Peritoneal dialysis (PD) frequently contributes to peritoneal damage which cannot be easily identified without invasive techniques, implying the urgent need for biomarkers and revealing mechanisms. Chronic glomerulonephritis (CGN) is one of the leading causes of receiving dialysis treatment. Here, we attempted to analyze the peritoneal dialysate collected from CGN patients when they receive continuous ambulatory peritoneal dialysis (CAPD) treatment for the first time and after a year to reveal the protein changes that resulted from PD. Proteins were displayed by two-dimensional gel electrophoresis (2DE). Altered gel spots were digested followed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis for protein identification. Eight proteins were found to have differential expression levels between two groups. Their differential expressions were validated by Western blots in other sets of peritoneal dialysates. Proteins identified with higher levels in the first-time dialysate suggested their dominant appearance in CGN patients, while those that showed higher levels in peritoneal dialysate collected after one year may result from initial peritoneal inflammation or changes in the permeability of the peritoneum to middle-sized proteins. All the identified proteins may provide a perceptiveness of peritoneal changes caused by PD and may function as potential biomarkers or drug targets.
Peritoneal dialysis (PD) is a widely used method of renal replacement therapy which also can lead to pathologic damage of the peritoneum and patient fatality. One potential mechanism is glucotoxicity to the peritoneal membrane [
In recent decades, analyses of peritoneal dialysate by proteomic approaches have opened up progresses in this field. A descriptive study was performed on the dialysate of nine paediatric PD children patients by SDS-PAGE [
Chronic glomerulonephritis (CGN), nephroangiosclerosis, and diabetes were identified as the most common causes of chronic renal failure [
Ten CGN patients were recruited in this study. Their peritoneal dialysates were collected first time at their early stage of receiving maintenance dialysis treatment (E1–10). After one year of CAPD treatment, their peritoneal dialysates were collected again to present the middle stage of treatment (M1–10). The clinical data of these patients were provided in Supplementary Table
Peritoneal dialysates from seven of the ten patients were used for protein profiling by two-dimensional gel electrophoresis (2DE) analysis, while thoes from the other three patients were used for further validation of the altered proteins by Western blotting analysis. Peritoneal dialysates collected from two stages, E1–E7 and M1–M7 (each containing 140
Protein samples were loaded onto immobilized pH gradient (IPG) gel strips for the first-dimensional gel electrophoresis. To minimize evaporation and urea crystallization, IPG cover oil (0.8 mL) was applied on top of each gel strip. Isoelectric focusing (IEF) was run following a stepwise incremental voltage program on an IPGphor Isoelectric Focusing System (GE Healthcare Life Sciences, Taipei, Taiwan). Samples were run under the following settings: 30 V for 16 h, 500 V for 1 h, 500 V for 1 h, and 8000 V for 4 h. After IEF, the strips were first placed in equilibration buffer with 1% dithiothreitol w/v for 15 min followed by in 2.5% w/v iodoacetamide for another 15 min. The IPG strip was placed on a 1.0 mm thick 12% polyacrylamide gel and was run at 300 V for 4-5 h for the second-dimensional gel electrophoresis. The visualization of the gels was achieved by using silver staining. The stained gels were scanned at a resolution of 300 dpi using an ImageScanner operated by LabScan 3.00 software (GE Healthcare Life Sciences). Image spots were detected and matched by using ImageMaster 2D software (GE Healthcare). The spot detection was manually checked. Each spot intensity was normalized by the total intensity volume of all spots detected on the gel. To identify the differentially expressed protein spots during the CAPD treatment, spot intensities of the sample from middle-stage groups were compared to those from the early-stage groups. Ratios were statistically analyzed using Student’s
The protein spots were washed with 25 mM ammonium bicarbonate and 50% acetonitrile/25 mM ammonium bicarbonate. After being washed three more times with 25 mM ammonium bicarbonate and 50% acetonitrile/25 mM ammonium bicarbonate, the gel fragments were placed at 56°C for 1 h in a solution containing 10 mM dithiothreitol, 55 mM iodoacetamide, and 25 mM ammonium bicarbonate for protein reduction and alkylation. Gel fragments were added with 10
Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used for protein identification. Tryptic peptides were fractionated in a C18 microcapillary column (75
Samples, each contained 25
To reveal the protein profile change due to CAPD treatment, 2DE was used in this study. Peritoneal dialysates were collected from seven CGN patients at the beginning of receiving CAPD treatment, which was used to present the early stage of peritoneal dialysate sample (E1–E7). These patients were treated with CAPD, and the middle-stage peritoneal dialysate samples were collected after a year. Each sample was analyzed individually and repeated three times.
To evaluate the reproducibility of the 2DE gels, twenty-one gels were subjected to image analysis. We found that approximately 95% of the spots can be detected within each triplicate experiment. Since the discrepancy of the gel images was determined by the reproducibility of the presence of each protein spots, here we counted the number of gels in which each spot appeared to evaluate the variance within each sample. Owing to the high similarity between the triplicate gels, a submaster gel can be generated from each patient sample at two different states. By combining all submaster gels (14 images), a master gel containing 368 spots was obtained. Among them, 163 (44.3%), 101 (27.4%), and 63 (17.1%) spots can be detected in 7, 6, and 5 gels of the early-stage gels (E1–E7), respectively, consisting of 88.8% of all the datable spots (Figure
Evaluation of the individual difference within each of the 7 peritoneal dialysates collected from (a) early stage and (b) middle stage. A total of 386 spots were detected in the master gel of those 14 gels. Among the E1–E7 gels, 163 (44.3%), 101 (27.4%), 63 (17.1%), 14 (3.8%), 7 (1.9%), 11 (2.9%), 5 (1.4%), and 4 (1.1%) spots were detected in 7, 6, 5, 4, 3, 2, 1, and 0 of the early-stage gels, respectively. Among the M1–M7 gels, 189 (51.4%), 96 (26.1%), 44 (12.0%), 11 (3.0%), 10 (2.7%), 6 (1.6%), 5 (0.8%), and 9 (2.4%) spots were detected in 7, 6, 5, 4, 3, 2, 1, and 0 of the middle-stage gels, respectively.
After evaluation of the similarity between samples, E1–E7 and M1–M7 (each with triplication) consisting of 42 gels were comparatively analyzed to identify CAPD induced protein change. Gel spots were analyzed by image analysis software for spot detection, spot matching, intensity normalization, and result output. Figure
Representative 2DE gels of the (a) early-stage and (b) middle-stage samples. Each sample containing 140
To identify these 10 altered protein spots, P01–P10 gel spots were excised, enzymatically digested, and subjected to LC-MS/MS analysis. Each spot was identified by searching the raw data against human protein database. Detailed information about each spot was listed in Table
Identification of 10 altered spots, of which 4 showed higher levels in the early-stage PD samples, while 6 had higher levels in the middle stage.
Spot numbera | Protein name | Swiss-Prot accession no. | Mr (E/T)b (kDa) | pI (E/T)c | No. of identified peptides | Sequence coverage (%) | Mascot score | Fold change ± SDd | Alteration (M/E) |
---|---|---|---|---|---|---|---|---|---|
P01 | Ig mu chain C region | P01871 | 68/49 | 6.5/6.4 | 5 | 10.2 | 215 |
|
Down |
P02 | Fibrinogen gamma chain | P02679 | 45/51 | 5.7/5.4 | 6 | 11.8 | 258 |
|
Down |
P03 | C-reactive protein | P02741 | 25/25 | 5.5/5.5 | 4 | 16.2 | 145 |
|
Down |
P04 | C-reactive protein | P02741 | 25/25 | 5.4/5.5 | 5 | 20.1 | 156 |
|
Down |
P05 | Ig delta chain C region | P01880 | 70/47 | 6.2/6.8 | 3 | 6.3 | 124 |
|
Up |
P06 | Alpha-1-antitrypsin | P01009 | 46/47 | 4.7/5.4 | 6 | 12.8 | 301 |
|
Up |
P07 | Histidine-rich glycoprotein | P04196 | 53/59 | 5.6/7.0 | 4 | 6.7 | 186 |
|
Up |
P08 | Histidine-rich glycoprotein | P04196 | 53/59 | 5.7/7.0 | 5 | 8.4 | 211 |
|
Up |
P09 | Apolipoprotein A-I | P02647 | 25/31 | 5.3/5.6 | 4 | 12.9 | 201 |
|
Up |
P10 | Serum amyloid P-component | P02743 | 27/25 | 5.7/6.1 | 6 | 24.5 | 298 |
|
Up |
aThe spot numbers are designated in Figures
bMr (E/T): experimental observation of apparent molecular weight in 2-DE/theoretical molecular weight calculated from protein sequence database.
cpI (E/T): experimental observation of pI in 2-DE/theoretical pI calculated from protein sequence database.
dFold change and standard deviation calculated from the protein spot intensity in middle-stage samples (M) versus that in early-stage samples (E).
Gene ontology analysis of 8 differentially expressed proteins.
Spot numbera | Protein name | Subcellular location | Biological process | Molecular function |
---|---|---|---|---|
P01 | Ig mu chain C region | Plasma membrane | Immune response | Antigen binding |
| ||||
P02 | Fibrinogen gamma chain | Extracellular space | Platelet activation, platelet degranulation, protein polymerization, response to calcium ion, and signal transduction | Cell surface binding and metal ion binding |
| ||||
P03/P04 | C-reactive protein | Extracellular space | Acute-phase response, complement activation, regulation of lipid storage, regulation of macrophage, opsonization, protein polymerization, response to hypoxia, and response to lead ion | Cell surface binding, cholesterol binding, choline binding, low-density lipoprotein particle binding, and metal ion binding |
| ||||
P05 | Ig delta chain C region | Extracellular space | Immune response | Antigen binding |
| ||||
P06 | Alpha-1-antitrypsin | Extracellular space | Acute-phase response, platelet activation, platelet degranulation, and regulation of proteolysis | Serine-type endopeptidase inhibitor activity |
| ||||
P07/P08 | Histidine-rich glycoprotein | Extracellular space | Angiogenesis, chemotaxis, regulation of angiogenesis, regulation of cell adhesion, regulation of cell growth and proliferation, regulation of fibrinolysis, platelet activation and degranulation, regulation of immune, response to tumor cell, and regulation of gene expression | Cell surface binding, cysteine-type endopeptidase inhibitor activity, heme binding, heparan sulfate proteoglycan binding, heparin binding, immunoglobulin binding, metal ion binding, serine-type endopeptidase inhibitor activity, and zinc ion binding |
| ||||
P09 | Apolipoprotein A-I | Extracellular space | Cholesterol metabolism, lipid metabolism, lipid transport, steroid metabolism, sterol metabolism, and transport | Cell surface binding, heme binding, heparin binding, immunoglobulin binding, metal ion binding, and zinc ion binding |
| ||||
P10 | Serum amyloid P-component | Extracellular space | Acute-phase response and protein folding | Metal ion binding and unfolded protein binding |
Proteins rarely act alone. Most of the time, they team up to carry out biological functions. Mapping of protein-protein interaction can help unravel the complicated molecular relationships in living systems. Here, in order to elucidate the main molecular interactions and biological connections, a protein-protein interaction network was constructed based on the 8 altered proteins by MetaCore (GeneGo, Inc., St Joseph, MI, USA). (illustrated in Figure
Interaction network analysis of 8 proteins that showed altered levels between early- and middle-stage groups. Proteins identified in this study, alpha-1-antitrypsin, serum amyloid P-component (APCS), C-reactive protein (CRP), fibrinogen gamma, apolipoprotein A-I (APOA1), histidine-rich glycoprotein (HPRG), Ig delta chain C region (IGHD), and Ig mu chain C region (IGHM), were marked with a circle. Full protein names for the abbreviation and legend for molecules used in MetaCore networks are provided in Supplementary Table B and Supplementary Figure A.
To further confirm the results of 2DE analysis, Western blotting was used to evaluate the expression levels of altered proteins in peritoneal dialysates from three other CGN patients (E8–E10 versus M8–M10). The fold change in the expression of each identified protein was obtained by dividing the mean band density of middle-stage sample by that of the early-stage samples (M/E). The expression levels of the three proteins, which were considered with higher levels in early-stage samples by 2DE, were confirmed by using Western blotting in three separate samples. As shown in Figure
Western blotting of the identified proteins in peritoneal dialysates collected from other three CGN patients at their first time (E08–E10) and after a year (M08–M10) of receiving CAPD treatment to further validate these protein candidates. The mean band density of E08–E10 and M08–M10 was calculated and used to obtain the fold change between the two groups (M/E). The fold is indicated in the bracket to the right of each set of bands. Western blotting confirmed that (a) Ig mu chain C region, fibrinogen gamma, and C-reactive protein show higher expression levels in the early-stage samples than in the middle-stage samples, while (b) Ig delta chain C region, alpha-1-antitrypsin, histidine-rich glycoprotein, apolipoprotein A-I, and serum amyloid P-component were observed to have higher levels in the dialysate after one year of CAPD treatment.
In order to comprehensively realize the protein group involved, a protein network analysis of the 8 altered proteins was done to deduce their interacting proteins. Six of the 8 identified proteins can be linked through protein-protein interaction, suggesting their close association in the system. Some interesting proteins were also recruited in the network and were found to be related to the process of inflammation or immune response. ABCA1 deficiency can lead to impaired cholesterol efflux to HDL or apolipoprotein A-I (APOA1), decreased apoE secretion, and increased secretion of inflammatory cytokines and chemokines [
We speculated that proteins identified with higher levels in the first-time dialysate suggested their prevailing expression in CGN patients. Ig mu chain C region (P01) was identified with higher levels in the initial peritoneal dialysate. According to a previous work, patients with primary glomerulonephritis showed IgM as dominant molecules in glomerular by immunofluorescence [
Proteins that have higher levels in peritoneal dialysate after one year of CAPD treatment may shed light on the mechanism or potential marker for initial peritoneal damage. The main role of Ig delta chain C region (P05) in immune system has been believed to be B-cell receptor. The expression of Ig delta chain C region is required in mature B cells and a key modulator of the humoral immune response (reviewed in [
This study has its limitations. It is very difficult to identify peritoneal damage unless invasive methods are used. Therefore, we cannot define the condition of peritoneal change, but we can assume that the peritoneum has slightly changed after the long-term CAPD treatment. This is a major obstacle in our experimental research. In addition, the small number of patients may contribute to the limited biological and clinical relevance of our findings. To address this issue, the differentially expressed proteins were further validated in another set of 3 CGN patients. This work has provided potential targets and mechanisms of the change due to peritoneal dialysis processes, which have been confirmed by mass spectrometry data and Western blotting. Before being applied clinically, further experiments are considered necessary to evaluate those targets in a sufficient number of participants.
In conclusion, to our knowledge, this is the first study to show differential proteomic profiling between peritoneal dialysates from CGN patients at the early and middle stage of CAPD treatment. The identified proteins may provide clues to the PD-induced loss of proteins from the peritoneum and assist the identification of potential biomarkers for noninvasive monitoring of peritoneal damage. Further studies of these potential proteins are needed for accessing their roles in both basic research and clinical monitoring.
The authors declare that they have no conflict of interests.
This research was supported by Grants CMFHR 10114 and CMFHR 10211 from the Chi-Mei Medical Center, Tainan, Taiwan. The authors thank the National Cheng-Kung University Proteomics Research Core Laboratory for providing assistance with two-dimensional gel electrophoresis, image analysis, mass spectrometry analysis, and database searching.