Systemic lupus erythematosus (SLE) is an autoimmune disease that predominantly affects women and typically has manifestations in multiple organs. Immune-system aberrations, heritable, hormonal, as well as environmental factors, contribute to the progression of organ damage (1). Lupus nephritis (LN) is a form of glomerulonephritis considered as a severe manifestation of the SLE and continues to be a major cause of morbidity and mortality for SLE patients (2, 3). Routine laboratory examinations such as proteinuria, anti–double-stranded DNA antibodies, and protein-to-creatinine ratio cannot fully reflect disease activity and present with the low diagnostic performance of LN (3–6). To date, renal biopsy was regarded as the gold standard for the diagnosis and classification of scarring and the extent of renal inflammation, but the invasiveness of this approach is not conducive to dynamic monitoring (7). Therefore, it is necessary to search and identify effective non-invasive biomarkers for the detection of LN in SLE.

Exosomes are extracellular vesicles (EVs) secreted by cells mainly consists of phospholipids bilayer structures, with a diameter range from 40 to 160 nm (average ~100 nm), existing in various body fluids including blood, urine and saliva, and plays important roles in remodeling extracellular matrix and transmitting signals and molecules to other cells (8, 9). Endogenous microRNAs (miRNAs), long non coding RNAs (lncRNAs), and tRNA-derived small noncoding RNA (tsRNAs) can be frequently transferred through exosomes from donor cells to recipient cells, displaying important immunomodulatory function in the pathogenesis of various autoimmune diseases, including SLE (10). Recently studies showed that abnormal expression of serum exosomal-miRNAs and lncRNAs was described in LN/SLE patients and both of them are associated with the activity of SLE, histological alterations, and the involvement of renal (11–13). Despite the growing interest in exploring serum exosome derived biomarkers for evaluating activity of disease and predicting the involvement of LN, few biomarkers have been used in clinical practice. Unlike other sample types like tissue or serum, urine testing can be truly non-invasive. Furthermore, the urine is physically close to the site of activity of the renal disease which may be promising specimens for monitoring the patients of SLE.

tsRNAs are novel noncoding small RNAs (14∼40 nt in length), has rarely been described in SLE, including tRNA-derived fragments (tRFs) and tRNA halves (tiRNAs) originate from mature tRNAs or their precursors under the circumstance of enzymatic lysis and stress condition, and have been found stable in the exosome circulating in the biofluids, including urine (11). Furthermore, tsRNAs exert an essential role in the pathophysiology of biological processes by connecting with proteins or mRNA, regulating gene expression, and inhibiting translation (13). Mounting evidence has indicated that the aberrated expression of tsRNAs are deeply implicated in cancer, parasitic disease (12–16). Recently, a few studies including our previous research pay attention to the relationship between SLE and tsRNA (14, 17–19). However, the diagnostic values and biological functions of tsRNAs in urinary exosome, especially for LN, are still ambiguous and intriguing.

In this study, RNA sequencing, qRT-PCR validation, and ROC analysis were performed to identify LN-associated tsRNA signatures in urinary exosome. The urinary exosome derived tsRNA signatures showed enormous potentiality as novel non-invasive biomarkers for diagnosing and predicting LN in SLE.

SLE is an autoimmune disease characterized by abnormal activation of lymphocytes and abnormal accumulation of autoantibodies (22, 23). Lupus nephritis is one of the most devastating manifestations of SLE, during which 40% of them will develop chronic kidney disease and 10–20% of them will progress to end-stage renal disease within 5 years after diagnosis (14, 24). Thus, early detection and prediction of the progression of LN is critical for the management and treatment of SLE. As the main diagnostic criterion for LN, the 24-hour urinary protein provides good estimate of proteinuria. However, this method still has many drawbacks except for the inconvenience and the length time taken to undertake the test, such as inaccurate timing, partial urine sample loss during urine retention, poor patient compliance for urine protein test, physical activity, and not appropriate for outpatients or children (14, 24–28). Similarly, the usefulness of examining the urinary sediment has been the subject of debate because of the huge variation in quantifying the finding (3). In fact, protein-to-creatinine ratio in spot urine has replaced 24-hour urinary protein collections in many centers, while this approach has not been carefully validated (3). In addition, non-conventional biomarkers, like monocyte chemoattractant protein-1 (MCP-1) (29, 30), kidney injury molecule–1 (KIM-1) (30), vascular cell adhesion molecule-1 (VCAM-1) (31), neutrophil gelatinase-associated lipocalin (NGAL) (32), and microRNAs (33) have been described in the diagnosis and disease activity monitoring of LN, while the specificity and sensitivity are typically difficult to meet clinical practice standards. Therefore, reliable non-invasive biomarkers with higher sensitivity and specificity for the early detection and risk stratification of LN are urgently needed.

Urinary exosomes originated from glomerular podocytes or renal tubules epithelial cells enclose various species of RNA including tsRNAs (34, 35). Zhu et al. (36) found that the bladder, endothelial cell, basal cell, monocyte, and dendritic cell may closely participate in the formation of urine exosomes by investigating the genetic sources of urinary exosomes both at the levels of organs and cells. The evidence showed that the existence of CD47 on exosomes prolongs their half-life period in the circulation (37). In addtion, the tsRNAs were encapsulated in exosomes protecting them from degradation, and have a good deal of methylated modifications and terminal modifications making them stable in body fluids (38).The changes of urinary exosomal tsRNAs may reflect renal injury so that can be expected to be a potential clinical biomarker without need for an invasive tissue biopsy.

Recently, Park. Et al. reported that tsRNAs were capable of regulating the chromatin states in immune cells, indicating that tsRNAs involved critically in the occurrence and development of immune-related diseases (11). Chioyu et al. has described that tsRNA were enrich in exosomes released by T cells, which played a critical role in the organ involvement of SLE (39). More importantly, the content of tsRNAs in vesicles were significantly higher than miRNAs (39–41). tsRNAs in serum as biomarkers of SLE and LN patients were confirmed in our previous studies (14, 15).Based on the perspective of non-invasive liquid biopsy, this was the first report to explore whether urinary exosome derived tsRNA signatures could be considered as biomarkers for distinguishing LN from SLE patients

In our study, exosomes were characterized by TEM and NTA and double phospholipid bilayer cup-shaped vesicle structures with average particle about 129 nm were obtained, which was consistent with previous study (42). According to the sequencing results, we discovered 275 upregulated urinary exosomes derived tsRNAs in LN compared with SLE without LN patients. In training phase, 10 tsRNAs were verified increasing expression in LN. In validation phase, based on more clinical specimens, we further narrowed the candidates tsRNAs down to tRF3-Ile AAT-1 and tiRNA5-Lys-CTT-1. In addition, ROC curve analyses showed the high-precision discriminatory power of tRF3-Ile AAT-1 and tiRNA5-Lys-CTT-1 between SLE patients with and without LN. The evidence showed that angiogenin cleaves t-RNA into fragment as tsRNAs under conditions of cell stress, which contributes to immune system dysregulation, autoantibody production and fatal comorbidities of SLE patients (39, 40). tRFs and tiRNAs inhibit translation in a variety of ways (38), and possibly slow down the progression of some inflammatory reaction manifestation of SLE in the kidneys. In our study, tRF3-Ile-AAT-1 and tiRNA5-Lys-CTT-1 showed good diagnostic value in identifying LN in SLE patients, which are better than conventional biomarkers such as anti-dsDNA antibody, C3 and C4. We further created a panel consisted of both tsRNAs and clinical parameters such as Alb and EGFR, demonstrating higher diagnostic value in distinguishing LN from SLE patients. SLE-DAI is a reliable indicator reflecting SLE activity scored by clinical symptoms and auxiliary examination, while 24 hours proteinuria is considered as the main diagnostic criteria for LN contraindicated to renal biopsy (3, 21). The two tsRNAs were positively correlated with proteinuria and SLE-DAI score, implying that tsRNAs may be involved in the activity of SLE and progression of LN. In addition, LN often present with mild proteinuria (3), and our study analyzed the significance of tsRNAs in clinical practice and found that it could identify active LN who were often missed because the urine protein in routine examination did not reach strong positive. Besides, both tsRNAs in our study can distinguish active SLE from stable SLE patients, which can help rheumatologist judge the severity and risk of disease. Intriguingly, there was a strong correlation between the two tsRNAs, while mechanism of exactly how they cooperate in LN depends on further research. GO analysis showed tRF3-Ile AAT-1 was enriched in insulin-like growth factor binding, which influenced autoimmune by modulating signaling pathways relevant to Th17/Treg balance (42). tiRNA5-Lys-CTT-1 was found involved with platelet-derived growth factor (PDGF) binding, there were studies showed that PDGF family have been concerned as an autoantigen target in the serum of SLE patients and play role in the pathogenesis of renal fibrosis (43). On the strength of above, we can hypothesize that tsRNAs may be involved in the occurrence and development of LN by participating in these pathways.

In summary, we first identified novel urine exosome derived two-tsRNAs signature in SLE patients. We then demonstrated that urine exosome derived tRF3-Ile AAT-1, tiRNA5-Lys-CTT-1 could be used as novel promising biomarkers for distinguishing LN from SLE patients. Furthermore, our study also revealed the potential biological functions of the two novel urine exosome derived tsRNAs. However, because of the limited samples and single-center experience of our study, randomized clinical trials will be needed to evaluate the possible application of the two-tsRNAs signature in the diagnosis and prognosis of LN in the future.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

The studies involving human participants were reviewed and approved by Ethics Committee of the Affiliated Drum Tower Hospital of Nanjing University Medical School. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

SC and XZ contributed equally to this work. SC and PY writing the original draft. XZ: methodology and data validation. ZL, JL: conceptualization and supervision. PY is the chief designer of the whole experiment. All authors contributed to the article and approved the submitted version.