Paeoniflorin Upregulates Mitochondrial Thioredoxin of Schwann Cells to Improve Diabetic Peripheral Neuropathy Indicated by 4D Label-Free Quantitative Proteomics

Diabetic peripheral neuropathy (DPN) is a diabetic complication characterized by demyelination. The pathogenesis of DPN has not been fully elucidated, thus lacking therapies. In the current study, we aimed to confirm whether paeoniflorin (PF) could improve DPN by upregulating mitochondrial thioredoxin (Trx2) based on 4D Label-free proteomic experiments of Schwann cells (SCs) mitochondria. Firstly, PF increased the expression of mitochondrial processing peptidase α (Pmpca) and small ubiquitin-related modifier 1 (Sumo1) to increase mitochondrial protein processing of Trx2. Then, PF increased the protein expression of Trx reductase 2 (TrxR2) and peroxiredoxin 3 (Prx3), which belong to mitochondrial Trx systems. Accordingly, PF decreased mitochondrial reactive oxygen species (ROS) while increasing mtDNA and mitochondrial membrane potential to improve mitochondria function under high glucose environment. Furthermore, total glucosides of paeony capsules (TGP), containing more than 90% PF, increased the Trx2, TrxR2, and Prx3 levels in sciatic nerve of DPN rats, thus reducing demyelination as well as improving mechanical pain threshold, thermal pain threshold, motor nerve conduction velocity (MNCV), and sensor nerve conduction velocity (SNCV). Overall, these results suggest that PF could provide protection for DPN by upregulating Trx2.


Introduction
According to the International Diabetes Federation's prediction in 2021, diabetes is a significant global challenge to the health and well-being of individuals, families, and societies. More than 50% of diabetes will develop into DPN [1]. DPN can cause neuropathic pain and foot ulcers and is also the main cause of leg amputation. DPN seriously reduces the quality of life of patients and increases the mortality of diabetes. Despite decades of research, our understanding of the molecular mechanisms underlying is still in its infancy [2,3]. Elucidating the complex pathogenesis and looking for effective treatment are still the focus of attention.
Hyperglycemia-induced excessive formation of mitochondrial ROS is considered a unified pathogenesis of diabetic complications including DPN [4]. Demyelination is one of the pathological features of DPN. ROS leads to mito-chondrial dysfunction in SCs which has been proved to be one of the mechanisms of demyelination [5]. The previous studies of our research group confirmed that PF upregulated the endogenous antioxidant stress nuclear factor erythroid 2 related factor 2 (Nrf2) pathway of SCs in a high glucose environment to reduce mitochondrial apoptosis [6]. In addition, we have improved that PF could increase mitochondrial fusion and reduce mitochondrial fission to maintain the balance of mitochondrial dynamics (data was not shown).
PF can improve the mitochondrial function of SCs, suggesting that it may be an important substance to intervene in demyelination. However, the target of PF regulating mitochondrial function has not been fully understand. Whether PF can intervene DPN by improving mitochondrial function is an urgent scientific problem to be solved. Based on previous studies, we performed 4D Label-free proteomic experiments on the mitochondria of SCs. The mechanism and target of PF to improve mitochondrial function will be clarified, and the scientific basis for the development of new drugs for the intervention of DPN will be provided.

Materials and Methods
2.1. Cell Culture and Treatment. RSC96 cells (CRL-2765) were obtained from ATCC and cultured in DMEM (containing 25 mM glucose) supplemented with 10% FBS at 37°C in a humidified 5% CO 2 atmosphere. The cells were treated with 150 mM glucose and 150 mM glucose pulsed 10 μM paeoniflorin (PF) for 24 hours.

Mitochondrial Isolation.
In RSC96 cells in T75 flask, mitochondrial fractions were prepared using the mitochondrial isolation kit for cultured cells (89874, Thermo Fisher). Briefly, 800 μL reagent A and 10 μL reagent B were added to the cells (2 × 10 7 ) and then were incubated 5 minutes on ice. 800 μL reagent C was added and centrifuged at 700 × g for 10 minutes at 4°C. The supernatant was collected and centrifuged at 12,000 × g for 15 minutes at 4°C. The supernatant was removed, and the pellet (mitochondria) was washed with 500 μL of reagent C. Mitochondrial fraction was collected by centrifugation using a centrifuge at 12,000 × g for 5 minutes at 4°C.

4D Label-Free Proteomic Analysis of Mitochondria.
Mitochondrial sample was sonicated on ice using a highintensity ultrasonic processor (Scientz) in lysis buffer (8 M urea and 1% Protease Inhibitor Cocktail) followed by centrifugation at 12,000 × g for 10 minutes at 4°C. Protein concentration was determined with BCA kit according to the manufacturer's instructions. For digestion, the protein solution was reduced with 5 mM dithiothreitol for 30 minutes at 56°C and alkylated with 11 mM iodoacetamide for 15 minutes at room temperature in darkness. The protein sample was then diluted by adding 100 mM TEAB to a urea concentration less than 2 M. Finally, trypsin was added at 1 : 50 trypsin-toprotein mass ratio for the first digestion overnight and 1 : 100 trypsin-to-protein mass ratio for the second 4-hour digestion. The tryptic peptides were analyzed on NanoElute UPLC coupled to timsTOF Pro (Bruker Daltonics, USA) mass spectrometer. Mass spectrometric measurements were carried out using the parallel accumulation serial fragmentation (PASEF) acquisition method. The obtained data were analyzed using MaxQuant version 1.6.5.0. Proteins with a fold change ðFCÞ > 1:5 (or <0.6667) and p < 0:05 were regarded as differentially expressed proteins (DEPs). Detailed methods are described in Supplementary material 1. Then, the cells were fixed by 4% paraformaldehyde, permeabled membrane by 0.1% Triton X-100, and blocked by 3% BSA. The cells were incubated overnight with mtTFA antibody (1 : 500, ab176558, Abcam). In the next day, the cells incubated with the secondary antibody donkey anti-rabbit IgG-H&L (Alexa Fluor 488). DAPI was used to stain cell nuclei. Fluorescent images were captured by Zeiss LSM 880 with Airyscan. Images were analyzed using Image-Pro Plus 6.0 software.

2.7.
Diabetic Rat Treatment. The protocol for using animals was approved by the Ethics Review Committee for Animal Experimentation of Capital Medical University (Ethical Inspection Number: AEEI-2018-144). Male Sprague-Dawley (SD) rats obtained from Experimental Animal Center at the Capital Medical University were randomly divided into three groups: normal, DPN, and total glucosides of paeony capsule (TGP, H20055058, Liwah Pharmaceutical) groups. The rats were fasted for 12 h and intraperitoneal injected of 60 mg/kg STZ (dissolved in 0.1 M citrate buffer, pH 4.5 at 4°C), while the rats in the normal group were injected only with citrate buffer. Fasting blood glucose (FBG) concentrations greater than or equal 16.7 mM after 72 h were treated as diabetes and were used for further experiments. This study uses the method of Yang et al. and the method description partly reproduces their wording [7][8][9]. The TGP group rats were intragastrically administered with TGP suspension (5.6 mg/ kg/day), and the normal and DPN group rats were given equal volume distilled water everyday day for 12 weeks.

Mechanical Sensitivity Measurements. A series of Von
Frey filaments (NC12773, Exacta) were employed to stimulate the plantar surface of the left hind paw with pressure. Paw withdrawal in response to each stimulus was recorded, and more than or equal 60% paw withdrawals were recorded as positive. 3 Oxidative Medicine and Cellular Longevity 2.11. Ultrastructure Analysis. Ultrastructure myelination of sciatic nerve was analyzed by T transmission electron microscope (HT7700, Hitachi). This study uses the method of Yang et al. and the method description partly reproduces their wording [8].
2.13. Oxidative Phosphorylation (OXPHOS) Assay. Mitochondria of the cells were extracted using the mitochondrial isolation kit for cultured cells. Total OXPHOS rodent WB antibody cocktail (ab110413, Abcam) was used to analyze the relative levels of OXPHOS complex in mitochondria by Western blot.       Oxidative Medicine and Cellular Longevity Proteintech). Fluorescent images were captured by Zeiss LSM 880 with Airyscan. Images were analyzed using Image-Pro Plus 6.0 software.

Statistical Analysis.
Data presentation and statistical analyses were carried out using Prism 7.04 software (Graph-Pad) and graphed as means ± standard error of the mean (S. E.M). Differences were analyzed by one-way ANOVA followed by Tukey's multiple comparisons test. p < 0:05 were considered statistically significant.

Proteomic Analyses Showed That Paeoniflorin
Upregulated SC Mitochondrial Trx2 in a High Glucose Environment. Protein extracts from mitochondria of RSC96 cells were collected for 4D Label-free analysis (Figure 1(a)). In total, we found that there were 177 DEPs (HG_PF/HG), including 152 upregulated proteins and 25 downregulated proteins (Figure 1(b)). Among them, there were 16 upregulated proteins and 2 downregulated protein subcellular located in mitochondria (Figure 1(d) and Table 1). Among these differentially expressed proteins, Trx2 was remarkably upregulated in mitochondria (FC = 3:112, p = 0:005) (Figure 1(c)) while downregulated in 25 mM glucose group compared to 150 mM glucose groups (FC = 0:664, p = 0:042). DEPs were classified by Gene Ontology annotation based on three categories: biological process, cellular component, and molecular function. Trx2 was related to cellular process, metabolic process, biological regulation, single-organism process, and response to stimulus. The molecular functions of Trx2 were known as binding and catalytic activity (Figure 1(e)). In this experiment, only 18 mitochondrial proteins met the criteria for DEPs of this proteomic analysis. This subcellular localization information given by proteomics is based on the subcellular structure prediction software wolfpsort. In Gene Ontology (GO) assignments "mitochondrion" and MitoCarta3.0, we identified 1,072 mitochondrial proteins (Supplementary material 2). This is consistent with others showing that widely used data repositories taken together can identify more mitochondrial proteins [10]. We found that the extracted proteins also contained proteins from other organelle sources. The high sensitivity of mass spectrometry-(MS-) based approaches leads to the identification of even minor contaminations of organelle preparations, and thus, the numerous lists of identified proteins contain a mixture of authentic mitochondrial proteins, loosely attached proteins, and contaminants [11]. In addition to proteomic analysis based on purified mitochondrial fraction, subtractive proteomic experiments and integrative mitochondrial proteomic methods were reported to try to get a highconfidence mitochondrial proteome [12].
From the previous results, we observed that PF increased Nrf2 protein level and nuclear translocation to reduce mitochondrial oxidative stress and damage [6]. Recent research indicated that disrupting mitochondrial thiol redox homeostasis leads to Nrf2 activation [13]. The mitochondrial Trx system is one of the mitochondrial thiol redox systems. The above analysis indicates that PF may increase Nrf2 activation by upregulating Trx2.
Moreover, Trx2 plays important roles in mitochondrial functions in view of the proposed specific interactions [14].

Oxidative Medicine and Cellular Longevity
In the analysis of the reported mitochondrial interactome of Trx2 and our Trx2-interacting proteins by proteomics of mitochondria in SCs, PF could upregulate 11 proteins and downregulate 6 proteins ( Table 2). These proteins involve in TCA cycle, glycolysis, amino acid metabolism, mitochondrial membrane integrity, mitochondrial protein synthesis, protein mitochondrial import and folding, and oxidative stress [14][15][16].
Based on the results of our previous study and these reports, we believe that PF can improve mitochondrial function by upregulating Trx2 as well as regulating its mitochondrial interactome. To test our hypothesis, we performed experiments on measuring the expression of Trx2.

Paeoniflorin Enhances Mitochondrial Function in SCs by
Upregulating Trx2. Trx2 exerts antioxidant stress and maintains mitochondrial function, which depends on the small ubiquitin-related modifier (SUMO)ylated mitochondrial processing peptidase (MPP) including MPPα, also known as Pmpca, to regulate its protein processing in mitochondria [17,18]. The unprocessed Trx2 (about 23 kDa) in the cytoplasm is transported to the mitochondrial matrix through the mitochondrial outer membrane transposase complex and the mitochondrial inner membrane transposase complex under the action of the N-terminal guiding sequence mitochondrial targeting signal peptide (MTS) [19].   [15]. We first examined whether PF affected the mitochondrial protein processing of Trx2. Our results showed that PF increased the expression of mature Trx2 in SCs under high glucose environment, consistent with increasing the expression of Pmpca and Sumo1, thus PF increasing the protein processing of Trx2 in mitochondria (Figure 2(a)). The mitochondrial Trx system consists of Trx2 and Trx reductase 2 (TrxR2), which maintains Trx2 in a reduced state by using mitochondrial NADPH as a substrate. Trx2 maintains the activities of the peroxidase Peroxiredoxin 3 (Prx3) [19]. To determine whether PF increases the mitochondrial Trx system, we further assessed the levels of TrxR2 and Prx3. The data showed that PF increased the protein expression of TrxR2 and Prx3 (Figure 2(b)).
Mitochondria are involved in ATP production through oxidative phosphorylation (OXPHOS) [20]. Trx2 was involved in the functional and structural integrity of the electron transport chain (ETC) related to OXPHOS. Mitochondrial genomes (mtDNA) encode proteins in ETC. Furthermore, the ETC makes mitochondria an important organelle involved in producing ROS [21]. Trx2 can directly act as a ROS scavenger and increases the mitochondrial membrane potential (MMP) [22]. Our results showed that PF increased OXPHOS under high glucose environment (Figure 2(c)). Meanwhile, PF increased mtDNA and MMP while decreased mtROS (Figure 2(d)). We conclude that PF enhanced mitochondrial function through upregulating Trx2 due to increased mitochondrial protein processing of Trx2, direct antioxidant stress, and mitochondrial interactome of Trx2.

TGP Improvement Nerve Function in DPN Rats Is
Related to the Upregulation of Trx2. Trx2 plays an important role in antioxidative stress and maintenance of mitochondrial function. It has been confirmed that Trx2 can provide protection for diabetes and complications such as diabetic cardiomyopathy and diabetic retinopathy [23,24]. Whether Trx2 can protect DPN has not been reported. Schwann cells (SCs) are the most abundant cells in the peripheral nervous system and belong to myelin forming cells. Mitochondrial dysfunction in SCs induces demyelination, which is one of the main pathogenesis of DPN [25]. This study has confirmed that PF upregulates Trx2 of SCs in a high glucose environment to improve mitochondrial function. We further studied whether PF can improve DPN and whether it is related to the upregulation of Trx2.
In this part of the experiment, we used TGP to intervene DPN rats. TGP is an effective component extracted from the dried root of Paeonia lactiflora Pall. PF accounts for more than 90% of the total glycosides and is the main active component of TGP [26]. Our results showed that TGP increased the myelin sheath of DPN rats (Figure 3(a)). It is suggested that TGP can improve mitochondrial dysfunction of SCs to alleviate demyelination. We further examined nerve function represented by mechanical pain threshold, thermal pain threshold, motor nerve conduction velocity (MNCV), and sensor nerve conduction velocity (SNCV). The data showed that TGP increased the mechanical pain threshold, MNCV, and SNCV while decreased thermal pain threshold, thus improving peripheral nerve function in DPN rats (Figure 3(b)).
To investigate whether the protection of nerve function was associated with regulating Trx2, we analyzed Trx2, TrxR2, and Prx3 levels in the sciatic nerve of DPN rats. Trx2, TrxR2, and Prx3 were increased in the TGP group compared to the model group (Figure 3(c)). In summary, we have shown that TGP can provide protection for DPN by upregulating Trx2.

Conclusions
In conclusion, the current study revealed that PF upregulated the Trx2 through increasing its mitochondrial protein processing. Enhanced mitochondrial Trx2 improved mitochondrial function of SCs under high glucose environment. This may be one of the mechanisms of PF intervention in DPN. In addition, Trx2 possesses the potential to be a useful molecular target in the clinical treatment of DPN. The inadequacy of our experiments is lacking of research on mitochondrial interactome of Trx2 in DPN, which would be the goal of our subsequent studies.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.