Isocitrate dehydrogenase 1 upregulation in urinary extracellular vesicles from proximal tubules of type 2 diabetic rats

Diabetic nephropathy (DN) is a major cause of chronic kidney disease. Microalbuminuria is currently the most common non‐invasive biomarker for the early diagnosis of DN. However, renal structural damage may have advanced when albuminuria is detected. In this study, we sought biomarkers for early DN diagnosis through proteomic analysis of urinary extracellular vesicles (uEVs) from type 2 diabetic model rats and normal controls. Isocitrate dehydrogenase 1 (IDH1) was significantly increased in uEVs from diabetic model rats at the early stage despite minimal differences in albuminuria between the groups. Calorie restriction significantly suppressed the increase in IDH1 in uEVs and 24‐hour urinary albumin excretion, suggesting that the increase in IDH1 in uEVs was associated with the progression of DN. Additionally, we investigated the origin of IDH1‐containing uEVs based on their surface sugar chains. Lectin affinity enrichment and immunohistochemical staining showed that IDH1‐containing uEVs were derived from proximal tubules. These findings suggest that the increase in IDH1 in uEVs reflects pathophysiological alterations in the proximal tubules and that IDH1 in uEVs may serve as a potential biomarker of DN in the proximal tubules.


| INTRODUCTION
Type 2 diabetes (T2D) has become an important global epidemic disease affecting both developing and developed countries.T2D can lead to macrovascular complications like stroke and cardiovascular diseases and microvascular complications such as retinopathy, neuropathy, and nephropathy.Diabetic nephropathy (DN) is characterized by glomerular hypertrophy and sclerosis, basement membrane thickening, mesangial dilation, and interstitial fibrosis. 1,2Many kinds of enzymes play important roles in the kidney in DN.Dipeptidyl peptidase-4 (DPP-4), a member of serine proteases, involves various pathways related to DN progression, such as glucagon-like peptide-1 signaling pathways.DPP-4 inhibitors are widely used as drugs for diabetes patients that may have renoprotective effects in DN. 3 Isocitrate dehydrogenase 1 (IDH1) catalyzes the oxidative decarboxylation of isocitrate to αketoglutarate and produces NADPH, an essential electron donor to important cellular antioxidant processes.IDH1 can prevent hyperglycemia-mediated reactive oxygen species production, cellular oxidative stress, and extracellular matrix accumulation in proximal tubular cells. 4N is a major cause of morbidity and mortality in patients with diabetes. 5Therefore, developing novel biomarkers to predict or evaluate the severity of DN is essential.Because the profiles of urinary protein excretion reflect functional alterations in the kidney, such as glomerular filtration rate, urine is thought to represent a desirable matrix for the discovery of biomarkers of nephropathy.Urinary albumin and total urinary protein excretion are clinically used as biomarkers for early DN detection.However, recent studies have suggested that renal structural damage may have advanced when these biomarkers emerge, often leading to inappropriate diagnosis and management. 6Moreover, urine contains various proteins at trace levels, many of which may reflect the pathophysiological alterations caused by diabetic urogenital complications. 7,8Therefore, novel diagnostic biomarkers are required for therapeutic interventions in DN.
Extracellular vesicles (EVs) are lipid bilayer-enclosed nanoscale structures released from almost all cell types into various biological fluids, such as urine and blood.0][11][12] Recent studies have demonstrated that EVs are involved in various pathological and physiological processes, including ischemia-reperfusion, 13 immune responses, 14 cancer cell migration, 15 and neurodegeneration. 16Compared to free proteins and nucleic acids not enveloped in the lipid bilayer of body fluids, biomolecules in EVs are protected from degradative enzymes such as proteases and nucleases. 17Additionally, EVs contain various proteins, such as endosomal pathway-associated and tissue-unique proteins. 18Owing to the stability and diversity of proteins in EVs, EVs derived from biological fluids have emerged as promising sources of biomarkers for the diagnosis and prognosis of various diseases.
In this study, we used spontaneous type 2 diabetic Otsuka Long Evans Tokushima Fatty (OLETF) rats and normal control Long-Evans Tokushima Otsuka (LETO) rats to discover novel biomarkers for the early diagnosis of DN because OLETF rats are very similar to T2D and DN in humans. 19,20We performed a comparative proteomic analysis of urinary EVs (uEVs) from OLETF and LETO rats at an early stage to annotate the differentially expressed proteins and identify potential biomarker candidates.The origin of uEVs containing potential biomarkers was also investigated by lectin affinity enrichment of uEVs.

| Animal experiments
Four-week-old male OLETF and LETO rats were obtained from Hoshino Laboratory Animals Inc. (Ibaraki, Japan).
For the proteomic analysis, LETO (n = 4) and OLETF rats (n = 4) were fed ad libitum a standard CE-2 diet (CLEA Japan Inc., Tokyo, Japan) between 4 and 24 weeks of age.For the CR experiments, OLETF rats were divided into two groups and fed either ad libitum a standard CE-2 diet (OLETF group; n = 5) or 70% of the amount of the diet consumed by OLETF group (OLETF-CR group; n = 5), whereas LETO rats were fed ad libitum a standard CE-2 diet (LETO group; n = 5) between 4 and 25 weeks of age.Their food consumption was measured daily.The body weights of the rats were measured once a week.Twentyfour-hour urine was collected from the rats in metabolic cages once a week, and sodium azide was added to a final concentration of at least 10 mM to prevent bacterial overgrowth in the urine.The collected urine samples were centrifuged immediately at 1000× g for 10 min at room temperature (RT).The supernatants were stored at −80°C after adding a protease inhibitor cocktail (Nacalai Tesque).The rats were maintained under controlled temperature and humidity with a 12-hour light/dark cycle.Rats for the proteomic analysis and CR experiments were sacrificed at 24 and 25 weeks of age.Blood was collected under anesthesia, and renal tissues were harvested after perfusion with phosphate-buffered saline (PBS).The tissues were fixed in 4% paraformaldehyde in PBS, embedded in paraffin, sectioned for staining, or snap-frozen in liquid nitrogen and stored at −80°C.All animal experiments were approved by the Animal Experiment Committee of the Graduate School of Bioagricultural Sciences, Nagoya University.

| Isolation of EVs
Urinary EVs for proteomic analysis were isolated using a size-exclusion chromatography column (EVSecond L70; GL Science, Tokyo, Japan) according to the manufacturer's protocol.In brief, rat urine after centrifugation at 3000× g for 10 min followed by 17 000× g for 15 min was concentrated to about 500 μL with Amicon® Ultra 100 K (Millipore, MA, USA).The concentrated urine was loaded onto the column and separated into 22 fractions in 100 μL (1-14 fractions) or 200 μL (15-22 fractions) of PBS.The fractions were then subjected to immunoblotting to detect CD9 expression.The CD9-positive fractions were pooled and used as uEVs.For other experiments, uEVs were isolated by differential centrifugation.Rat urine samples were centrifuged at 3000 × g for 10 min at 4°C to remove cellular debris and then at 17 000× g for 15 min at 4°C to remove large EVs.The supernatants were ultracentrifuged at 200 000× g for 70 min at 4°C.The pellets were resuspended in PBS and ultracentrifuged.The resulting pellets were resuspended in 100 μL of PBS and used as uEVs.Protein content was measured using the bicinchoninic acid (BCA) assay (Thermo Fisher Scientific, Waltham, MA, USA).

| Size analysis of the EVs
The nanopore-based system qNano (IZON Science) was used to determine the particle size distribution of uEVs with nanopore NP150.The particle concentration was calibrated with 230 nm carboxylated polystyrene beads at a concentration of 9.2 × 10 11 particles/mL.

| Electron microscopy
Morphology of the EVs was observed with HT7820 transmission electron microscope (TEM) (Hitachi, Tokyo, Japan) at Nitto Analytical Techno-Center Company Co., Ltd. as previously described. 21

| Immunoblot analysis
The EV proteins were subjected to reducing and nonreducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (10% gel).Immunoblot analysis was performed as previously described. 21hemi-Lumi One Super (Nacalai Tesque, Kyoto, Japan) was added and the bands were visualized using the WSE-6100 LuminoGraph I chemiluminescence imaging system (ATTO, Tokyo, Japan).The intensities of bands were determined using CS Analyzer 4 (ATTO, Tokyo, Japan).

| Proteomic analysis by mass spectrometry
2][23] The gels were fixed, washed with water and 1:1 acetonitrile/50 mM ammonium bicarbonate, and then dehydrated in acetonitrile.Subsequently, 50 mM tris (2-carboxyethyl) phosphine (TCEP) was added, and the gels were incubated for 10 min at 60°C before they were alkylated with 100 mM iodoacetamide for 1 h at room temperature in the dark.The gels were washed twice, dehydrated with acetonitrile, and then rehydrated with Trypsin/Lys-C Mix (0.01 mg/mL; Promega, Madison, WI, USA).Subsequently, 50 mM ammonium bicarbonate (20 μL) was added, and the samples were incubated for 16 h at 37°C.The digestion was stopped by adding 20% trifluoroacetic acid (TFA).The resulting solution was filtered through Durapore PVDF membrane (0.1-μm pore size), vacuum dried, and resuspended in 0.1% TFA in 2% acetonitrile to extract the peptides.
Nanoscale liquid chromatography coupled with tandem mass spectrometry (nano LC-MS/MS) was performed using a Dionex U3000 gradient pump (Thermo Fisher Scientific) coupled with Q Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Fisher Scientific).The samples were concentrated on a C 18 trap column (5-μm particle size, 300 μm inner diameter, 5 mm length; Chemical Evaluation and Research Institute, Tokyo, Japan) and separated on a C 18 column (3-μm particle size, 100-μm inner diameter, 125 mm length; Nikkyo Technos, Tokyo, Japan) at a flow rate of 0.5 μL/min.The mobile phases were solvent A (0.5% acetic acid) and solvent B (0.5% acetic acid in 80% acetonitrile).The elution gradient for solvent B was as follows: 5% to 40% B over 100 min, then 40% to 100% B for 1 min, holding at 100% B for 3 min, then back to 5% B over 1 min, and finally re-equilibrating at 5% B for 10 min.Electrospray ionization was performed in positive-ion mode.The instrument was operated in the DDA mode.Xcalibur 4.1.50(Thermo Fisher Scientific) was used to record peptide spectra.The full scan was acquired from 350 to 1800 m/z with a resolution of 17 500, automatic gain control (AGC) as 3 × 10 6 , and maximum injection time as 60 ms.MS/MS scans were performed with a resolution of 35 000, AGC target as 1 × 10 5 , and maximum injection time as 60 ms.The 10 highest intensity precursor ions were isolated using the quadrupole analyzer in a window of 2.0 m/z and fragmented by higher energy collisional dissociation (HCD) fragmentation with normalized collision energy (NCE) of 27%.Multiply charged peptides were chosen for MS/MS experiments.Dynamic exclusion time was set to 20 s.

| Data analysis
Uniprot Rattus norvegicus (TaxID = 10116) protein sequence database and cRAP for contaminants (http:// www.thegpm.org/ crap/ ) were used.MS/MS spectra were interpreted, and peak lists were generated using Proteome Discoverer 2.2.0.382 (Thermo Fisher Scientific).Searches were performed using SEQUEST (Thermo Fisher Scientific).Search parameters were set as follows: enzyme selected with two maximum missing cleavage sites, a mass tolerance of 10 ppm for peptide tolerance, 0.02 Da for MS/ MS tolerance, fixed modification of carbamidomethyl (Cys), and variable modification of oxidation (Met).Peptide identifications were based on significant Xcorr values (high confidence filter).Peptide identification and modification information returned from SEQUEST were filtered at a false discovery rate (FDR) of 1% using the Percolator node of Proteome Discoverer to obtain confirmed peptide identification and modification lists of HCD MS/MS.

| Bioinformatic analysis
Gene ontology (GO) enrichment analysis of proteins identified by nano LC-MS/MS was performed using the DAVID v2021 database (Database for Annotation, Visualization, and Integrated Discovery) (http:// david.abcc.ncifc rf.gov/ ).

| Biochemical measurements
Urinary albumin concentration was measured using an LBIS rat albumin ELISA kit (Shibayagi, Gunma, Japan).Urinary total protein concentration was measured using a Micro TP-test kit (Wako, Osaka, Japan).Urinary creatinine concentration was assessed using the LabAssay™ Creatinine Kit (Wako, Osaka, Japan).Serum triglyceride level was measured using the LabAssay™ Triglyceride Kit (Wako, Osaka, Japan).

| Immunofluorescence
After heating paraffin sections to 60°C for 2 h, they were deparaffinized with xylene (three treatments, 10 min each), rehydrated sequentially with 100% (two treatments), 90%, 70% ethanol for 8 min each, and washed with PBS for 8 min.The tissue sections were incubated in PBS containing 0.1% sodium dodecyl sulfate (SDS) for 5 min and washed with PBS for 8 min.For antigen unmasking, the sections were boiled in 10 mM sodium citrate buffer (pH 6.0) for 15 min in a microwave oven, cooled at RT, and washed with PBS thrice for 3 min.The slides were blocked with 1% bovine serum albumin (BSA) in PBS for 30 min at RT, incubated with anti-IDH1 antibody (1:200), anti-SGLT2 antibody (1:100), anti-NHE3 antibody (1:100), anti-AQP1 antibody (1:200), and anti-AQP2 antibody (1:200)  in Can Get Signal solution B (TOYOBO, Osaka, Japan) overnight at 4°C, and washed with PBS thrice again.The sections were then exposed to donkey anti-rabbit IgG (H + L) highly cross-adsorbed secondary antibody-Alexa Fluor™ Plus 488 (1:500) and donkey anti-mouse IgG (H + L) highly cross-adsorbed secondary antibody-Alexa Fluor™ Plus 594 (1:500) in Can Get Signal solution B for 1 h at RT and washed with PBS thrice.Autofluorescence quenching was performed using the Vector TrueVIEW Autofluorescence Quenching Kit (Vector Laboratories, Burlingame, CA, USA), followed by mounting with VECTASHIELD Vibrance Antifade Mounting Medium with DAPI (Vector Laboratories), according to the manufacturer's instructions.
For staining of kidney sections with biotinylated LCA, sections were blocked with 0.1% w/v streptavidin solution in PBS and avidin solution using an avidin-biotin blocking kit (Vector Laboratories) after blocking with BSA.They were then incubated with 10 μg/mL biotinylated LCA in Can Get Signal solution B overnight at 4°C, followed by incubation with anti-IDH1 antibody (1:200), anti-NHE3 antibody (1:100), or anti-AQP2 antibody (1:200) for 4 h at RT. Donkey anti-mouse IgG (H + L) highly cross-adsorbed secondary antibody-Alexa Fluor™ Plus 488 (1:500) and Streptavidin Alexa Fluor™ 594 conjugate were used as secondary antibodies.

| Lectin affinity enrichment
Lectin selection-For this experiment, uEVs in 1000 μL of PBS were prepared from 180 mL of pooled urine of 10-week-old OLETF rats using ultracentrifugation as described above.To identify the lectin that enabled the enrichment of uEVs released from proximal tubular cells or collecting duct cells, eight biotinylated lectins (DBA, LCA, LTL, PNA, RCA I, SBA, UEA I, and WGA) were prepared.Each of the lectins or an equal volume of PBS (50 μL) was added to uEVs in 100 μL of PBS and incubated on a rotator at 4°C for 1 h.Dynabeads M-280 Streptavidin (50 μL, Invitrogen) were washed with PBS thrice, and a 50 μL Dynabeads M-280 Streptavidin solution in PBS was prepared.Subsequently, these beads in PBS were added to each sample and incubated on a rotator at 4°C for another 1 h.The unbound uEV fraction was removed and stored as the flow-through fraction.The beads were washed twice with PBS.Furthermore, uEVs precipitated by lectins were eluted with 80 μL of 0.2 M eluting sugar each, with mixing for 30 min at RT (L-Fuc for LTL and UEA I; D-Gal for PNA and RCA I; D-GalNAc for DBA and SBA; D-Man for LCA; D-GlcNAc for WGA).The eluates were collected in new tubes and stored as pull-down fractions.
LCA affinity enrichment-In total, 14 mL of urine samples collected from each of the four OLETF and four LETO rats at 17 weeks of age were used to isolate uEVs for LCA affinity enrichment.The procedure for the LCA affinity enrichment of uEVs was the same as Lectin selection.

| Characterization of uEVs isolated from type 2 diabetic model rats
Both OLETF and LETO rats gained body weight in an age-dependent manner.The body weight of OLETF rats was higher than that of age-matched LETO rats throughout their lifetime (Figure 1A).Although 24 h urinary albumin levels in OLETF rats were higher than that in LETO rats at 10 weeks of age or later, minimal differences were observed between LETO and OLETF rats at 5 weeks of age (Figure 1B).Therefore, we isolated uEVs from both OLETF and LETO rats at 5 weeks of age using a sizeexclusion chromatography column.The particle diameter of the isolated uEVs was measured using tunable resistive pulse sensing analysis.The mean particle diameters of uEVs from LETO (LETO-uEVs) and OLETF rats (OLETF-uEVs) were 140.5 ± 2.4 and 145.6 ± 3.6 nm, respectively (Figure 1C,D).The ultrastructures of LETO-EVs-and OLETF-EVs were examined by transmission electron microscopy (TEM).Most membrane-enclosed particles were 80-200 nm in diameter (Figure 1E).Immunoblot analysis detected the EV markers (CD9, Alix, and TSG101) in both LETO-uEVs and OLETF-uEVs, whereas Lamin A did not (Figure 1F).

| Identification of a biomarker candidate protein for early DN diagnosis
Comparative proteomic analysis of uEVs from 5-week-old LETO and OLETF rats (n = 4 each) was performed to identify candidate biomarkers for early DN diagnosis.In total, 959 and 961 proteins were identified in LETO-uEVs and OLETF-uEVs, respectively (Table S1).Further, 61 of the top 100 marker proteins from Vesiclepedia, a public data repository for EV cargo, overlapped with the proteins identified in both groups.Gene ontology enrichment analysis showed that proteins associated with the membrane were enriched in both uEV samples (Figure 1G).These results indicated that EVs were successfully isolated from the urine of both rat types.We applied the following stringent criteria to narrow down the possible candidate proteins: First, we selected proteins with a false discovery rate <0.01.Second, among the upregulated proteins in OLETF-uEVs, proteins whose abundance and PSMs were detected in all OLETF-uEV samples were selected.In contrast, proteins whose abundance and PSMs were detected in all LETO-uEV samples were selected among the downregulated proteins in OLETF-uEVs.A total of 408 proteins met these criteria, including 359 upregulated and 49 downregulated proteins in the OLETF-uEVs.Among these proteins, the immunoglobulin gamma-2b chain C region showing the largest fold change with p-values <.01 was often identified as contamination in EV preparations from urine. 24Isocitrate dehydrogenase 1 (IDH1) showed the second-largest fold change with p-values <.01 (Figure 1H).Therefore, we selected IDH1 as a candidate biomarker for further study.

OLETF-uEVs at 5 and 25 weeks of age
To validate the levels of IDH1 in LETO-uEVs and OLETF-uEVs, uEVs isolated by ultracentrifugation were subjected to immunoblot analysis.We found that IDH1 expression in OLETF-uEVs was significantly higher than that in LETO-uEVs at 5, 10, 20, and 25 weeks of age (Figure 2A-D).The levels of IDH1 in uEVs from OLETF rats were increased in an age-dependent manner (Figure 2E).
To investigate the presence of IDH1 in its free form in the urine, we fractionated centrifuged urine using sizeexclusion column chromatography, followed by immunoblot analysis.IDH1 was detected only in the fractions containing CD9, indicating that IDH1 did not exist in its free form in the urine (Figure 2F).

| Calorie restriction significantly suppressed the increase of IDH1 in OLETF-uEVs
6][27][28] Supplementation with 70% of the volume of food consumed by OLETF rats fed ad libitum prevented diabetes mellitus and its complications in OLETF rats. 25Hence, the effect of 70% CR on the amount of IDH1 in uEVs was examined.The body weights of the rats in the OLETF-CR group were suppressed to approximately 71% of those in the OLETF group, aligning with the weights observed in the LETO group (Figure 3A).CR also suppressed serum triglyceride, total urinary protein, and 24 h urinary albumin levels (Figure 3B-D).The level of IDH1 in the uEVs from all groups was determined using immunoblot analysis.The results showed that CR significantly attenuated the increase in IDH1 in OLETF rats, and the level of IDH1 in uEVs had a strong correlation with 24-h urinary albumin excretion (Figure 3E,F).

| IDH1 was mainly expressed in proximal tubular cells and collecting duct cells in the kidney
The distribution of IDH1 in rat kidneys was investigated to identify candidate host cells releasing IDH1-containing EVs.Immunofluorescence staining with an anti-IDH1 antibody showed that IDH1 was mainly expressed in the cortex and inner medulla of the kidneys (Figure 4A).The expression level of IDH1 in the cortex of OLETF rats was significantly higher than that of LETO rats.Similar results were observed in the immunoblot and real-time quantitative polymerase chain reaction (qPCR) analyses (Figure 4B-D).Furthermore, IDH1 colocalized with proximal tubule markers (SGLT2, NHE3, and AQP1) in the cortex and with the collecting duct marker AQP2 in the inner medulla of OLETF rats (Figure 4E).However, IDH1 was not detected in the glomeruli.IDH1 in the kidney was mainly located in proximal tubular cells and collecting duct cells.

| IDH1-containing uEVs were mainly released from proximal tubular cells in the kidney
0][31][32] The immunofluorescence study suggested that IDH1-containing uEVs originated from proximal tubular cells, collecting duct cells, or both, as IDH1-containing uEVs are released from cells expressing IDH1.Therefore, a pull-down assay of uEVs from the pooled urine of OLETF rats with eight biotinylated lectins was performed to identify the lectin that recognized the surface sugar chain of uEVs released from either the proximal tubular or collecting duct cells (Figure 5A).uEVs from proximal tubular cells expressing SGLT2, NHE3, and AQP1 were enriched in the pull-down fractions after Lens culinaris lectin (LCA).In contrast, uEVs from collecting duct cells expressing AQP2 were enriched in the flow-through fractions (Figure S1).Thus, we found that LCA was a suitable lectin to separate uEVs from proximal tubular and collecting duct cells.Furthermore, a pull-down assay of uEVs from four OLETF and four LETO rats by LCA was performed to confirm the differences between OLETF and LETO rats and between individuals.Proximal tubule markers were enriched in all LCA pull-down fractions, whereas collecting duct markers were enriched in all LCA flow-through fractions.IDH1 was enriched in LCA pull-down fractions (Figure 5B).No differences were observed in the patterns between species and individuals.Furthermore, staining of kidney sections showed that LCA colocalized with NHE3 and IDH1 in the cortex (Figures 5C, S2 and S3).These results indicated that IDH1-containing uEVs originated from the proximal tubules rather than from the collecting ducts in the kidney.

| DISCUSSION
This study showed that IDH1 expression significantly increased in uEVs from OLETF rats at 5 and 25 weeks of age.CR in OLETF rats significantly attenuated the increase in IDH1 in uEVs, suggesting that IDH1 in uEVs was at least partially regulated by the induction of DN and might serve as an early biomarker for DN.
IDH1 is an enzyme located in the cytoplasm that plays a protective role against oxidative stress and ultraviolet radiation stress. 33Interestingly, IDH1 is induced in renal proximal tubular OK cells by hyperglycemia, and both the expression and enzymatic activity of IDH1 are upregulated in the renal cortex of streptozotocin-induced diabetic rats and diabetic db/db mice. 4Additionally, Li et al. reported that IDH1 is significantly upregulated in the renal cortex of OLETF rats at 36 weeks of age. 19However, to our knowledge, IDH1 expression in uEVs has not been examined in detail.Moreover, these studies did not focus on the early stages.Here, we showed that the expression of IDH1 was upregulated in OLETF rats at both early and severe albuminuria stages.Although our results were observed only in OLETF rats, the increased IDH1 in uEVs may potentially be observed in other diabetic models based on previous studies.
Diabetes and its complications were induced by spontaneous overeating in OLETF rats.OLETF rats display characteristics similar to those of human T2D.Therefore, rats are commonly used in studies of T2D and its complications.The cumulative incidence of diabetes was 0% in 20-week-old OLETF rats supplied with 70% of the amount of food consumed by OLETF rats fed ad libitum. 25Our results showed that the LETO group also consumed 60%-75% of the amount of food consumed by the OLETF group throughout their lifetime.Collectively, 70% was presumed to be the optimal calorie restriction level.The finding that CR attenuated the increase in IDH1 in OLETF-uEVs indicated that the increase in IDH1 could be attributed to the factor related to the induction of DN in OLETF rats.However, further studies are required to verify the detailed mechanisms.
The amount of CD9, a common EV marker protein, was used in our study to normalize the quantity of IDH1, although several proteins are commonly used as EV markers.Salih et al. reported that CD9 concentration in human urine is strongly correlated with urinary creatinine, allowing CD9 to be used to normalize spot urine. 34Moreover, Blijdorp et al. reported that urinary creatinine concentration is highly correlated with the concentration of EVs in human urine. 35Based on these reports, CD9 can be used as a potential loading control for uEVs.duct cells, and intercalated collecting duct cells, respectively. 36However, a method to enrich rat uEVs with lectins depending on their original segments has not been established because the expression pattern of the surface sugar chain varies among species.Here, we found that the LCA lectin enabled the enrichment of uEVs from proximal tubular cells and not from collecting duct cells.Using biotinylated LCA lectin, the proximal tubular cells were identified as the original segment of IDH1containing uEVs in both OLETF and LETO rats.The pull-down of uEVs using antibodies could be another way to enrich uEVs, depending on their origin.Notably, antibodies must target the outer regions of membrane proteins, and their effectiveness may vary depending on the specific makers or lot.However, lectins are less expensive than antibodies, and their reactivity does not usually vary depending on their makers or lot, although the avidity of lectin-carbohydrate binding is relatively low.Overall, depending on their origin, lectins appear to be good tools for enriching uEVs.
Proximal tubules are part of the nephron and play a pivotal role in the kidney.Most glomerular-filtered substances, such as nutrients and ions in primary urine, are reabsorbed into the proximal tubular cells.The metabolic function of the proximal tubular cells is strongly associated with renal, cardiovascular, and metabolic diseases. 37eportedly, at the early stage of DN, proximal tubular damage is more closely linked to kidney dysfunction than glomerular defects, challenging the long-standing belief that DN is primarily characterized by glomerular issues. 38his report supports our finding that IDH1, a protein that increases in uEVs in the early stages of DN, originates from proximal tubular cells.
Although IDH1 was expressed in proximal tubular and collecting duct cells, IDH1-containing uEVs were exclusively derived from proximal tubular cells.We hypothesized that EVs from proximal tubular cells likely constituted the largest proportion of uEVs or that proximal tubular cells had a preference for sorting IDH1 into EVs.However, confirmation was challenging due to the  In summary, we identified IDH1 as a potential biomarker protein in uEVs.We found that IDH1 was derived from proximal tubular cells using the LCA lectin enrichment established in this study and staining of kidney sections.Our results may provide some clues for research on DN in the severe and early stages.Moreover, the method of enriching rat uEVs derived from proximal tubular cells using lectins will be valuable in biomarker research involving rat uEVs.

F I G U R E 1
Characterization and proteomic analyses of urinary extracellular vesicles (uEVs) from LETO and OLETF rats.(A, B) Body weight (A) and 24 h-urinary albumin excretion (B) of LETO and OLETF rats.Different letters above the columns indicate a significant difference (p < .05).(C, D) Particle diameter distribution of uEVs from LETO (C) and OLETF (D) rats assessed by tunable resistive pulse sensing analysis.(E) Representative transmission electron microscope images of uEVs from LETO (left) and OLETF (right).The scale bar is 200 nm.(F) Immunoblots of EV markers (CD9, Alix, and TSG101) and negative marker (Lamin A) in uEVs and kidney homogenate.(G) Gene ontology (GO) enrichment analysis of uEV proteins with top six terms for the cellular component.(H) Volcano plot of uEV proteins that met our criteria.Proteins are plotted according to their −log 10 p-values and log 2 fold change (OLETF/LETO).p-values were calculated by a two-tailed Student's t-test of log 2 -transformed normalized protein abundances.Horizontal dashed lines indicate p-values of .01,and vertical dashed lines indicate fold changes of 2. Red dots represent proteins that are significantly increased in OLETF-EVs.

F I G U R E 3
Calorie restriction (CR) significantly attenuates the progression of diabetic nephropathy and the upregulation of IDH1 in OLETF-EVs.(A-D) Effects of CR on body weight (A), serum triglyceride (B), 24 h-urinary albumin excretion (C), and urinary total protein to creatinine ratio (D) of OLETF rats.(E) Immunoblots of uEVs from LETO, OLETF, and OLETF-CR rats at 25 weeks of age.Bar graph showing IDH1/CD9 ratio determined by densitometry.(F) Correlation between IDH1/CD9 ratio in uEVs and 24 h-urinary albumin.Identifying the source of biomarker-containing EVs is crucial when using uEVs for diagnosis or prognosis because uEVs are a mixture of EVs derived from the urinary tract.Svenningsen et al. reported that LTL, DBA, and PNA lectins can isolate human uEVs containing markers of proximal tubular cells, principal collecting

F I G U R E 4
IDH1 is upregulated in proximal tubular cells of OLETF rats.(A) Representative images of the renal cortex, outer medulla, and inner medulla subjected to immunohistochemistry for IDH1 from LETO and OLETF rats.The scale bar is 100 μm.(B, C) Immunoblots of IDH1 in the cortex of rat kidney.Bar graphs showing IDH1 protein level determined by densitometry (C).(D) Real-time qPCR analysis of Idh1 mRNA level in the cortex of rat kidneys.(E) Representative images of kidney sections subjected to immunohistochemistry for IDH1 and then for proximal tubule markers (NHE3, SGLT2, and AQP1) or collecting duct marker (AQP2) from OLETF rats.The scale bar is 100 μm.Red, NHE3, SGLT2, AQP1, and AQP2.Green, IDH1.Blue, DAPI.

F I G U R E 5
Lectin affinity enrichment of uEVs from LETO and OLETF rats.(A) Schematic representation of lectin affinity enrichment.The uEVs isolated by ultracentrifugation were mixed with biotinylated lectin or PBS (as control) and then mixed with streptavidin magnetic beads.The uEVs attached to the beads were eluted by eluting sugar.(B) Immunoblots of IDH1, proximal tubule markers (NHE3, SGLT2, and AQP1), collecting duct marker (AQP2), and EV markers (Alix, TSG101, and CD9) in uEVs enriched by Lens culinaris lectin (LCA) from LETO and OLETF rats.(C) Staining of the renal cortex, outer medulla, and inner medulla with IDH1, LCA, and DAPI.The scale bar is 100 μm.Red, LCA.Green, IDH1.Blue, DAPI.
difficulty in measuring and comparing the exact number of EVs released from each nephron segment.