*3.7. CY-09 Relieved Cognitive Impairment and Pathological Injury in 3*×*Tg-AD Mice*

Previous studies showed that NLRP3 inflammasome activation contributes to the pathology of AD. The results of this study demonstrated that inhibition of the NLRP3 inflammasome by CY-09 significantly recovered glucose metabolism. Next, we focused on the effects of CY-09 on cognitive impairment and classically pathological biomarkers in AD and confirmed whether CY-09 is a potential therapeutic drug for AD.

The Morris water maze was used to evaluate the learning and memory abilities of the four groups of mice. As shown in Figure 7, on the fifth day, the escape latency of 3×Tg-AD mice was remarkably longer than that of the NTg mice, but notably reduced after CY-09 treatment, with the *p*-values all lower than 0.05. Moreover, the time in the target quadrants of the 3×Tg-AD mice was less than that of the NTg mice, while increasing after CY-09 treatment in 24 h and 72 h space exploration experiments, with a *p*-value lower than 0.05 and a *p*-value of 0.1282, respectively. No differences were found in swimming speed between the four groups of mice. The results exhibited that inhibition of the NLRP3 inflammasome by CY-09 helped to relieve the cognitive impairment of the 3×Tg-AD mice.

Afterward, we detected the expression and distribution of pathological proteins in AD model mice. As shown in Figure 8, compared with the NTg mice, expressions of APP, betasite app cleaving enzyme 1 (BACE1), sAPPβ, and Aβ1-42 were significantly increased in 3×Tg-AD mice, with the *p*-value lower than 0.01, 0.05, 0.05, and 0.05. Expressions of sAPPα, post-synaptic density protein 95 (PSD95) and synaptophysin were greatly decreased in 3×Tg-AD mice, with the *p*-values all lower than 0.05. However, in CY-09 treated 3×Tg-AD mice, the expressions of these proteins were reversed, with increased expression of sAPPα, PSD95, and synaptophysin; decreased levels of APP, BACE1, sAPPβ and Aβ1-42 were found in comparison to those in the untreated 3×Tg-AD mice, with the *p*-value lower than 0.05 and 0.01, respectively. Immunostaining of 6E10 confirmed the decrease in the expression and distribution of Aβ in CY-09 treated 3×Tg-AD mice (Figures 8i and S4).

**Figure 6.** CY-09 reversed the expression and distribution of metabolic enzymes in 3×Tg-AD mice. (**a**,**c**–**h**) Western blot analysis of cHK1 (HK1 in the cytoplasm), cHK2 (HK2 in the cytoplasm), mHK1 (HK1 in mitochondria), mHK2 (HK2 in mitochondria), pyruvate dehydrogenase α 1 (PDHE1α) and cytochrome c oxidase subunit IV (COX4) in NTg, NTg + CY-09, 3×Tg-AD, and 3×Tg-AD + CY-09 mice. (**b**) Detection of hexokinase activity in the brain tissue of the four groups of mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05, \*\* *p* < 0.01 vs. NTg mice, & *<sup>p</sup>* < 0.05 vs. NTg + CY-09 mice, # *<sup>p</sup>* < 0.05, ## *<sup>p</sup>* < 0.01 vs. 3×Tg-AD mice).

Furthermore, as demonstrated in Figures 9 and S5, we observed increased expression and distribution of pS404-tau in 3×Tg-AD mice, but treatment with CY-09 greatly reduced the expression and distribution, with the *p*-values lower than 0.05 and 0.01. There were no differences in the expressions of Tau5 among the four groups of mice. However, remarkably increased total human tau was found in the 3×Tg-AD mice and the 3×Tg-AD + CY-09 mice, with a *p*-value lower than 0.001, while no reduction was found in the CY-09-treated 3×Tg-AD mice. By contrast, expression of p-GSK3β-Ser9 relative to GSK3β exhibited a decrease in 3×Tg-AD mice (*p* = 0.0834), but it significantly increased after CY-09 treatment, with a *p*-value lower than 0.05. Together, these results demonstrated that CY-09 can reverse the expression and distribution of pathological proteins and alleviate cognitive impairment in the 3×Tg-AD mice.

**Figure 7.** CY-09 improved the learning and memory ability of 3×Tg-AD mice. (**a**) Escape latency and (**b**) swimming speed during training of the four groups of mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *<sup>p</sup>* < 0.05: 3×Tg-AD mice vs. NTg mice, & *<sup>p</sup>* < 0.05: 3×Tg-AD mice vs. NTg + CY-09 mice, # *p* < 0.05: 3×Tg-AD + CY-09 mice vs. 3×Tg-AD mice). (**c**,**d**) Time in target quadrants in 24 h and 72 h space exploration experiments. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05 vs. NTg mice, # *p* < 0.05 vs. 3×Tg-AD mice). (**e**) Schematic diagram of the swimming paths of the four groups of mice.

## *3.8. CY-09 Decreased Oxidative Stress in 3*×*Tg-AD Mice*

Finally, we explored the effects of CY-09 on oxidative stress and ferroptosis. First, we detected ROS and malondialdehyde (MDA) in the four groups of mice. As shown in Figure 10, significantly increased ROS levels and MDA were found in 3×Tg-AD mice, with the *p*-value lower than 0.001 and 0.01. After the CY-09 treatment, ROS and MDA were notably decreased, with the *p*-value lower than 0.001 and 0.05. These data reflected that CY-09 can reduce oxidative stress in 3×Tg-AD mice.

**Figure 8.** CY-09 reversed the expression and distribution of Aβ related proteins in 3×Tg-AD mice. (**a**–**h**) Western blot analysis of amyloid precursor protein (APP), beta-site app cleaving enzyme 1 (BACE1), sAPPα, sAPPβ, Aβ1-42, post-synaptic density protein 95 (PSD95), and synaptophysin in NTg, NTg + CY-09, 3×Tg-AD, and 3×Tg-AD + CY-09 mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05, \*\* *p* < 0.01 vs. NTg mice, & *p* < 0.05, && *p* < 0.01 vs. NTg + CY-09 mice, # *p* < 0.05, ## *p* < 0.01 vs. 3×Tg-AD mice). (**i**) Distribution of Aβ in the CA3 region and cortex of the four groups of mice (scale bar: 100 μm and 250 μm; arrows indicate Aβ in 3×Tg-AD mice).

**Figure 9.** CY-09 reversed the expression and distribution of tau-related proteins in 3×Tg-AD mice. (**a**–**e**) Western blot analysis of Tau5, HT7, p-Tau-Ser404, GSK3β, and p-GSK3β-Ser9 in NTg, NTg+CY-09, 3×Tg-AD, and 3×Tg-AD + CY-09 mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \*\*\* *p* < 0.001 vs. NTg mice, & *p* < 0.05, && *p* < 0.01, &&& *p* < 0.001 vs. NTg + CY-09 mice, # *p* < 0.05 vs. 3×Tg-AD mice). (**f**) Distribution of p-Tau-Ser404 in the CA3 region and cortex of the four groups of mice (scale bar: 100 μm).

Then, we tested the ferroptosis-related proteins, including TF, TFR, FPN, NCOA4, ACSL4, GPX4, and SLC7A11. TF and TFR were responsible for the transport of Fe2+ to the cell, while FPN transported Fe2+ outside the cell. NCOA4 participated in the ferritinophagy. GPX4, SLC7A11, and ACSL4 were critical enzymes in lipid peroxidation. As reported in Figure 11, compared with NTg mice, expressions of TFR and ACSL4 were increased, while expressions of FPN, NCOA4, GPX4, and SLC7A11 were decreased in the 3×Tg-AD mice, with the *p*-values all lower than 0.05. Furthermore, when compared with the NTg + CY-09 mice, the expressions of NCOA4, GPX4, and SLC7A11 were significantly

decreased in the 3×Tg-AD mice, with the *p*-values all lower than 0.05. However, except for ACSL4, there were no differences in these proteins between the 3×Tg-AD mice and the 3×Tg-AD + CY-09 mice. The expression of ASCL4 was reduced in the 3×Tg-AD + CY-09 mice, with a *p*-value lower than 0.05. No differences in TF expression were found between the four groups of mice except for NTg mice and 3×Tg-AD + CY-09 mice. The results implied that the inactivation of the NLRP3 inflammasome by CY-09 cannot reverse the ferroptosis in the 3×Tg-AD mice.

**Figure 10.** CY-09 decreased reactive oxygen species (ROS) levels and malondialdehyde (MDA) expression in the four groups of mice. (**a**) ROS levels in NTg, NTg + CY-09, 3×Tg-AD, and 3×Tg-AD + CY-09 mice; (**b**) expression of MDA in the four groups of mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \*\* *p* < 0.01, \*\*\* *p* < 0.001 vs. NTg mice, && *p* < 0.01, &&& *p* < 0.001 vs. NTg + CY-09 mice, # *p* < 0.05, ### *p* < 0.001 vs. 3×Tg-AD mice).

Reduced expressions of MDA and ACSL4 in the 3×Tg-AD + CY-09 mice prompted us to detect the fatty metabolism changes in the four groups of mice. Data presented in Figure 12, reveal that in comparison to NTg mice, the expressions of acetyl coA carboxylase (ACC), 3-hydroxy-3-methylglutaryl-coenzyme A synthase 1 (HMGCS1), and 3-hydroxy-3 methylglutaryl-coenzyme A reductase (HMGCR) were greatly increased in the 3×Tg-AD mice, with the *p*-values all lower than 0.05. A significant decrease was found in the expression of p-ACC in the 3×Tg-AD mice, with a *p*-value lower than 0.05. After CY-09 treatment, expressions of ACC, p-ACC, HMGCS1, and HMGCR were reversed. Fatty acid synthase (FAS) was also increased in the 3×Tg-AD mice, but with no significant difference in protein levels. Similarly, there were no differences in low-density lipoprotein receptor related protein 1 (LRP1) and sterol regulatory element binding protein 2 (SREBP2) between the four groups of mice, except for LRP1 in CY-09-treated NTg and 3×Tg-AD mice. Increased ACC and FAS and decreased p-ACC indicated increased fatty acid synthesis, consistent with the increased MDA and ACSL4. HMGCS1 and HMGCR are responsible for the synthesis of cholesterol. LRP1 and SREBP2 regulate lipid metabolism homeostasis and cholesterol levels, respectively. The above results suggested that increased fatty acid synthesis may be the main cause of increased lipid peroxidation. CY-09 could reduce the synthesis of fatty acid and lipid peroxidation by inhibiting NLRP3 inflammasome activation in the 3×Tg-AD mice.

127

**Figure 11.** CY-09 could not reverse the ferroptosis in 3×Tg-AD mice. (**a**–**h**) Western blot analysis of transferrin (TF), transferrin receptor (TFR), ferroportin (FPN), nuclear receptor coactivator 4 (NCOA4), long-chain acyl-CoA synthetase 4 (ACSL4), glutathione peroxidase 4 (GPX4), Solute Carrier Family 7, and Member 11 (SLC7A11) in the four groups of mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05 vs. NTg mice, & *p* < 0.05 vs. NTg + CY-09 mice, # *p* < 0.05 vs. 3×Tg-AD mice).

**Figure 12.** CY-09 decreased the synthesis of fatty acids in 3×Tg-AD mice. (**a**–**h**) Western blot analysis of fatty acid synthase (FAS), acetyl CoA carboxylase (ACC), p-ACC, 3-hydroxy-3-methylglutaryl-Coenzyme A synthase 1 (HMGCS1), 3-hydroxy-3-methylglutaryl-Coenzyme A reductase (HMGCR), low-density lipoprotein receptor related protein 1 (LRP1), and sterol regulatory element binding protein2 (SREBP2) in the four groups of mice. (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *<sup>p</sup>* < 0.05 vs. NTg mice, & *<sup>p</sup>* < 0.05 vs. NTg + CY-09 mice, # *<sup>p</sup>* < 0.05 vs. 3×Tg-AD mice).

#### **4. Discussion**

Neuroinflammation has been recognized as a key event in inducing the onset and progression of AD, and it is closely linked to Aβ accumulation, tau hyperphosphorylation, and decreased cerebral glucose metabolism [3]. NLRP3 inflammasome is the most important inflammasome involved in neuroinflammation. In this study, CY-09 was used to inhibit the activation of NLRP3 inflammasome, leading to the restoration of glucose metabolism, as demonstrated by the increases in the expression and distribution of glucose transporters and related enzymes and the attenuation of insulin resistance. Moreover, inhibition of NLRP3 inflammasome activation by CY-09 could also reduce oxidative stress and improve cognitive ability in AD model mice.

In recent years, several studies have shown that NLRP3 inflammasome activation contributes to the progress of AD pathologies, specifically Aβ and tau proteins [42,43]. Inhibition of NLRP3 inflammasome activation helps to reduce Aβ deposition and improve cognitive impairment [43]. Most inhibitors of NLRP3 inflammasome activation do not specifically target NLRP3, except for CY-09, which combines directly with the ATP-binding motif of the NLRP3 NACHT domain [32]. Thus, CY-09 was selected in this study to inhibit NLRP3 inflammasome activation. As crossing the BBB and binding to NLRP3 in the brain is the prerequisite for inhibition, we used LC-MS/MS to characterize the molecular structure of CY-09 and detected its level in the brains of the mice. These results proved that CY-09 crossed the BBB to the brain to exert its biological function. Then, we further detected the expression and distribution of Aβ and tau proteins in CY-09-treated and non-treated 3×Tg-AD mice. Consistent with previous studies [43], our results showed reversed expression levels of pathological proteins in CY-09 treated 3×Tg-AD mice. In addition, CY-09 could also relieve cognitive deficits in the AD mice.

Reduced cerebral glucose metabolism has been reported in the AD brain due to the decrease in glucose transport, insulin resistance, and reduced metabolic enzymes [44–46]. We used 18F-FDG PET to analyze the alteration of glucose metabolism in CY-09-treated and non-treated 3×Tg-AD mice. Our results showed increased glucose uptake and metabolism rates in CY-09-treated 3×Tg-AD mice. GLUT1, GLUT3 and GLUT4 are responsible for glucose uptake and transport in the brain. Decreased expressions of GLUT1 and GLUT3 in AD indicated reduced glucose uptake and transport [47–50]. In this work, GLUT1 and GLUT3 expression levels were found to be increased in CY-09-treated 3×Tg-AD mice, while GLUT4 also exhibited higher expression and distribution. These results implied that inhibiting NLRP3 inflammasome activation by CY-09 may improve insulin resistance. Insulin is a critical hormone that regulates blood glucose [20,51]. It is transported into the brain and binds with insulin receptors, which are expressed in the cell membranes to activate the insulin signaling pathway [21,52,53]. In AD, insulin levels increase in the blood and decrease in the brain, the insulin signal pathway is suppressed, and GLUT4 cannot localize to the cell membrane, leading to reduced glucose uptake. Intranasal insulin injection helps to recover the insulin signaling pathway, stimulate the transfer of GLUT4 to the cell membranes, and thus increase the uptake of glucose in AD [54]. CY-09 has been reported to inhibit NLRP3 inflammasome activation in order to improve insulin resistance in obesity and non-alcoholic fatty liver disease. In this study, elevated insulin levels and recovered insulin signal pathways were detected in the CY-09-treated AD mice. Increased insulin sensitivity was also measured by GTT and ITT in the CY-09-treated AD mice. We also detected the recovered insulin signal pathway in CY-09-treated 3×Tg-AD mice. These data suggested that the inhibition of NLRP3 inflammasome activation by CY-09 helps to alleviate insulin resistance in AD.

Insulin resistance and glucose metabolism dysfunction are two hallmarks of diabetes mellitus. As they were also found in AD, most researchers considered AD as type 3 diabetes mellitus. However, it should be noted that in diabetes, decreased insulin secretion or insulin resistance leads to increased blood glucose [20,47]. Abnormal insulin level in the blood is the main cause of diabetes. While in AD, the higher insulin levels in the blood and lower levels in the brain indicated a damaged insulin signaling pathway. Therefore, maintaining the stability of insulin levels and blood glucose is the key point for diabetes, but increase the insulin levels and glucose metabolism in the brain is more important in AD.

Glucose was transported into the cells by GLUTs and phosphorylated by mitochondriabound HK to initiate glucose metabolism. Decreased expression and abnormal distribution of HK were found in AD. Glycolysis and oxidative phosphorylation (OxPhos) are two main pathways of glucose metabolism [55]. Glycolysis generates little ATP, while OxPhos generates a large amount of ATP to meet the energy needs of the neurons [56]. The dividing point between the two pathways is the metabolic selection of pyruvate. Pyruvate is catalyzed by lactate dehydrogenase to generate lactate in glycolysis, while it is catalyzed by PDHE to generate Acetyl-CoA in OxPhos. An increasing number of studies have

demonstrated that the metabolic pattern of neurons changes from OxPhos to aerobic glycolysis in the AD brain [25,56,57]. Reduced PDHE expression may be one reason for this shift. Here, we detected the increased expression levels of HK1 and PDHE1α in CY-09 treated 3×Tg-AD mice, which are beneficial for maintaining glucose metabolism and ATP production in the brain.

Oxidative stress is another important pathological characteristic that is closely related to NLRP3 inflammasome and glucose metabolism in AD. ROS was released from the oxidative respiratory chain and can activate the NLRP3 inflammasome [58,59]. Here, we found that the inactivation of the NLRP3 inflammasome by CY-09 decreased ROS levels in CY-09-treated 3×Tg-AD mice. Oxidative stress causes iron metabolism disorders and leads to ferroptosis, which is a new area in the research of AD pathogenesis [25,26]. Ferroptosis manifests as increased cellular Fe2+, decreased GPX4, and increased lipid peroxidation [27]. Elevated Fe2+ results from increased transferrin transport and ferritinophagy [28]. Consistent with the studies, we detected increased TF, TFR, and FPN and decreased GPX4 and SLC7A11 in the 3×Tg-AD mice. Unfortunately, there was no significant difference in these proteins between the AD and CY-09-treated AD mice. These results indicate that the inactivation of NLRP3 inflammasome does not reduce ferroptosis. Moreover, a contradiction between decreased ferritinophagy and increased ferroptosis was found in AD. Generally, decreased NCOA4 represents decreased ferritinophagy [28,60]. Thus, the expression and function of this protein remain to be clarified.

As a marker of ferroptosis, ACSL4 also participates in lipid peroxidation. Some studies showed that lipid peroxidation is related to Aβ deposition and the hyperphosphorylation of tau [29–31]. In this paper, the levels of ACSL4 and MDA were increased in 3×Tg-AD mice and decreased after CY-09 treatment. The results prompted us to explore the effect of CY-09 on lipid metabolism, mainly the synthesis and metabolism of fatty acids and cholesterol. FAS and ACC are rate-limiting enzymes of fatty acid synthesis. In AD mice, the expression of FAS and ACC increased, whereas the phosphorylation level of ACC deceased, thus indicating an increase in fatty acids synthesis. HMGCS1 and HMGCR are important for cholesterol synthesis [61]. Increased HMGCS1 and HMGCR were also found in AD mice. Surprisingly, expressions of ACC, HMGCS1, and HMGCR were decreased in CY-09-treated AD mice. This suggests that decreased MDA and lipid peroxidation may be related to the decreased synthesis of fatty acids and cholesterol in CY-09-treated AD mice. LRP1 plays an important role in lipid metabolism, glucose metabolism, insulin signaling, and the elimination of Aβ in AD [26,62]. Increased LRP1 helps to improve cognitive ability in AD [63]. SREBP2 is a negative regulator of LRP1 [64]. However, in our results, no difference in LRP1 and SREBP2 was found between AD mice and CY-09-treated AD mice. Here, a limitation should be noted. In this study, we used only one strain of AD model mice to study the effect of NLRP3 inflammasome activation on glucose metabolism and the role of CY-09. Using two or more strains of mice would help to validate the experimental results. Therefore, the conclusion drawn in this paper is restricted to only the 3×Tg-AD mice, and it should be verified using additional AD models in the near future.

#### **5. Conclusions**

Summarily, inhibiting NLRP3 inflammasome activation by CY-09 helps to restore cerebral glucose metabolism, improve memory and learning ability, and reduce fatty acid synthesis and lipid peroxidation in the 3×Tg-AD mice. Thus, CY-09 has the potential to be developed for the treatment of AD. Further studies are required regarding the shift between glycolysis and OxPhos pathways in AD in order to better understand the mechanism of neuroinflammation and glucose metabolism in the development of AD pathology.

**Supplementary Materials:** The following are available online at https://www.mdpi.com/article/ 10.3390/antiox12030722/s1, Figure S1: Content of CY-09 in NTg and NTg + CY-09 mice brains, Figure S2: Fluorescence intensity of GLUT4 in NTg, NTg + CY-09, 3×Tg-AD and 3×Tg-AD + CY-09 mice (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05, \*\* *p* < 0.01 vs. NTg mice, & *<sup>p</sup>* < 0.05, &&& *<sup>p</sup>* < 0.001 vs. NTg + CY-09 mice, ## *<sup>p</sup>* < 0.01 vs. 3×Tg-AD mice), Figure S3: Fluorescence intensity of (**a**) IR and (**b**) p-IR in NTg, NTg + CY-09, 3×Tg-AD and 3×Tg-AD + CY-09 mice (n = 6, mean <sup>±</sup> SD, one-way ANOVA and Bonferroni post hoc test; \*\* *<sup>p</sup>* < 0.01 vs. NTg mice, && *<sup>p</sup>* < 0.01 vs. NTg + CY-09 mice, # *<sup>p</sup>* < 0.05 vs. 3×Tg-AD mice), Figure S4: Fluorescence intensity of Aβ (6E10) in (**a**) CA3 region and (**b**) cortex of NTg, NTg + CY-09, 3×Tg-AD and 3×Tg-AD + CY-09 mice (n = 6, mean <sup>±</sup> SD, one-way ANOVA and Bonferroni post hoc test; \* *<sup>p</sup>* < 0.05 vs. NTg mice, & *<sup>p</sup>* < 0.05 vs. NTg + CY-09 mice, # *<sup>p</sup>* < 0.05, ## *<sup>p</sup>* < 0.01 vs. 3×Tg-AD mice), Figure S5: Fluorescence intensity of p-tau-S404 in (**a**) CA3 region and (**b**) cortex of NTg, NTg + CY-09, 3×Tg-AD and 3×Tg-AD + CY-09 mice (n = 6, mean ± SD, one-way ANOVA and Bonferroni post hoc test; \* *p* < 0.05 vs. NTg mice, & *<sup>p</sup>* < 0.05 vs. NTg + CY-09 mice, ## *<sup>p</sup>* < 0.01 vs. 3×Tg-AD mice).

**Author Contributions:** Conceptualization, S.H. and Q.L.; data curation, Q.L.; formal analysis, S.H.; funding acquisition, J.N. and Q.L.; investigation, S.H.; methodology, S.H.; project administration, S.H., K.Z. and Y.H.; resources, Z.H. and X.L.; software, X.H.; supervision, J.N. and Q.L.; writing—original draft, S.H.; writing—review and editing, P.X., Q.X. and Q.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Key Research and Development Program of China, grant No. 2018YFE0118900; the Shenzhen Science and Technology Innovation Commission, grant No. JCYJ20200109110001818; the Guangdong Provincial Key S&T Program, grant No. 2018B030336001, and the Shenzhen-Hong Kong Institute of Brain Science-Shenzhen Fundamental Research Institutions, grant No. 2022SHIBS0003.

**Institutional Review Board Statement:** The animal study protocol was approved by the Institutional Animal Care and Use Committee of Tongji Medical College of Huazhong University of Science and Technology (IACUC Number: 2390; approved date: 27 February 2018). All the animal experiments complied with the animal care protocol.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author.

**Acknowledgments:** We would like to thank the Instrumental Analysis Center of Shenzhen University (Xili Campus) and the Central Research Facilities of the College of Life Science and Oceanography of Shenzhen University for providing access to the experimental instruments (laser scanning confocal microscope, LSM710, ZEISS, Oberkochen, Germany).

**Conflicts of Interest:** The authors declare the following financial interest or personal relationships that may be considered as potential competing interests: Q.X. reports that he is the chief scientist and co-founder of Suzhou RAYCAN Technology Co., Ltd. (Suzhou, China). S.H., Z.H., X.H., X.L., K.Z., Y.H., P.X., J.N. and Q.L. have nothing to disclose.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
