**3. Discussion**

In the present study a wide variety of proteins was identified and quantitated using two mass-spectrometric approaches in urine samples of patients with FSGS and MCD. No significant differences were found between the proteomic profiles of patients with MCD and a general FSGS group. However, we found that the differences in the urinary protein profiles in FSGS patients were highly dependent on the severity of the disease—thus the patients were subdivided into those with mild FSGS—normal kidney function and steroid-sensitive NS, and those characterized by severe proteinuria, impaired renal function and steroid-resistant NS. These two groups showed significant differences in the levels of 68 proteins. The group with a severe progressive FSGS showed a number of abundant kidney-derived proteins in urine.

Apolipoproteins, vitronectin, hemopexin and gelsolin reflect the process of active damage to podocytes [24–26,32,34,38–40,46–51]. Apolipoprotein A-1 (ApoA-1), a small 28 kDa high-density lipoprotein (HDL) component, is considered as a putative permeability factor [32,34,50]. In a study by Puig-Gay N. et al. an increase in ApoA-1b in the urine was noted immediately before the onset of proteinuria in patients with FSGS recurrence in the transplant [32]. This protein is absent in the urine of healthy individuals and in most patients with glomerular proteinuria caused by other glomerulopathies [34,50]. Our data also suggests a potential role of ApoA in the pathogenesis of primary FSGS. In terms of response to therapy, our data is consistent with that of Kalantari et al. who differentiated steroid-sensitive and steroid-resistant patients with a confirmed FSGS by urine proteome analysis. Among 21 proteins, ApoA-1 was one of the most significant marker between steroid-sensitive and steroid-resistant forms. An increase in ApoA-1 is associated with hyperlipidemia and low-density lipiprotein oxidation [51].

Like Kalantari et al. and other groups of researchers, we established the role of ApoA-1 and some other urinary proteins in diagnosis of FSGS [32,51,52], however, we took a different approach in the study and introduced a specific index that allows us to evaluate not only the response to steroid therapy, but also and the level of proteinuria and kidney function. This approach has been more effective in separating FSGS patients by urinary proteome to two groups. In addition, the group of patients with mild FSGS is comparable to those with MCD, while the proteomic profile of the FSGS group 2 is significantly different, and apparently, this model can be used to assess the severity of the disease.

Among the biomarkers of podocyte damage, we found an increase in the level of hemopexin, which is a glycoprotein with the highest affinity for hem [24]. Hemopexin is currently considered as one of the possible circulating factors of idiopathic FSGS. Hemopexin binds free methemoglobin, further recognized by CD91 on hepatocytes or macrophages in the spleen, liver, and bone marrow [25]. Cell culture studies have shown that hemopexin can induce the redistribution of the actin cytoskeleton in podocytes and development of proteinuria [24–26].

An increase in the urinary excretion of vitronectin in FSGS patients is also of interest, since vitronectin activates β3-integrins, molecules that ensure the fixation of podocytes to the GBM. In case of FSGS, the loss of vitronectin may be associated with the process of podocyte detachment from the glomerular basement membrane [38]. Elevated serum gelsolin aggravates the development of proteinuria and renal dysfunction by F-actin rearrangement, foot process effacement and cell movement [39,40]. These markers of podocyte dysfunction were elevated only in severe steroid-resistant FSGS.

In group 1, patients with mild FSGS, other factors of podocyte dysfunction are detected in the urine—galectin-3—binding protein. It is suggested that this protein is elevated in the kidney tissue in MCD and in the urine in lupus nephritis, and it appears to have a pro-inflammatory function [44,45].

Compared to other glomerulopathies, complement activation processes are not well understood in FSGS. In the second FSGS group we found a significant increase in urinary excretion of complement components C4b, C9, as well as factor B, I and a decrease in CD59, an inhibitor of the membrane attack complex, which indicates the possible role of complement activation associated with stronger damage. Our data is consistent with the results of a study by Huang J. et al., who showed the possibility of systemic complement activation in FSGS patients with increased levels of C3a, C5a, and C5b-9 in blood plasma and urine [43]. Activation of the alternative complement pathway in FSGS may be associated with a poor prognosis [30,31]. On the other hand, an increase in the level of some complement components in the urine may be the result of loss due to severe glomerular filter damage. For a more accurate asessement of complement system disorders in FSGS, the study these components in the blood of these patients is required.

The intensity of accumulation of ECM components is reflected by the excretion of fetuin A, β2 microglobulin, and immunoglobulin chains in the urine of patients with FSGS [37]. The same changes in the protein profiles indicating the damaging processes on tubular cells and accumulation of ECM in the interstitium can be noticed in patients with FSGS with impaired renal function resistant to steroid therapy in our study. An equally important component of disease progression and lack of response to steroid therapy is tubulointerstitial fibrosis and tubular damage in FSGS, which are reflected by an increase in the level in the urine of lumican and cystatin C. These processes are also evidenced by an increase in vitamin D-binding protein and retinol-binding protein 4 in the urine of FSGS patients with impaired renal function [18–21,37]. Simultaneously with the processes of interstitial inflammation, the mechanisms of accumulation of ECM components are activated [18–20]. Our data confirms the results of experimental studies [23,27–29,52]. A study of urinary proteins dynamic changes was conducted on a focal segmental glomerulosclerosis rat model (adriamycin-induced nephropathy) and showed that levels of Fetuin-A and alpha 1 microglobulin may be promising markers for early detection of FSGS. Thus, some proteins or their combinations can change with the disease progression [52]. Interalpha-trypsin inhibitor heavy chain H1 and H2 activate the CD44+ parietal epithelial cells, the main profibrogenic cells in the glomeruli, and thus are a powerful stimulus for glomerulosclerotic processes [27–29].

Many urine proteins are of serum origin and enter the urine through the damaged glomerular filter. However, these proteins can cause additional damage and pathology progression. For example, components of the complement system or plasminogen. The conversion of plasminogen to plasmin in urine can activate the epithelial sodium channels and cause sodium retention in the renal tubules—one of the mechanisms of renal edema [23]. Interestingly, in case of MCD and mild FSGS with a favorable prognosis, some of the proteins found in the urine seem to be of a protective nature, such as for example plasma protease C1 [35].

Perez et al. analyzed urinary peptide profiles using magnetic bead-based technology, combined with MALDI-TOF mass spectrometry, in 44 patients diagnosed with MCD and FSGS. In this work the low molecular weight fraction of urine, containing peptides of proteins were analyzed, while in our work the high-molecular protein fraction was isolated. The authors showed that FSGS patients had higher levels of uromodulin fragments and lower concentrations of fragments of A1AT [53]. In 2017 Perez V. et al. also ran a study that included 24 patients with MCD and 25 patients with FSGS and analyzed their urine proteome using two-dimensional gel electrophoresis in combination with MS to detect urinary biomarkers capable of differentiating MCD and FSGS. They showed that urine concentrations of alpha-1 antitrypsin, transferrin, histatin-3, and 39S ribosomal protein L17 were decreased in the FSGS group, and the calretinin level was increased as compared to the MCD group [54]. We did not find any significant differences between the MCD and the general FSGS groups. However, we found an increase in alpha-1 antitrypsin, and other proteins in patients with severe FSGS as compared with those with MCD and mild FSGS, who showed similar profiles of urine proteins.

Thus, we have identified a wide range of proteins that differ in patients with mild course of steroid-sensitive FSGS/MCD and FSGS patients with a progressive steroidresistant NS. Proteins excreted in the urine reflect damage to several parts of the nephron podocytes, tubulo-interstitium, accumulation of ECM, complement activation. Patients

with MCD and FSGS with steroid-sensitivity and a favorable course of progress showed similar urine protein profiles, while severe progressive FSGS with steroid-resistant nephrotic syndrome were characterized by the early activation of the complement system and profibrogenic mechanisms and accumulation of extracellular matrix components at the onset of the disease.
