Renal microvascular damaged in diabetes mellitus with hyperlipidemia through the regulation of MAMs by PACS2 in endothelial cell

: Background: The prevalence of diabetic vascular complications is rapidly increasing, especially in the diabetes mellitus with hyperlipidemia. Consistent hyperglycemia and hyperlipidemia impairs microvascular, but lack of effective intervention target to prevention or reduced the risk of serious bad ending. Methods: A mouse model of diabetes combined with hyperlipidemia were established by STZ injection and high fat diet to observe the possible damage of HGHF to renal blood vessels include vascular permeability, fibrosis and subcellular structure. Then, we replicated an in vitro endothelial cell injury model treated by 30mm Glucose and 0.1mm palmitic acid to verify its main functional changes. Proteomics and metabolomics were used to explore the molecular mechanisms behind diabetic microvascular damage. The mechanisms were further verified at siRNA interference and transgenic knockout mice. Results: We found that renal vascular permeability impaired and fibrosis increased significantly in the stz+HFD mice. In human umbilical vein endothelial cells (HUVECs) treated with high glucose/high fat (HGHF), the number of mitochondrial-associated membranes (MAMs) and the expression of phosphofurin acidic cluster sorting protein 2 (PACS2) increased. In particular, gene manipulation of PACS2 altered endothelial cell MAMs. Knocking down PACS2 restored the barrier function of HUVECs. In vivo, knocking out PACS2 ameliorated the kidney injury in diabetic mice induced by streptozotocin and fed with high-fat diet for up to 20 weeks. PACS2-/- mice leaked less vascular Evan’s blue and impro ved glomerular fibrosis in the kidney tissue of hyperglycemia and hyperlipidemia mouse model. We further observed the reduction of fatty acid β - oxidation (FAO), CPT1α expression, and NADPH production in endothelial cells induced by HGHF. These changes in fatty acid metabolism were rescued by silencing PACS2, but were blocked by the FAO inhibitor, etomoxir. Conclusion: PACS2 impacts the metabolic response of endothelial cells to through MAMs. Loss of PACS2

Diabetes mellitus is a chronic metabolic disorder that can cause severe microvascular and macrovascular complications. Current diabetes treatment uses a variety of strategies to control blood glucose levels to reduce the occurrence of complications [1]. However, according to the results of large population studies, simple glucose control can effectively delay the onset of diabetic nephropathy, but even if glucose control is strengthened, the final incidence of vascular complications in diabetic patients has not been reduced [2,3]. And prevalence of dyslipidemia associated with complications in diabetes are common and sever. The lipid-lowering treatment has proved can reduces cardiovascular events and mortality in sever cohort study [4,5]. The combined effects of high glucose (HG) and free fatty acid levels [6] may be the main cause of vascular complications of diabetes. As a relatively common and serious high-glucose and high-fat (HGHF) condition, diabetes mellitus combined with hyperlipidemia also has a higher incidence of rapidly progressing diabetic vascular complications [7,8]. Therefore, revealing the mechanism of HGHF on blood vessels may provide a new way to prevent and treat diabetic vascular damage.
In our research, the renal function and microvascular of kidney are main vascular research object. Recent studies have shown that the mitochondrial function of glomerular endothelium is crucial in diabetic nephropathy [9]. In addition, endothelial cell proliferation, apoptosis, angiogenesis, nitric oxide secretion, and permeability are also important in the progression of diabetic vascular complications [10]. According to reports, endothelial cells can transform into different morphologies or subtypes with their state or the presence of varying stimulating factors [11]. It was found that the expression of genes related to glycolysis, tricarboxylic acid cycle, nucleotide purine and pyrimidine synthesis pathways in resting endothelial cells (QECs) was lower than that in proliferating endothelial cells (PECs), while the expression of genes related to the FAO pathway in the former cells is 3-4 times higher [12]. In addition, studies have reported that when QECs switch to an angiogenic phenotype, higher levels of glycolysis will occur to promote endothelial cell migration. As a high energy nutritional substrate, HGHF may mediate the damage and repair of endothelial function through the metabolic changes of endothelial cells. Therefore, we examined the mechanism of endothelial cell metabolism changes under HGHF.
In this study, we focused on diabetic microvascular dysfunction, especially the renal microvascular and glomerular function in hyperglycemia and hyperlipidemia condition. We first established a mouse model of diabetes combined with hyperlipidemia to observe the possible damage of HGHF to renal blood vessels. Then, we replicated an in vitro endothelial cell injury model of HGHF to verify its main functional changes. In addition, we used proteomics and metabolomics to explore the molecular mechanisms behind diabetic microvascular damage. Finally, the uncovered mechanisms were further verified at both the cellular and animal levels.

Animal model
All animal manipulations were approved by the Institutional Animal Care and Use Committee of Central South University, Changsha, China (2019sydw0021). All efforts were made to minimize mouse suffering. Eight-week-old male mice C57BLKS/Nju (wildtype, WT) and phosphofurin acidic cluster sorting protein 2 (PACS2) knockout mice (B6/JGpt-PACS2 em2Cd771/Gpt, PACS2-/-) were obtained from the Nanjing Biomedical Research Institute of Nanjing University and housed under standard pathogen-free conditions at the Xiangya Medical School of Central South University.
After one week of acclimatization, the mice were randomly assigned to the normal chow diet (ND) group and the streptozotocin (STZ) plus high-fat diet (HFD) (STZ + HFD) group. The latter was fed HFD (D12109C, Research Diets, USA) after grouping, and intraperitoneally injected with 10 mg/kg STZ (S0130, Sigma-Aldrich) once a week in the 9th and 10th week to simulate hyperglycemic and hyperlipidemic conditions [13]. STZ was dissolved in sodium citrate-hydrochloric acid buffer solution, pH 4.5 (SSC). The ND group was injected with the same volume of SSC at the 9th and 10th weeks. In the 11th week, blood glucose was measured, and mice with blood glucose > 11.1 mmol/L was considered to have diabetes. The flow chart of establishing HGHF-fed diabetic mouse model is showed in Fig. 1A.
The 24-h urine microalbumin (u-mALB) was measured once a week from the 8 th week, but the mouse urine was not collected next week after blood sampling to avoid the effect of transient blood loss on renal function. All mice were sacrificed at 20 weeks, and kidney tissues were fixed in 4% paraformaldehyde and stained with hematoxylin-eosin (HE) and Masson trichrome.

Vascular leakage assay
As an indicator of vascular permeability, albumin extravasation was evaluated in the Miles assay by measuring the extravasation of albumin-bound Evans blue. Briefly, mice were inhalational anesthetized with Isoflurane. Five min later, 150 μL of Evans Blue (30 mg/ml; Sigma-Aldrich, San Louis, MO, USA) was administered via the tail vein and allowed to circulate for 30 min, then extravasated Evans blue dye was evaluated [14]. Take pictures to show differences in Evans Blue extravasation of kidney. And dry one half of kidney at 150°C, on foil, for 48 h. Add 500 μl formamide to each tissue in 1.5ml tubes and incubate to 55 °C for 72h to extract Evans Blue from tissue. Measure OD620 of each sample and calculate plasma extravasation= OD620nm/g dry weight.

Fitc-dextran permeability assay
HUVECs were grown to confluence and treated with or without HGHF (30 mM glucose& 0.1mM palmitic acid) for 48 h on type I collagen-coated culture inserts with 0.4-µm pores (BD bioscience, USA). After that, 15 μL of 5 mg/mL FITC-dextran 40 (Sigma-Aldrich) was added to the upper chamber and incubated at 37°C for 15min, 30min and 60min. The fluorescence intensity of FITC-dextran diffused into the lower chamber was measured at Ex/Em = 485/590 nm.
Endothelial cell permeability was detected by using an In Vitro Vascular Permeability Assay kit following the manufacturer's protocol (Millipore, Burlington, MA, USA). After forming a tighter monolayer and exposure to HGHF, the cells were stained with FITC-labeled bovine serum albumin (BSA) for 20 min, and the amount of FITC-BSA [15] diffused across the endothelial monolayer was observed under a fluorescent microscope with FITC filter and quantified on a fluorescence plate reader.

Oxygen consumption assay
The kinetic changes of extracellular oxygen levels, an indicator of aerobic metabolism of living cells, are measured by using the MitoXpress Xtra oxygen consumption assay kit (Luxel Bioscience, USA) according to the manufacturer's manual [16]. Antimycin A, a potent electron transport chain complex III inhibitor, was used to shut down mitochondrial respiration (zero oxygen consumption control).
The uncoupling agent carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP) that disrupts the mitochondrial proton gradient was used to drive its maximal mitochondrial respiration rate. Basically, the phosphorescent signal of MitoXpress Xtra is quenched by oxygen and produces a signal that is inversely proportional to the amount of oxygen present. Time-resolved fluorescence (TR-F) was measured at

Immunohistochemistry
The kidney tissues were fixed in 4% glutaraldehyde at 4°C for 48 h, and then postfixed in 1% osmium tetroxide at room temperature for 2 h. After pre-staining with barbitone acetate for 10 min, histological samples were dehydrated with acetone, and embedded in paraffin wax. Immunohistochemical staining was performed to detect PACS2 and CD31 (ab28364, Abcam, Cambridge, UK) in mouse kidney endothelial cells as described previously, and visualized using fluorochrome-conjugated goat antimouse (ab150115, Abcam) or anti-rabbit IgG H&L (ab150077, Abcam). The numbers of PACS2-positive cells were counted and normalized to the total numbers of endothelial cells in the same field with the use of computer-assisted morphometric analysis.

Electron microscope
Cells were isolated on nickel fitters, stained with 2% uranyl acetate for 10 min, and then stained with Reynold's lead citrate for 5 min. The endoplasmic reticulum (ER)mitochondria contacts of endothelial cells were evaluated by a transmission electron microscopy (TEM; Hitachi-7650, Tokyo, Japan) at 60 kV. The ER mitochondrial contacts were quantified as described previously [18]. The images were analyzed using ImageJ (National Institutes of Health). The mitochondrial and ER membranes were delineated using the freehand tool. The selected areas were converted to masks and perimeters of ER were calculated. Two independent investigators quantified the images blindly. For the MAM quantification, we normalized the total ER connected to mitochondria to total ER perimeter.

Label-free quantitative proteomics
Total proteins were extracted from HUVECs in HGHF or control. Peptide samples (2 μg) were separated using an EASY-nLC1200 system and then analyzed using a Q Exactive mass spectrometer (120 min/sample), which was performed at Kangchen Biotech (Shanghai, China) [19]. Identification of the differentially expressed peptides was set at P < 0.05, fold-change > 2, following gene ontology (GO) enrichment analysis using Blast2Go 4.0.7 software.

Statistical analysis
The SPSS 22.0 software (SPSS, Chicago, IL, USA) was used for statistical analysis. Data are expressed as mean ± standard deviation (SD). Statistical differences were assessed by Student t-test of the means between two groups or by one-way analysis of variance of the means between multiple groups. P < 0.05 was considered a significant difference.

Renal vascular dysfunction in mice with diabetes and hyperlipidemia
In order to explore the effect of HGHF in vivo, we used low-dose STZ injection and a long-term high-fat diet to establish a mouse model of diabetes with hyperlipidemia (the STZ+HFD group) (Fig. 1A). Serum biochemical analysis and u-mALB quantitative results showed that after 20 weeks of continuous high fat feeding, the STZ+HFD group obtained a stable internal environment of hyperglycemia and hyperlipidemia compared with the ND group fed with normal diet (Fig. 1B). u-mALB increased significantly from the 16 th week, indicating that the renal function had been impaired at this time. The blood CRE and BUN increased significantly at the 20 th week, which also indicated that the renal function of the STZ+HFD group was abnormal.
Consistent with serum and urinary biochemical analysis, HE and Masson staining showed pathological changes in renal tissue, manifested as atrophy of glomerular vascular loops (Fig.1C) and renal microvascular fibrosis ( Fig.1D and F).
Furthermore, evaluation of the extravasation of Evan blue showed that the kidney tissue in the STZ+HFD group was significantly blue stained, and extravasation index was increased ( Fig. 1E and G). These results support that HFD induced renal vascular dysfunction in diabetic mice.

HGHF induces endothelial hyperpermeability in HUVECs
Due to the definite damage of renal blood vessels, we hypothesized that endothelial cell function changes significantly in diabetic nephropathy. To test this, we further constructed a HGHF-treated HUVEC cell model. We observed that fluorescently labeled dextran permeated a single layer of endothelial cells in a 0.4-μm pore size chamber and compared the upper and lower fluorescence values of the permeable chamber to reflect the endothelial barrier function. After measuring the leakage of FITC-labeled dextran through the endothelial cell monolayer, we found that the endothelial leakage in the HGHF group increased significantly at 15 and 30 min after adding FITC-dextran, and almost reached permeability equilibrium within 60 min ( Fig. 2A). Furthermore, the immunofluorescence of VE-cadherin, a key regulator of endothelial adhesion junctions, was obviously internalized under HGHF treatment, and its distribution on the cell membrane was reduced (Fig. 2B). All these data indicate that long time (48 h) exposure to HGHF does significantly produce barrier dysfunction.

HGHF increases mitochondrial-associated membranes (MAMs) in endothelial cells
In order to study potential target proteins that can regulate endothelial cell dysfunction induced by HGHF, we evaluated proteome changes of endothelial cells treated with HGHF. By mining the mass spectrometry reference database, protein identification was performed on data-independent acquisition (DIA) data results (identification criteria: precursor threshold, 1.0% false discovery rate (FDR) and protein threshold, 1.0% FDR). In the two sets of samples, we identified 26,201 peptides and 3,678 protein molecules. The identified proteins were screened for differentially expressed proteins between the ND group and the HGHF group, and the relative fold change of expression > 1.2 and the Student t test q-value < 0.05 were used as filter criteria to process DIA quantitative data. On this basis, 320 differentially expressed proteins were selected out (Supplemental Fig. 1A). Among them, 138 were up-regulated and 182 were down-regulated after HGHF treatment. Gene Ontology (GO) annotation analysis showed that they are mainly related to the protein localization and unfolded protein response of the ER as well as the translational elongation and termination of mitochondria (Fig. 3A). Among all differential proteins, the localized mitochondria and ER were occupied respectively (Supplemental Fig. 1).
Proteomics analysis showed that the mitochondrial and ER proteins of endothelial cells treated with HGHF changed significantly, the adhesion proteins changed significantly, and the active renal damage indicator lipocalin-2 (LCN2) was significantly increased (Supplemental Fig. 1).
The double staining of mitochondria and ER revealed higher co-localization in the presence of HGHF in HUVECs under a fluorescence microscopy ( Fig. 3B and D).
The same cell samples were sent to TEM, which is observed enrichment of ER and MAMs ( Fig.3C and E). Next, we verified the changes in mitochondria-ER contact in vivo. TEM observation of renal endothelial cells showed a significant increase in MAMs of renal cortical glomerular endothelial cells, accompanied by mitochondrial swelling and mitochondrial cristae degeneration in the renal vascular endothelial cells (Fig. 3F). These results indicate that mitochondria are involved in endothelial cell dysfunction induced by HGHF.

Based on the TEM and immunofluorescence observations of endothelial cells in
HGHF-treated HUVECs and HFD-treated kidney tissues in vivo, we further explored the expression changes of three key proteins that regulate MAMs, acyl-CoA synthetase 4 (FACL4), glucose-regulated protein 75 (GRP75), and PACS2 in response to FFA treatment. Among these proteins, PACS2 was found to be significantly increased in HUVECs (Fig. 4A and B) and mouse renal vascular tissues (Fig. 4C). To further understand the role of PACS2, we silenced PACS2 in HUVECs using siRNA (Fig. 4D and E). Using confocal imaging technique, we found that knocking down PACS2 attenuated HGHF-enhanced co-localization of mitochondria and ER (Fig. 4F). In addition, we observed a decrease in the MAM interface of mitochondria and ER and an increase in the ER signal, indicating that the secretion and detoxification functions of the cell is enhanced. Besides, after knocking down PACS2 in HUVECs, the leaked FITC-dextran ratio was significantly reduced (Fig.   4H). These results indicate that PACS2 plays a role in the mitochondria-ER connectivity.

Loss of PACS2 expression improves kidney injury induced by HFD
To confirm the effect of PACS2 on vascular endothelium under HGHF in vivo, we constructed PACS2 knockout mice (PACS2-/-) and used the same STZ injection and HFD treatment to produce the glucolipotoxic conditions depicted in Fig. 1A. The efficiency of gene knockout was verified by immunofluorescence microscopic observations, showing that PACS2 was lost in CD31-labeled vascular endothelial cells (Fig. 5B). As showed in Fig. 1B, we observed that the STZ+HFD treatment altered the biochemical parameters of renal function in WT mice, however, PACS2 knockout eliminated the increase in serum parameters caused by STZ injection and high fat diet, including blood glucose, TG, CRE, and BUN, but had no effect on serum CHO at week 20 (Fig. 5A). Compared to the WT mice with the STZ+HFD treatment, PACS2-deficient mice showed a moderate reduction in glomerular fibrosis in Masson staining ( Fig. 5B and C). In addition, we found that the glomerular cortex endothelial cells and vascular endothelial cell MAMs in the kidney tissue of PACS2-/-+STZ+HFD mice were significantly reduced (Fig. 5D), as was the glomerular Evans blue leakage index ( Fig. 5E and F), suggesting that PACS2 protects kidney tissue against HGHF by regulating MAMs

HUVECs
To further understand the downstream changes after the increase of MAMs, we performed metabolomics testing. KEGG pathway analysis results found that most of the differential metabolites were enriched in the metabolism pathways (Fig. 6A).
To test the changes in fatty acid metabolism of endothelial cells under long-term HGHF treatment, we measured HUVECs' mitochondrial respiration using PA as an energy substrate. By adding FCCP or ETO into cultured cells, we evaluated FAOdependent OCR (PA-OCR). The results showed that both FAO-dependent OCR were significantly reduced after HGHF exposure. However, silencing PACS2 reversed the PA-OCR inhibited by HGHF (Fig. 6B). Quantitative NADPH/NADP + (Fig. 6C) and ATP (Fig. 6D) analysis showed that the NADPH/NADP + ratio and ATP production in HUVECs were significantly reduced after HGHF treatment. However, PACS2 knockdown prevented the effect of HGHF on the NADPH/NADP + ratio, but did not affect ATP production.
Western blot analysis found that carnitine palmitoyl-transferase 1α (CPT1α), a key enzyme for FAO, decreased under HGHF stress, but was rescued by PACS2 knockdown (Fig. 6E-G). We also examined the CPT1α in kidney tissues of diabetic mice fed with HGHF ( Fig. 1) and found that it was obviously reduced (Fig. 6H) in Oil red-positive tissues (Fig. 6I).
These results suggest that PACS2 can regulate FAO, which may disturb the balance of endothelial redox homeostasis through the production of NADPH, thereby affecting the endothelial barrier function.

ETO block FAO to damage the barrier function of endothelial cells.
To further understand the role of PACS2 and FAO in HGHF treatment processing, we observed the synergistic effect of PACS2 silencing and FAO inhibition on the barrier function of endothelial cells. We found that regardless of PACS2 knockdown and HGHF treatment, ETO alone increased the internalization of VE-cadherin and expanded the distance between endothelial cells (Fig. 7A), and that had the most destructive effect on the barrier function of HGHF-treated cells. It was found that PACS2 siRNA inhibited the effect of HGHF on endothelial barrier function, but had no effect on ETO activity. This finding was consistent with the FITC-dextran leakage ratio, which was significantly higher in all ETO treatment groups (Fig. 7B). These results suggest that the changes in FAO may be downstream of MAMs, which will affect the response of PACS2 to energy conditions.

Discussion:
The purpose of this study was to determine the effect of hyperglycemia and hyperlipidemia condition on endothelial cell function and its mechanism in the occurrence and development of diabetic vascular complications. We observed mitochondria and ER at the cellular and animal levels through confocal and electron microscope imaging, and found that MAMs increased significantly under the pressure of HGHF ( Fig. 3B and C, Fig. 4F, and Fig. 5C), accompanied by an increase in PACS2 expression ( Fig. 4A and C and diabetic smooth muscle cell phenotypic transformation [25]. In this study, we first discovered the relationship between HGHF and MAMs in endothelial cells, and explored the physiological significance of MAMs. We found that MAMs may affect the initialization of VE-cadherin, which sustains intercellular adherent junctions of endothelial cells [26]. Moreover, PACS2 downregulation can improve the barrier function of endothelial cells under HGHF (Fig. 7A). These findings may indicate the MAMs more important in the diabetes with hyperlipidemia patients, which already have definite dyslipidemia.
As previous studies have showed, HGHF can trigger unfolded protein response and cellular inflammation, leading to decreased insulin biosynthesis, impaired insulin secretion, or cell death [27,28]. And lymphatic vessel integrity dysfunction in mice models of obesity and type 2 diabetes [29]. In our study, we found that mice with STZ+HDF treatment have high renal microvascular permeability (Fig. 1), HGHF induces endothelial barrier dysfunction in HUVECs (Fig. 2) and the VE-cadherin decreased in outer cell membrane were direct cause of high permeability of endothelial cells.
It was reported that HUVECs will increase FAO and decrease glycolysis when assembled into a formed network [30]. And when glucose and lipids both overload, the increased glucose will lead to the accumulation of malonyl-CoA, which affects the activation of CPT1α and limits the oxidation of fatty acids in the mitochondria [27,31]. Therefore, we examined the metabolomics of HUVECs under HGHF and found that the metabolic pathways changed the most in the KEGG analysis ( Fig. 6A and Supplemental Fig. 2). In particular, long chain fatty acids, such as glycerol 3phosphate, myristate, palmitate, and oleate, were increased in HGHF-treated cells, which may cause fatty acid coenzyme A to accumulate in the cytoplasm in the form of complex lipids or ceramides, resulting in cell damage. This was supported by Oil Red staining, which showed significantly positive kidney tissues in the STZ+HDF group (Fig. 6I), indicating a decrease in fatty acid utilization. Therefore, in the OCR assay, we further verified that the FAO activity of endothelial cells treated with PA ( Fig. 7A).
According to recent reports, CPT1α is present in MAMs [32] and may have an important regulatory role in MAMs. We examined the expression of CPT1α in HUVECs and mouse kidneys, both of which fully support that CPT1α mediates the reduction of FAO under HGHF. Recent study also showed FAO is essential for endothelial function [33] [34]. It has been reported that endothelial cells rely on FAO to increase the regeneration of NADPH to maintain redox homeostasis [12].
Consistent with this, we found that blocking FAO by ETO reduced the endothelial cell barrier function under HGHF, which may due to the redox imbalance (Fig. 7).
Interestingly, this damage caused by FAO inhibition could not be offset by PACS2 knockdown, which means that FAO and NADPH reduction are downstream of PACS2. Thus, endothelial cells on hyperglycemia and hyperlipidemia condition forms a chain of evidence from PACS2, MAMs to FAO and VE-cadherin (Fig. 8).
Our study shows that knocking down PACS2 led to a decrease in MAMs in HUVECs and vascular endothelial cells of mouse kidney tissues, which is in agreement with previous results in cancer cells [35], smooth muscle cells [36].

Conflict of interest
The authors confirm that there are no conflicts of interest.

Ethics approval
The animal experimental processes were approved by the Ethnic Committee of Third Xiangya hospital and conducted in strict accordance to the standard of the

Availability of data and material
The datasets generated/analyzed during the current study are available.

Author's Contributions
Zhihao Shu wrote the paper and conceived and designed the experiments.
Professor Alex F Chen Help modify and optimize experimental design. Professor Shuhua Chen and HongXiang collected and provided the cell samples for this study.
Ruoru Wu, Shaoli Zhao, Jie Xiao, Hengdao Liu, Xuewen Wang, Jie Ouyang, Yulan Liu and Wenfang Xiong contributed to drafting the article. All authors have read and approved the final submitted manuscript. Professor Hongwei Lu is the GUARANTOR for the article who accepts full responsibility for the work and/or the conduct of the study, had access to the data, and oversaw the decision to publish.       shown or data are represented as the mean ± SD, n = 3. *P < 0.05.  PACS2 is a key component of the mechanism behind diabetic microvascular dysfunction caused by HGHF. Under high glucose/high fat conditions, PACS2 is upregulated and increases the number of MAMs, followed by a decrease in CPT1α expression, free fatty (FA) β-oxidation, and NADPH production in renal vascular