Patients and healthy volunteers provided written consent to participate in the study, which was approved by the Local Ethics Committee of the Sechenov University (protocol No. 10–19 dated 17 July 2019). All handlings and use of material were provided according to the WMA Declaration of Helsinki on Ethical Principles for Medical Research Involving Human Subjects (the revision approved in Fortaleza, 2013). All the participants were aware of the research purpose. Informed consent was obtained from all the participants.

The study involved 75 subjects, including 51 patients having a verified diagnosis of a urologic disease: kidney calculus, kidney cyst, and kidney cancer, as well as 24 conditionally healthy subjects (control group). The inclusion criteria for the group of participants diagnosed with a urological disease were as follows:

– patient’s age ≥18 years;

The non-inclusion criteria were as follows:

– patients diagnosed with cancer of other organs and systems;

The kidney disease (KD) group consisted of patients having kidney calculus and kidney cyst. The final diagnosis in the kidney cancer group (KC) was made based on multislice computed tomography (MSCT) and histologic examination.

Triple-phase helical computed tomography of the abdomen and the peritoneal cavity (kidneys) using the bolus-tracking technique was conducted regardless of disease stage in order to differentiate between benign and malignant tumors. Scans were acquired before and after administration of a contrast agent to assess its accumulation. Contrast enhancement of the neoplasm was determined by comparing the Hounsfield scale parameters (in Hounsfield units, HU) before and after administration of the contrast agent. Changes in contrast by at least 15 HU were regarded as contrast enhancement and viewed as a functionally active tumor region (EAU, 2024a). Malignancy and tumor grade were verified by postoperative histological examination.

Cystic structures were detected and classified using the Bosniak system. The group of patients with kidney disease comprised patients with Bosniak I or II cysts as these cystic structures are benign.

The number of subjects in the kidney disease (KD) group was 25; in the kidney cancer (KC) group, 26; and in the healthy (CNTR) group, 24 (Table 2). The comparison groups were matched with respect to the number of subjects and their age, sex, and height distribution. Differences in these parameters between comparison groups were non-significant among males and females (p ⩾ 0.05).

The body mass index (BMI) parameters come under notice, which differed significantly in the KD and KC groups compared to patients in the control group. The mean BMI among all the female urologic patients was 32.1 ± 4.95 kg/m², which corresponded to overweight or class 1 obesity. Among male patients, this parameter was 26.62 ± 3.76 kg/m², mostly corresponding to normal body weight or overweight. Patients diagnosed with kidney cyst or kidney calculus had BMI = 28.3 ± 5.18 kg/m² (females) and 25.43 ± 3.86 kg/m² (males). For study subjects diagnosed with kidney cancer, BMI was 35.9 ± 4.72 kg/m² among females (corresponding to class 1 and 2 obesity) and 27.8 ± 3.65 kg/m² among males (mostly overweight and class 1 obesity). The control group consisted of conditionally healthy subjects with BMI lying in the normal range.

Blood samples in study subjects were collected strictly after fasting, at 8.00–10.00 a.m., in a clinical diagnostic laboratory. For proteomic analysis, blood samples were collected from the cubital vein into vacutainer tubes containing 3.8% sodium citrate anticoagulant (Improvacuter, Guangzhou Improve Medical Instruments Co., Ltd., Guangzhou, China). The samples were centrifuged at 3,000 rpm during 6 min at room temperature. Each plasma sample (500 µL) was collected into two dry Eppendorf polypropylene microtubes, frozen, and stored at −80°C until analysis.

Blood chemistry tests were conducted on a Torus 1,200 semi-automatic chemistry analyzer (Moscow, Russia). Reaction mixtures were prepared in accordance with the manufacturer’s instructions. The measurements were performed using the kinetic method, the endpoint method, and fixed time. Quality control was performed by measuring control “normal” (TruLab N) and “pathological” (TruLab P) sera (DiaSys Diagnostic Systems GmbH, Germany). The reference values from the application sheets are summarized in Table 3.

Preliminary preparation of blood plasma for proteomic analysis was thoroughly described in Stepanov et al. (2023). The protein fraction was precipitated with methanol (JT Baker, Landsmeer, the Netherlands). Alkylation was performed in the presence of 2% 4-vinylpyridine solution (Aldrich, Gillingham, United Kingdom) in 30% isopropanol solution (Fisher Chemical, Laughborough, United Kingdom). Trypsinolysis (∼400 ng trypsin per sample, Promega, Madison, Wisconsin, United States) involved two steps and was performed in 75 mM triethylammonium bicarbonate buffer (pH 8.2). The hydrolysates were vacuum dried before storage (Concentrator Plus, Eppendorf, Hamburg, Germany). Before mass spectroscopy measurements, the dry residue was re-dissolved using 0.1% formic acid (Acros Organics, Geel, Belgium).

Spectra were acquired on a Xevo™ G2-XS Q-TOF quadrupole time-of-flight mass spectrometer (Waters, Wilmslow, United Kingdom) in the positive ionization mode within a normal dynamic range of masses. The capillary voltage was 3 kV; the cone voltage was 67 V. The drying gas flow rate was 680 L/min; the focusing gas velocity was 50 L/min. The source temperature was set at 150°C; desolvation temperature was set at 350°C. Ions were surveyed in the data-independent acquisition mode (SONAR). The full MS scanning was performed in a range of 300–1,300 m/z with a low-energy CID fragmentation at 6 eV, followed by scanning in the SONAR mode with quadrupole mass start at m/z = 178 and quadrupole mass stop at m/z = 1,287 with a 22 Da mass window and high-energy CID fragmentation range from 15 to 37 eV. The total time for a complete scan cycle was set at 0.418 s. Mass correction using leucine–enkephalin (m/z = 556.2771, z = 1+) injected every 30 s at a concentration of 50 pg/mL was active at scanning times from 0.5 to 59 min.

Chromatographic separation was performed on an Acquity™ UPLC H-Class Plus chromatography system (Waters, United Kingdom) using a BEHC18 column (1.7 µm particle size, geometry 2.1 × 50 mm; Waters, United Kingdom) at 40°C and a flow rate of 0.3 mL/min in gradient mobile phases A (water) and B (acetonitrile) supplemented with 0.1% formic acid and 0.03% acetic acid. The following elution gradient was applied: 0–1.5 min, 3% B; up to 26.5 min, 19% B; up to 42 min, 32% B, and up to 43.5 min, 97% B; the column was then exposed to the isocratic mode up to 47.5 min and 3% B at 49 min. The post-run column equilibration lasted 6 min.

The obtained data were analyzed using the PLGS (Protein Lynx Global Server, version 3.0.3, Waters, United Kingdom) software employing the UniProt KB database with preset parameters for the SONAR/MS^E scanning mode (Stepanov et al., 2023). The data were processed using the Apex 3D mode with a lock-mass correction of leucine–enkephalin (m/z = 556.2771¹⁺) within 0.2 Da detection window. Low-energy and high-energy intensity thresholds were adjusted to 60 and 10 counts, respectively, between 0.5 and 49 min of the acquired data. Spectral data after processing were browsed against Uni Prot KB (release September 2023) protein sequences with a concatenated reverse-sequences database. Ions were surveyed with a peptide tolerance of ±25 ppm and fragment ion tolerance of ±75 ppm using at least three fragment ion matches per peptide and at least two peptide matches per protein after digestion with trypsin with a maximum of one missed cleavage. S-pyridylethylation was applied as a fixed type of modification, while S, T and Y phosphorylation (SEP, TPO and TPR) was applied as a variable modification. Peptide and protein identifications were percolated using the 2% false discovery rate. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE (Perez-Riverol et al., 2021) partner repository with the dataset identifier PXD051799 “Kidney disease proteome research” (Reviewer account details: Username: reviewer_pxd051799@ebi.ac.uk; Password: R858MIv3).

PSMs detected with PLGS were processed with two steps:

1. Filtering: only samples with more than 70 unique proteins (by UniProt ID) identified in sample were chosen for next steps;

Differentially expressed proteins were identified for a sample (N = 57) filtered according to the following criteria:

1. Protein is found in ≥50% samples within a group;

The fold change value was calculated as follows:

If C c t r l > C t e s t , then f C h = C t e s t C c t r l ; else f C h = C c t r l C t e s t ∗ − 1 , where C_cntr is the median protein concentration in the control group; C_test is the median protein concentration in the comparison group; fCh is the fold change.

The p-value was determined by Mann–Whitney U rank test (Mann and Whitney, 1947) in the scipy. stats package (SciPy, 2024b). False discovery control was performed employing the Benjamini–Hochberg procedure (Benjamini and Hochberg, 1995) also using the scipy. stats package (SciPy, 2024a).

Principal component analysis (PCA) and Sparse partial least-squares discriminant analysis (sPLS-DA) with 0.95 ellipse confidence were conducted using the MixOmics software package for R programming language (mixOmics, 2024) based on the examples of using these methods (sPLSDA SRBCT Case Study, 2024; mixOmics Vignette, 2024) for a group of proteins (N = 104) that were identified in no less than 20 samples.

The acceptance criteria for protein identification were based on the Human Proteome Project Mass Spectrometry Data Interpretation Guidelines 3.0 (Malsagova et al., 2024): a candidate feature in the proteomic data had to meet the unicity criterion, meaning that certain proteins had to be covered by more than one unique protein-specific peptide without interference from any other protein.

In all study subjects, blood chemistry parameters were either normal or lay within the lower limit of the normal range. Slight decline in albumin level as well as reduced creatinine level and alanine aminotransferase activity were observed in the group of patients with kidney cancer. Blood glucose, triglyceride, and albumin levels were either normal or slightly decreased in the group of conditionally healthy subjects (Table 4).

For all thee selected parameters, the resulting values lie within the normal range. However, triglyceride and glucose levels in the KD and KC groups are significantly elevated compared to the CNTR group. The normal ranges of blood chemistry parameters in the KC group are possibly caused by the fact that in almost all the cases, there was a malignant tumor structure sized ≤7 cm, correlating with stage T1 according to the UICC TNM Classification of Malignant Tumors 2017, eighth edition (Brierley et al., 2017).

A total of 644 proteins were identified in the analyzed samples; 163 of them were shared among comparison groups (Figure 1A). Only few proteins were specific to groups of patients with kidney pathologies KC or KD. For this reason, no analysis of group-specific proteins was conducted.

The identified proteins shared by the analyzed comparison groups are expectedly involved in important biological processes (Figure 1B): classical antibody-mediated complement activation (R-HSA-173623, n = 46); FCGR activation (R-HSA-2029481, n = 45); binding and uptake of ligands by scavenger receptors (R-HSA-2173782, n = 54). A full list of shared proteins is provided in Supplementary Table S1.

Table 5 summarizes proteins whose blood level (fold change) differs for the selected comparison groups (KD and KC) with respect to control. The blood levels of proteins involved in immune response are simultaneously elevated in two comparison groups (kidney disease (KD) and kidney cancer (KC)) with respect to the control group. Thus, the blood level of complement factor B (P00751) in patients with kidney cancer increased 1.7-fold (р = 0.004); similar elevation was observed in patients in the kidney disease group (1.4-fold; p = 0.13). Furthermore, the blood levels of such proteins as alpha-1-acid glycoprotein 2 (P19652) (р = 0.02) and ceruloplasmin (P00450) (р = 0.04) in patients in the kidney disease and kidney cancer groups were also simultaneously increased 1.4-fold and 1.3-fold, respectively. The maximum change in protein level was observed for hemoglobin subunit beta (P68871): its level increased more than twofold (p = 0.08) in the KD group and more than fourfold, in the KC group (р = 0.0008).

The comparison groups could not be segmented by principal component analysis (PCA): the proteomic profiles were found to densely overlap (Figure 2A). For achieving discrimination, we performed selection and classification of descriptors using sparse partial least-squares discriminant analysis (sPLS-DA) (Figure 2B). The 12.5% and 15% variance in the consolidated proteome was interpreted by discriminant analysis (Figure 2B). Figure 2B demonstrates that the proteomic profiles corresponding to the control group differ to the maximum extent from the group of patients having a pathology. Meanwhile, the profiles of the KD and KC groups overlap as indicated by the large area of proteomic profile overlaps. Figure c shows that in general, the analysis provides a fair result of data segmentation into comparison groups (70%–80%).

In all three images in Figure 2A–C, the values referring to the control group, the KD group, and the KC group are shown in red, blue, and green, respectively. The absolute weights of the sparse loading vectors for each sPLS-DA dimension in the data set are presented in Supplementary Table S2; Figure 3 shows proteins with the greatest weights, including those whose level varied significantly.

Proteins making the greatest contribution to segmentation of comparison groups can either be classified into the four key biological processes, including lipid transport and metabolism, complement system activation, iron and heme transport, and implementation of adaptive immunity or are acute phase proteins (Figure 3).

Post-translational modifications (PTMs) often affect protein function (Nikolsky et al., 2023). In this study, we analyzed the possible phosphorylated proteoforms, one of the most common native protein modifications. Table 6 lists the results of searching for phosphorylated proteoforms, including those carrying phosphorylated serine (SEP), phosphorylated threonine (TPO) and phosphorylated tyrosine (PTR).

Table 6 shows that the phosphorylated proteoform of apolipoprotein B-100 was observed in eight blood biosamples in the KC and KD groups. Such proteoforms as complement factor B (n = 9), prothrombin (n = 6), and transthyretin (n = 5) are also more common in groups with a kidney pathology compared to the control group. Observation of these phosphorylated proteoforms in the investigated blood samples is interesting in terms of their abundance in the comparison groups. Meanwhile, these results do not provide information enabling precise classification of each group of patients with a pathology (KC and KD): occurrence of PTM-containing proteoforms is approximately identical in both groups, being indicative of the synergistic pathogenesis of the selected groups of urologic diseases (kidney calculus, kidney cyst, and kidney cancer). The results also suggest that this study needs to be continued using a significantly larger number of study subjects (patients having a kidney pathology (KC, KD) and control group (CNTR).

Analysis of the location of a modifying moiety with respect to the interface of binding to native partner proteins (the Interface column) helps assess the potential effect of a PTM on protein structure and functions. Table 7 lists the data on the topology and geometrical parameters of the identified proteoforms. Experimentally determined 3D structures in Protein data bank, including those forming complexes with partner proteins, were identified for four proteins listed in Table 7. For these proteins, it was found that the region carrying modified amino acid residues is located on the protein globule surface and is accessible for interacting with a partner protein. In three cases, the regions carrying modified amino acid residues are located directly at the interface of binding to the partner protein or near the binding site (see Table 7, K_SASA ≤ 1). The K_SASA parameter characterizes changes in the solvent-accessible surface area for a peptide in the protein chain and the protein–protein complex. This parameter indirectly indicates the location of peptide (in our case, carrying a modifying moiety) with respect to the interface of binding to the partner protein.

The type of structural motif of the protein comprising the peptide carrying a modifying moiety was identified in each case. Thus, for CFB complexes, the partner of complement factor C4a (PDB ID 2XWJ) modified peptide is located in the dense β-hairpin of the beta-containing ribbon functional domain (Figure 4A). In proteins of different organisms, this domain type is involved in formation of protein–protein and protein–nucleic acid complexes (e.g., in zinc finger proteins, where the ribbon domain is stabilized by coordination bonds with a zinc cation) (Li et al., 2022). The K_SASA parameter of the CFB/C4a complexes is 0.6, indicating that the revealed CFB peptide carrying a modifying moiety is engaged in organization of the binding interface of the complex. Other members involved in activation of the complement cascade C4a/C4b also form complexes with partner proteins (PDB ID 6YSQ, Figure 4B). The identified modified peptides in the complexes are a part of β-hairpins of the sandwich functional domain. The sandwich functional domain in proteins is also an essential structural participant of protein complex formation (Nikolsky et al., 2024). In our study, the identified peptides carrying a modifying moiety also resided directly in the binding interface (K_SASA = 0.5) of two complement factors, C4a and C4b. Figure 4C illustrates the prothrombin protein in complex with coagulation factors Va and Xa; the region carrying modified amino acid residues is located in an unstructured domain residing on the protein globule surface. However, the modification site in this complex is remote from the protein binding interface (K_SASA = 1).

No full-atom models are available for three proteins (APOB, TTR, and NRAP) in PDB. The lack of structural data on these proteins in the PDB and AlphaFold databases makes it difficult to map the phosphorylation site with respect to the interface of binding to the partner protein. However, potential partners (presented in Table 7) were identified for the remaining proteins.

Structural analysis of proteins associated with development of diseases (primarily, cancer) is interesting for researchers (e.g., a recent study of the occurrence of α-helix- and β-strand-carrying structural domains in proteins encoded by over- and underexpressed genes) for 21 cancer types. The authors called this study “pan-cancer structurome” and demonstrated an increased amount of β-sandwich domains and insufficient amount of α-helices in proteins (Medvedev et al., 2023).

The complement system is the key part of the immunity. Disbalancing of activation regulation of this system may cause inflammation and further development of different diseases (Lin et al., 2018). Patients with various tumor pathologies have increased expression of some complement factors in the classical complement pathway, which contribute to immune surveillance through cytotoxicity and lysis of tumor cells (Afshar-Kharghan, 2017). However, overexpression of the membrane and soluble forms of inhibitors of complement components was observed in the case of solid tumors, protecting the tumor against complement-mediated lysis (Reese et al., 2020). Interestingly, our study revealed increased levels of proteins of the C3/C5 axis, namely, CFB (FC = 1.7, p = 0.004), and C3 (FC = 1.3, uncertain value), C4B (FC = 1.5, uncertain value), Clu (FC = 1.4, p = 0.065), which is consistent with the literature data on prognostic significance of these factors in human kidney cancer (Xi et al., 2016; Netti et al., 2021; Reese et al., 2020). Proteins CFB, С3, C4A, and C4B are associated with the development of kidney diseases (see Supplementary Table S5). Complement activation may induce or promote glomerular injury in multiple kidney diseases (Vivarelli et al., 2024) (Figure 5). However, conspicuous is the fact that the role of the axis in oncogenesis is rather ambiguous, since a number of studies have demonstrated that tumor grows under C3/C5 activation conditions through recruitment and activation of myeloid suppressor cells in tumors (Reis et al., 2018; Pio et al., 2013; Macor et al., 2018; Berraondo et al., 2016). Acute-phase protein alpha-2-macroglobulin (A2M) is another circulating component potentially associated with the C3/C5 axis of complement activation (the classical pathway). It is a multi-functional high-molecular-weight homotetrameric glycoprotein interacting with proteinases, which induces conformational changes in the monomer molecules and modulates their function (Arimura and Funabiki, 2021; Lagrange et al., 2022). The inhibitory mechanism of A2M involves formation of a tetrameric structure around active proteases and physically hampers the interaction between proteases and substrates (the so-called snap-trap mechanism) (Vandooren and Itoh, 2021a). There are parallels between functioning of C3 and C4 factors and A2M. The pivotal element of the system is the cleavage of C3 and C4 factors followed by exposure of internal thioester binding neighboring glycoproteins (Vandooren and Itoh, 2021a). Similar to C3 and C4, A2M undergoes proteolysis and conformational changes to expose internal thioester that binds and entraps active protease (Vandooren and Itoh, 2021b). Along with C3 and C4, A2M is classified as a member of the thioester-containing protein (TEP) family (Vandooren and Itoh, 2021a). All these factors are eventually involved in formation of the membrane attack complex (MAC), which leads to cell lysis and forces cells release inflammatory cytokines. MAC formation is regulated (namely, inhibited) by clusterin protein, which is essential for protecting own cells against being attacked by the MAC during complement activation (Lin et al., 2018).

Protein serine proteinase complement factor B (CFB) is involved in complement activation (Riihilä et al., 2019). CFB binds to C3b and forms the C3bB complex, participates in formation of C3 convertase and further C5 convertase, acting as an activator of the complement system via the alternative pathway (Figure 5). Dysregulation of these processes may lead to hypercoagulation and thrombosis development. Hyperactivation of the alternative complement pathway with involvement of CFB results in the development of C3 glomerulopathy, which leads to accumulation of the C3 component and glomerular injury. Vitronectin (VTN, FC = 1.3, uncertain value), whose expression is upregulated during complement activation (Ruiz-Molina et al., 2023) is a defense protein and an inhibitor of complement activation via the alternative pathway (Figure 5).

The proteolytic cascades (the immune system, the complement and the coagulation system) are finely mutually balanced by a network of regulatory factors. The complement–coagulation crosstalk can influence activation, amplification and regulatory functions in both systems (Heurich and McCluskey, 2023). In our study, we noted that proteins of the coagulation system (fibrinogen chain A, B and G (FGA, FGB, FGG, respectively)), as well as phosphorylated prothrombin (F2), contribute to segmentation of the comparison groups. Figure 5 shows the potential mechanism of the crosstalk between the coagulation system and complement factors, the so-called “coagulo-complementome”. Prothrombin and thrombin-activatable fibrinolysis inhibitor (TAFI) mediate this interaction (Danckwardt et al., 2013). Blood clotting factors play an important role in cancer biology (Bauer et al., 2022). Coagulation abnormalities are common among cancer patients who have an increased procoagulant activity due to the release of tumor-derived procoagulants into the bloodstream (Danckwardt et al., 2013). Elevated plasma levels of fibrinogen in patients with renal cell carcinoma were reported in the literature (Selby et al., 2018; Chen et al., 2024; Tupikowski et al., 2023; Wang et al., 2020).

Plasma protease C1 inhibitor (SerpinG1) belongs to the family of serine protease inhibitors and is an important regulator of the complement, coagulation, and fibrinolytic systems. The blood levels of the protease/SerpinG1 complexes (C1/SerpinG1, Thrombin/SerpinG1) are proportional to the level of activation of these enzyme cascades in blood plasma (Kajdácsi et al., 2020) (Figure 5). In our study, we observed that the SerpinG1 level in renal patients was moderately increased (FC = 1.5, p = 0.003). The “coagulo-complementome” factors show great promise as therapeutic targets and diagnostic factors (Danckwardt et al., 2013).