2020, Vol. 31 
b Shaanxi Institute of Flexible Electronics (SIFE) & Institute of Biomedical Materials & Engineering (IBME), Northwestern Polytechnical University (NPU), Xi'an 710072, China
Parkinson's disease (PD) is a serious neurodegenerative disease. The brains of afflicted patients show reduced dopamine-secreting neurons in the substantia nigra (SN), which results in resting tremor, bradykinesia, and rigidity in patients [1, 2]. Mitochondria, which are important organelles, provide energy for cells. Increasing evidence also shows that mitochondrial dysfunction is closely related to PD [3, 4]. Cellular viscosity is a key factor in controlling the interaction between substances, signaling, and biomacromolecules [5, 6]. Because of its important influence on intracellular substance interactions, cellular viscosity has been proven to be an important marker of diseases such as PD [7]. Furthermore, the viscosity in the mitochondrial matrix changes with the mechanically or osmotically induced mitochondrial network organization, which is closely related to the respiratory state of the mitochondria. As such, viscosity can be effectively observed via the state and vitality of mitochondria [8]. Hydrogen sulfide (H2S), a critical signaling gasotransmitter, is extensively recognized as a participant in multiple physiological processes [9], such as neural activity regulation, insulin release regulation, angiogenesis induction, and inflammation inhibition and is present in cells free from oxidative stress damage [10-12]. Most of the endogenous H2S in the body is converted into sulfate and thiosulfate through the oxidative metabolism of mitochondria. Under normal physiological conditions [13-15], H2S neither accumulates in the body nor causes damage to cells. As a cytoprotective factor, H2S in mitochondria upregulates the expression of superoxide dismutase (SOD) and downregulates the expression of reactive oxygen through ischemia/reperfusion injury, thereby inhibiting the activity of cytochrome oxidase [16, 17]. In particular, studies have shown that mitochondrial dysfunction in Parkinson's disease and the antioxidative stress of H2S have protective effects on PD [18].
Fluorescence imaging is widely used in biological detection because of its simple operation, high sensitivity, and high resolution [19-21]. Recently, many excellent fluorescent probes have made good progress in mitochondrial viscosity and H2S imaging [22-24]. Li et al. reported a fluorescent molecule (Mito-VS) that could simultaneously detect viscosity and H2S through different channels [25]. However, since the probe had a short excitation and emission wavelength, the detection was greatly interfered by biological autofluorescence, thereby largely limiting the imaging depth in the living body. Compared with one-photon (OP) fluorescent probe, the two-photon (TP) fluorescent probe is characterized by greater penetration depth, weaker biological autofluorescence, etc., [26, 27]. Therefore, based on previous studies, we reported a TP dual-detection fluorescent probe Mito-HS, which could simultaneously detect mitochondrial viscosity and H2S in living organisms.
In the past, some H2S probes had improved targeting effects with the addition of mitochondrial targeting groups, such as triphenylphosphine [28, 29], which does not undergo structural changes before and after reacting with H2S. Therefore, the probe could be targeted to mitochondria before and after the reaction with H2S. However, it has not been proven whether the probe reacted with H2S on mitochondria or with H2S in the cytoplasm and then targeted the mitochondria. Here, we designed and reported Mito-HS with a targeting group that would become invalid as the probe reacted with H2S, and therefore, the probe lost its ability to target mitochondria after reacting with H2S. This meant that the fluorescence on the mitochondria was a result of the reaction with H2S on the mitochondria alone. The Mito-HS reported by us reacted slowly with H2S, and therefore, H2S in mitochondria was more easily detected. The synthetic route of Mito-HS is described in Scheme 1 and Supporting information. All intermediates and final compounds were unambiguously characterized by 1H NMR, 13C NMR and high-resolution mass spectrometry as shown in Supporting information.
|
Download:
|
| Scheme 1. Overall design of the strategy for dual-detection of viscosity and H2S levels with mitochondria-targeted TP fluorogenic probe (Mito-HS). | |
In our study, the "D-π-A" molecular configuration in the preMito we designed was similar to that in other TP fluorophores [30]. The N, N-disubstituted unit acted as an electron donor, and the pyridine cation served as an electron-withdrawing moiety, emitting green fluorescence through the intramolecular charge transfer (ICT) effect [31]. Indicated by free intramolecular rotation between dimethylaminonaphthalene and pyridine, the twisted intramolecular charge transfer (TICT) state was observed. Mito-HS is non-fluorescent, but when viscosity increases, the rotation in its molecules is restricted, and Mito-HS releases a strong red fluorescence emission. As such, we used density functional theory (DFT) to calculate HOMO-LUMO gaps of Mito-HS and pre-Mito molecules. The calculation results (Fig. S1 in Supporting information) showed that the HOMO-LUMO gap of Mito-HS was smaller than that of pre-Mito, resulting in longer absorption and emission wavelengths of Mito-HS. By comparing the rate of targeting mitochondria and the rate of reaction between specific reaction sites (war head) and H2S, we quantitatively detected the changes of H2S in mitochondria and cytoplasm [32, 33]. Therefore, two situations can occur. (1) When VH2S-R < Vmito-T, the probe first targets the mitochondria and then reacts with H2S to detect the H2S content in the mitochondria; (2) When VH2S-R > Vmito-T, the probe first reacts with H2S in the cytoplasm, and the unreacted probe then targets the mitochondria and reacts with H2S in the mitochondria to detect H2S in both the cytoplasm and mitochondria (VH2S-R: reaction rate of probe with H2S; Vmito-T: response rate of probe targeting mitochondria). This feature allowed us to build a platform, connect different warheads to the dye, evaluate its reaction rate, and create a database. The Mito-HS reported by us reacted slowly with H2S, and therefore, H2S in mitochondria was more easily detected.
The pre-Mito/Mito-HS pair is an excellent TP fluorescent system that can emit strong fluorescence during structure restriction/conversion. As shown in Table S1, Figs. S2 and S3 (Supporting information), pre-Mito shows a maximum absorption at 380 nm (ε = 25400 L mol-1cm-1) and a maximum emission band at 585 nm in PBS containing 30% DMSO at pH 8.0. In the same buffer, Mito-HS had a maximum absorption at 480 nm (ε = 6700 L mol-1cm-1) and a maximum emission band at 750 nm. Compared with pre-Mito, the wavelength of radiation from Mito-HS increased from 100 nm to 165 nm. The OP brightness (εΦ) of pre-Mito was as high as12, 446, and the value of the probe was relatively low (εΦ = 670). The maximum TP action cross-section (σΦ) and the quantum yield in PBS containing 30% DMSO at pH 8.0 of pre-Mito was 485.61 GM and 0.49, respectively, and the corresponding values for Mito-HS were 58.87 GM and 0.1, respectively. These experimental data demonstrated that pre-Mito was an outstanding fluorophore in OP/TPFM for fluorescent probes and bioimaging.
Next, we tested the sensitivity of pre-Mito/Mito-HS to viscosity. In various ratios of methanol (M)-glycerol (G) solutions, the maximum absorption of Mito-HS was not sensitive to viscosity changes (Fig. 1A). As shown in Fig. 1B, there was almost no fluorescence of Mito-HS in pure methanol. However, the fluorescence intensity of Mito-HS at 750 nm was significantly enhanced as the viscosity increased with increasing glycerol concentration. Moreover, we found a strong linear relationship between the viscosity value and fluorescence intensity (R2 = 0.99) (Fig. 1C), which indicated that Mito-HS could quantitatively detect changes in viscosity values. In contrast, the fluorescence intensity of pre-Mito was almost unchanged in solutions with different viscosities (Figs. S5 and S6 in Supporting information).
|
Download:
|
| Fig. 1. (A) Absorption spectra of Mito-HS (5 μmol/L) in the solution with different ratio of M (methanol):G (glycerol). (B) Fluorescence spectra of Mito-HS (5 μmol/L, λex = 480 nm) in the solution with different ratio of M:G. (C) The relationship between fluorescence intensity (750 nm) and viscosity values. (D) Absorption and (E) fluorescence spectra of Mito-HS (5 μmol/L, λex = 405 nm) reacted with different concentrations of H2S (0-10 equiv.) in PBS buffer contains 50% DMSO at pH 8.0 for 2 h. (F) Relationship between fluorescence intensities and H2S (0-1 equiv.). | |
Thereafter, we evaluated the optical response of Mito-HS to H2S in vitro. First, probe spectroscopy experiments with varying concentrations of H2S were employed. After 2 h incubation with different concentrations of H2S, absorption near 480 nm gradually decreased (Fig. 1D), and the red fluorescence at 585 nm significantly increased (Fig. 1E). There was a distance of 165 nm between the maximum emission channel of H2S and the maximum emission channel of viscosity, which was larger than that reported in previous studies. In effect, it was extremely difficult for the detection of viscosity and H2S to affect or interfere with each other by using Mito-HS as a fluorogenic probe. Additionally, when the equivalent concentration of H2S was between 0 and 1, the change in fluorescence intensity showed a strong linear relationship (Fig. 1F). Using this relationship, the detection limit of Mito-HS for H2S determined to be 11.66 nmol/L (Fig. S6 in Supporting information), which was far lower than the concentration of H2S in the cells. Hence, Mito-HS could have a good detection capability in live cells. The time response of Mito-HS to different concentrations of H2S proved that in high concentrations of H2S, the reaction between the probe and H2S completes within 2 h and reaches stability. Under physiological conditions of low concentrations of H2S, the reaction did not plateau within 2 h, which revealed that the speed of the reaction between Mito-HS and H2S could be much slower than that of targeting mitochondria (Fig. S7 in Supporting information). Pre-Mito and Mito-HS both showed stable fluorescence signals in the pH range of 4.0–10.0 and at different temperatures (Figs. S8 and S9 in Supporting information). Next, we evaluated the selectivity of Mito-HS for H2S by incubating MitoHS with a variety of competing species, including metal ions, ROS/RNS, and amino acids. The measurement results showed that only H2S could significantly enhance the fluorescence signal at 585 nm, and the fluorescence changes caused by other biological species could be ignored (Fig. S10 in Supporting information). Compared with the group supplemented with H2S only, the signal of the MitoHS group with different added species showed no significant enhancement, which proved that no reaction occurred between Mito-HS and other chemical molecules. According to the reaction mechanism, the azide group of Mito-HS reacts with H2S to form an amino group and then undergoes an elimination reaction to obtain pre-Mito [34]. Therefore, we observed and proved the aforementioned process by liquid chromatography-mass spectrometry (LC–MS) successfully (Fig. S12 in Supporting information). Next, to verify the TP imaging ability of Mito-HS in cells, we investigated the responses of Mito-HS to viscosity and H2S, respectively, by using TP excited fluorescence technology. The results showed that upon increasing the viscosity or H2S level, fluorescence intensity of Mito-HS in red/green channel was significantly enhanced at 750/550 nm (Fig. S13 in Supporting information). The above results showed that Mito-HS had high sensitivity and good selectivity for H2S and adequate responsiveness to changes in viscosity.
Upon in vitro analysis of the responsiveness of Mito-HS to viscosity and H2S, we tested the response of Mito-HS to H2S and viscosity in living cells. First, we evaluated the cytotoxicity through a standard MTT assay using the HepG2 cell line. It was verified that the cytotoxicity was considerably low when the probe concentration was below 20 μmol/L and could be imaged in living cells (Fig. S14 in Supporting information). We then investigated the response of Mito-HS to changes in HepG2 cell viscosity. According to literature reports, nystatin is an ionophore that regulates mitochondrial viscosity. Nystatin can induce mitochondrial structural changes by interrupting the ion balance, thereby causing a substantial increase in mitochondrial viscosity [35]. We mixed Mito-HS and nystatin in vitro, and the results showed that the fluorescence intensity did not change, which indicated that the probe was not affected by nystatin and could detect viscosity changes in living cells (Fig. S15 in Supporting information). As shown in Figs. 2A and B, HepG2 cells were treated with Mito-HS (10 μmol/L) for 1 h exhibited weak fluorescence signals in OP/TPFM. After pre-treatment with nystatin (10 μmol/L) and with the same measurement conditions, the fluorescence in the red channel was significantly enhanced and there was a 15-fold difference with the former. Under the high viscosity induced by nystatin (10 μmol/L), Mito-HS was co-stained with Mito Tracker Red, a commercially available mitochondrial tracer, and the Pearson's r value between Mito-HS and Mito Tracker Red channels was 0.81, far beyond the threshold required for correlation 0.5, which proved that the Mito-HS could specifically detect mitochondrial viscosity changes (Figs. 2C and D).
|
Download:
|
| Fig. 2. (A) Fluorescence imaging of viscosity in HepG2 cells under OP/TPFM. HepG2 cells were incubated with Mito-HS (10 μmol/L, 1 h) w/o pre-treated with nystatin (10 μmol/L, 1 h). Control: untreated cells. (B) The intensity of images (A). ***P < 0.001, n = 3, student's test. (C) Fluorescence images of HepG2 cells (pre-treated with 10 μmol/L nystatin for 1 h) stained with Mito-HS and Mito Tracker Red. (D) The intensity profile of the linear region of interest in HepG2 cells co-stained with Mito-HS and Mito Tracker Red as being indicated in last panel of (C). λex = 488 nm for OPFM and 800 nm for TPFM; PMT = 700–750 nm; Scale bar = 20 μm. | |
From previous experimental data, we know that Mito-HS slowly reacts with H2S and can target mitochondria. Due to the inability of Mito-HS to target mitochondria after reacting with H2S, we used Mito-HS to detect H2S content in living cells in situ by OP/TP fluorescence imaging. H2S in cells is mainly obtained from sulfur-containing amino acids such as Cystein (Cys) through cystathionine-γ-lyase (CES) and cystathionine synthase (CBS) [36]. Therefore, by adding exogenous Cys into cells, H2S content could be increased in cells. As shown in Figs. 3A and B, HepG2 and LO2 cells were incubated with Mito-HS with only LO2 cells showing weak fluorescence in the green channel. The fluorescence intensity of HepG2 cells was 6-fold of that of the LO2 cells. Upon addition of exogenous Cys into HepG2 and LO2 cells, a 2-fold increase was observed. Similarly, the fluorescence intensity of HepG2 cells was 6-fold of that of the LO2 cells. The results showed that H2S levels in cancer cells were higher than normal.
|
Download:
|
| Fig. 3. (A) Fluorescence imaging of Mito-HS response to H2S in LO2 cells and HepG2 cells under OP/TPFM. LO2 cells and HepG2 cells were treated with Mito-HS (10 μmol/L, 1 h) w/o pre-treated with Cys (10 μmol/L, 1.5 h). (B) The calculated intensity of images in (A). ***P < 0.001, n = 3, student's t-test. λex = 405 nm for OPFM and 760 nm for TPFM; PMT = 550–600 nm; Scale bar = 20 μm. | |
We then explored the relationship between endogenous H2S levels and viscosity in a PD cell-based model. Rotenone (ROT) can induce an increase in ROS and cause apoptosis [37]. Meanwhile, Brillian-VB3 (VB3) can improve the symptoms of PD by regulating the vulnerability of mitochondria [38]. The SH-SY5Y were incubated with and without VB3 (10 μmol/L, 3 h) before ROT (1 μmol/L, 12 h) treatment. Cells were then treated with Mito-HS (10 μmol/L, 1 h). As shown in Figs. 4A and B, SH-SY5Y cells incubated with ROT showed weak fluorescence in the green channel (PMT = 550-600 nm) and strong fluorescence in the red channel (PMT = 700-750 nm). The fluorescence intensities of SHSY5Y cells without VB3 treatment were almost consistent. The results indicated that Mito-HS can assess H2S levels in PD cell models. Moreover, endogenous H2S levels negatively correlated with viscosity in the PD model. To verify that Mito-HS had good TP fluorescence imaging ability, we applied Mito-HS to tissue imaging experiments in detecting H2S in live mouse tumors by using TPFM. As shown in Fig. 5, the tumor tissue slices treated with Mito-HS showed green fluorescence up to a depth of 180 μm. This experiment proved that Mito-HS was an excellent TP fluorogenic probe that detects H2S levels in living tissues.
|
Download:
|
| Fig. 4. (A) Fluorescence images of Mito-HS response to H2S/viscosity in SH-SY5Y cells by using OPFM. SHSY-5Y cells were treated with Mito-HS (10 μmol/L, 1 h) w/o treated with ROT (1 μmol/L, 12 h), VB3 (10 μmol/L, 3 h), ROT (1 μmol/L)/VB3 (10 μmol/L) for 15 h. (B) Fluorescence intensity of images (A). λex = 405 nm in green channel and 488 nm in red channel for OPFM; PMT = 550–600 (green)/700–750 (red) nm; Scale bar = 20 μm. | |
|
Download:
|
| Fig. 5. TP fluorescence imaging of H2S in tumor tissue sections. Fluorescence images of tumor tissue section incubated with Mito-HS (50 μmol/L) for 1 h. λex = 760 nm for TPFM; PMT = 550–600 nm; Scale bar = 100 μm. | |
By using a PD cell-based model, we demonstrated that Mito-HS was able to assess the relationship between H2S and viscosity at the cellular level and that Mito-HS had good TP fluorescence imaging ability. Therefore, we evaluated the response of Mito-HS to changes in H2S and viscosity in an insect PD model with TPFM. The PD features of Drosophila are obtained using parkin gene knockout technology [39]. DL-Propargylglycine (PAG) is a specific inhibitor of CES agents that can inhibit the activity of CES, thereby reducing the content of H2S in the cell [40]. As shown in Figs. 6A and B, Drosophila brains were incubated with Mito-HS (50 μmol/L, 1 h) before imaging under TPFM in the two channels. In wild-type Drosophila brains, the intensity of green fluorescence was high and that of red fluorescence was low. In contrast, in Parkin-null Drosophila brains, the intensity of green fluorescence was low but the intensity of red fluorescence was high. Similarly, after removing H2S with PAG, both the Parkin-null and wild-type Drosophila brains showed reduced fluorescence intensities in the green channels. On the contrary, the red channel fluorescence intensity exhibited improvement. Thus, we proved that the viscosity in PD Drosophila increased while the H2S level decreased.
|
Download:
|
| Fig. 6. (A) Fluorescence images of Mito-HS response to H2S and viscosity in drosophila brains by using TPFM. Drosophila were incubated with Mito-HS (50 μmol/L) for 1 h w/o treated with PAG (1 mmol/L, 1 h). (B) Fluorescence intensity of images (A). λex = 760 nm in green channel and 800 nm in red channel for TPFM; PMT = 550–600 (green)/700–750 (red) nm; Scale bar = 100 μm. | |
In conclusion, the mitochondrial-targeted TP fluorogenic probe Mito-HS could detect mitochondrial H2S and viscosity levels simultaneously. The probe emits red fluorescence at a wavelength of 750 nm and shows excellent linearity as the viscosity increases. Additionally, Mito-HS has high sensitivity and selectivity for H2S detection. Compared with previous studies, the probe had better penetration depth and could detect the viscosity and H2S content in mitochondria in situ as well as the interaction between them. The results showed that the PD model showed lower H2S content and higher viscosity than did the normal. In addition, there is an inverse correlation between the H2S content and viscosity. Therefore, based on this result, we can do more in-depth work to deepen our understanding of PD and find potential treatments.
Declaration of competing interestThe authors declare that there are no conflicts of interest.
AcknowledgmentsThis work was financially supported by the National Natural Science Foundation of China (Nos. 81672508, 21675085), Jiangsu Provincial Foundation for Distinguished Young Scholars (Nos. BK20170041, BK20170042), Natural Science Foundation of Shaanxi Province (No. 2019JM-016), China-Sweden Joint Mobility Project (No. 51811530018) and Fundamental Research Funds for the Central Universities.
Appendix A. Supplementary dataSupplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2020.03.063.
| [1] |
S. Przedborski, Nat. Rev. Neurosci. 18 (2017) 251-259. DOI:10.1038/nrn.2017.25 |
| [2] |
X. Cao, L. Cao, L. Ding, J.S. Bian, Mol. Neurobiol. 55 (2018) 3789-3799. |
| [3] |
A. Navarro, A. Boveris, Front. Aging Neurosci. 2 (2010) 34. |
| [4] |
V.A. Morais, B. de Strooper, J. Alzheimers Dis. 20 (2010) S255-S263. DOI:10.3233/JAD-2010-100345 |
| [5] |
D. Su, C.L. Teoh, L. Wang, X. Liu, Y.T. Chang, Chem. Soc. Rev. 46 (2017) 4833-4844. DOI:10.1039/C7CS00018A |
| [6] |
M.K. Kuimova, S.W. Botchway, A.W. Parker, et al., Nat. Chem. 1 (2009) 69-73. DOI:10.1038/nchem.120 |
| [7] |
S. Varanese, Z. Birnbaum, R. Rossi, A. Di Rocco, Parkinsons Dis. 2010 (2010) 480260. |
| [8] |
Z. Yang, Y. He, J.H. Lee, et al., J. Am. Chem. Soc. 135 (2013) 9181-9185. DOI:10.1021/ja403851p |
| [9] |
R. Wang, Antioxid. Redox Signal. 12 (2010) 1061-1064. DOI:10.1089/ars.2009.2938 |
| [10] |
Y.D. Wen, H. Wang, S.H. Kho, et al., PLoS One 8 (2013) e53147. DOI:10.1371/journal.pone.0053147 |
| [11] |
C. Szabo, Nat. Rev. Drug Discov. 6 (2007) 917-935. DOI:10.1038/nrd2425 |
| [12] |
L. Li, P. Rose, P.K. Moore, Annu. Rev. Pharmacol. Toxicol. 51 (2011) 169-187. DOI:10.1146/annurev-pharmtox-010510-100505 |
| [13] |
J. Furne, J. Springfield, T. Koenig, E. DeMaster, M.D. Levitt, Biochem. Pharmacol. 62 (2001) 255-259. DOI:10.1016/S0006-2952(01)00657-8 |
| [14] |
M.H. Stipanuk, I. Ueki, J. Inherited Metab. Dis. 34 (2011) 17-32. |
| [15] |
C.Q. Chen, H. Xin, Y.Z. Zhu, Acta Pharmacol. Sin. 28 (2007) 1709-1716. DOI:10.1111/j.1745-7254.2007.00629.x |
| [16] |
Q. Yuan, S. Hong, S. Han, et al., PLoS One 6 (2011) e25811. DOI:10.1371/journal.pone.0025811 |
| [17] |
Y. Kimura, Y.I. Goto, H. Kimura, Antioxid. Redox Signal. 12 (2010) 1-13. DOI:10.1089/ars.2008.2282 |
| [18] |
H.J. Choi, C.S. Lim, M.K. Cho, et al., Sens. Actuators B:Chem. 283 (2019) 810-819. DOI:10.1016/j.snb.2018.12.087 |
| [19] |
S. Feng, D. Liu, W. Feng, G. Feng, Anal. Chem. 89 (2017) 3754-3760. DOI:10.1021/acs.analchem.7b00135 |
| [20] |
L. Yu, Y. Qiao, L. Miao, Y. He, Y. Zhou, Chin. Chem. Lett. 29 (2018) 1545-1559. DOI:10.1016/j.cclet.2018.09.005 |
| [21] |
L. Wang, W. Du, Z. Hu, et al., Angew. Chem. Int. Ed. 58 (2019) 14026-14043. DOI:10.1002/anie.201901061 |
| [22] |
N. Gupta, S.I. Reja, V. Bhalla, et al., Chem. Commun. 51 (2015) 10875-10878. DOI:10.1039/C5CC02984H |
| [23] |
M.D. Hammers, M.J. Taormina, M.M. Cerda, et al., J. Am. Chem. Soc. 137 (2015) 10216-10223. DOI:10.1021/jacs.5b04196 |
| [24] |
L.L. Li, K. Li, M.Y. Li, et al., Anal. Chem. 90 (2018) 5873-5878. DOI:10.1021/acs.analchem.8b00590 |
| [25] |
S.J. Li, Y.F. Li, H.W. Liu, et al., Anal. Chem. 90 (2018) 9418-9425. DOI:10.1021/acs.analchem.8b02068 |
| [26] |
Y. Ni, H. Zhang, C. Chai, et al., Adv. Healthc. Mater. 8 (2019) e1900212. DOI:10.1002/adhm.201900212 |
| [27] |
D. Chen, W. Qin, H. Fang, et al., Chin. Chem. Lett. 30 (2019) 1738-1744. DOI:10.1016/j.cclet.2019.08.001 |
| [28] |
Y.L. Pak, J. Li, K.C. Ko, et al., Anal. Chem. 88 (2016) 5476-5481. DOI:10.1021/acs.analchem.6b00956 |
| [29] |
S.K. Bae, C.H. Heo, D.J. Choi, et al., J. Am. Chem. Soc. 135 (2013) 9915-9923. DOI:10.1021/ja404004v |
| [30] |
L. Li, C.W. Zhang, G.Y. Chen, et al., Nat. Commun. 5 (2014) 3276. DOI:10.1038/ncomms4276 |
| [31] |
H.J. Park, C.S. Lim, E.S. Kim, et al., Angew. Chem. Int. Ed. 51 (2012) 2673-2676. DOI:10.1002/anie.201109052 |
| [32] |
J. Xu, Y. Zhang, H. Yu, X. Gao, S. Shao, Anal. Chem. 88 (2016) 1455-1461. DOI:10.1021/acs.analchem.5b04424 |
| [33] |
M. Ren, B. Deng, K. Zhou, et al., Anal. Chem. 89 (2017) 552-555. DOI:10.1021/acs.analchem.6b04385 |
| [34] |
X. Wang, L. An, Q. Tian, K. Cui, RSC Adv. 9 (2019) 33578-33588. DOI:10.1039/C9RA06698E |
| [35] |
M.J. Hansson, S. Morota, M. Teilum, et al., J. Biol. Chem. 285 (2010) 741-750. DOI:10.1074/jbc.M109.017731 |
| [36] |
S. Singh, D. Padovani, R.A. Leslie, T. Chiku, R. Banerjee, J. Biol. Chem. 284 (2009) 22457-22466. DOI:10.1074/jbc.M109.010868 |
| [37] |
P. Caboni, T.B. Sherer, N. Zhang, et al., Chem. Res. Toxicol. 17 (2004) 1540-1548. DOI:10.1021/tx049867r |
| [38] |
P.A. Williams, J.M. Harder, N.E. Foxworth, et al., Science 355 (2017) 756-760. DOI:10.1126/science.aal0092 |
| [39] |
Y. Liu, L. Bai, Y. Li, et al., Sens. Actuators B:Chem. 279 (2019) 38-43. DOI:10.1016/j.snb.2018.09.107 |
| [40] |
Q. Sun, R. Collins, S. Huang, et al., J. Biol. Chem. 284 (2009) 3076-3085. DOI:10.1074/jbc.M805459200 |