is representative of the level of inflammation. As shown in Fig. 4d, the PGE2 level in the ALI mouse model was significantly high, whereas it was reduced in the serum of the mice treated with either TA1 or IMC. To investigate anti-inflammatory effects in the ALI mouse model, we further analyzed the production of pro-inflammatory cytokines, such as TNF-α and IL-1β in serum. The levels of TNF- α and IL-1β increased for 24 h in the ALI mouse model; however, the levels were significantly decreased in the group treated with TA1 or IMC (Fig. 4e, f and S18). We also confirmed that TA1 overcomes inflammatory responses, suppressing plasma levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in ALI mouse models (Fig. 4g, h). These results strongly suggest that the theranostic agent TA1, developed for the first time in this study, can selectively treat inflammation-related diseases by releasing the corresponding drug to the inflammatory site exclusively upon reaction with H2Sn and, simultaneously, aid in the diagnosis of the inflammation by fluorescence imaging in vivo.