Photoinduced Carbene for Effective Photodynamic Therapy Against Hypoxic Cancer Cells

Abstract

Open AccessCCS ChemistryRESEARCH ARTICLES14 Nov 2022Photoinduced Carbene for Effective Photodynamic Therapy Against Hypoxic Cancer Cells Zhanghua Sun, Yuncong Chen, Shankun Yao, Hao Yuan, Dongfan Song, Zijian Guo and Weijiang He Zhanghua Sun State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Google Scholar More articles by this author , Yuncong Chen *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Nanchuang (Jiangsu) Institute of Chemistry and Health, Nanjing 210000, Jiangsu Google Scholar More articles by this author , Shankun Yao State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Google Scholar More articles by this author , Hao Yuan State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Google Scholar More articles by this author , Dongfan Song State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Google Scholar More articles by this author , Zijian Guo *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Nanchuang (Jiangsu) Institute of Chemistry and Health, Nanjing 210000, Jiangsu Google Scholar More articles by this author and Weijiang He *Corresponding authors: E-mail Address: [email protected] E-mail Address: [email protected] E-mail Address: [email protected] State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Chemistry and Biomedicine Innovation Center (ChemBIC), Nanjing University, Nanjing 210023, Jiangsu Google Scholar More articles by this author https://doi.org/10.31635/ccschem.022.202202324 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Photodynamic therapy (PDT) has attracted much attention because of its advantages over chemotherapy, such as broad spectrum, high selectivity, and low drug resistance. However, most photosensitizers (PSs) used in PDT are O2-dependent and show limited therapeutic efficacy toward hypoxic solid tumors. Therefore, developing PSs that produce reactive oxygen species (ROS) in an O2-independent manner is highly demanded. Herein, we constructed a novel O2-independent PS ( TPA-N) based on α-diazo-aryl acetate, which could generate reactive carbene under visible light irradiation. Photoinduced carbene could react with biosubstrates and cause significant carbene stress and ROS, effectively killing tumor cells even under hypoxic conditions. TPA-N showed much less O2 dependence compared with traditional PDT reagent protoporphyrin IX. A mechanistic study suggested that TPA-N could induce mitochondrial membrane potential collapse and upregulate apoptosis-related proteins upon light irradiation. This work provided a reliable strategy for developing O2-independent PSs against hypoxic tumor cells through photoinduced carbene. Download figure Download PowerPoint Introduction The nonspecific damaging ability of reactive oxygen species (ROS) and precise photoirradiation provide photodynamic therapy (PDT) the advantages of minimal resistance, confined invasiveness, and a wide spectrum of anticancer activity in fighting tumors. The capacity of PDT to treat cancer by delivering accurate photoirradiation and ROS is a major benefit of this type of therapy.1–6 But, the efficacy of PDT is hindered by insufficient O2 supply.7,8 Many strategies, such as perfluorocarbon delivering oxygen into tumors and catalase promoting O2 generation, were developed to reduce negative consequences of hypoxia in this hot field,9–11 besides PDT technique innovation.12,13 Unlike the O2 supply strategies, the dyes of type I PDT exhibit the analogous possibility of diminishing O2-dependence.14,15 One major and most poisonous ROS is the superoxide radical (O2−•), the primary oxidant to consider during cancer treatment.16 These electron-bearing free radicals undergo rapid chemical oxidation,17 and are prone to seize electrons from nearby substances such as proteins, DNA, and lipids inducing a domino effect that conducts oxidative chain reactions leading to cancer cell death.18,19 However, due to the hypoxia condition in solid tumors, there are considerable obstacles in an attempt to generate adequate ROS to achieve satisfactory anticancer effects. Although type I photosensitizers (PSs) are less O2 dependent, they still need O2 for ROS generation required in tumor cell destruction.20,21 In this regard, exploring other oxygen-independent free radicals might be a viable option for effective tumor therapy.22 Diazo compounds are in widespread use in synthetic organic chemistry.23–26 The simplest-diazomethane was readily catalyzed to carbene with molecular nitrogen as the only byproduct. What is more, compared to •OH, carbene displays more powerful and rapid reactions to some electron-rich species through nucleophilic addition.27–29 Therefore, due to the existence of plentiful functional groups and nucleophilic atoms in biomolecules, carbene can assault cells more efficiently. On the other hand, phototherapeutic agents utilize tissue oxygen, excited by light to produce harmful ROS, commonly singlet oxygen (1O2), that destroy unhealthy cells through necrosis or apoptosis.30–32 Regrettably, the hypoxic tumor microenvironment caused by unusual tumor blood vessels and fast tumor propagation slows down the O2-dependent PDT procedure and actually prevents its realistic utilization in clinical cancer therapy.10,33,34 Free radicals generated in situ by diazo compounds are independent of oxygen and disrupt intracellular redox homeostasis to kill cells. Therefore, it is highly desirable to construct visible light-activated diazo compounds for highly effective hypoxic tumor treatment. Herein, we designed and synthesized a carbene precursor ( TPA-N) for antitumor therapy via carbene stress (Scheme 1). By modifying the structure of diazo compounds upon blue light irradiation, the carbene precursor produced a strongly oxidized, oxygen-independent electrophilic carbene. The results yielded in vitro experiments showed that the formation of carbene could effectively cause carbene stress under hypoxic and normoxic conditions. A mechanistic study demonstrated that, upon light irradiation, the reactive carbene generated by TPA-N could induce the collapse of mitochondrial membrane potential (MMP) and upregulation of apoptosis-related proteins, leading to effective apoptosis of MCF-7 human breast cancer cells. Moreover, TPA-N showed distinct inhibition on the growth of three-dimensional (3D) multicellular spheroids (MCSs), suggesting its potential for combating hypoxic solid tumors. This study offers a powerful strategy for the development of novel PSs with carbene generation ability, which is of great importance for improving the therapeutic effect of PDT against hypoxic tumor cells. Scheme 1 | Schematic diagram of the apoptosis pathways caused by TPA-N and the carbene formation reaction of TPA-N under hypoxia conditions by light irradiation. Download figure Download PowerPoint Experimental Methods The apparatus and materials used are supplied in the Supporting Information. Synthesis of TPA-N Supporting Information Scheme S1 illustrates the synthetic process of TPA-N using intermediates obtained from the literature. The TPA-N, TPA-Br, TPA-E, and TPA-C were well identified by high-resolution mass spectrometry (HRMS) and nuclear magnetic resonance (NMR) techniques ( Supporting Information Figures S1–S8). Cell culture The MCF-7 cells were cultured in RPMI-1640 medium in an incubator set at 37 °C in a 5% CO2 atmosphere. The 1640 medium contained streptomycin (50 units/mL), penicillin (100 units/mL) and fetal bovine serum (10%, v/v). For 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) cell viability and proliferation assay, intracellular imaging for photoinduced ROS, apoptotic study, and western blot assay, the following four MCF-7 cell-treated groups were investigated for comparison: 1. Normoxic light group: After the cells were incubated with TPA-N in the dark for 6 h, the cells were placed under light irradiation (λ = 450 nm, power density = 180 J/cm2, 360 s) and then returned to the incubator; 2. Hypoxic light group: Before irradiation (λ = 450 nm, power density = 180 J/cm2, 360 s), the cells incubated with TPA-N in the dark for 5 h were placed in an anaerobic bag for 1 h to form anoxic conditions and irradiated. 3. Normoxic dark group: The cells were cultured with TPA-N in the dark. 4. Hypoxic dark group: Cells were cultured with TPA-N in a hypoxic bag under dark conditions. Results and Discussion Design, synthesis, and photochemical properties of TPA-N Diazo compounds have been widely used for synthesis and protein cross-linking under UV light via carbene generation.35–37 However, the toxicity of photoinduced carbene intermediates to cancer cells has been rarely investigated. Since UV light could show severe damage to biological samples, we sought to introduce triphenylamine and benzothiazole in the molecular scaffold to extend the absorption wavelength in the visible range. The synthesized TPA-N ( Supporting Information Scheme S1) was confirmed by 1H NMR, HRMS, and 13C NMR ( Supporting Information Figures S1–S6). Then the photophysical property of TPA-N was investigated, which showed a strong absorption band in the visible range (∼440 nm) and a decent fluorescence emission with distinct solvatochromism ( Supporting Information Figures S9 and S10 and Table S1). Furthermore, the stability of TPA-N under physiological conditions (phosphate buffer solution, PBS) and in vitro cell culture systems were studied ( Supporting Information Figures S11 and S12). TPA-N showed little change in fluorescence under physiological conditions and in cell culture medium over a period, indicating that TPA-N had good stability. However, due to the poor water solubility of TPA-N, the fluorescence of TPA-N decreased upon precipitation over time. The photocatalytic property of TPA-N was investigated in dimethyl sulfoxide with or without blue light irradiation. As shown in Supporting Information Figures S13 and S14, only under blue light irradiation (500 mW/cm2) that the emission peak decrease steadily and was sustained within 4 min. Moreover, carbene production by TPA-N under blue light irradiation was confirmed by electron paramagnetic resonance (EPR). 5,5-Dimethyl-1-pyrroline N-oxide (DMPO) was used as a collector for carbene. As shown in Figure 1a, the powder EPR spectrum of TPA-N showed the significant coupling of the radicals with g = 2.0067 at 89 K low temperature, similar to the g factor of a free electron (g = 2.0023), intuitively showing the generation of a free electron by this blue light-photocatalysis of TPA-N.38,39 In addition, the photocatalytic reaction of phenylacetylene with TPA-N was carried out (Figure 1b), and high-performance liquid chromatography (HPLC) was used to monitor the reaction for 60 min (Figure 1c). Consumption of TPA-N, along with the appearance of TPA-C, was rapidly observed, and the conversion was completed within 30 min. Additional products of carbene were determined by mass spectrometry (MS) and NMR, which further demonstrated its production under light irradiation (Figure 1d and Supporting Information Figures S7 and S8). What is more, the photoinduced 1O2 quantum yields (ΦΔ, λex 450 nm) of TPA-N in acetonitrile were determined to be 0.004, which is very low compared with commonly used type II PSs ( Supporting Information Figure S15).40 The above results provided a faithful experimental basis for TPA-N to successfully generate carbene by blue light irradiation. The reactive carbene species generated by light irradiation of TPA-N could react with nearby proteins, causing the dysfunctioning of these biosubstrates and the subsequent cell death (Figure 1e). We chose bovine serum albumin (BSA) as a model protein to conduct a photo-cross-linking reaction with TPA-N. Strong fluorescence was observed when BSA was treated with TPA-N and blue light irradiation for 10 min, suggesting efficient cross-linking of BSA and TPA-N. However, almost no fluorescence was observed when BSA was treated with TPA-N in the absence of light irradiation, indicating a negligible noncovalent interaction of the BSA and TPA-N. This result confirmed that TPA-N possessed photo-cross-linking ability with biosubstrates to induce carbene stress via photoirradiation, suggesting a high potential for O2-independent PDT. Figure 1 | Verification of the ability of TPA-N to produce carbene. (a) The EPR spectra of TPA-N-generated carbene under blue light excitation were detected by DMPO. (b) The photocatalytic reaction of TPA-N (0.1 M) with phenylacetylene (1 M, 10 equiv) in methylene dichloride upon blue laser irradiation (1 W/cm2). (c) HPLC traces of photocatalytic reaction of TPA-N under blue laser irradiation at given time points. (d) The mass spectra characterization of TPA-C was obtained after the photocatalytic reaction of phenylacetylene with TPA-N. Supporting Information Figures S7 and S8 are the associated NMR and MS. (e) Visible light-dependent photo-cross-linking of BSA with TPA-N. Download figure Download PowerPoint Cellular uptake and subcellular accumulation of TPA-N The TPA-N possessed intense emission, noted when cells were imaged by confocal laser scanning microscopy (CLSM). With the prolongation of incubation time, the fluorescence of TPA-N was uniformly distributed and remarkably increased in the cytoplasm of MCF-7 cells (Figure 2a,b). It is worth mentioning that the optimal time for TPA-N entry into MCF-7 cells was 6 h. Colocalization experiments of TPA-N with different commercial dyes for subcellular localization of organelles were investigated in MCF-7 cells. The confocal emission images of the TPA-N channel overlapped poorly with the image obtained from the commercial organelle dye channel, indicating that TPA-N showed no specific accumulation in any organelle (Figure 2c). Figure 2 | (a and b) Intracellular fluorescence and quantitative analysis of MCF-7 cells incubated with TPA-N after different incubating times. (c) Colocalization images of MCF-7 cells co-stained with TPA-N and ER-Tracker Red, Mito-Tracker Red, Golgi-Tracker Red, Lyso-Tracker Red, and Hoechst 33342 stain, with the intensity profile of synchrony for the white line. (d) Intracellular oxidative stress was detected by a DCFH-DA probe under normoxia (O2 = 21%) and hypoxia (O2 < 0.1%). The MCF-7 cells were treated with TPA-N (5 μM) in normoxic and hypoxic conditions and irradiated by 450 nm laser (180 J cm−2, 360 s) (Scale bar: 50 μm). Download figure Download PowerPoint Intracellular oxidative stress induced by carbene generation 2′,7′-Dichlorodihydrofluorescein diacetate (DCFH-DA),41 a fluorescent probe for ROS, was used to detect changes in intracellular oxidative stress and indirect assessment of intracellular oxidative carbene in MCF-7 cells. An initiation of intracellular oxidative stress is caused by a dramatic increase in intracellular oxidative species that destroys the original redox balance.42 In the meantime, the highly reactive carbene would rob electrons from adjacent lipids, water, proteins, and other species to form more new free radicals, which would reinforce intracellular oxidative stress. As shown in Figure 2d, after the MCF-7 cells were treated with TPA-N upon 450 nm laser irradiation for 360 s, the fluorescence of the probe could be detected under both hypoxic and normoxic conditions, indicating that the production of carbene from TPA-N did not depend on oxygen. When no light was applied, almost no fluorescence of the probe was observed in the experimental group in which cells were incubated with TPA-N. The above results showed that in the absence of a laser, TPA-N could not be activated to form carbene, proving that the ability of TPA-N to release carbene was controlled by light. In vitro cytotoxicity of the TPA-N Next, the standard MTT cell viability method was applied to study the antitumor effect of TPA-N under hypoxic and normoxic conditions. As proved in Figure 3a,b, whether under normoxia or hypoxia conditions, MCF-7 cells cultured with TPA-N exhibited good cell viability under dark conditions, indicating that TPA-N possessed good biocompatibility. However, with 450 nm laser irradiation (180 J cm−2, t = 360 s), TPA-N showed a noticeable inhibition effect on MCF-7 cells under both normoxia and hypoxia conditions (Figure 3a,b). The IC50 of TPA-N under light was determined to be 0.71 μM under normoxia (pO2: 21%) and 0.98 μM under hypoxia (pO2 < 0.1%) conditions, indicating potent cytotoxicity with minimum dependence on oxygen. As shown in Figure 3c, under normoxic conditions, 62.5% of cancer cells were damaged by the commonly used PS protoporphyrin IX (PPIX) after light exposure, while the cell killing rate dropped to 24.5% under hypoxia, indicating a strong oxygen-dependent PDT effect of PPIX. In sharp contrast, the cell killing rate of TPA-N slightly decreased from normoxia (55.0%) to hypoxia (48.2%), confirming its much less oxygen-dependent PDT characteristic. Moreover, the cytotoxicity of TPA-N towards normal human breast epithelial cells (HBL100) was ( Supporting Information Figure S16) found to be similar to MCF-7 cells. Figure 3 | (a and b) The effects of different concentrations of TPA-N on cell viability under normoxia and hypoxia were investigated with or without 450 nm laser (180 J cm−2, 360 s) illumination. (c) MCF-7 cells were co-incubated with TPA-N (0.8 μM) and PPIX (1.6 μM) for 6 h, then MCF-7 cells were treated with 450 nm (180 J cm−2, 360 s) or 630 nm (29 J cm−2, 360 s) laser irradiation in normoxia and hypoxia. (d) Confocal images of live and dead MCF-7 cells stained with calcein AM (green color) and PI (red color), respectively, under different treatments (Scale bar: 50 μm). Download figure Download PowerPoint The inhibitory ability of TPA-N on cancer cells was further investigated using CLSM to visually differentiate dead cells and live cells stained with the cell-permeant dye, calcein AM to determine cell viability (green fluorescence), and propidium iodide (PI) to detect dead cells (red fluorescence). As demonstrated in Figure 3d, green fluorescence was distributed in almost all cancer cells, and no red fluorescence appeared in three groups of control, control + light, and TPA-N – light, suggesting negligible cell inhibition in these three groups. However, the TPA-N + laser group showed completely quenched fluorescence and distinct red fluorescence in both normoxia and hypoxia, adding further evidence for strong anticancer PDT efficiency of TPA-N without oxygen-dependency. Cell death mechanism in both normoxia and hypoxia The above studies demonstrated that under hypoxia, TPA-N could efficiently generate carbene, with resultant successful cell death. In order to investigate the cell death pathway involving PDT induced by TPA-N, different cell death inhibitors indicated below were used. TPA-N and various cell death inhibitors were applied for the co-culture with MCF-7 cells (24 h, Figure 4a), and the cell viability was investigated after 6 h with light exposure. Compared with no cell death inhibitor group, the MTT assay showed that no significant changes in cell viability occurred in autophagy inhibitors (3-methyladenine, 3-MA), necrosis inhibitors (necrostatin-1, Nec-1), and ferroptosis inhibitors (ferrostatin-1, Fer-1) groups. However, the viability of cells in the apoptosis inhibitor z-VAD-fmk group increased by 1.7-fold, indicating that TPA-N induced cell death mainly by apoptosis under light conditions. Figure 4 | (a) Different inhibitors and TPA-N (1 μM) were co-cultured with MCF-7 cells, and cell viabilities were tested after 24 h. MCF-7 cells were first co-incubated with four different inhibitors [Nec-1 (50 mM), 3-MA (100 mM), z-VAD-fmk (50 mM), and Fer-1 (50 mM)] for 1 h before cocultured with TPA-N. All experiments were performed under hypoxia (O2 < 0.1%), and MCF-7 cells were incubated with TPA-N for 6 h before light exposure (λ, 450 nm, 180 J cm−2, 360 s). (b) Apoptosis-related proteins were detected by Western-Blot in MCF-7 cells cultured with TPA-N (1 μM, 24 h at 37 °C). (c) Confocal imaging of MMP was tested by JC-1 staining. MCF-7 cells were co-cultured with TPA-N (5 μM, 6 h at 37 °C) for 6 h and illuminated with blue light (λ, 450 nm; 30 J cm−2, 300 s). A hypoxic environment was achieved using a hypoxia bag for 1 h. Download figure Download PowerPoint Cell viability tests under different conditions implied that TPA-N induced apoptosis by photoirradiation under hypoxia. This appeared reasonable because previous studies have shown that ROS-mediated photodamage induced cell death by upregulating the expression of the apoptosis regulator Bax protein.43,44 Cytochrome C is ultimately released from mitochondria by activated Bax, and the apoptosis complex, generated by cytochromes, procaspase-9, and Apaf-1, results in activated caspase-9, critical to the apoptotic pathway. We used a western-blot assay to detect apoptosis-related proteins (Figure 4b and Supporting Information Figure S17) to demonstrate the TPA-N-induced apoptotic death mode. The expression of Bax protein was significantly upregulated under hypoxic conditions following irradiation of TPA-N-treated MCF-7 cells. Furthermore, the expressions of procaspase-9 and procaspase-3 proteins were notably downregulated. These data confirmed that TPA-N induced cell damage through the apoptotic pathway. Due to the oligomerization of Bax, the permeability of the mitochondrial outer membrane would increase. In addition, the released cytochrome C would lead to apoptosis and a decrease in MMP (ΔΨm).43,44 Confocal imaging of MMP was obtained with JC-1 dye. MCF-7 cells incubated with TPA-N in the absence of light irradiation showed strong red fluorescence and very weak green fluorescence under both hypoxia and normoxia, indicating intact MMP (Figure 4c) of the MCF-7 cells. However, MCF-7 cells treated with TPA-N upon light irradiation exhibited quenched red fluorescence and intense green fluorescence, suggesting the collapse of MMP under both normoxia and hypoxia conditions. These results confirmed that the reactive carbene generated by photo irradiation could cause mitochondria dysfunction, leading to apoptosis, with minimum O2 dependence. PDT activity in 3D MCSs Inspired by TPA-N’s ability to produce carbene in the in vitro experiments and adherent-cell imaging, we attempted to examine the PDT effects of TPA-N in 3D MCSs. Since MCSs could effectively block the diffusion limit of oxygen, which is about 150–200 μm, and knowing that MCS with a radius of about 250 μm is generally used to simulate the microenvironment of solid tumors,45 we treated MCF-7 MCSs with TPA-N and dosed every 2 days while light exposure was given for a short time (180 Jcm−2, 360 s). As shown in Figure 5, the volume of MCSs in the control group without TPA-N increased remarkably after four days of culture, either under light or dark conditions. Although TPA-N did not inhibit the growth of MCSs under dark conditions, light exposure to different concentrations of TPA-N-treated MCSs significantly inhibited the growth of MCSs in a dose-dependent fashion, as the highest dose of 20 μM TPA-N caused the MCSs to collapse under light irradiation. The unique ability of TPA-N to inhibit the growth of MCSs under light exposure suggested that TPA-N has the potential to inhibit the growth of hypoxic tumors in vivo. Figure 5 | Microscopic images of MCF-7 MCSs. MCSs were cultured with different concentrations of TPA-N (10 μM and 20 μM) and irradiated with 450 nm (180 J cm−2, 360 s) laser light (Scale bar: 500 μm). Download figure Download PowerPoint Conclusion TPA-N was designed and synthesized to promote carbene production to overcome the O2-dependence of traditional PDT. The fabrication of this strongly nucleophilic and oxidative free radical led to carbene stress in tumor cells under both normoxia and hypoxia conditions. In addition, due to its powerful reactivity towards biomolecules bearing electrons, MCF-7 cells were severely damaged and eventually died. Carbene-induced intracellular oxidative stress by robbing nearby substances of electrons damaged the MMP and led to cell apoptosis with minimum O2-dependence. TPA-N distinctly suppressed the growth of 3D MCSs under light irradiation. This work offered a powerful approach to the development of novel PDT agents against hypoxic tumor cells with high efficiency. This strategy could also be applied readily for carbene-based PDT with improved penetration depth by extending the excitation wavelength to red or near-infrared region through structural modifications, which is currently under investigation by our group. Supporting Information Supporting Information is available and includes (1) synthesis routes and methods, (2) experimental methods, (3) 1H and 13C NMR spectra, (4) HR-MS spectra, (5) UV and fluorescence spectra, (6) photoinduced 1O2 generation ability in acetonitrile solution, (7) intracellular fluorescence and quantitative analysis, and (8) colocalization images with different commercial dyes. Conflict of Interest There is no conflict of interest to report. Funding Information The work was under financial support from the National Natural Science Foundation of China (grant nos. 22122701, 21977044, 21731004, 91953201, 92153303, and 21907050), the Natural Science Foundation of Jiangsu Province (grant no. BK20202004), the Excellent Research Program of Nanjing University (grant no. ZYJH004) and the Open Research Fund of the National Center for Protein Sciences at Peking University in Beijing (grant no. KF-202201). Acknowledgments The authors wish to acknowledge technicians Jun Cai and Yufei Jiang for their helpful suggestions on the experimental tests.