Up-regulating Lipin1 Attenuates High Glucose-Induced Apoptosis in RSC96 cells Through Regulating Mitochondrial Dynamics

Diabetic peripheral neuropathy, one of the major complications resulting from diabetes, affects diabetic patients with high morbidity and mortality. It has emerged as a severe public health problem. However, its underlying pathogenesis and effective treatment strategies have not been fully studied. In the present study, we explored the role of lipin1 on mitochondrial function of Schwann cells under diabetic conditions. With high glucose stimulation or lipin1 down-regulation, the cell viability of Schwann cells was inhibited,accompanied with mitochondrial dysfunction and morphological abnormality. Besides, we found that high glucose stimulation or lipin1 silencing also disturbed the balance of mitochondrial dynamics, presented as increased levels of mitochondrial ssion-related proteins (including DRP1 and FIS1) and decreased levels of mitochondrial fusion-related proteins (including MFN1 and OPA1). Furthermore, we demonstrated that up-regulating lipin1 ameliorated high glucose-induced disorder of mitochondrial dynamics and functions, and ultimately improved cell viability. Our results suggest that lipin1 may play a protective role on high glucose-stimulated Schwann cells through regulating the balance of mitochondrial dynamics. alleviated HG-induced peripheral neuropathy, emerged as enhanced motor nerve conduction velocity, improved nerve pathological morphology, and increased nerve ber diameter and nerve ber density (Xu . These observations were consistent with previous studies showing that lipin1 deciency animals develop peripheral neuropathy, characterized by myelin degradation, decreased motor nerve conduction velocity and compound muscle action potentials (Nadra et al., 2008) . Taken together, these observations demonstrated that lipin1 may have a relationship with the development of diabetic peripheral neuropathy.


Introduction
Diabetes mellitus is a complex metabolic disorder which is characterized by abnormal glucose and lipid metabolism (Huang et al., 2017) . Diabetic peripheral neuropathy (DPN), one of the most common microvascular complications resulting from diabetes, affects more than 50% of diabetic patients over the course of the disease (Feldman et al., 2017) . The clinical symptoms of DPN include numbness, burning, tingling sensation, and intractable pain, seriously affect the quality of life in patients (Sandireddy et al., 2014) . Earlier works demonstrated that DPN was induced by multifactorial metabolic disorders, including nitrogen and oxidative stress, increased advanced glycation end-products, neuroin ammation, accumulation of sorbitol and so on (Juster−Switlyk and Smith, 2016) . However, the underlying pathophysiologic mechanism of DPN remains incompletely understood and there is no effective therapy that could permanently reverse neuropathy. diacylglycerol (DAG) improves mitochondrial ssion by recruiting ssion-related proteins (Alshudukhi et al., 2017) . As we know, blood glucose elevation could in uence the lipid metabolism in cells. Previous studies demonstrated the alterations of lipid composition, including total-cholesterol (TC), triglycerides (TG), lowdensity lipoprotein cholesterol (LDL-C), and fatty acid (FFA) levels in diabetic animal models (Jiang et al., 2019; Wang et al., 2020) . Since lipid metabolism are associated with high glucose and mitochondrial dynamics, we hypothesis that high glucose stimulation may affect properties of mitochondrial membrane lipids and further disturb mitochondrial dynamics.
Lipin1 is the key phosphatidate (PAP) enzyme in lipid metabolism. It is well-known that DAG, function as a crucial second messenger in lipid metabolism, is the precursor for triacylglycerol (TAG), neutral phospholipid, phosphatidyllecithin, and cephalin. Lipin1 catalyzes the dephosphorylation of phosphatidic acid (PA) to DAG by its PAP enzyme activity, and further mediates the biogenesis of phospholipids and transduction of signaling pathway. Studies have shown that activity declines of PAP1 could control the pathway by which membrane lipids are synthesized, and in uence the related cellular biological functions, including cell membrane stability, signal pathway transmission, adipocyte and nerve cell differentiation, and respiratory de ciency (Nadra et al., 2008;Csaki et al., 2013) . Lipin1, serves as a key regulator in lipid metabolism, is crucial for myocyte injury in skeletal muscle though regulating mtDNA and total mitochondrial protein level (Schweitzer et al., 2019) . Furthermore, previous study suggested that lipin1 may impact on mitochondrial function, presented as regulating mtDNA and total mitochondrial protein level in skeletal muscle (Schweitzer et al., 2019) . Taken together,we propose that lipin1 modulate mitochondrial membrane lipid composition and thus in uence mitochondrial dynamics and functions.
Until now, most investigations have tended to focus on the function of lipin1 in liver, adipose tissue, skeletal muscle and breast cancer (Arai et al., 2018;Rashid et al., 2019;Song et al., 2020) , surprisingly little information is available concerning the functional status of lipin1 in peripheral nerves. Schwann cells are the most numerous glial cells of peripheral nerves and in uence many aspects of nerve functions, such as neurotrophic factors secretion, nervous impulses along axons, extra cellular matrix production and repairment of injured nerves (Yan et al., 2018) . Numerous researches demonstrated that Schwann cell . Our previous research showed that the level of lipin1 protein in cultured Schwann cells was signi cantly decreased when exposed to high glucose, while up-regulating the expression of lipin1 attenuated neuronal damage in diabetic animal models (Xu et al.,2015) . These results suggested that lipin1 may be related to the development of DPN.
Taken together, we intend to identify the relationship between lipin1 and DPN, and further to explore the underlying mechanisms. Our results showed that with exposure to high glucose or lipin1 silencing, the cell viability of Schwann cells was declined, concurrent with mitochondrial dysfunction and mitochondrial dynamics imbalance. However, over-expression of lipin1 could relieve the impairment of cell viability and mitochondrial functions in Schwann cells with high glucose stimulation. Our study suggested that lipin1 could modulate the cell function of Schwann cells by affecting mitochondrial dynamics homeostasis,providing a novel approach for the prevention and treatment of DPN.

Lentivirus transfection
Lipin1 silencing or over-expressing lentivirus transfection of Schwann cells were performed. Lentiviral vectors and the transfection enhancer were purchased from Genchem Biotechnology Company (Shanghai, China). Transduction RSC96 Schwann cells were seeded into 96-well plates at a density of 4 × 10 3 cells for each well and incubated for the experiment with conditioned media. Cells were treated with lentiviral vectors and the transfection enhancer HitransG A according to the manufacture structures. In addition, empty lv-GFP lentivirus was also transfected and acts as a control. The cells were collected for quantitative PCR or western blotting assay to assess the transfection e cacy.

ATP content
The ATP content levels of schwann cells in each group were measured using ATP bioluminescent assay kit (Solarbio, Beijing, China) by following the manufacturer's instructions. In brief, the supernatant was discarded and the cells were homogenized in hot double distilled water at a density of 1 × 10 6 cells. Then the cells were mixed with detection reagent coming with the kit and centrifuged at 4000 rpm for 5 minutes. The optical density of the cells was assessed using a microplate luminometer (RZ-9618; Ruiz). The cellular ATP content was determined from the ATP standard curve, according to the manufacturer's instructions.

Mitochondrial membrane potential
Mitochondrial membrane potential (MMP) was assessed with JC-1(5,50,6,60-tetrachloro-1,10,3,30tetraethyl-benzimidazolyl-carbocyanine iodide) staining. Brie y, Schwann cells with different stimuli were rinsed with ice-cold PBS, and then stained with 2 µL/mL JC-1 for 15 min at 37°C in the dark. After washing with PBS to remove the unbound dye, the results were examined using a ow cytometer equipped with an argon laser. The ratio of red and green uorescence was measured as an indicator of membrane potential change. Fluorescence intensity was quanti ed using Image J software. The green uorescence intensities from the JC-1 monomer leaked out to the cytoplasm and the red uorescence intensities from the aggregated form of JC-1 within the mitochondria.

Western blot analyses
Schwann cells in different groups were collected and homogenized in ice-cold lysis buffer. Protein concentrations were determined using a bicinchoninic acid protein assay kit (Thermo Scienti c, USA). Equal amount of protein was separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto polyvinylidene di uoride (PVDF) membrane (Millipore, USA).
Subsequently, the membranes were blocked with 5% fat-free milk (Solarbio, Beijing, China) and then incubated overnight at 4°C with primary antibodies: rabbit monoclonal anti-total-DRP1 (dynamic-related protein 1), rabbit monoclonal anti-MFN1 (mitofusin-1), rabbit monoclonal anti-OPA1 (optic atrophy-1) antibodies (Cell Signaling Technology, Boston, MA, USA) and rabbit polyclonal anti-FIS1 (mitochondrial ssion protein-1) antibody (Abcam, Cambridge, UK). After being incubated with the appropriate horseradish peroxidase-conjugated secondary antibodies (1:5,000) in a shaker at 37°C for 1 h, the membrane was washed four times with PBS for 10 min. The protein complexes were visualized using an Enhanced Chemiluminescence kit and analyzed using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). Finally, the relative protein expression levels were normalized to that of total actin (Bioss, Beijing, China).

Electron microscopy
The samples used for transmission election microscopy were collected, xed with 3% glutaraldehyde at 4°C for 2 h. After 3 times rinsed in phosphate buffer solution, the samples were dehydrated in a graded series of ethyl alcohol and embedded in epon. Then, the ultrathin sections were stained with uranyl acetate and lead citrate. Finally, the samples were examined under a transmission electron microscope (JEOL-1200EX, Osaka, Japan).

Statistical analysis
All data are expressed as mean ± standard deviation values. The normality of the distribution of variables was assessed using the non-parametric Kolmogorov-Smirnov test. For data with a normal distribution, one-way ANOVA were applied, while for data that are not normally distributed, Kruskal-Wallis test were used. Post hoc comparisons were performed with a level of signi cance set at P ≤ 0.05. Statistical analysis was performed using SPSS® version 17.0.

Lipin1 affects cell viability in high-glucose-stimulated Schwann cells
To investigate appropriated experimental condition, RSC96 Schwann cells were incubated with 50-100 mM glucose to simulate an in vitro model of diabetic neuropathy, with 25 mM concentration of glucose serving as the normal concentration. The MTT assay was performed to assess the cell proliferation of Schwann cells in different groups. As shown in Fig. 1a, with the increase of glucose concentration, the cell viability was inhibited in a dose-dependent manner. Compared with the normal group, treatment of 50 mM glucose resulted in no obvious difference(P > 0.05). However, stimulation with 75-and 100-mM glucose signi cantly decreased cell proliferation, with 52% inhibition at 100 mM glucose treatment. Hence, 100 mM was selected as the optimal stimulation concentration for the following experiments.
Earlier studies from our laboratory have provided the insights that protein level of lipin1 was reduced in RSC96 Schwann cells upon high glucose stimulation, accompanied with cell viability inhibition (Xu et al., 2014) . Hence, to explore the relationship between lipin1 and cell viability in RSC96 Schwann cells, lipin1 silencing and overexpressing lentivirus transfection was constructed to further clarify this relationship between Lipin1 and cell viability. Cells overexpressing lipin1 (Lv-Lipin1) and the (Lv-Lipin1-Con) were subjected to HG conditions (HG + Lv-Lipin1 group and HG + Lv-Lipin1-Con group). Cells of lipin1 silencing (Lv-Lipin1 shRNA) and the corresponding controls (Lv-Lipin1 shRNA-Con) were cultured under normal glucose condition (NG + Lv-Lipin1shRNA group and NG + Lv-Lipin1shRNA-Con group). Successful genetic modi cation was visualized using uorescence microscopy, with > 90% of these RSC96 Schwann cells were detected positive for green uorescence (Fig. 1b).
The Schwann cells in all groups were tested by MTT assay and ow cytometry to assess the effect of Lipin1 on glucose -induced cell injury. As shown in Fig. 1c and 1d, compared with cells in the NG + LV-Lipin1shRAN-Con group, the cell viability within both HG + LV-Lipin1-Con and NG + LV-Lipin1shRNA group were signi cantly decreased (P < 0.05). And there is no signi cant difference between NG + LV-Lipin1shRAN and HG + LV-Lipin1-Con group. Moreover, the cell viability in HG + LV-Lipin1 group was signi cantly higher than that in HG + LV-Lipin1-Con group cells (P < 0.05). Lipin1 overexpression ameliorated cell viability under high glucose condition. These results suggest that lipin1 may have relationship with the alteration in Schwann cells viability exposed to high glucose.
3.2 Lipin1 enhances mitochondrial function in Schwann cells exposed to high glucose concentration.
To assess mitochondrial function, we measured mitochondrial membrane potential which is a vital parameter of mitochondrial metabolism (Alshudukhi et al., 2018) and ATP content in RSC96 Schwann cells under different conditions. Stimulation of Schwann cells with high glucose weakened mitochondrial function, re ected in decreased MMP (Fig. 2a, 2b) and ATP content (Fig. 2c) (P < 0.05). Moreover, the effect of lipin-1 silencing on mitochondrial function was similar to those of high glucose exposure in Schwann cells. On the contrary, lipin1 overexpressing in Schwann cells ameliorated high glucose-induced mitochondrial dysfunction(P < 0.05). These results indicated that high glucose-aroused impaired mitochondrial function and weakened mitochondrial metabolism was ameliorated upon lipin-1 overexpressing in Schwann cells.

Lipin1 protected against high glucose-induced mitochondrial morphological disorder in Schwann cells
Using transmission electron microscopy (TEM), the mitochondria ultrastructure under different conditions were observed. As shown in Fig. 2d, mitochondria from NG + LV-Lipin1shRNA-Con group displayed intact morphology with parallel cristae. By contrast, mitochondria in the HG + Lv-Lipin1-Con group exhibited irregular, dense cristae and even leaving empty vesicle-like structure, which were also observed in the NG + LV-Lipin1shRNA group. After lipin1 up-regulating, disorganized cristae were ameliorated in HG + Lv-Lipin1 groups. These results demonstrate that Lipin1 reversed the mitochondrial morphological de cit observed in Schwann cells with high glucose stimulation.

Lipin1 ameliorates the imbalance of mitochondrial dynamics induced by high glucose stimulation
To measure the effect of lipin1 on mitochondrial dynamics in Schwann cells exposed to hyperglycemia, western blot analysis was performed to assess the expression levels of mitochondrial fusion -and ssion -related proteins. As shown in Fig. 3, the level of ssion-related protein, such as total-DRP1, were elevated in the HG + LV-Lipin1-Con group and the NG + LV-Lipin1shRNA group,compared with those in NG + LV-Lipin1shRNA-Con group (P < 0.05). However, lipin1 overexpression reduced the protein level of total-DRP1 exposed to high glucose (P < 0.05). As that observed for total-DRP1 similar changes in FIS1 protein level ( ssion -related protein) were obtained in these conditions, but there was no signi cant statistical difference. In contrast, lipin1 silencing inhibited the level of MFN1 under normal conditions, but there was no signi cant statistical difference. The expression level of MFN1 was suppressed after high glucose stimulation, while lipin1 overexpression reverse the inhibited MFN1 protein level exposed to high glucose (P < 0.05). The level of OPA1 another mitochondrial fusion protein, was decreased in the HG + Lv-Lipin1-Con group and NG + Lv-Lipin1shRNA group, as compared with that observed in the NG + Lv-Lipin1shRNA-Con group. On the contrary, with lipin1 upregulation in the HG + Lv-Lipin1 group, the level of OPA1 was also increased, compared to the HG + Lv-Lipin1-Con group. However, there was no signi cant statistical difference, which might be related to the insu cient sample size. These results demonstrate that lipin1 could rectify high glucose -induced imbalance of mitochondrial dynamics in Schwann cells.

Discussion
Type 2 diabetes mellitus (T2DM) is a complex metabolic disorder that is characterized by abnormal glucose and lipid metabolism. Hence, exploring the mechanisms of lipid metabolism is vital and may offer clues to the discovery of new strategies for diabetes mellitus. Lipin1, acts as the phosphatidate phosphates, play a key role in the regulation of lipid metabolism at multiple nodal points. Lipin1 usually locates in the cytosol and translocate to the endoplasmic reticulum (ER) to catalyze the conversion of PA to DAG, which is a key substrate for the synthesis of TAG, PE and PC. Lipin1 is highly expressed in liver, adipose, skeletal muscle, and is also present in endoneurium of peripheral nerve of adult mice (Nadra et al., 2008) . Earlier studies from our laboratory have found that the expression of lipin1 in sciatic nerve and hippocampus were signi cantly decreased in diabetic neuropathy rat models (Xu et al, 2015(Shang et al., 2020) . Meanwhile, over-expression of lipin1 alleviated HG-induced peripheral neuropathy, emerged as enhanced motor nerve conduction velocity, improved nerve pathological morphology, and increased nerve ber diameter and nerve ber density (Xu et al, 2015) . These observations were consistent with previous studies showing that lipin1 de ciency animals develop peripheral neuropathy, characterized by myelin degradation, decreased motor nerve conduction velocity and compound muscle action potentials (Nadra et al., 2008) . Taken together, these observations demonstrated that lipin1 may have a relationship with the development of diabetic peripheral neuropathy.
Peripheral nerves are composed of three distinct tissue compartments: the epineurium, perineurium, and endoneurium. The outermost epineurium surrounds perineurium, which in turn surrounds the fascicle of axons, Schwann cells, and broblasts. Schwann cells, derived from the neural crest, are the most abundant cells of the endoneurium. In the peripheral neural system, the myelin sheath allows for e cient action potential transmission and provides trophic support for related axons (Nadra et al., 2008) . Nerve myelin sheath is the extended plasma membrane of the Schwann cell, most of the myelin lipid is provided from Schwann cell. Therefore, Schwann cells viability could play a crucial role in the development of peripheral nerve growth. Since diabetic peripheral neuropathy is characterized by nerve demyelination and degeneration of nerve bers in dorsal rot ganglion and peripheral nerves (Hong et al., 2008) , the viability of Schwann cells is closely associated with diabetic peripheral neuropathy.
In our study, the cell viability of Schwann cells was signi cantly weakened with high glucose treatment or with lipin1 silencing. In contrast, over-expression of lipin1 ameliorated cell viability with HG stimulation. It suggested that the loss of lipin1 suppressed cell viability of Schwann cells, while lipin1 treatment could alleviate HG-aroused cell apoptosis. Our gain-of-function and loss-of-function studies uniformly revealed that lipin1 alleviated HG-aroused decline in cell viability of Schwann cells. However, the underling molecular mechanisms remain elusive.
Mitochondria are semi-autonomous organelles that maintain the bilateral structure of the membranes. Mitochondria, served as bioenergetic powerhouses, provide the cell with energy in the form of ATP generated by oxidative phosphorylation. Hence, mitochondria are the powerhouses of the cell (Spinelli and Haigis, 2018) and are critical for cells survival. Previous studies demonstrated that mitochondrial disorder have close relationship with DPN, presented as hyperglycemia decreased mitochondrial respiration and ATP production in cultured dorsal root ganglion neurons (DRGs) (Rumora et al., 2019) . Peripheral nerves have long axons that are wrapped with myelin produced by Schwann cells. Because of its morphological complexity and rapid changes in metabolic requirements, neurons are critically dependent on energy metabolism, predominantly supplied with mitochondrial oxidative phosphorylation (Pellerin and Magistretti, 2003) . Besides, due to the limited capacity for self-renewal, the nervous system has the lowest capacity to maintain healthy after impairment of mitochondrial bioenergetics (Sajic, 2014) . Therefore, peripheral nerves are particularly dependent on effective mitochondrial function and distribution.
In our study, we tested the mitochondrial membrane potential and ATP content in Schwann cells to assess mitochondrial metabolism. In our experiments, mitochondrial dysfunction occurs in Schwann cells insulted to high glucose and lipin1 silencing under normal glucose, presented as decreased levels of MMP and ATP content. By contrast, over-expression of lipin1 alleviate high glucose aroused mitochondrial dysfunction. In addition, in order to see if metabolic disorder induced by hyperglycemia are also re ected in the morphological changes, mitochondrial structures were also observed using electron microscopic. Mitochondrial after high glucose stimulation and lipin1 silencing exhibited morphological disorder, manifested as displayed irregularly and dense cristae, whereas lipin1 up-regulation reversed high glucose-aroused mitochondrial swelling and chromatin condensation. Our observations are consistent with previous reports that EDL muscle of lipin-1 de ciency d mice presented impaired mitochondrial clearance and loss of membrane potential mitochondrial (Alshudukhi et al., 2018) . Glycolysis muscle bers in patients with heritable Lipin1-null mutations displayed mitochondrial aggregates and declined mitochondrial cytochrome c oxidase activity (Alshudukhi et al., 2018) . Overall, these observations demonstrate that lipin1 reversed high glucose-induced mitochondrial dysfunction and morphological disorder. Lipin1 overexpression might improve cell viability through regulating mitochondrial functions in Schwann cells.
Mitochondria are highly dynamic organelle that constantly undergo fusion and ssion, which are pivotal for the distribution of mtDNA and the maintenance of mitochondrial function (Friedman and Nunnari, 2014) .
Mitochondrial fusion is a process in which two separate mitochondria are fused into one (Sajic, 2014) . Mitochondrial fusion mediates material mixing between damaged mitochondria and healthy mitochondria to ensure the integrity of the entire mitochondrial network. Mitochondrial ssion is the division of one mitochondrial divide into two, which facilitates segregation of damaged mitochondria and enhance the number of mitochondria (Wu et al., 2016) . It has previously been observed that the alterations of mitochondrial membrane lipid composition, such as CL and DAG, could change the balance of mitochondrial dynamics (Alshudukhi et al.,2017) . Furthermore, mitochondrial dynamin-related family of large GTPases are all located at mitochondrial membranes, which are composed of proteins and lipids. Therefore, we speculate that lipin1, the key regulator of membrane lipid composition, maybe associated with mitochondrial dynamics by regulating phospholipid synthesis of the mitochondrial membrane (He et al., 2017) . The fusion of mitochondria is regulated by the proteins belonging to the dynamin-related family of large GTPases. MFN1 located in the outer membrane (OMM) and OPA1 located in the inner membrane (IMM). MFN1 serves as a tether between fusing mitochondria and contributed to arti cial membrane clustering (Giacomello et al., 2020) , OPA1 function as coordinate OMM and IMM, and remodel mitochondrial cristae (Giacomello et al., 2020) . The ssion of mitochondria is primarily regulated by DRP1, which translocate from the cytosol onto mitochondria outer membrane, more precisely onto ER-mitochondria contact sites (Lee and Yoon, 2016) . This can be mediated by the OMM anchoring adaptors, including FIS1, MFF, Mid49 (Sabouny and Shutt, 2020) . Although FIS1 is not essential for mitochondrial division, it could inhibit the mitochondrial fusion machinery and its over-expression leads to mitochondrial fragmentation (Giacomello et al., 2020) . Studies have demonstrated that lipin1 catalytic domain co-localized with DRP1 at constricted regions of the tubules. Through the same localized domain, lipin1 could participate in the ssion process and promotes ssion in a PA-independent manner (Huang et al., 2011) .
In our experiments, the levels of mitochondrial ssion related proteins, including DRP1 and FIS1 were increased, whereas fusion related protein, such as MFN1 and OPA1 were decreased under high glucose conditions. The role of lipin1 silencing is similar to that under high glucose conditions. In contrast, lipin1 treatment inhibited DRP1 and FIS1 protein expression and promoted MFN1 and OPA1 protein expression. These ndings suggested that lipin1 may have an antagonist effect to high glucose inducement on mitochondrial dynamic equilibrium of Schwann cells. Lipin1 treatment may protect against cells from mitochondrial dynamics imbalance induced by high glucose, and then keep mitochondria healthy and functional.
Our results are all based on in vitro experiments, which will be further veri ed by improving in vivo experiments in future studies. In addition, we only detected the in uence of lipin1 on the expression of mitochondrial dynamic -related proteins, and lacked the analysis of changes in the lipid composition of mitochondrial membrane, which will be improved in the subsequent experiments.
Taken together, our study demonstrates for the rst time that lipin1 could ameliorate cell viability of Schwann cells by affecting mitochondrial dynamics and functions,providing a novel approach for the prevention and treatment of DPN.

Declarations
Ethics approval and consent to participate RSC96 rat Schwann cells were obtained from Committee on Type Culture Collection of Chinese Academy of Sciences (Shanghai, China).

Consent for publication
Informed consent was obtained from all authors.
Availability for data and materials All data generated or analyzed during this study are included in this published article   HG+LV-Lipin1-Con group. The results are representative of three independent experiments.