Peptidomic analysis
This exploratory urinary peptidomic analysis resulted in a list of 329 sequenced peptides to be further investigated. Statistical assessment including Benjamini-Hochberg FDR adjustment yielded a list of 70 statistically significant peptides (adjusted p -value < 0.05) affected by the GLP-1R agonist treatment, detailed information of the peptide identifications is provided in Table S1. The distribution of peptide intensity of these 70 statistically significant peptides (red spots) amongst the 329 peptides was also examined (Figure 2A). Volcano plot assessing the regulation of all the 329 peptides in response to the treatment (Figure 2B), highlighted the downregulation of urinary peptides in majority of the statistically significant peptides (66 out of 70 peptides), the vast majority in downregulation could also be visualized on comparison of the urinary peptide profiles from pre-treatment (Figure 2C) and post-treatment (Figure 2D), as obtained from the CE-MS analysis.
The 70 statistically significant peptides were identified as fragments of 26 proteins, as shown in Table 2. Most of the peptides (59 out of 70) originated from the collagen family of proteins and 41 out of the 59 collagen peptides belonged in majority to three collagen proteins, namely, collagen alpha-1(III) chain (P02461; COL3A1; n =16), collagen alpha-1(I) chain (P02452; COL1A1; n =15) and collagen alpha-2(I) chain (P08123; COL1A2; n =10). All the remaining identified 11 (out of 70) urinary peptides came from different proteins, alpha-1-antitrypsin (P01009; SERPINA1), apolipoprotein C-III (P02656; APOC3), CD99 antigen (P14209; CD99), cleavage and polyadenylation specificity factor subunit 6 (Q16630; CPSF6), Cornulin (Q9UBG3; CRNN), corticosteroid-binding globulin (P08185; SERPINA6), hemoglobin subunit alpha (P69905; HBA1; HBA2), myoglobin (P02144; MB), neurosecretory protein VGF (O15240; VGF), polymeric immunoglobulin receptor (P01833; PIGR) and transthyretin (P02766; TTR). Four out of the 70 urinary peptides showing a statistically significant upregulation on treatment with GLP-1R agonists in T2DM patients, as observed in Figure 2B, were characterized as COL3A1 (n =3) and COL1A2 (n =1) peptides (Table 2).
Since several peptides from COL1A1 and COL3A1 were found significantly associated with treatment (Figure 2E and 2F, respectively), we investigated the alignment of the peptides in their protein structures. The identified peptide sequences were aligned with the primary structure of the proteins, as shown in Figure 2G and Figure 2H, respectively for COL1A1 and COL3A1. For both proteins, peptides appeared equally distributed and no specific hot spot became apparent.