Decreased in n‐3 DHA enriched triacylglycerol in small extracellular vesicles of diabetic patients with cardiac dysfunction

Abstract Purpose Diabetic cardiomyopathy is the leading cause of death in diabetic patients, and the mechanism by which factors other than hyperglycemia contribute to the development of diabetic cardiomyopathy is unknown. Serum small extracellular vesicles (sEVs) carry bioactive proteins or nuclei, which enter into remote tissues and modulate cell functions. However, in diabetic conditions, the changes of lipids carried by sEVs has not been identified. Our study aims to explore the changes of lipids in sEVs in diabetic patients with cardiovascular disease, we hope to provide new ideas for understanding the role of lipid metabolism in the pathogenesis of diabetic cardiomyopathy. Methods SEVs samples derived from serum of health controls (Ctrl), diabetic patients without cardiovascular diseases (DM), and diabetic patients with cardiovascular diseases (DM‐CAD) were used for lipidomics analysis. Because AC16 cells are also treated with those sEVs to confirm the entrance of cells and effects on insulin sensitivity, a lipidomics analysis on cells was also performed. Results and Conclusions In this study, we found that docosahexaenoic acid (DHA)‐triacylglycerides of sEVs from serums of DM‐CAD patients decreased significantly, and those sEVs could enter into AC16 cells and diminish insulin sensitivity. In addition, DHA‐triacylglycerides were also decreased in cells treated with sEVs from DM‐CAD. Therefore, DHA‐triacylglycerides carried by sEVs may mediate intercellular signaling and be associated with the incidence of diabetic cardiovascular complications.


DM-CAD sEVs diminished activation of AKT in cultured cardiac cells, and
showed the same decrease in n-3 DHA-TG, suggesting n-3 DHA-TG may be associated with diabetic cardiac dysfunction.

| INTRODUCTION
The number of patients with type 2 diabetes (T2D) is estimated to reach 780 million by 2045. 1 Cardiovascular complication is the main reason for death in T2D patients.3][4] Thus, it is important to explore the mechanisms that mediate the development of cardiovascular disease in diabetic patients.Dysregulation of glucose metabolism and lipid metabolism is the main pathological manifestation of T2D.Excess glucose and lipid are proven to diminish normal endothelial function, destruct pancreatic cells, and induce insulin resistance. 5,6In clinical practice, hypoglycemic agents and lipid-lowering agents are first-line drugs for the treatment of diabetes. 7owever, even in patients with good blood sugar control, the incidence rate of cardiovascular complications is still high, indicating more sophisticated regulation metabolism is involved in the progress of diabetic cardiomyopathy. 8mall extracellular vesicles (sEVs) are doublemembrane vesicles that are released into extracellular spaces by almost all types of cells.They cannot replicate and do not have functional nuclei.sEVs are defined by size as 50-200 nm, and they are mostly produced by the fusion of multivesicular body (MVB) with the plasma membrane, which is also known as exosomes.However, some EVs are produced by direct budding from the plasma membrane, which also has size of 50-200 nm, and they cannot be distinguished by current isolation methods, 9 so the term of sEVs represents vesicles (50-200 nm) that are generated by both pathways.In recent years, many studies have shown that sEVs carry bioactive cargoes, including lipids, proteins, and nucleic acids, which mediate cell-cell communication and regulate biological functions in recipient cells. 10,11In diabetic condition, serum sEVs were shown to carry sonic hedgehog, a protein that modulates immunity, or arginase1 to induce endothelial dysfunction. 12In addition, numbers of noncoding RNA are shown to carry by sEVs and are associated with incidence of diabetic complications. 13he changes in lipid profiles of sEVs are proven to be associated with the progress of diseases.For example, levels of phosphatidylcholine (PC), phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and lyso-PC of plasma sEVs from psoriatic patients were found to be increased compared to healthy control. 14Unsaturated fatty acid-enriched lyso-PC concentration of sEVs derived from intestinal epithelial cells increased after hypoxia and reoxygenation injury. 15Additionally, sEVs from the nonmetastatic cell line displayed a marked increase in PC (34:1), PE (36:2), sphingomyelin (SM; d18:1/16:0), hexosylceramide (HexCer; d18:1/24:0), and HexCer (d18:1/24:1) compared with controls, whereas these same lipids species were decreased in the metastatic cell line. 16herefore, the determination of lipid alterations of plasma sEVs is useful for the diagnosis.However, lipid profiles of sEVs in patients with diabetes or diabetic cardiovascular disease are rarely reported.In our experiment, we found that some species of triglycerides in serum sEVs of diabetes patients changed significantly with the lipidomics method, and we confirmed the changes in cell experiments.These results could provide new ideas for the early diagnosis of diabetes cardiovascular complications.

| Serum sample preparation for lipidomic analysis
We collected 20 blood samples from each of three groups: health controls (Ctrl), diabetic patients without cardiovascular diseases (DM), and diabetic patients with cardiovascular diseases (DM-CAD).Then we selected six cases from each group of these samples for subsequent lipidomics analysis.The selection criteria included (1) all groups were of the same age (56-62 years); (2) all groups had no hepatitis, cancer, or genetic diseases; (3) DM group patients had no complications of kidney disease or eye disease; and (4) the cardiovascular disorders in DM-CAD group patients are mainly characterized by decreased diastolic function.The body mass index (BMI) values and antidiabetic prescriptions that were used for those patients are summarized in Table 1.

| Isolation and analysis of sEVs
Plasma-derived sEVs were isolated using the Exoeasy Maxi kit (Qiagen, Cat#76064) according to the manufacturer's protocol.The protein concentration of sEVs was measured using a Qubit Protein Assay kit (Molecular Probes from Thermo Fisher Scientific).The particle number of sEVs was measured by NanoSight LM10 with NTA2.3 Analytical software (NanoSight, Wiltshire, UK).
For electron microscopy, the isolated sEVs were dropped onto the carbon support membrane copper mesh and let stand for 3-5 min, then 2% phosphotungstic acid was dropped onto the carbon support membrane copper mesh for 2-3 min, then observed under a transmission electron microscope and images collected for analysis by transmission electron microscope (HITACHI, HT7700).

| Lipid extraction
To samples of EVs 100 μL each, 200 μL ice 75% methanol was added.The samples were placed in an ice bath ultrasound for 15 min, 1 mL of ice MTBE was added for sufficient vortex oscillation, then they were rotated and mixed in a 4 C refrigerator for 1 h.After continuing ice bath ultrasound for 15 min, 200 μL water were added into the tube, standing at room temperature for 10 min, centrifuged at 4 C 14 000 g for 15 min.The layered upper liquid was removed, then the samples were dried with nitrogen gas.
For mass spectrometry conditions: electron eneneba spray ionization positive and negative modes were used for detection respectively.The sample was tested in both positive and negative modes.After separation by UHPLC, the sample was analyzed using a Q Active plus mass  For insulin stimulation assays, AC16 was supplemented with 10 μg protein of unstained control-sEVs or DM-CAD-sEVs and incubated for 24 h.After washing with PBS, cells were stimulated by 100 nM insulin (M9194, AbMole, USA) for 15 mins, then lysed by Mammalian Protein Extraction Reagent (Pierce from Thermo Fisher Scientific).The supernatants after centrifugation at 14 000 g, 5 min of cell lysates were used for protein assays.

| Statistical analysis
Software MSDAIL (Version 4.0.9) was used for lipid identification and quantification processing.One-way analysis of variance was used for multiple comparisons.A value of p < .05 was considered indicative of significance.

| Characterization of sEVs isolated from the serum of T2D patients with cardiovascular disease
We performed a lipidomic analysis of sEVs of Ctrl, DM, and DCM.SEVs were collected from the serum of patients separately by sEVs extraction kit as described in the methods.As shown by the electronic microscope, isolated sEVs of all groups have double membrane structure (Figure 1A), and western blot indicates the sEVs are CD9, TSG101 positive (Figure 1B), indicating some of the sEVs are of MVB origin.In addition, they are distributed in particle diameter mostly around 100 nm (Figure 1C).

| UHPLC-MS/MS-based lipidomic analysis in sEVs derived from the serum of patients with diabetic cardiomyopathy
Lipidomic analysis of sEVs of three groups has been performed.A total of 69 lipid classes are identified in sEVs in all three groups, including 2069 lipid species.Of all the lipid classes, we analyzed high abundance glycerides, including(PC), PE, phosphatidylserine (PS), PG, SM, and triacylglycerol (TG), and diacylglycerol (DG) (Figure 1D).
The results showed no significance in all these highabundance lipid classes in all three groups (Figure 2A-F).
As reports showed that n-3 unsaturated fatty acid (PUFA) bond to glycerides is important for biological functions, 17,18 we further analyze whether there are changes in lipid species with n-3 PUFA.Surprisingly, the specific lipid classes, docosahexaenoic acid (also known as DHA, 22:6) enriched lipids, have shown different changes, as shown in Figure 3, DHA-enriched TG and PC are significantly decreased in the DM-CAD group compared to the DM group (Figure 3A), but this is not observed in DHA-enriched PE and DG.Moreover,

| sEVs diminished the insulin signaling pathway of AC16 cells
In the following study, we examined if the plasma sEVs can enter into cardiomyocytes, first, we labeled the sEVs with PKH67, which is a membrane fluorescence agent, and added the labeled sEVs into culture media of human origin cell line AC16, the results showed that sEVs from plasma can enter into cardiac cells (Figure 5A), and sEVs from DM-CAD patients can diminish insulin sensitivity in normal AC16 cells, as indicated in Figure 5B, both ser473 and thr308 phosphorylation of AKT response to insulin stimuli are decreased by sEVs-DM-CAD treatment, but not by sEVs-control treatment.In addition, 2-NBDG glucose uptake assay showed glucose uptake stimulated by insulin is decreased by sEVs-DM-CAD treatment but not sEVs-control treatment (Figure 5C).

| DISCUSSION
In recent years, the ability of sEVs to mediate cell-cell communication and crosstalk between organs has attracted great interest in exploring the biological function of sEVs.In diabetic conditions, sEVs derived from different tissues affected cellular functions in different ways.Experimental study shows that sEVs from bone marrow stem cells increase the gene expression of insulin signal pathway (insulin, Pdx1, Smad2, Smad3, and transforming growth factor β) in the type 1 diabetes condition and alleviate insulin resistance in the T2D rat model. 19,20n addition, sEVs derived from adipose-derived stem cells have the potential to be used in the clinic to promote wound healing in patients with diabetes. 21Circulating sEVs are mostly derived from hemopoietic cells. 22argoes carried by circulating sEVs are found to be modulators or biomarkers for diabetes and its complications, for example, miR-122-5p shuttling by EVs regulates the viability and apoptosis of cardiomyocytes in diabetic rodent models. 23Arginase1 carried by serum sEVs induces endothelial dysfunction in diabetes. 12In addition, some insulin-signaling proteins, such as p70S6K, p-GSK3β, p-Akt, and p-insulin receptor, are contained in serum sEVs, indicating its association with incidences of insulin resistance. 24ipid profiles are an important mechanism for maintaining cellular insulin activity.Lipids of lower carbon number and double bond content were associated with an increased risk of diabetes, whereas lipids of higher carbon number and double bond content were associated with decreased risk.This pattern was strongest for TG and persisted after multivariable adjustment for age, sex, BMI, fasting glucose, fasting insulin, total triglycerides, and high-density lipoprotein (HDL) cholesterol. 25EPA (20:5) and DHA (22:6) are well-known n-3 PUFAs that are normally ingested by consumption of fish oils.EPA and DHA were shown to have health benefits in cardiovascular diseases by reprogramming triglyceride-rich lipoproteins metabolism, reducing inflammatory mediators (cytokines and leukotrienes), and modulating cell adhesion molecules. 26Moreover, evidence suggests that DHA is more efficient in decreasing blood pressure, heart rate, and platelet aggregation compared to EPA. 27 Recent studies suggest that n-3 PUFA in fish oil is primarily esterified as TG, 28 which was shown to have higher bioavailability and lower oxidation rate than free style.n-3 PUFA-TG in the diet significantly decreased the serum total cholesterol and non-HDL cholesterol. 29In diabetic condition, dysfunction of lipid TGs are associated with the incidence of diabetes.TGs can be carried by sEVs and promote or protect against diabetes and its complications directly or indirectly.In particular, TG (12:0_18:2_22:6), TG (16:0_11:1), TG (49:0), (TG 51:1), and DG (18:2_22:6) were independently associated with increased T2D risk. 30n our study, we found that total TG and PC content has no significant difference between the control, DM, and DM-CAD groups.DHA-enriched TG and PC are significantly decreased in the DM-CAD group compared to the DM group.More important, after being treated with DM-CAD-sEVs in AC16 cells, DHA-enriched TG, including TG (18:1, 18:2, 22:6), TG (18:1, 18:1, 22:6), TG (18:1, 16:1, 22:6), TG (18:1, 22:5, 22:6), and TG (17:0, 18:1, 22:6), are also decreased compared to the control group, suggesting the component of TGs in sEVs constitute an important part of DHA-TG in AC16 cells.

| LIMITATIONS AND CONCLUSIONS
We performed a widely targeted quantitative lipidomic analysis of serum sEVs of healthy controls and diabetic patients with or without cardiovascular disease.In addition, we performed a lipidomic analysis of AC16 cells that were treated with sEVs derived from DM-CAD patients and revealed differential lipid compositions between groups.Comparing sEVs from DM and DM-CAD, sEVs from DM-CAD lacked with DHA-TG species.We further demonstrated the sEVs from DM-CAD could diminish insulin-stimulated phosphorylation of AKT in AC16 cells.However, due to the limitation of relative quantitative nature in nontargeted lipodomics, and lipid classes detection by nontargeted lipodomics are not comprehensive, a targeted lipidomic analysis on TG species should be performed.Also, these potential biomarkers need to be validated in a larger cohort.Additionally, further in vitro experimental studies should be conducted to confirm the biological function of DHA-TG in cell signal transduction.Overall, our findings make explicit the concept that the DHA-TG component of sEVs from serums may mediate intercellular signaling and be associated with the incidence of diabetic cardiovascular complications.ORCID Wenguang Chang https://orcid.org/0000-0002-4151- Abbreviations: BMI, body mass index; DM-CAD, diabetes patients with cardiovascular disease.

F I G U R E 1
Isolation and characterization of small extracellular vesicles (sEVs) from serum of diabetic patients with or without cardiac dysfunction.(A) sEVs were isolated from serum of control, DM, and DM-CAD patients by commercial kits with 6 independent replicates.Electron microscopy images of sEVs isolated from serum of control, DM, and DM-CAD.(B) CD9 and TSG101 of sEVs from different groups were detected by western blots.Scale bar: 100 nm.(C) Particle distributions of sEVs by nanosight particle tracking system.(D) Lipid classes and relative abundance detected in lipidomic analysis.DM, diabetic patients without cardiovascular diseases; DM-CAD, diabetic patients with cardiovascular diseases.oxidized PC (PC-O) and PG (PG-O) also showed a decrease in DHA-enriched species (Figure 3E,F), but not for oxidized PS (PS-O) (Figure 3G).As shown in the figures, TG, PC, and PC-O has relatively high abundance in sEVs, we further analyze whether other essential fatty acid-enriched TG, PC, and PC-O species are