Urinary extracellular vesicle proteins for biomarker discovery in chronic kidney disease
Article information
Abstract
Chronic kidney disease (CKD) is a progressive condition characterized by declining kidney function principally driven by diabetes mellitus, glomerulonephritis, and hypertension. Although renal biopsy is the gold standard for diagnosing CKD, its invasive nature restricts its use for early detection and routine clinical applications. Current noninvasive biomarkers, including the estimated glomerular filtration rate and albumin-to-creatinine ratio, are useful indicators of kidney dysfunction but fall short in specificity and sensitivity required to distinguish between CKD subtypes, including diabetic kidney disease and glomerulonephritis. Recently, urinary extracellular vesicles (uEVs), nano-sized lipid bilayer entities ranging from 30 to 1,000 nm in diameter and secreted by renal cells, have emerged as promising biomarkers for CKD. uEVs encapsulate a diverse array of proteins, nucleic acids, and lipids that mirror kidney pathophysiology, presenting a noninvasive means to assess disease progression, inflammation, fibrosis, and oxidative stress within the urinary system. Furthermore, uEVs offer a molecular fingerprint of kidney health, positioning them as potential tools for precision medicine. This review explores the diagnostic potential of uEVs, underscoring the need for standardization in urine collection, normalization techniques, and uEV isolation methodologies. We also highlight uEV-based biomarkers that distinguish various CKD subtypes and mirror pathological changes within the kidneys and urogenital system. As molecular proxies of their cells of origin, uEVs hold significant promise in enhancing CKD diagnostics to enable early detection, disease classification, and the development of novel therapeutic strategies.
Introduction
Chronic kidney disease (CKD) is a progressive disorder characterized by the gradual loss of kidney function [1]. It is indicated by an estimated glomerular filtration rate of less than 60 mL/min per 1.73 m2, along with increased markers of renal damage, such as albuminuria and hematuria [1]. CKD is primarily caused by conditions such as diabetes mellitus, glomerulonephritis, and immunoglobulin A (IgA) nephropathy, or hypertension, all of which lead to ongoing damage to the renal structure and function [2]. Without effective management, CKD can advance to end-stage renal disease (ESRD), which is marked by fluid retention, anemia, disruptions in bone and mineral metabolism, acidosis, and cardiovascular disease [3]. The prevalence of CKD was estimated to exceed 800 million individuals globally in 2017 [4]. The global burden of the disease rose from 19th in 2013 to 12th in 2017 and is projected to climb to 5th by 2040 [4]. Clinical indicators such as the estimated glomerular filtration rate and the albumin-to-creatinine ratio from urine biopsies are commonly utilized to diagnose kidney dysfunction [2]. However, these indicators lack high specificity and sensitivity in distinguishing various types of CKD, such as diabetic kidney disease (DKD). While renal biopsies are highly accurate for diagnosing CKD, they are not practical for early diagnosis or for monitoring disease progression. Therefore, it is imperative to identify more sensitive biomarkers and develop effective diagnostic techniques to accurately diagnose and monitor CKD progression. Particularly, effective management of CKD is essential to reduce its progression and minimize kidney damage, thereby preventing the development of ESRD through the meticulous regulation of hypertension, careful monitoring of diabetes mellitus, and prevention of metabolic acidosis [5].
Extracellular vesicles (EVs), naturally occurring lipid bilayer vesicles typically ranging from 30 nm to 1 µm in size, have recently been acknowledged as a biomarker source that reflects the pathophysiological status of the originating cells [6]. EVs contain diverse cargoes, such as proteins, genetic materials, metabolites, and lipids, derived from their originating cells or tissues [7]. They are stably present in virtually all biological fluids, including blood, urine, cerebrospinal fluid, and lymph, distributed via systemic circulation [8]. Notably, plasma contains approximately 1010 particles per mL of EVs [9]. In contrast, urinary EVs (uEVs) are found at lower concentrations, with an observed concentration of about 109 particles per mL in urine [10]. Given their association with representative molecules from the parental cells, EVs have shown significant diagnostic potential in minimally invasive or noninvasive liquid biopsies, facilitating the monitoring of disease occurrence, progression, and response to treatment [11]. Specifically, EV-related biomarkers have been increasingly identified for a broad range of diseases, including cancer [8], cardiovascular diseases [12], central nervous system disorders [13], and urinary system disorders [14]. Importantly, numerous studies have demonstrated the presence of diverse uEV-related biomarkers that characterize the status and regulate pathways involved in oxidative stress, apoptosis, inflammation, fibrosis, and renal injury within the urinary system [11]. The identification of uEV biomarkers enables the detection of previously unrecognized molecules in urine, given that uEVs contain a unique pool of components at lower concentrations compared to the bulk major components in urine [15,16]. In this review, we address the general aspects of EV subtypes, isolation, and diagnostic biomarkers for CKD uEVs.
Subtypes of extracellular vesicles
EVs are generally classified based on their biogenesis, which includes exosomes, ectosomes, and apoptotic vesicles (Table 1, Fig. 1) [17]. Exosomes are derived from multivesicular bodies that merge with the plasma membrane, thereby releasing intraluminal vesicles into the extracellular space [18]. The development of multivesicular bodies from early endosomes is facilitated by the endosomal sorting complexes required for transport (ESCRT) machinery [19] or through membrane lipid alterations via the sphingomyelinase-ceramide pathway [20]. Ectosomes, also known as microvesicles, are larger EVs ranging from 100 nm to 1 µm that are budded directly from the plasma membrane [21]. Their formation is closely linked to the rearrangement of the actin cytoskeleton, which enables membrane budding and fission [22]. Moreover, ADP-ribosylation factor 6 has been implicated in the release of ectosomes in cancer cells [23]. Apoptotic vesicles, released as a result of cellular turnover and programmed cell death, are detected in biological fluids as a subclass of EVs [24–26].
Classification of EVs
Biogenesis and origin of uEVs in the pathophysiology of the urinary system.
This schematic illustrates the biogenesis, molecular composition, and potential diagnostic applications of uEVs in kidney diseases. uEVs originate from various renal cells, including podocytes, tubular epithelial cells, and endothelial cells, and harbor a diverse array of bioactive molecules such as proteins, messenger RNA (mRNA), microRNA (miRNA), and DNA, which reflect kidney function and pathophysiological changes. The kidneys and urinary tract act as primary sites for uEV secretion, with these uEVs subsequently excreted into the urine, carrying molecular signatures indicative of renal health and disease. Representative uEV-associated biomarkers have been identified in relation to kidney tissues and diseases [15,44,49,61–64].
AGP1, alpha-1-acid glycoprotein 1; ALIX, programmed cell death 6-interacting protein; AQP1, aquaporin-1; AQP2, aquaporin-2; ARF6, ADP-ribosylation factor 6; BSG, basigin; CD133, prominin-1; CD24, cluster of differentiation 24; DPP4, dipeptidyl peptidase 4; EGFR, epidermal growth factor receptor; Elf3, ETS-related transcription factor Elf-3; INF2, inverted formin 2; MASP2, mannan-binding lectin serine protease 2; MMP2, matrix metallopeptidase 2; TRPC6, short transient receptor potential channel 6; TSG101, tumor susceptibility gene 101; uEVs, urinary extracellular vesicles.
In terms of canonical EV markers, tetraspanins such as CD9, CD63, and CD81 are widely recognized as key identifiers of EVs [26]. Exosomes, specifically, are enriched with tumor susceptibility gene 101 (TSG101), programmed cell death 6-interacting protein (ALIX), and syntenin-1, which are associated with endosomes and the ESCRT pathway, whereas ectosomes chiefly contain integral membrane proteins such as integrins [27]. Moreover, annexin A1 and basigin (BSG) have been identified as specific markers for ectosomes [27–29]. However, these molecular markers are not exclusive to any single EV subtype due to partial overlap [27]. Additionally, the substantial heterogeneity of EVs in terms of size, molecular composition, and charge complicates the specific isolation of individual EV subtypes following their release [30]. Given these challenges, the International Society for Extracellular Vesicles (ISEV), in the Minimal Information for Studies of Extracellular Vesicles (MISEV) guideline, recommends using the generic term “extracellular vesicle” to encompass all secreted vesicle subtypes and discourages biogenesis-based classifications unless the EV population has been rigorously validated [31].
Diagnostic potential of extracellular vesicles as disease biomarkers
The lipid bilayer nature of EVs allows them to incorporate integral membrane proteins and to be with a protein corona on their surface [32]. Moreover, their intravesicular space contains soluble proteins, genetic material, and various metabolites [6]. These physiological properties contribute to their potential as disease biomarkers [7,33,34]. Secreted by nearly all cell types, EVs carry selectively enriched cargo from their parent cells, serving as robust indicators of cellular pathophysiological states. Additionally, their lipid bilayer structure provides stability in biological fluids, protecting their contents against enzymatic degradation by proteases and nucleases [35]. EVs also contain multiple antigens, potentially enhancing the sensitivity and specificity of diagnostics compared to traditional single biomarkers, such as carcinoembryonic antigen, known as CEA, for colorectal cancer or prostate-specific antigen, known as PSA, for prostate cancer [8]. Additionally, genetic mutations are frequently selectively sorted into EVs, and their higher prevalence in EVs compared to whole-cell lysates offers greater diagnostic utility [36]. Another advantage of using EVs as biomarkers is their enrichment of relatively low-abundance plasma membrane proteins, which can indicate the state of the parent cell and reflect tissue- or cell-type-specific signatures, particularly under pathological conditions [37].
Despite these advantages, the application of EVs in omics-based biomarker discovery faces challenges due to contamination from bulk soluble proteins, which can hinder the identification of low-abundance EV-associated membrane proteins [33,38,39]. For instance, blood-derived EVs are frequently co-isolated with highly abundant serum proteins, such as albumin and immunoglobulins, which complicates downstream analyses [40]. Likewise, uEVs must be meticulously separated from Tamm-Horsfall protein (THP), also known as uromodulin, the most abundant protein in urine, which can interfere with the identification of EV-specific components [41]. Overcoming these challenges requires precise collection of biological fluids and improvement of isolation techniques to enhance EV purity and specificity, thereby improving their reliability as disease biomarkers.
Collection and normalization of urinary extracellular vesicles
Urine is a readily accessible biological fluid that can be obtained noninvasively, making uEVs a promising source of biomarkers for monitoring kidney function, urogenital tract conditions, and systemic disorders [42]. However, the concentration and composition of uEVs are influenced by the method of urine collection, necessitating careful consideration of standardized protocols [42]. Urine collection methods can be broadly categorized into three primary types: 1) timed urine, 2) spot urine, and 3) first or second morning urine. Timed urine collection involves sampling at specific intervals, such as 24-hour or hourly collection, while spot urine is obtained at a single time point. Morning urine collection is divided into first and second morning voids, each offering advantages in standardization. Additionally, urine can be collected via full voiding or midstream sampling, where the initial portion of urine is discarded, although a clear distinction between these methods has not been well established [42].
Among these methods, first morning urine is widely regarded as the most suitable sample for uEV analysis due to its higher EV concentration compared to spot urine, making it the preferred choice in research [42]. Proteomic comparisons between first and second morning urine showed minimal differences in protein levels and crucial uEV markers, such as aquaporin-2 (AQP2), sodium/hydrogen exchanger 3 (NHE3), ALIX, and TSG101 [43]. Furthermore, 96.3% of the proteins identified in EVs from first and second morning urine were consistent, indicating that both sample types are suitable for EV analysis [44]. Despite its frequent use, spot urine demonstrates significant variability in uEV concentration due to factors like diet, hydration status, and circadian rhythms, which complicates direct comparisons among individuals. To mitigate these inconsistencies, normalization techniques, including adjustment for urinary creatinine levels, are often implemented [45]. Alternatively, timed urine collection, particularly 24-hour urine sampling, is utilized to minimize variability, offering a more consistent representation of molecular composition and vesicle concentration [42].
Given the impact of collection time on uEV characteristics, normalization is essential for accurately assessing changes in uEV release and composition. Current normalization strategies incorporate both absolute and relative measures, including total protein content, uEV concentration, and biomarker expression, alongside adjustments based on urine creatinine levels, osmolality, or time of collection [42]. Given the technical variations introduced during uEV processing, appropriate corrections are necessary before meaningful comparisons can be made [46]. A significant challenge in uEV research is the lack of standardized normalization methods, which are crucial for addressing inconsistent factors such as excretion rates and technical inconsistencies in sample handling [47]. Normalization approaches can be categorized into absolute and relative excretion rates. Relative excretion rates define uEV marker abundance in relation to other properties, such as uEV count, a molecular marker, or total amounts of EV protein, nucleic acid, or lipid content, making them widely applicable in omics studies and nephrological research [48,49]. Absolute excretion rates are typically approximated by normalizing to urine osmolality or creatinine in spot urine samples [46,50]. The establishment of standardized normalization strategies is critical for ensuring the reproducibility and accuracy of uEV-based biomarker studies.
Purity and isolation of urinary extracellular vesicles
As described above, EVs are complex structures comprised of proteins, lipids, nucleic acids, and other biomolecules. These diverse components bestow distinctive physical properties, including size, density, charge, and molecular markers, which distinguish EVs from single-component molecules [6]. These properties facilitate the isolation of EVs using various biochemical techniques, including ultracentrifugation, density gradient ultracentrifugation, polymer precipitation, size exclusion chromatography, immunoprecipitation, asymmetric flow-field-flow fractionation, and acoustic-based methods, to extract EVs from a range of biological fluids, such as urine [6,51,52]. However, the purity of the isolated EVs significantly depends on the protein concentration of the originating biological fluid. For instance, high-protein fluids such as blood and urine lead to substantial protein adherence to the EV surface, resulting in a protein corona comprising molecules like fibronectin, apolipoproteins, and immunoglobulins [53]. Consequently, the high degree of protein attachment in vivo poses a significant challenge for in vitro EV isolation. Hoshino et al. [54] demonstrated that although the same ultracentrifugation method with repeated washing steps was used to isolate EVs from cell lines, blood, other biological fluids, and tissues, EVs derived from plasma and serum exhibited lower detection rates of canonical EV markers, such as CD63 and CD81, compared to EVs isolated from cell culture media. This indicates that contamination levels significantly differ between EVs derived from biological fluids and those from conditioned media. Co-purified non-EV proteins in biological fluids tend to lead to preferential identification of highly abundant proteins, thereby obscuring the detection of canonical EV markers such as tetraspanins [55]. Therefore, to minimize contamination and ensure a reliable EV proteome, high-purity EV isolation methods should be carefully selected based on the biological fluid.
To assess the purity of isolated EVs, Webber and Clayton [56] proposed a metric based on the ratio of particle number to protein amount. The particle concentration can be determined using techniques such as nanoparticle tracking analysis or tunable resistive pulse sensing [57], while protein content is typically quantified using conventional protein assays, such as the Bradford or bicinchoninic acid assay. According to this metric, an EV preparation is considered high purity if the ratio exceeds 3 × 1010 particles/µg, while a ratio between 2 × 109 and 2 × 1010 particles/µg indicates low purity, and ratios below 1.5 × 109 particles/µg suggest significant contamination [27]. High-purity EVs can be efficiently isolated from conditioned culture media using cushion ultracentrifugation, while EVs of lower purity are typically isolated via simple ultracentrifugation [27]. Additionally, high-purity EVs have been successfully isolated from glioma cells using density gradient ultracentrifugation, achieving a purity ratio of 3.53 × 1010 particles/µg [58]. Although Clayton’s purity assessment method is applicable to in vitro culture media, its applicability to in vivo biological fluids remains to be fully validated. When the same ultracentrifugation method, including a washing step, was applied to urine and serum, the resulting EV purities were approximately 1 × 109 particles/µg and 7 × 109 particles/µg, respectively, both markedly lower than the 2 × 1010 particles/µg derived from culture media [58]. The purity of uEVs varies depending on the isolation method, ranging from approximately 1 to 7 × 108 particles/µg [59,60].
Likewise, a primary challenge in uEV research is the effective isolation of uEVs from urine. The MISEV and the Urine Task Force of the ISEV recommend standardized protocols for the collection, isolation, characterization, and normalization of uEVs [42]. Similar to the protein albumin in blood, urine contains proteins that are highly abundant and often co-isolate with uEVs, potentially compromising their purity. The most abundant protein in urine, THP, which exists as a polymer, can co-sediment with uEVs, thus reducing their purity and yield. To minimize THP contamination, reducing agents such as dithiothreitol are routinely employed to depolymerize THP, preventing its co-isolation with uEVs [42,51]. The elimination of THP is crucial for downstream analyses like mass spectrometry-based proteomics, where contamination can mask the detection of low-abundance proteins associated with EVs. Standardizing uEV isolation methods and implementing effective purification strategies are vital for enhancing the reliability and reproducibility of uEV-based biomarker studies.
Origins of urinary extracellular vesicle
The analysis of uEVs provides a noninvasive means to investigate the status of kidney and urogenital tract diseases. In 2004, Pisitkun et al. [49] conducted pioneering research in this area by isolating uEVs via ultracentrifugation and characterizing them using mass spectrometry-based proteomics. A total of 295 proteins, predominantly derived from nephron epithelial cells and urothelial cells of the urinary system, were identified [49]. Their findings indicated that several of these proteins were associated with ESCRT proteins, suggesting that exosomes constitute a significant subpopulation of uEVs [49]. The composition and subtypes of uEVs are influenced by both physiological stress and pathological conditions. Despite their relative scarcity in bulk urine, these EVs are intricately associated with specific cell types and tissues within the urinary system, reflecting the status of the parental cells [42]. This ability to detect low-abundance biomolecules, previously unidentified in urine, opens new avenues for disease diagnosis [15,16].
Most uEVs originate from cells of the urinary system [42], rendering them a critical source of molecular biomarkers for diseases affecting the kidneys, bladder, prostate, uterus, testes, and other organs in the urogenital tract (Fig. 1) [15,44,49,61–64]. These EVs may also play a functional role in both physiological and pathological processes within this system [65]. For example, uEVs from renal cells, such as podocytes, proximal tubular cells, and collecting duct cells, frequently contain AQPs, which are commonly found in kidney-derived EVs in healthy individuals. However, under pathological conditions, the abundance and proportions of AQPs can be altered, as evidenced by reduced urinary AQP1 levels in both rat models of renal ischemia-reperfusion injury and human kidney transplant patients [66]. These alterations underscore the potential of uEVs as indicators of kidney function and disease, including acute kidney injury and CKD. Furthermore, uEVs derived from podocytes show changes in complement receptor-1 expression in renal transplant patients with podocyte injury [67].
Beyond the kidney, other cells of the urogenital tract also contribute to the uEV population. Cells of the bladder and urethra release EVs containing uroplakins, proteins that are specifically expressed in the bladder [49], along with the epithelial marker CD24 [42]. Prostate epithelial cells secrete EVs containing prostate-specific antigen and other proteins derived from the prostate, which are of diagnostic significance for conditions such as prostate cancer [8]. Given that uEVs transport a diverse molecular cargo, which varies according to their cellular origin and disease state, they constitute a highly heterogeneous group. This heterogeneity augments their potential as biomarkers for a wide array of urogenital diseases, including CKD, rendering them a valuable tool for noninvasive disease detection and systemic health monitoring.
Biomarker potential of urinary extracellular vesicles in diabetic kidney disease
As previously mentioned, CKD is a progressive pathological condition characterized by the decline of kidney function and structural deterioration. The predominant causes of CKD include diabetes mellitus, glomerulonephritis, IgA nephropathy, and hypertension [2]. Given the complex and multifactorial nature of CKD, identifying reliable biomarkers for early detection and subsequent monitoring of the disease remains a significant challenge. In this context, uEVs have emerged as a promising biomarker source (Table 2), offering valuable insights into disease pathophysiology and potential therapeutic targets [42]. Proteins associated with uEVs play crucial roles in inflammation, fibrosis, oxidative stress, and metabolic dysregulation, all of which contribute to the progression of CKD [11]. Furthermore, advancements in high-throughput proteomics have significantly enhanced the characterization of the uEV proteome, facilitating the discovery of novel biomarkers with potential applications in the early detection, prognosis, and therapeutic response monitoring of CKD [68]. The stability of uEVs in biological fluids and their capacity to reflect dynamic changes in renal pathology further emphasize their clinical relevance [42].
uEV biomarkers in chronic kidney disease
Diabetic nephropathy, also known as DKD, is a significant complication of diabetes mellitus characterized by persistent albuminuria and progressive renal dysfunction [69]. Diabetic nephropathy impacts approximately 20% to 50% of patients with both type 1 and type 2 diabetes mellitus [69] and often progresses with minimal early clinical signs. This underscores the critical need for reliable biomarkers for the early detection and monitoring of the disease. The traditional noninvasive marker, microalbuminuria, has inadequate sensitivity and specificity, making it necessary to explore alternative biomarkers [70]. Recent investigations have considered uEVs as promising biomarkers for the early diagnosis of DKD. For example, levels of neutrophil gelatinase-associated lipocalin in uEVs have been observed to increase rapidly following kidney injury in patients with type 1 diabetes mellitus [71]. Moreover, expression of voltage-dependent anion-selective channel protein 1, known as VDAC1, in kidney tissues and uEVs is diminished in patients with diabetic nephropathy, indicative of its impaired role in calcium transport, gene regulation, and apoptosis [72]. In contrast, levels of α-microglobulin/bikunin and histone-lysine N-methyltransferase are elevated in uEVs from diabetic nephropathy patients, suggesting their roles in regulating kidney function, cellular survival, signaling, and apoptosis [72]. Additionally, the protein C-megalin, primarily expressed in proximal tubule cells, has been found to progressively increase in stage from normoalbuminuric to macroalbuminuric, endorsing its potential as an early diagnostic marker [73]. Podocytes play an essential role in maintaining the glomerular filtration barrier and preventing proteinuria, and podocyte dysfunction is a hallmark of diabetic nephropathy. Progressive injury to podocytes leads to disruption of this barrier, which accelerates the progression of diabetic nephropathy [74]. Recent studies have shown that uEVs may indicate podocyte dysfunction [75]. Notably, the release of podocyte-derived EVs is significantly elevated in patients with type 1 diabetes mellitus, particularly under hyperglycemic conditions [76]. Furthermore, increased expression of Elf3 protein in uEVs has been associated with Smad3 activation-dependent podocyte injury in patients with diabetic nephropathy, underscoring its potential as a disease biomarker [77].
Dysregulation of enzymes in parental cells of the urinary system is reflected in uEVs, which are implicated in the progression of DKD. Musante et al. [78] demonstrated that uEV protease and protease inhibitor profiles can serve as biomarkers for kidney damage in DKD. Their study revealed a gradual increase in the activation of cathepsins A, C, and D from normoalbuminuric to macroalbuminuric stages, while matrix metalloproteinase 2 (MMP2) exhibited a progressive decline in uEVs [78]. Furthermore, protease inhibitors such as cystatin B, serpin A8, and serpin B5 were found upregulated in uEVs during the early stages of albuminuria, indicating their potential protective role against renal injury and inflammation [78]. Ceruloplasmin, a protein involved in iron and copper metabolism, was significantly elevated in uEVs from DKD patients, indicating its involvement with oxidative stress and inflammation [79]. Additionally, Sun et al. [80] reported a progressive increase in dipeptidyl peptidase 4 (DPP4) expression in uEVs as DKD advanced, implicating DPP4 in the promotion of inflammation and fibrosis. Also, α1-antitrypsin, a key protein involved in responses to inflammation and oxidative stress, was observed to gradually increase in uEVs during DKD progression [81].
Biomarkers linked to inflammation and fibrosis have also been identified in uEVs from patients with diabetic nephropathy. Cappelli et al. [82] reported increased CD73 expression, implicating its involvement in the transforming growth factor beta signaling pathway, a key driver of renal fibrosis and disease progression. Similarly, Ding et al. [83] demonstrated that phytanoyl-CoA dioxygenase domain-containing protein 1, which is associated with lipid metabolism, exhibited increased expression among diabetic nephropathy patients, suggesting a role in alterations of lipid metabolism that contribute to inflammation and fibrosis. Moreover, epidermal growth factor receptor and p21-activated kinase 6 are significantly upregulated in uEVs from diabetic nephropathy patients, indicating their roles in renal inflammation, fibrosis, and cellular stress [84].
Advancements in high-throughput mass spectrometry-based proteomics have greatly improved characterization of the uEV proteome, leading to the identification of novel DKD biomarkers. Increased levels of mannan-binding lectin serine protease 2 (MASP2) and calbindin were observed with the progression of DKD, while levels of S100 calcium-binding protein A8 (S100A8) and S100A9 decreased [85], indicating disruptions in inflammatory and calcium homeostasis pathways. This elevation in MASP2 expression in uEVs correlated with an enhanced inflammatory response via the complement system, and the increased calbindin levels suggested alterations in calcium homeostasis [85]. Additionally, phosphoproteomic studies revealed phosphorylated AQP2 and phosphorylated glycogen synthase kinase 3 as potential DKD biomarkers, indicating disturbances in water balance regulation and inflammatory signaling [86]. Overall, significant advances have been made in elucidating the role of uEVs in DKD, enabling more precise identification of biomarkers. These findings underscore the potential of uEVs to reflect inflammatory and metabolic changes in the kidney and urinary system.
Biomarker of urinary extracellular vesicles in glomerulonephritis, immunoglobulin A nephropathy, and hypertensive nephropathy
Beyond DKD, glomerulonephritis is another principal cause of CKD, characterized by immune-mediated glomerular inflammation and injury [87]. Glomerulonephritis, an inflammatory condition affecting the glomeruli, has etiologies ranging from autoimmune diseases such as IgA nephropathy to infections and toxic exposures [87]. Diagnosing glomerulonephritis involves assessing albuminuria, classified into normoalbuminuria, microalbuminuria, and macroalbuminuria based on levels of urinary albumin excretion [87]. Recently, uEVs have emerged as a promising noninvasive source of molecular biomarkers for glomerulonephritis, providing critical insights into glomerular injury, immune responses, and disease progression (Table 2) [88]. For example, the expression of CD133 in uEVs correlates with acute and chronic glomerular damage, showing declining levels in patients with chronic glomerulonephritis and varying degrees of albuminuria, suggesting its potential as a biomarker for glomerular injury [62]. Furthermore, the enrichment of disease-specific proteins such as nephrin, short transient receptor potential channel 6 (TRPC6), inverted formin 2 (INF2), and phospholipase A2 receptor in uEVs underscores their utility in glomerulonephritis diagnostics [61].
In glomerulonephritis, hematuria, characterized by the presence of blood in the urine, frequently occurs when red blood cells enter the urine due to glomerular damage [89]. The passage of hemoglobin through the capillary wall can damage podocytes, leading to changes in the glomerular filtration barrier, especially evident in patients experiencing episodes of macroscopic hematuria, such as those with IgA nephropathy [89]. Particularly, IgA nephropathy, the most prevalent cause of glomerular hematuria, involves the deposition of IgA in the glomerular mesangium. In cases of IgA nephropathy, the secretion of uEVs is linked with injury or changes in the status of tubular and renal epithelial cells [90]. For example, nidogen-1, a critical basement membrane protein for structural integrity, is conspicuously diminished in uEVs from patients with IgA nephropathy. Moreover, uEV proteins related to inflammation and oxidative stress, such as mannosyl-oligosaccharide 1,2-alpha-mannosidase IA, haptoglobin, and monocyte chemoattractant protein 1, are significantly increased [91]. These findings suggest that podocyte dysfunction and inflammatory responses contribute to the progression of IgA nephropathy [91]. MASP2, a crucial activator of the complement system, is overexpressed in uEVs from patients with IgA nephropathy, indicating a potential role for complement-mediated glomerular injury [92]. Analysis of uEV biomarkers could help differentiate subtypes of glomerulonephritis. For example, vasorin levels in uEVs differ between IgA nephropathy and membranous glomerulonephritis, with elevated expression in membranous glomerulonephritis and reduced levels in IgA nephropathy, indicating variations in glomerular endothelial-podocyte interactions [93]. However, both disorders are characterized by increased ceruloplasmin and decreased aminopeptidase N in uEVs, suggesting a common underlying pathology influenced by inflammation and oxidative stress [93]. Furthermore, ceruloplasmin and α1-antitrypsin are upregulated in IgA nephropathy, while aminopeptidase N and vasorin precursor show increased expression in thin basement membrane nephropathy alone [41]. These findings underscore the role of uEV biomarkers in distinguishing glomerular diseases and elucidating their pathophysiological mechanisms.
Hypertension is a significant contributor to CKD progression, frequently exacerbating glomerular damage and leading to ESRD [94]. Moreover, hypertension can induce renal impairment, further contributing to the development of CKD [95]. The bidirectional relationship between hypertension and CKD underscores the importance of identifying biomarkers to differentiate between subtypes of hypertensive nephropathy. AGP1 is significantly elevated in uEVs from patients with primary aldosteronism compared to those with essential hypertension [96]. Furthermore, AQP1, AQP2, and AGP1 in uEVs have been identified as key biomarkers distinguishing primary aldosteronism from essential hypertension, reflecting aldosterone-driven dysregulation of water balance and electrolyte homeostasis in secreted uEVs [97]. Long-chain fatty acid transport protein 2 and amnionless in uEVs have also been proposed as promising biomarkers for hypertensive kidney injury [68]. Distinct protein expression patterns in uEVs differentiate de novo hypertensive albuminuria from established hypertensive nephropathy, with myeloperoxidase, olfactomedin 4, and antithrombin III showing differential expression in early versus advanced disease stages [98]. Collectively, these findings highlight the potential of uEVs in advancing CKD diagnostics by enabling noninvasive tracking of kidney-specific molecular changes.
Conclusions
uEVs have emerged as crucial mediators of intercellular communication and valuable biomarkers for CKD, including DKD and glomerulonephritis. The molecular composition, influenced by renal pathophysiology, and secretion of uEVs reflect dynamic alterations associated with the progression of CKD, such as inflammation, fibrosis, oxidative stress, and metabolic dysregulation. In DKD, specific uEV-associated proteins, including DPP4, ceruloplasmin, and α1-antitrypsin, are linked to hyperglycemia-induced renal injury, serving as potential biomarkers for early disease detection and monitoring of progression. Similarly, in glomerulonephritis, uEV profiling has identified key glomerular-specific proteins such as CD133 and nidogen-1, which provide insights into podocyte dysfunction and complement-mediated glomerular injury. Additionally, hypertension-associated CKD exhibits distinct uEV protein signatures, such as AGP1 and AQPs, distinguishing primary aldosteronism from essential hypertension and highlighting their potential role in CKD subtype classification. The MISEV guidelines [31], established by the ISEV, provide standardized criteria for EV isolation, characterization, and reporting, ensuring reproducibility and rigor in uEV research. Adherence to MISEV recommendations enhances data comparability across studies, facilitates biomarker discovery, and promotes the translational potential of uEV-based diagnostics. In particular, to achieve confident uEV biomarker discovery, high-purity isolation methods such as density gradient ultracentrifugation have enhanced the reliability of uEV biomarker discovery. However, traditional bulk uEV analyses face limitations in capturing the heterogeneity of EV populations, as different CKD subtypes may be associated with distinct uEV subpopulations. Addressing this challenge requires the development of advanced EV analysis approaches, enabling precise characterization of uEV dynamics across CKD stages. As molecular representatives of their parental cells in the kidney and urogenital tract, uEVs could be utilized for the development of innovative diagnostic tools and precision medicine approaches in CKD and related disorders. Future research should focus on refining and optimizing urine storage conditions, normalization strategies, and uEV isolation techniques, while also integrating multi-omics approaches and validating biomarker candidates in large-scale clinical studies to facilitate their clinical translation.
Notes
Conflicts of interest
All authors have no conflicts of interest to declare.
Funding
This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2022R1C1C1010979, RS-2024-00348103, and RS-2023-00219563) and supported by the Soonchunhyang University Research Fund.
Data sharing statement
The data presented in this study are available from the corresponding author upon reasonable request.
Authors’ contributions
Conceptualization: CWS, JSO, DC
Data curation, Investigation, Visualization: CWS, EYL
Funding acquisition: DC
Writing–original draft: All authors
Writing–review & editing: EYL, JSO, DC
All authors read and approved the final manuscript.