Table 1.
EVs as diagnostic resources in solid organ transplantation.
| Research Topic | Organ | Study Population | EVs Origin | EVs Component | Reported Outcomes | Reference |
|---|---|---|---|---|---|---|
| Preoperative donor organ assessment | Kidney | LD | Urinary EVs | Various membrane protein signatures | Kidneys with nephrosclerosis correlated to fewer podocyte, parietal, or tubular cell EVs, among others. Kidneys with nephron hypertrophy correlated to fewer mesangial or descending limbs of Henle’s loop cell EVs. | Turco A.E. et al. [58] |
| LD vs. DD | miRNAs analysis, miR-326 |
No overall differences were found in EVs miRNA profiles of living and deceased donors in normofunctioning grafts at 1 year. Only miRNA-223, which targets pro-apoptotic protein Bcl-2, was found overexpressed in living donors. |
Lozano-Ramos S.I. et al. [59] | |||
| Prediction of postoperative graft function | Kidney | Human TR with vs. without DFG | EVs in preservation fluid | Multiple miRNAs | Differences between the two groups were found in 10 miRNAs upon basic analysis, but differences were lost upon multiple testing correction. Groups were not successfully identified via unsupervised clustering in PCA. | Gremmels H. et al. [60] |
| Human TR with vs. without DFG | Urinary EVs | Membrane protein (CD133+ EVs) | Patients with DFG had a significant increase in the CD133+ extracellular vesicle subpopulation compared with patients with early graft function. CD133+ may reflect the activity of progenitor cells in damage repair. | Dimuccio V. et al. [62] | ||
| Human TR with vs. without DFG (from DD vs. LD) | NGAL | NGAL levels were higher in kidney recipients from DDs. NGAL levels are significantly higher in patients with DGF compared with early graft function. |
Alvarez S. et al. [63] | |||
| Human TR | NGAL, cystatin C, and IL-18 mRNA |
Free urinary NGAL and cystatin C were correlated with serum creatinine at day 7 post-transplant. However, a poor correlation between EV-contained NGAL, cystatin C, or IL-18 mRNA and serum creatinine was found. | Peake P.W. et al. [64] | |||
| Human TR | Proteomic analysis, phosphoenolpyruvate carboxykinase 2 (PCK2) |
Proteomic profile 1-day post-transplant correlated with renal function at 1 year. PCK2 1-day post-transplant in uEVs, but not in renal tissue, could predict renal function at 1 year. |
Braun F. et al. [119] | |||
| Human TR with vs. without DFG (from DD) | Plasma-derived EVs | hsa-miR-33a-5p, hsa-miR-98-5p, and hsa-miR-151a-5p |
A total of 52 miRNAs were found to be overexpressed in DGF compared with early graft function; of them, the mentioned 3 miRNAs were coexpressed. hsa-miR-151a-5p was positively correlated with first-week markers of graft function. | Wang J. et al. [65] | ||
| Human TR with good vs. poor outcome based on eGFR | Proteomic analysis | Proteomic profile could differentiate patients with good outcomes from those with poor outcomes based on eGFR at 1 month. | Al-Nedawi K. et al. [66] | |||
| Human TR with vs. without DFG (eGFR < 60 mL/min/1.73 m2) vs. healthy controls | miRNA analysis (miR-21-5p, miR-210-3p, and miR-4639-5p) | The panel could accurately differentiate between subjects with chronic allograft dysfunction and normal graft function, with better performance than simple or double indicators (ROC-AUC 0.89). | Chen Y. et al. [67] | |||
| Diagnosis of acute graft rejection | Kidney | Human TR with vs. without acute rejection | Urinary EVs | mRNA analysis (CXCL11, STAT1, SERPINA1, BMP7, NAMPT, IFNGR1, and IL18BP, among others) | This panel outperforms eGRF in early diagnosis of acute graft rejection (ROC-AUC 0.93). Moreover, it can differentiate mechanism of rejection (TCMR vs. ABMR). | El Fekih R. et al. [80] |
| Human TR with vs. without ABMR vs. TCMR | Proteomic analysis, cystatin C (CST3), and lipopolysaccharide-binding protein (LBP) |
The combination of EV-contained CST3 and LPS can accurately identify ABMR patients versus non-rejection patients (ROC-AUC 0.879 and 0.901, respectively), as well as to differentiate them from TCMR. | Kim M. et al. [77] | |||
| Human TR with vs. without TCMR | Proteomic analysis Tetraspanin 1 (TSPAN1) and hemopexin (HPX) |
TSPAN1 and HPX were significantly overexpressed in TCMR patients. | Lim J.H. et al. [78] | |||
| Human TR with vs. without TCMR | T-cell-derived urinary EVs | Membrane protein (CD3) | Presence of T-cell-specific membrane marker CD3 could accurately predict TCMR (ROC-AUC 0.911). | Park J. et al. [81] | ||
| Human TR with vs. without AR |
Whole urine urinary EVs |
Proteomic analysis | Eleven proteins were overexpressed in AR, three of which (CLCA1, PROS1, and KIAA053) were specific to the EV fraction. | Sigdel T.K. et al. [79] | ||
| Human TR with vs. without ABMR |
Plasma-derived EVs | Membrane markers (C4d+, CD144+, and annexin V+) |
C4d+/CD144+ and C4d+/annexin V+ EV subpopulations were significantly increased in AR patients. Upon treatment, C4d+/CD144+ EVs significantly decreased. | Tower C.M. et al. [82] | ||
| Human TR with vs. without ABMR vs. TCMR |
mRNA analysis | Six genes were overexpressed in ABMR patients. A combination of 4 genes (gp130, SH2D1B, TNFα, and CCL4) can accurately predict ABMR. | Zhang H. et al. [83] | |||
| Liver | Human TR with vs. without TCMR |
Plasma-derived EVs | Multiple miRNAs | Expression of miRNAs was significantly different between AR and non-AR patients. miR-223 and let-7e-5p were up-regulated in AR patients, whereas miR-199a-3p was down-regulated. | Wang W. et al. [120] | |
| Human TR with vs. without TCMR |
Plasma-derived EVs | Galectin-9 | Levels of galectin-9 were higher in patients with acute TCMR. | Zhang A. et al. [87] | ||
| Lung | Human TR with vs. without AR or BOS |
Plasma-derived EVs BALF-derived EVs |
Donor HLA; lung-associated self-antigens (collagen V [Col-V] and K alpha 1 tubulin [Kα1T]) miRNA analysis |
EV-contained donor HLA and collagen V were significantly overexpressed in AR and BOS compared with healthy patients (p < 0.05). Collagen V was detected 3 months before AR and 6 months before BOS diagnosis. Differentially expressed immunoregulatory miRNAs were found for AR (miR-92a and miR-182) and BOL (previous ones and miR-142-5p and miR-155) compared with control. |
Gunasekaran M. et al. [100] | |
| Human TR with vs. without AR |
BALF-derived EVs | RNA analysis | Transcriptomic signatures were significantly different between patients with and without AR. Patients with AR showed overexpression of antigen-processing immune activation pathways. | Gregson A. et al. [99] | ||
| Heart | Mice TR with vs. without AR | Plasma-derived EVs | Total plasma EV concentration | Total plasma EV concentration remained stable in control group, while it significantly decreased in the AR group at grade 0R rejection on histology. The model proved accurate for early prediction of AR (ROC-AUC 0.934) before any histology changes are detected. | Habertheuer A. et al. [93] | |
| Human TR with vs. without ABMR vs. TCMR |
Membrane proteins (37 proteins) | AR EVs had increased concentration and decreased diameter. AR overexpressed HLA-I, CD41b, ROR-1, and SSEA-4 compared with controls. TCMR overexpressed CD2 and CD3, while ABMR overexpressed HLA-II, CD-326, CD19, CD20, and CD-25. The diagnostic model built on these markers reached a high accuracy (ROC-AUC 0.865) |
Castellani C. et al. [94] | |||
| Human TR with vs. without ABMR vs. TCMR |
Proteomic analysis | A total of 45 EV-derived proteins were identified to differentiate 3 groups: control/heart failure group, heart transplant without rejection and, ABMR and TCMR. A total of 15 of them were differentially expressed between the 2 last groups (p < 0.05). Most of these proteins play a role in the immune response (complement activation, adaptive immunity, and coagulation). | Kennel P. et al. [95] | |||
| Pancreatic islets | Mice TR from human islets, with and without induced AR | Plasma-derived EVs from donor’s islets and recipient T cells | EVs concentration, proteomic analysis, and miRNA analysis |
AR led to a decrease in donor EVs and an increase in T cell recipient EVs. Four proteins were differentially expressed in AR versus control: angiopoietin 1, HSC70, C3, and hemopexin. Changes in microRNA and proteomic profiles were detected in AR prior to clinical effects (hyperglycemia). |
Vallabhajosyula P. et al. [89] | |
| Diagnosis of chronic graft rejection | Kidney | Human TR with chronic ABMR vs. healthy and other-cause damage (calcineurin inhibitors toxicity and interstitial fibrosis) |
Urinary EVs | SYT17 | Chronic ABMR patients had significantly higher SYT17 than the other groups. SYT17 could predict chronic ABMR with higher accuracy than traditional laboratory parameters (ROC-AUC 0.82). | Takada Y. et al. [84] |
| Human TR with vs. without chronic ABMR |
Proteomic analysis, APOA1, TTR, PIGR, HPX, AZGP1, and CP |
Expression of the six proteins was increased in chronic rejection compared with long term graft survival. | Jung H.Y. et al. [85] | |||
| Human TR with chronic allograft disfunction, with vs. without ABMR |
T helper cells and plasma-derived EVs | Membrane proteins (CD4, CXCR5, CXCR3, and CTLA4) | The CD4+ CXCR5+ CXCR3- EV subpopulation was higher in ABMR patients, while expression of CTLA-4 was lower in this group. | Yang J. et al. [86] | ||
| Lung | Human TR with vs. without BOS Mice immunized with EVs from patients with vs. without BOS |
Plasma-derived EVs | Kα1T; Col-V MHC-II; costimulatory molecules, CD40, CD80, and CD86; and transcription factors (NF-κB, hypoxia-inducible factor 1-α, and IL-1R–associated kinase 1, among others) | The aforementioned proteins were overexpressed in BOS versus control patients. Mice treated with EVs from BOS patients developed a specific proinflammatory phenotype. |
Gunasekaran M. et al. [121] | |
| Diagnosis of ischemia reperfusion injury | Kidney | Mice with vs. without IRI Mice with vs. without IRI |
Plasma-derived renal EVs | miRNA-23a | IRI increased miRNA-23a, which plays a role in macrophage activation. Inhibition of miRNA-23a ameliorated inflammation in the renal parenchyma. | Li Z. et al. [106] |
| miR-374b-5p | Levels of miR-374b-5p were increased after IRI. Inhibition of miR-374b-5p would alleviate kidney injury, showing its role in the damage cascade. | Ding C. et al. [107] | ||||
| Rats with vs. without unilateral IRI | Urinary EVs | Aquaporin 1 (AQP1); fetuin-A |
Glycosylated AQP1 secretion was significantly reduced in the first 6 h after IRI compared with controls or other causes of renal injury. AQP1 was also reduced in a TR patient 48 h after transplantation. |
Sonoda H. et al. [104] | ||
| Diagnosis of infection | Kidney | Human TR with vs. without BK virus nephropathy | Urinary EVs | bkv-miR-B1-5p; bkv-miR-B1-5p/miR-16 |
Levels of viral miRNA (bkv-miR-B1-5p and bkv-miR-B1-5p/miR-16) showed a significant correlation with urinary BK viral load, as well as to plasma BK viral load, and could accurately predict viruria (ROC-AUC 0.989 and 0.985, respectively). | Kim M. et al. [116] |
| Lung | Human TR with vs. without symptomatic respiratory tract infection Mice as recipients for EVs treatment |
Plasma-derived renal EVs | Lung-associated self-antigens (collagen V [Col-V], K alpha 1 tubulin [Kα1T]), 20S proteasome, and viral antigens |
EV-contained self-antigens and viral antigens were higher in recipients of symptomatic respiratory viral infections. Mice immunized with those EVs developed immune responses to self-antigens, such as fibrosis, small airway occlusion, and cellular infiltration. | Gunasekaran M. et al. [118] | |
| Immunosuppressive drug monitoring | Kidney | Human TR under calcineurin inhibitor treatment with vs. without chronic calcineurin inhibitor toxicity vs. interstitial fibrosis and tubular damage from other causes | Urinary EVs | Proteomic analysis, CTSZ, RAB8A and SERPINC1 |
Members of the uroplakin and plakin families were significantly overexpressed in the group with calcineurin inhibitor toxicity. CTSZ, RAB8A, and SERPINC1 were significantly overexpressed in patients with toxicity compared with normally functioning ones. |
Carreras-Planella L. et al. [109] |
| Human TR under various immunosuppressive therapies and tacrolimus therapy | miRNA analysis | Expression of miR-155-5p and miR-223-3p showed significant correlation with tacrolimus dose and could be used to monitor toxicity. miR-223-3p also correlated with serum creatinine. |
Costa de Freitas R. et al. [110] |
All transplant recipients received allogenic grafts. All changes in the “reported outcomes” column were measured in EVs from the aforementioned origins. ABMR: antibody-mediated rejection, AR: acute rejection, BALF: bronchoalveolar lavage fluid, BOS: bronchiolitis obliterans syndrome, DD: deceased donor, DGF: delayed graft function, eGFR: estimated glomerular filtration rate, IRI: ischemia-reperfusion injury, NGAL: neutrophil gelatinase-associated lipocalin, LD: living donor, PCA: principal component analysis, TR: transplant recipient, and TCMR: T-cell-mediated rejection.