Secondary Logo

Journal Logo

Original article

Serum peptidome profiling for identifying pathological patterns in patients with primary nephrotic syndrome

Lan-ting, HUANG; Qiong, WEN; Ming-zhe, ZHAO; Zhi-bin, LI; Ning, LUO; Yong-tao, WANG; Xiu-qing, DONG; Xue-qing, YU

Author Information
doi: 10.3760/cma.j.issn.0366-6999.2012.24.017
  • Free


Primary nephrotic syndrome (NS) characterized by massive proteinuria, hypoalbuminemia, edema and hyperlipidemia is a common kidney disease with an estimated annual incidence of three in every 100 000 adults and two in every 100 000 children.1,2 Three types of glomerular diseases are the most common: minimal change disease (MCD), membranous nephropathy (MN), and focal segmental glomerulosclerosis (FSGS). Renal biopsy is necessary to guide the appropriate immune therapy strategies and predict the prognosis.3 However, renal biopsy is an invasive procedure with potential risk of complications, such as hematuria, local infections, hematoma or embolism. These complications may be very serious and result in nephrectomy.4,5 Moreover, renal biopsy is not suitable for patients with comorbidities or higher risks, including bleeding disorders, obesity, diabetes, advanced renal insufficiency, or a solitary kidney.6–8 In addition, histological diagnosis takes time, and occasionally a biopsy may fail to obtain adequate tissue for accurate diagnosis. Thus, renal biopsy is conditionally limited, time-consuming and can not be performed multiple times, which also restricts its application in long-term follow-up in these patients.

Compared with renal tissue, serum is much easier to obtain and contains a large array of peptides derived from numerous functional proteins. When blood flows through organs, each cell from the surrounding tissue sheds protein fragments or peptides into the serum, either as metabolites or as signals to other cells. These cellular products may reflect ongoing physiological and pathological events.9,10 Pathological changes in the kidneys are likely to have a complex correlation with alteration of serum peptidome profiles; it would be useful to search for serum biomarkers and establish a non-invasive, rapid and reproducible test for predicting the pathologic pattern of glomerular diseases.

In recent years, proteomic tools and bioinformatics analyses have been used to identify serum or plasma peptide markers which are the characteristic or indicators of specific diseases;11–15 these systematic proteomic strategies have unveiled diagnostic information and have brought about innovation in biomarker research from a single molecule to a panel of molecules.16,17 Magnetic bead-based matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MB-based MALDI TOF MS), as one of these promising methods, is appropriate for the rapid detection of low-concentration peptides with high resolution, reproducibility, and throughput.18–20 This technology has been improved and explored for its potential clinical applications.21–25

In this study, serum peptidome spectra of primary NS patients (including MCD, MN and FSGS) and healthy controls were obtained by MB-based MALDI TOF MS. A serum peptidome pattern capable of identifying different primary NS and healthy controls was generated by ClinProt software (Bruker Daltonics, Germany) using a training set of 31 MCD, 15 MN, 11 FSGS and 30 normal individuals. Peptides from the serum peptidome pattern were sequenced by liquid chromatography electrospray ionization quadrupole time-of-flight (LC-ESI-Q-TOF).


Patients and sample collection

Serum samples were collected from patients who were diagnosed with primary NS both clinically and pathologically in the Department of Nephrology, First Affiliated Hospital, Sun Yat-sen University, China. All patients underwent renal biopsy to obtain at least two strips of kidney tissue for evaluation by light microscopy, immunofluorescence and electron microscopy. Professional renal pathologists made all the pathological diagnoses. Any cases of clinical or pathological features indicating a secondary cause, such as autoimmune disease, infections, cancer or exposure to nephrotoxic drugs, were excluded. Ethical Committee approved this study and informed consent forms were obtained from all the participants.

Serum samples were collected one day before drug administration from patients undergoing steroid treatment. Briefly, serum was drawn in drying tubes after overnight fasting and incubated at 4°C for at least 1 hour, then centrifuged at 4°C and 1800 ×g for 30 minutes and stored in aliquots at -80°C until analysis. The same strict, routine protocols were also applied to sample collections from healthy volunteers.

One hundred and seventy-four pairs of serum samples were collected. Samples were derived from 114 patients with primary NS (62 with MCD, 30 with MN, and 22 with FSGS) and 60 healthy volunteers. Clinical features for all the participates are shown in Table 1. There were no significant differences in gender (P=0.535), age (P=0.167), serum albumin (P=0.140), serum total cholesterol (P=0.089), serum creatinine (P=0.100), or 24-hour urine protein (P=0.840) among the different groups.

Table 1
Table 1:
Demographics and clinical information of NS patients and control subjects

Sample preparation

Each sample was processed using ClinProt purification reagent sets (Bruker Daltonics) for proteome fractionation. Magnetic bead weak cation-exchange (WCX) was used for serum preparation to yield more peaks and to lower the background noise. Briefly, after thoroughly mixing 10 μl magnetic beads with 10 μl binding buffer, 5 μl serum was added and incubated for 5 minutes. After washing to remove all unbound protein, chromatographically retained proteins were eluted using 5 μl elution solution. This procedure required approximately 45 minutes.

Finally, 1 μl of a mixture containing the eluted peptide fraction and 0.3 g/L a-cyano-4-hydroxy-cinnamic acid in 2:1 ethanol/acetone (volume/volume) were spotted onto the Anchor-ChipTM 600-μm target (Bruker Daltonics). Duplicates or triplicates were used to minimize sample preparation bias.

MALDI TOF mass spectrometry analysis

Air-dried targets were measured by MALDI TOF MS (Ultraflex; Bruker Daltonics). Spectral detection was carried out in the linear mode, and external calibration with peptide/protein standards (Bruker Daltonics) was performed in the range of 1000 to 15 000 Dalton (Da). Eight-hundred shots were acquired (100 laser shots at eight different spot positions) at a power of 45% and signals with a signal to noise ratio (S/N) >3 were recorded in automatic mode.

LC-ESI MS/MS for peptide identification

The serum samples containing a significant majority of the discriminating peaks were separately filtered through ultra-filtration tubes with a molecular weight cut-off of 3000 Da. Peptides were then separated and analyzed with MALDI TOF or LC-ESI-Q-TOF. For ESI-Q-TOF analysis, peptides were separated and analyzed on an Ultimate 3000 nanoHPLC (Dionex, Sunnyvale, USA) coupled to a micrOTOF-Q (Bruker Daltonics). A 10-μl solution of each sample was injected into the trap column with the autosampler of the Ultimate 3000. The trap column was then washed with 0.1% formic acid at a flow rate of 20 μl/min for 5 minutes to desalt the samples. Subsequently, the 10-port valve was switched to direct the flow to the separation column. The desalted peptides were then separated on a C18 column (packed in-house; Synergi C18; 150 mm × 0.075 mm) and analyzed on a micrOTOF-Q mass spectrometer with a nanoelectrospray ionization ion source. The flow rate was maintained at 400 nl/min. The gradient was started at 3% acetonitrile (ACN) with 0.1% formic acid and linearly increased to 43% ACN over 40 minutes, then to 73% ACN over 5 minutes, and finally to 95% ACN over another 5 minutes. The gradient was then decreased to 3% ACN over 1 minute and maintained at 3% ACN for 14 minute. The mass spectrometer was operated in the positive ion mass spectrometric (MS) mode, and data-dependent analysis was employed for survey scans mass-to-charge ratio (m/z) 350–1500 to choose up to three of the most intense precursor ions. For collision-induced dissociation (CID) tandem MS/MS analysis, collision energies were chosen automatically as a function of m/z and charge. The collision gas was argon. The temperature of the heated sample source was 180°C and the electrospray voltage was 1400 V. External mass calibration in quadratic regression mode using sodium formate resulted in typical mass errors of 5 ppm in the m/z range of 50–2000. Dynamic exclusion was continued for 1 minute.

Database searching

Mass spectra were processed with Data Analysis v4.0 (Bruker Daltonics), and the result in file format in the materials and geometry (MGF) documents were searched against the Swiss Prot 57.15 human database and its randomized sequence databases using Mascot software (Matrix Science Ltd., UK), with no specific enzyme. For MALDI TOF data, mass tolerances of peptide and MS/MS were set as 100 ppm and 0.5 Da, respectively, while for Q-TOF data, both the peptide and MS/MS mass tolerance were set as ± 0.05 Da.

Statistical analysis

The spectra from each serum sample were separately analyzed by the same procedure. The spectra from 174 serum samples were grouped into disease-related classes and processed by ClinPro Tools v2.2 software (Bruker Daltonics). A standard workflow was used for spectra preparation including the following steps: baseline subtraction, normalization, average calculation, and average peak list calculation. The degree of variation was evaluated by calculating the between-day and within-day coefficient of variation (CV) for all of the main peaks of spectra from pooled serum samples.22

Before statistical analysis, serum samples from each group were randomly and equally subdivided into a training set and a testing set. Both sets included 31 MCD patients, 15 MN patients, 11 FSGS patients, and 30 healthy controls.

A genetic algorithm (GA) implemented in the ClinPro Tools v2.2 software (Bruker Daltonics) was used for spectral analysis. This algorithm mimics natural evolution and is used to select the peak combinations that are most relevant for separation. Pattern determination is used to identify an optimal set of peaks, which gives the best separating model determined upon the model generation spectra used, and validated on test spectra or by a cross-validation procedure. The parameter settings for model generation were as follows: the maximal number of best peaks was 15 and maximal number of generations was 50; automatic detection of the initial number of peak combinations was applied; the mutation and crossover rates were 0.2 and 0.5, respectively; use of varying random seed was not applied; and the number of neighbors was three. The training set spectra were used to construct the classification model, while the testing set spectra were used to validate the model.

The ClinProTools v2.2 software was used for spectra analysis and GA implemented in the ClinProTools v2.2 software was used for generation of peptidome pattern. The differences in peak values were considered statistically significant when P <0.05. Other statistical analyses were performed using the SPSS13.0 (SPSS Inc., USA) software. Quantitative data were described as mean ± standard deviation (SD) and receiver operating characteristic curve (ROC) analysis was used to evaluate the diagnostic capability of the selected peak.


Reproducibility of serum peptidome profiling

A previously described method was used to confirm the reproducibility of our approach.22 Peptide profiles derived from the same serum sample were compared in triplicate and spotted on a chip for MALDI TOF MS in triplicate or duplicate. Replicate mass spectra showed highly reproducible peaks. For analysis of the serum spectra, within-day CVs were 0.11%-13.61% and between-day CVs were 1.83%-17.73%.

Characterization of the peptidome profile

Peptidome spectra of 174 serum samples were generated by MALDI TOF MS combined with magnetic bead WCX, and 203 protein peaks were detected. The serum overall sum spectra of each group are acquired. Within mass range from 1500 to 12 500 Da, a large number of differentially expressed peptides could be detected.

Construction of the classification model

The training set spectra were analyzed by the GA to generate a peptidome pattern. The serum peptidome pattern was based on 14 significantly different peaks, with masses ranging from 1500 Da to 9200 Da (Table 2), which could clearly distinguish primary NS patients from healthy controls. Moreover, the different types of NS were also clearly distinguished. In the training set, the serum model accurately recognized spectra from all MCD, all MN, 10 of 11 FSGS and all healthy control patients.

Table 2
Table 2:
Serum peptide peaks included in the GA-based model to distinguish primary patients with MCD, MN and FSGS from normal controls

Evaluation of the classification model

The testing set spectra were used for external validation of the GA-based model. Based on the serum testing set, the model evaluation and clinical diagnosis for each sample were compared (Table 3). Sensitivities and specificities were calculated to evaluate the accuracy of the model (Table 4).

Table 3
Table 3:
Evaluation of the serum peptidome pattern for the testing set
Table 4
Table 4:
Sensitivity and specificity of serum model in the testing set (% (n/N))

Identification of peptides from the classification model

The peak from the serum model at m/z 1538.1 in the linear mode was detected at m/z 1536.76 in the reflector mode, and its sequence was determined as ADSGEGDFLAEGGGVR. The peak was identified as fibrinogen alpha chain, with an Ions Score of 55, an Expect score of 0.0024 and a matches score of 16/276 fragment ions, using the 20 most intense peaks.

Peak areas of peptide m/z 1538.1 in the spectra from four groups were calculated. Healthy subjects were statistically different from those of the disease groups, while there were no statistically significant among disease groups (Figure 1). Additionally, the diagnostic capability of each peak was evaluated by ROC curve analysis (Figure 2), which showed ideal accuracy for discriminating healthy subjects from disease groups with an area under the curve (AUC) of 0.783–0.864.

Figure 1.
Figure 1.:
Box plot of calculated peak areas of the serum peptide m/z 1538.1 used in the cluster for the four groups. * P <0.001 vs. control.
Figure 2.
Figure 2.:
The area under curve determined by receiver operating characteristic curve analysis for serum peak m/z 1538.1 to identify four different groups. A: MCD vs. healthy control. B: FSGS vs. healthy control. C: MN vs. healthy control. D: MCD vs. FSGS. E: MCD vs. MN. F: FSGS vs. MN.


In the present study, highly reproducible serum peptide profiles were detected using MB-based MALDI TOF MS. A serum peptidome pattern was constructed and could discriminate MCD, MN, FSGS and healthy control with high accuracy. Using LC-ESI MS/MS, the serum peptide at m/z 1538.1 was identified as fibrinogen alpha chain.

Screening urine for predictive biomarkers of different glomerular diseases has been proposed on the basis of proteome analysis, such as two-dimensional electrophoresis (2DE) or capillary electrophoresis (CE)-MS.26–28 Our findings confirmed that MB-based MALDI TOF MS was effective for serum peptidome analysis to differentiate pathological patterns and achieved high reproducibility, throughput and resolution as former studies reported.13,29–31 The peptidome pattern by itself may qualify as a potential tool for indicating disease state.16,19,29

We tried to identify and analysis all the peptides in the peptidome pattern. Finally, one of the serum peptide at m/z 1538.1 was identified as fibrinogen alpha chain, which exhibited a lower relative intensity in NS patients than in healthy controls, and defects in the fibrinogen alpha chain have been observed in NS because of an amyloidogenic mutation.32 There is probably a correlation between the expression of fibrinogen alpha chain and initialization of NS. Moreover, AUC determined by ROC analysis of the serum peak at m/z 1538.1 showed an excellent ability to discriminate between healthy subjects and the different disease groups. To further define the potential predictive value of this peptide, we are in the process of conducting follow-up studies on patients, involving detecting expression levels of peptides in different types of NS, which may provide opportunities for research into the pathogenesis, prognosis or response to therapy.13,33,34

During peptidome profiling, minor changes in pre-analytical conditions or analytical procedures may influence reproducibility or even lead to bias, which is a significant challenge.35–37 Previous studies have generated a well-defined platform for peptidome profiling.22–24,38,39 In this study, strict criteria for case inclusion, a uniform procedure for specimen collection, a standardized protocol for specimen processing, and constant operating conditions were used to minimize variations and achieve high reproducibility. However, further studies with a larger sample size are needed to confirm the effectiveness of the peptidome pattern for identifying different pathologic patterns in primary NS.

In summary, MB-based MALDI TOF MS is an effective screening tool with high reproducibility for serum peptidome profiling. These findings indicate that peptidome profiling is a promising strategy for a non-invasive method in identifying pathologic pattern in patients with NS.


1. Hull RP, Goldsmith DJ. Nephrotic syndrome in adults. BMJ 2008; 336: 1185-1189.
2. Traum AZ, Schachter AD. Urinary proteome profiling to search for biomarkers in steroid-resistant nephrotic syndrome. Contrib Nephrol 2008; 160: 29-36.
3. Orth SR, Ritz E. The nephrotic syndrome. N Engl J Med 1998; 338: 1202-1211.
4. Whittier WL, Korbet SM. Renal biopsy: update. Curr Opin Nephrol Hypertens 2004; 13: 661-665.
5. Whittier WL, Korbet SM. Timing of complications in percutaneous renal biopsy. J Am Soc Nephrol 2004; 15: 142-147.
6. Eiro M, Katoh T, Watanabe T. Risk factors for bleeding complications in percutaneous renal biopsy. Clin Exp Nephrol 2005; 9: 40-45.
7. Stiles KP, Yuan CM, Chung EM, Lyon RD, Lane JD, Abbott KC. Renal biopsy in high-risk patients with medical diseases of the kidney. Am J Kidney Dis 2000; 36: 419-433.
8. Seigneux S, Martin PY. Management of patients with nephrotic syndrome. Swiss Med Wkly 2009; 139: 416-422.
9. Liotta LA, Ferrari M, Petricoin E. Clinical proteomics written in blood. Nature 2003; 425: 905.
10. Villanueva J, Shaffer DR, Philip J, Chaparro CA, Erdjument-Bromage H, Olshen AB, et al. Differential exoprotease activities confer tumor specific serum peptidome patterns. J Clin Invest 2006; 116: 271-284.
11. Pietrowska M, Polanska J, Marczak L, Behrendt K, Nowicka E, Stobiecki M, et al. Mass spectrometry-based analysis of therapy-related changes in serum proteome patterns of patients with early-stage breast cancer. J Transl Med 2010; 8: 66.
12. Musharraf SG, Hashmi N, Choudhary MI, Rizvi N, Usman A, Atta-Ur-Rahman. Comparison of plasma from healthy nonsmokers, smokers, and lung cancer patients: pattern-based differentiation profiling of low molecular weight proteins and peptides by magnetic bead technology with MALDI-TOF MS. Biomarkers 2012; 17: 223-230.
13. Gianazza E, Mainini V, Castoldi G, Chinello C, Zerbini G, Bianchi C, et al. Different expression of fibrinopeptide A and related fragments in serum of type 1 diabetic patients with nephropathy. J Proteomics 2010; 73: 593-601.
14. Kim JY, Song HJ, Lim HJ, Shin MG, Kim JS, Kim HJ, et al. Platelet factor 4 is an indicator of blood count recovery in acute myeloid leukemia patients in complete remission. Mol Cell Proteomics 2008; 7: 431-441.
15. Xiao D, Zhao LJ, Chu SL, Jing H, Wang C. Proteomic analysis for detecting serum biomarkers related to smoking in humans. Chin Med J 2012; 125: 3027-3032.
16. Chinello C, Gianazza E, Zoppis I, Mainini V, Galbusera C, Picozzi S, et al. Serum biomarkers of renal cell carcinoma assessed using a protein profiling approach based on ClinProt technique. Urology 2010; 75: 842-847.
17. Mor G, Visintin I, Lai Y, Zhao H, Schwartz P, Rutherford T, et al. Serum protein markers for early detection of ovarian cancer. Proc Natl Acad Sci U S A 2005; 102: 7677-7682.
18. Tirumalai RS, Chan KC, Prieto DA, Issaq HJ, Conrads TP, Veenstra TD. Characterization of the low molecular weight human serum proteome. Mol Cell Proteomics 2003; 2: 1096-1103.
19. Du J, Yang S, Lin X, Bu L, Nan Y, Huo S. Use of anchorchip-time-of-flight spectrometry technology to screen tumor biomarker proteins in serum for small cell lung cancer. Diagn Pathol 2010; 5: 60.
20. Villanueva J, Lawlor K, Toledo-Crow R, Tempst P. Automated serum peptide profiling. Nat Protoc 2006; 1: 880-891.
21. Vitzthum F, Behrens F, Anderson NL, Shaw JH. Proteomics from basic research to diagnostic application: a review of requirements needs. J Proteome Res 2005; 4: 1086-1097.
22. Hsieh SY, Chen RK, Pan YH, Lee HL. Systematical evaluation of the effects of sample collection procedures on low molecular weight serum plasma proteome profiling. Proteomics 2006; 6: 3189-3198.
23. Albrethsen J. Reproducibility in protein profiling by MALDI TOF mass spectrometry. Clin Chem 2007; 53: 852-858.
24. Fiedler GM, Baumann S, Leichtle A, Oltmann A, Kase J, Thiery J, et al. Standardized peptidome profiling of human urine by magnetic bead separation and matrix assisted laser desorption ionization time of flight mass spectrometry. Clin Chem 2007; 53: 421-428.
25. Zhang Q, Li SZ, Feng CS, Qu XD, Wang H, Zhang XN, et al. Serum proteomics of early postoperative cognitive dysfunction in elderly patients. Chin Med J 2012; 125: 2455-2461.
26. Weissinger EM, Wittke S, Kaiser T, Haller H, Bartel S, Krebs R, et al. Proteomic patterns established with capillary electrophoresis and mass spectrometry for diagnostic purposes. Kidney Int 2004; 65: 2426-2434.
27. Varghese SA, Powell TB, Budisavljevic MN, Oates JC, Raymond JR, Almeida JS, et al. Urine biomarkers predict the cause of glomerular disease. J Am Soc Nephrol 2007; 18: 913-922.
28. Shui HA, Huang TH, Ka SM, Chen PH, Lin YF, Chen A. Urinary proteome and potential biomarkers associated with serial pathogenesis steps of focal segmental glomerulosclerosis. Nephrol Dial Transplant 2008; 23: 176-185.
29. Schwamborn K, Krieg RC, Grosse J, Reulen N, Weiskirchen R, Knuechel R, et al. Serum proteomic profiling in patients with bladder cancer. Eur Urol 2009; 56: 989-996.
30. Huang Z, Shi Y, Cai B, Wang L, Wu Y, Ying B, et al. MALDI-TOF MS combined with magnetic beads for detecting serum protein biomarkers and establishment of boosting decision tree model for diagnosis of system. Rheumatology 2009; 48: 626-631.
31. Yoo BC, Kong SY, Jang SG, Kim KH, Ahn SA, Park WS, et al. Identification of hypoxanthine as a urine marker for non-Hodgkin lymphoma by low mass ion profiling. BMC Cancer 2010; 23: 10-55.
32. Picken MM, Linke RP. Nephrotic syndrome due to an amyloidogenic. J Am Soc Nephrol 2009; 20: 1681-1685.
33. Knepper MA. Proteomics and the kidney. J Am Soc Nephrol 2002; 13: 1398-1408.
34. Kasthuri RS, Verneris MR, Ibrahim HN, Jilma B, Nelsestuen GL. Studying multiple protein profiles over time to assess biomarker validity. Expert Rev Proteomics 2006; 3: 455-464.
35. Check E. Proteomics and cancer running before we can walk. Nature 2004; 429: 496-497.
36. Cheng AJ, Chen LC, Chien KY, Chen YJ, Chang JT, Wang HM, et al. Oral cancer plasma tumor marker identified with bead based affinity fractionated proteomic technology. Clin Chem 2005; 51: 2236-2244.
37. Hortin GL. The MALDI-TOF mass spectrometric view of the plasma proteome and peptidome. Clin Chem 2006; 52: 1223-1237.
38. Baumann S, Ceglarek U, Fiedler GM, Lembcke J, Leichtle A, Thiery J, et al. Standardized approach to proteome profiling of human serum based on magnetic bead separation and matrix-assisted laser desorption ionization time-of-flight mass spectrometry. Clin Chem 2005; 51: 973-980.
39. Bruegel M, Planert M, Baumann S, Focke A, Bergh FT, Leichtle A, et al. Standardized peptidome profiling of human cerebrospinal fluid by magnetic bead separation and matrix assisted laser desorption ionization time of flight mass spectrometry. J Proteomics 2009; 72: 608-615.

nephrotic syndrome; serum; matrix-assisted laser desorption/; ionization time-of-flight mass spectrometry; diagnostic model

© 2012 Chinese Medical Association