Georg Martin Fiedler, Sven Baumann, Alexander Leichtle, Anke Oltmann, Julia Kase, Joachim Thiery, * and Uta Ceglarek.

Transcription

1 Clinical Chemistry 53: (2007) Proteomics and Protein Markers Standardized Peptidome Profiling of Human Urine by Magnetic Bead Separation and Matrix-Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry Georg Martin Fiedler, Sven Baumann, Alexander Leichtle, Anke Oltmann, Julia Kase, Joachim Thiery, * and Uta Ceglarek Background: Peptidome profiling of human urine is a promising tool to identify novel disease-associated biomarkers; however, a wide range of preanalytical variables influence the results of peptidome analysis. Our aim was to develop a standardized protocol for reproducible urine peptidome profiling by means of magnetic bead (MB) separation followed by matrix-assisted laser desorption/ionization time-of-flight (MALDI- TOF) mass spectrometry (MS). Methods: MBs with defined surface functionalities (hydrophobic interaction, cation exchange, and metal ion affinity) were used for peptide fractionation of urine. Mass accuracy and imprecision were calculated for 9 characteristic mass signals (M r, ). Exogenous variables (instrument performance, urine sampling/storage conditions, freezing conditions, and freeze-thaw cycles) and endogenous variables (ph, urine salt and protein concentrations, and blood and bacteria interferences) were investigated with urine samples from 10 male and 10 female volunteers. Results: We detected 427 different mass signals in the urine of healthy donors. Within- and between-day imprecision in relative signal intensities ranged from 1% to 14% and from 4% to 16%, respectively. Weak cationexchange and metal ion affinity MB preparations required adjustment of the urinary ph to 7. Storage time, Institute of Laboratory Medicine, Clinical Chemistry and Molecular Diagnostics, University Hospital, Leipzig, Germany. These authors contributed equally to this work. * Address correspondence to this author at: Institute of Laboratory Medicine, Clinical Chemistry and Molecular Diagnostics, University Hospital Leipzig, Liebigstr. 27, D Leipzig, Germany. Fax ; thiery@medizin.uni-leipzig.de. Received August 7, 2006; accepted December 21, Previously published online at DOI: /clinchem storage temperature, the number of freeze-thaw cycles, and bacterial and blood contamination significantly influenced urine peptide patterns. Individual urine peptide patterns differed significantly within and between days. This imprecision was diminished by normalization to a urinary protein content of 3.5 g. Conclusion: This reliable pretreatment protocol allows standardization of preanalytical modalities and facilitates reproducible peptidome profiling of human urine by means of MB separation in combination with MALDI-TOF MS American Association for Clinical Chemistry Peptidome analysis based on mass spectrometric screening methods, such as surface-enhanced laser desorption/ ionization time-of-flight (SELDI-TOF) 1 and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS), is a promising high-throughput approach for identifying new potential biomarkers in various body fluids (1 6). By virtue of its noninvasiveness and availability, peptidome profiling of human urine is now becoming an important method for investigating kidney physiology and detecting novel disease-associated markers of renal and bladder diseases (7 13). Analyses based on Western blotting or immunologic methods have been successfully used for identifying proteins in human urine (14). Studies with 2-dimensional gel electrophoresis techniques have been carried out in the fields of metabolism, toxicology, and renal diseases (15 19). Liquid chro- 1 Nonstandard abbreviations: SELDI-TOF, surface-enhanced laser desorption/ionization time-of-flight; MALDI-TOF, matrix-assisted laser desorption/ ionization time-of-flight; MS, mass spectrometry; MB, magnetic bead; MB-HIC C8, magnetic bead hydrophobic-interaction chromatography; MB-IMAC Cu, magnetic bead immobilized metal ion affinity chromatography; MB WCX, magnetic bead weak cation-exchange chromatography; S/N, signal-to-noise. 421

2 422 Fiedler et al.: Magnetic Bead Separation and MALDI-TOF MS of Urine matography tandem MS has also been applied for the identification of urinary peptides after tryptic digestion (20). In contrast to these complex methods, SELDI-TOF MS and magnetic bead (MB) separation followed by MALDI-TOF MS enable high-throughput investigations of proteins and peptides without the laborious samplepretreatment procedures (3, 7, 21). In addition, SELDI- TOF MS and MALDI-TOF MS especially focus on the region of low M r ( ), which includes the small proteins, protein/peptide fragments, and peptides that constitute the peptidome or fragmentome and that are overlooked by traditional techniques (5). Before urinary peptidomics can proceed from the bench to the bedside, however, a large number of important barriers need to be overcome (2, 22). The preanalytical phase especially is affected by a wide range of preanalytical variables, both exogenous (instrument settings, urine collection and storage methods, freezing conditions, and the number of freeze-thaw cycles) and endogenous (ph, urine concentrations of salts and proteins, and blood and bacterial interferences), that can markedly influence the results of peptide profiling (8, 21 26). Therefore, before clinical implementation of the method can be warranted, the importance of these variables on peptidome profiling has to be investigated, and the preanalytical and analytic steps have to be standardized (22, 27 30). The aim of our study was to investigate instrument performance and the effect of preanalytical variables on urinary peptidome profiling with MB separation and MALDI-TOF MS and to establish a reliable pretreatment protocol for collecting and storing urine samples to permit valid and reproducible analyses of urinary peptide patterns. Materials and Methods chemicals, standards, and consumables Gradient-grade acetonitrile and ethanol were obtained from J.T. Baker, and pro analysi grade trifluoroacetic acid, urea, sodium chloride, and acetone were purchased from Sigma-Aldrich. Peptide Calibration Standard I, Protein Calibration Standard I, and -cyano-4-hydroxycinnamic acid were purchased from Bruker Daltonics. We used 0.2-mL polypropylene tubes (8-tube strips) for manually operated MB preparations. Automated MB preparations were performed with 96-well plates and TubePlates from Biozym, low-profile polypropylene tubes from ABgene, and Modular Reservoir Quarter Modules from Beckman Coulter. For sample storage, 450- L CryoTubes were purchased from Sarstedt. urine collection and storage First- and 2nd-morning urine samples were collected from 20 healthy volunteers (10 women and 10 men, ages years) in 100-mL urine cups (Sarstedt). Multivariable strips (Combur10 Test ; Roche Diagnostics) were used to analyze samples for specific gravity, ph, leukocytes, nitrite, protein, glucose, ketones, urobilinogen, bilirubin, and blood, and a Hitachi 917 analyzer (Roche Diagnostics) was used for assays of electrolytes, total protein, creatinine (enzymatic test), and urea (for results, see Table 1 in the Data Supplement that accompanies the online version of this article at org/content/vol53/issue3). Unless otherwise stated, samples were centrifuged at g for 10 min at room temperature. Aliquots of 450 L were stored at 80 C. Urine samples were thawed at room temperature for 30 min, adjusted to ph 7, centrifuged again, and immediately processed for peptidome separation after the sample was adjusted for total protein content. Variations in sample pretreatment are described in the experimental section. The samples from the 10 female and 10 male volunteers were pooled for precision and mass accuracy analysis. peptidome separation We used ClinProt purification reagent sets (Bruker Daltonics) for peptidome separation of samples and magnetic particles (particle size, 1 m; mean pore size, 40 nm; specific surface area, 100 cm 2 /g) with defined surface functionalities [MB hydrophobic-interaction chromatography (MB HIC C8), MB immobilized metal ion affinity chromatography (MB IMAC Cu) beads, and MB weak cation-exchange chromatography (MB WCX)]. MB purifications were performed according to the manufacturer s protocol for serum and with the ClinProt liquid-handling robot. The protocol was adapted with regard to sample and binding-solution volumes. A 30- L volume of the urine sample was diluted in 60 L binding solution and added to the bead slurry (5 L MB HIC C8 beads, 5 L MB IMAC Cu beads, or 10 L MB WCX beads). After thorough stirring, samples were incubated for 1 min (MB HIC C8 and MB WCX beads) or 5 min (MB IMAC Cu beads) at room temperature. The protein content was normalized by increasing the loaded urine sample volume in a stepwise manner. After the 1st fractionation step, residual urine was removed, and the procedure was repeated with a 2nd urine fraction (60 L maximum). Following the stepwise application of sample and MB separation, the peptide fraction was eluted from MB HIC C8 beads with 5 L acetonitrile/ water (1:1, volume/volume). For MB WCX beads, we used 5 L elution solution and 4 L stabilization buffer, and we eluted the peptide fraction from MB IMAC Cu beads with 10 L of elution solution from Bruker Daltonics. To prepare the MALDI target, we spotted 1 L ofa mixture containing 10 L 0.3 g/l -cyano-4-hydroxycinnamic acid in 2:1 ethanol/acetone (volume/volume) and 1 L of the eluted peptide fraction onto the Anchor- Chip 600- m target (Bruker Daltonics). mass spectrometry A linear MALDI-TOF mass spectrometer (Autoflex; Bruker Daltonics) was used for peptidome profiling with the following settings: linear positive mode; ion source 1,

3 Clinical Chemistry 53, No. 3, kv; ion source 2, kv; lens, 9.00 kv; pulsed ion extraction, 120 ns. For matrix suppression, we used a high gating factor with signal suppression up to an M r of 500. Mass calibration was performed with the standard calibration mixture of peptides and proteins (M r range, ). Four MALDI preparations (MALDI spots) from each sample were measured. For each MALDI spot, 450 spectra were acquired (30 laser shots at 15 different spot positions). To increase detection sensitivity, we removed excess matrix with 6 shots at a laser power of 45% before data acquisition at a power of 25%. All signals with a signal-to-noise (S/N) ratio 3 inanm r range of were collected with the AutoXecute tool of the flexcontrol acquisition software (version 2.0; Bruker Daltonics), and Bruker Daltonics flexanalysis software was applied for data analysis. ClinProTools bioinformatics software (version 2.0 beta; Bruker Daltonics) was used for the recognition of peptide patterns. evaluation of exogenous variables Instrument performance. Mass accuracy and within-day and between-day reproducibility were determined by the relative peak intensities of 9 characteristic signals of the urine sample. MB HIC C8, MB WCX, and MB IMAC Cu beads were used for 10 bead preparations on 3 consecutive days. Urine storage and freeze-thaw cycles. The influence of storage temperature and time on the peptide profiles was investigated with the pooled urine sample after separation by MB HIC C8. Pooled urine was divided into 2 aliquots. One milliliter of the fresh urine aliquot and 1 ml of a deep-frozen aliquot (thawed at room temperature for 30 min) were incubated in 1.5-mL vials at 4 C, 25 C, and 40 C for 0, 1, 3, 6, 24, 72, and 168 h. At each defined time point, L aliquots were removed and prepared immediately. We also investigated the effect of up to 3 freeze-thaw cycles on the peptide pattern. A 1-mL aliquot of a pooled urine sample was frozen for 24 h and then thawed for 30 min at room temperature. Thirty microliters of urine was removed and processed immediately. The remaining sample was frozen again. The procedure was repeated twice. evaluation of endogenous variables Urine-collecting conditions. We evaluated differences between urine samples collected on the 1st and 2nd mornings with urine aliquots from 5 female and 5 male volunteers. Samples were prepared with MB HIC C8 beads according to the standard protocol. The peptide profiles were compared with respect to relative peak intensities. Influence of urea and salt. The influence of urea and salt content on peptide profiles was examined by adding urea and sodium chloride stock solutions to the pooled urine sample to achieve final concentrations of 50, 100, 250, 500, and 1000 mmol/l. A blank sample was diluted in the same way with water and analyzed to avoid misinterpretation of spectra caused by sample dilution. Blood and bacterial interferences. To investigate the effect of blood contamination on peptide pattern, we added whole blood to the pooled urine samples in blood urine ratios of 1:100, 1:500, 1:1000, 1:2500, 1:5000, 1:10 000, and 1: (vol/vol). Results were compared with uncontaminated urine samples. In addition, the influence of bacteria was determined in an ex vivo simulation. One hundred microliters of an Escherichia coli suspension ( colonyforming units/l) was added to 900 L of pooled urine samples without significant bacteriuria. These mixtures were incubated at room temperature for 0, 30, and 60 min. Urine samples were diluted 1:1 (volume/volume) with acetonitrile containing 5 g/l trifluoroacetic acid for inactivation of bacteria. The samples were then lyophilized, resuspended with 35 L water, and processed by MB MALDI-TOF MS (30 L). The results were compared with sterile urine samples. Creatinine, protein, and ph normalization. Urine samples collected midstream from 10 volunteers on the 1st and 2nd mornings were normalized to the creatinine concentration and to the protein content. The creatinine concentration was adjusted to the lowest measured creatinine concentration (4 mmol/l for the 1st morning s sample and 2.5 mmol/l for the 2nd morning s urine sample) by diluting each sample with 0.15 mol/l sodium chloride. The total protein amounts of the 1st and 2nd morning s urine samples were adjusted to 3.5 g by varying sample volume from 30 to 153 L. To evaluate the optimal ph value for MB purification, we adjusted the ph of pooled urine samples to ph 6, ph 7, ph 8, and ph 9 with 1 mol/l NH 4 HCO 3 or 1 mol/l (NH 4 ) 2 SO 4. Results evaluation of exogenous variables Instrument performance. After separation with MB HIC C8, MB WCX, and MB IMAC Cu beads, we detected 427 peaks in human urine with an S/N ratio 3 (MB HIC C8, 205 signals/spectrum; MB WCX, 217 signals/spectrum; MB IMAC Cu, 197 signals/spectrum) with an overlap frequency of 42.6%. We detected 244 signals overall with an S/N ratio 6 and 158 signals overall with an S/N ratio 10. The reproducibility of mass spectra generation was determined by evaluating relative peak intensities. For the pooled urine sample, we obtained CVs in the ranges of 1% 14% (within days) and 4% 16% (between days; see Table 2 in the online Data Supplement). A mass accuracy of 0.035% (350 ppm) was obtained in the M r range of This result corresponds well to our previous results with standardization of the serum proteome (29). We used MB HIC C8 beads for all further investigations of the effects of preanalytical variables because these MBs

4 424 Fiedler et al.: Magnetic Bead Separation and MALDI-TOF MS of Urine have a separation capability that is more versatile than the other bead modifications. Urine storage and freeze-thaw cycles. The influence of storage temperature and time on the stability of the peptide pattern was investigated by incubating fresh and once-frozen pooled urine samples. Once-frozen samples showed greater reproducibility and improved stability against increased temperatures compared with fresh urine (data not shown). We also investigated the effect of 3 freeze-thaw cycles on the peptide pattern. Generally, freezing only once decreased the relative peak intensity of some mass signals (e.g., M r s , , and ), whereas other mass signals (M r s , , , and ) increased. The variation in the relative intensities of mass signals in fresh and frozen urine is shown in Fig. 1. Compared with once-frozen samples, recurrent freeze-thaw cycles (3 ) did not further affect urine peptide patterns. The peptide patterns of once-frozen urine samples were stable up to 24 h at 4 C with no detectable degradation; however, we observed significant changes in peptide patterns at 25 C after 6 h of storage. At 40 C, peptide patterns showed significant decreases and increases in signal intensities for numerous signals by 3 h (Fig. 2). evaluation of endogenous variables Urine-collecting conditions. Urine samples collected on the 1st and 2nd mornings showed significant differences in relative signal intensities. The peptide profiles of 2nd-morning urine samples obtained with MB HIC C8 beads featured reduced intensities for several low-m r signals (e.g., , , , , and ) and by increases in other mass signals (e.g., M r s , , and ). Therefore, the peptide patterns of urine Fig. 2. Characteristic mass signals after MB HIC C8 purification following sample storage for different times and at different temperatures. Presented are results for storage at 4 C (A), 25 C (B), and 40 C (C) for the indicated times. Fig. 1. Two-dimensional mass plot comparing relative peak areas of 2 characteristic mass signals after MB HIC C8 purification with respect to the effect of freeze-thaw cycles on urine peptide pattern. The ellipses represent the standard deviations of the selected M r s of each group. samples collected on the 1st and 2nd mornings were different (Fig. 3). We also investigated the variation in individual urinary-peptide patterns by analyzing midstream urine samples collected on the 1st and 2nd mornings for 5 consecutive days. The variation in relative signal intensities for 9 characteristic signals ranged from 4% to 31% for samples collected on the 1st morning and

5 Clinical Chemistry 53, No. 3, Fig. 3. Two-dimensional mass plot comparing relative peak areas of 2 characteristic mass signals after MB HIC C8 purification with respect to 1st- and 2nd-morning urine samples. The ellipses represent the standard deviations of the selected M r s of each group. from 5% to 43% for samples collected on the 2nd morning. Compared with these peptidomic data, the median between-day variations in creatinine concentration for samples collected on the 1st and 2nd mornings were 31.9% and 28.2%, respectively, and those for total protein were 41.0% and 39.5%, respectively. Influence of urea and salt. Increasing urea and sodium chloride concentrations did not disturb sample preparation or MS analyses up to a concentration of 1000 mmol/l. Blood and bacterial interferences. Blood contamination influenced the peptide patterns of urine samples. New mass signals were detected up to a urine blood ratio of 2500:1 (see Fig. 1 in the online Data Supplement). For samples with a urine blood ratio of 100:1, we detected both completely new signals and ion-suppression effects on previously detected signals. We evaluated the effect of bacterial interference on peptide profiles with an ex vivo simulation. Comparison of the 4 sample groups (control urine and urine incubated at room temperature for 0, 30, and 60 min with colony-forming E. coli units/l) showed decreases in intensity for several signals in all urine samples incubated with a bacterial suspension for 30 min. The most significant differences were observed for mass signals of M r s and (Fig. 4). This effect was not observed immediately after the addition of the bacteria. ph, creatinine, and protein normalization. Urine ph influenced protein separation with the MB WCX and MB IMAC Cu beads. The best signal yield and highest signal intensities were obtained at ph 7 with MB WCX beads (217 signals with an S/N ratio 3) and MB IMAC Cu beads (197 signals with an S/N ratio 3). The numbers and intensities of signals with M r s 3500 were decreased Fig. 4. Characteristic mass signals (A and B) after incubating urine samples with E. coli suspension and subsequent MB HIC C8 purification. above ph 7. We observed no ph-dependent differences in peptide patterns with MB HIC C8 beads. We normalized creatinine concentrations by diluting urine samples to a particular creatinine concentration. Creatinine concentrations varied from 4.0 to 29.2 mmol/l for 1st-morning samples and from 2.5 to 15.9 mmol/l for 2nd-morning samples. We adjusted creatinine concentrations to 4 mmol/l (1st-morning samples) and 2.5 mmol/l (2nd-morning samples). A comparison of the spectra for normalized and nonnormalized urine samples after MB purification showed a decline in signals with an S/N ratio 10 of up to 15% in diluted samples, compared with undiluted samples. Signal losses of up to 35% were detected after normalizing the creatinine concentrations of 2nd-morning samples. Generally, adjustments in creatinine concentration via dilution did not decrease the variation in individual signals but produced a loss of mass signals. Concentrations of total protein for the 20 analyzed urine samples ranged from 29.0 to mg/l for 1stmorning samples and from 8.2 to mg/l for 2ndmorning samples. Sample dilution to the lowest protein concentration produced a loss of mass signals similar to that observed for creatinine. Therefore, we varied the urine sample volume from 30 to 60 L to obtain a maximal loading of the MBs. The capacity of the pipetting robot limited the maximum sample volume to 60 L.

6 426 Fiedler et al.: Magnetic Bead Separation and MALDI-TOF MS of Urine Fig. 5. Normalization effects on MB HIC C8 urine peptide patterns for 2 healthy volunteers (A and ). Without normalization, 3.4 mmol/l creatinine, 22.8 mg/l total protein, 30- L sample volume (0.7 g adsorbed; A1); without normalization, 14.4 mmol/l creatinine, mg/l total protein, 30- L sample volume (3.3 g adsorbed; B1); normalized by dilution, 2.5 mmol/l creatinine, 16.8 mg/l total protein, 30- L sample volume (0.5 g adsorbed; A2); normalized by dilution, 2.5 mmol/l creatinine, 19.1 mg/l total protein, 30- L sample volume (0.6 g adsorbed; B2); normalized by applied sample volume, 3.4 mmol/l creatinine, 22.8 mg/l total protein, 153- L sample volume (3.5 g adsorbed; A3); normalized by applied sample volume, 14.4 mmol/l creatinine, mg/l total protein, 33- L sample volume (3.5 g adsorbed; B3). Therefore, we normalized the total protein content of urine samples to 3.5 g for the 1st- and 2nd-morning samples. With MB HIC C8 purification, the number of mass signals in the urine with an S/N ratio 10 increased from 60 before protein normalization to 90 after normalization. In addition, interindividual variation in the urine peptide patterns of the 20 healthy volunteers was minimized (Fig. 5). Discussion Urinary peptidome profiling with high-throughput methods such as MB MALDI-TOF MS or SELDI-TOF MS is a promising tool in nephrology research (9, 12, 13). Both methods are of particular interest because they enable rapid analyses of many individual samples in large-scale clinical studies, but the preanalytical and analytic steps require thorough validation before clinical implementation can be warranted (8, 21, 22, 26 29). Therefore, we used MB MALDI-TOF MS to investigate instrument performance and the effects of various exogenous and endogenous variables on urine peptide profiles and established a reliable pretreatment protocol. MB MALDI-TOF MS seems to be more sensitive than SELDI-TOF MS. With 3 different MB preparations (MB HIC C8, MB WCX, and MB IMAC Cu beads), we detected 427 mass peaks with an S/N ratio 3 (M r range, ). Previous investigations of urine with SELDI-TOF MS yielded fewer mass peaks. The WCX2 ProteinChip detected 180 peaks in 138 samples (7), the CM10 (cationic) and the IMAC30 (anionic) ProteinChips detected 133 peaks in the M r range of (26), and use of the NP20 ProteinChip demonstrated only 25 peaks/spectrum (21). One reason for the improved sensitivity might be the higher adsorption capability of the beads porous surface structure compared with on-layer chips (3). Analysis of urine with MB HIC C8 beads produced fewer mass peaks with an S/N ratio 3 than serum analysis (serum, 250 peaks; urine, 200 peaks) (29), but the 20% higher proportion of signals with an S/N ratio 10 in urine samples might improve data analysis. We also determined the reproducibility and mass accuracy of the MB MALDI-TOF MS method. The within-day and between-day imprecision of urine peptidome profiling was 16%. SELDI-TOF MS produced imprecision values of up to 30% (21). One reason for the improved precision of MB MALDI-TOF MS might be the application of 4-fold spotting of the MALDI target for each urine sample. The determination of the mass accuracy (M r range, ) produced absolute differences in M r of , which are comparable to our data for serum and in accordance with the known mass accuracy of MALDI-TOF MS (29, 31). Differences in collection, sample storage, blood and bacterial interferences, and modes of normalization (creatinine, protein, and ph) of urine samples had significant effects on the variation in the peptide patterns. The time of urine collection (1st or 2nd morning) significantly influenced the peptide pattern because of the varying concentrations of urine components; therefore, it is important that collection times be defined exactly for all samples according to the requirements of the clinical study. Compared with our peptide analysis of serum, the urine

7 Clinical Chemistry 53, No. 3, Fig. 6. Recommended sample pretreatment protocol for reproducible MB-based urinary peptidome profiling. peptide pattern is much more stable at room temperature (29); however, a storage time 6 h induced significant changes in peptide patterns and therefore should be avoided. In contrast to the results of Schaub et al. (21), we found differences between fresh and once-frozen samples. The peptide profiles of once-frozen urine samples showed better reproducibility. Therefore, we recommend that urine samples be frozen at 80 C before processing. We demonstrated that bacterial (E. coli) contamination affected the time-dependent alteration of several urine peptide mass signals (M r range, ). Contamination with blood also interfered with the peptide pattern down to a urine blood ratio of 2500:1, a contamination level that produces no change in urine color. It is important, therefore, that each urine sample be checked with a urine test strip for the presence of leukocytes, nitrite, and blood. The use of MB WCX and MB IMAC Cu beads requires adjusting the urine ph to ph 7. In contrast, no ph adjustment is necessary for MB HIC C8 separation because of the strongly acidic fractionation conditions and the hydrophobic interactions between analytes and the beads porous surface structure. A sensitive point in urine proteome profiling is the dependence of the urinary protein concentration on the fluid intake and whether urine samples should be normalized to the protein content. We showed that adjusting the protein concentration by varying the volume of the applied sample minimized the influence of variation in urine protein concentration. Sample preparation for the MBs enables stepwise sample loading on the bead surfaces up to 5 g total protein (the maximum loading capacity). In conclusion, we showed that MB separation in combination with MALDI-TOF MS is a sensitive and reproducible analytic platform for peptidome profiling of human urine. The influence of various preanalytical variables must be taken into account, however. The reliable pretreatment protocol we have presented (Fig. 6) allows standardization of the critical preanalytical phase and facilitates the reproducible peptidome profiling of human urine. This work was supported by a grant from the Sächsische Aufbaubank, by a formel-1 grant of the Medical Faculty of the University Leipzig, and by Grant Th 374/2-3 of the Deutsche Forschungsgemeinschaft. References 1. Aldred S, Grant MM, Griffiths HR. The use of proteomics for the assessment of clinical samples in research. Clin Biochem 2004; 37: Diamandis EP. Mass spectrometry as a diagnostic and a cancer biomarker discovery tool: opportunities and potential limitations. Mol Cell Proteomics 2004;3:

8 428 Fiedler et al.: Magnetic Bead Separation and MALDI-TOF MS of Urine 3. Villanueva J, Philip J, Entenberg D, Chaparro CA, Tanwar MK, Holland EC, et al. Serum peptide profiling by magnetic particleassisted, automated sample processing and MALDI-TOF mass spectrometry. Anal Chem 2004;76: Zhang X, Leung SM, Morris CR, Shigenaga MK. Evaluation of a novel, integrated approach using functionalized magnetic beads, bench-top MALDI-TOF-MS with prestructured sample supports, and pattern recognition software for profiling potential biomarkers in human plasma. J Biomol Tech 2004;15: Hortin GL. The MALDI-TOF mass spectrometric view of the plasma proteome and peptidome. Clin Chem 2006;52: Hortin GL, Jortani SA, Ritchie JC Jr., Valdes R Jr., Chan DW. Proteomics: a new diagnostic frontier. Clin Chem 2006;52: Rogers MA, Clarke P, Noble J, Munro NP, Paul A, Selby PJ, et al. Proteomic profiling of urinary proteins in renal cancer by surface enhanced laser desorption ionization and neural-network analysis: identification of key issues affecting potential clinical utility. Cancer Res 2003;63: Hortin GL, Meilinger B, Drake SK. Size-selective extraction of peptides from urine for mass spectrometric analysis. Clin Chem 2004;50: Thongboonkerd V. Proteomics in nephrology: current status and future directions. Am J Nephrol 2004;24: Haubitz M, Wittke S, Weissinger EM, Walden M, Rupprecht HD, Floege J, et al. Urine protein patterns can serve as diagnostic tools in patients with IgA nephropathy. Kidney Int 2005;67: Jurgens M, Appel A, Heine G, Neitz S, Menzel C, Tammen H, et al. Towards characterization of the human urinary peptidome. Comb Chem High Throughput Screen 2005;8: Thongboonkerd V, Malasit P. Renal and urinary proteomics: current applications and challenges. Proteomics 2005;5: Vidal BC, Bonventre JV, I-Hong Hsu S. Towards the application of proteomics in renal disease diagnosis. Clin Sci (Lond) 2005;109: McKee JA, Kumar S, Ecelbarger CA, Fernandez-Llama P, Terris J, Knepper MA. Detection of Na( ) transporter proteins in urine. J Am Soc Nephrol 2000;11: Marshall T, Williams K. Two-dimensional electrophoresis of human urinary proteins following concentration by dye precipitation. Electrophoresis 1996;17: Sarto C, Marocchi A, Sanchez JC, Giannone D, Frutiger S, Golaz O, et al. Renal cell carcinoma and normal kidney protein expression. Electrophoresis 1997;18: Celis JE, Wolf H, Ostergaard M. Bladder squamous cell carcinoma biomarkers derived from proteomics. Electrophoresis 2000;21: Thongboonkerd V, McLeish KR, Arthur JM, Klein JB. Proteomic analysis of normal human urinary proteins isolated by acetone precipitation or ultracentrifugation. Kidney Int 2002;62: Zerefos PG, Vougas K, Dimitraki P, Kossida S, Petrolekas A, Stravodimos K, et al. Characterization of the human urine proteome by preparative electrophoresis in combination with 2-DE. Proteomics 2006;6: Spahr CS, Davis MT, McGinley MD, Robinson JH, Bures EJ, Beierle J, et al. Towards defining the urinary proteome using liquid chromatography-tandem mass spectrometry. I. Profiling an unfractionated tryptic digest. Proteomics 2001;1: Schaub S, Wilkins J, Weiler T, Sangster K, Rush D, Nickerson P. Urine protein profiling with surface-enhanced laser-desorption/ ionization time-of-flight mass spectrometry. Kidney Int 2004;65: Hortin GL. Can mass spectrometric protein profiling meet desired standards of clinical laboratory practice? Clin Chem 2005;51: Kouri T, Harmoinen A, Laurila K, Ala-Houhala I, Koivula T, Pasternack A. Reference intervals for the markers of proteinuria with a standardised bed-rest collection of urine. Clin Chem Lab Med 2001;39: Kumps A, Duez P, Mardens Y. Metabolic, nutritional, iatrogenic, and artifactual sources of urinary organic acids: a comprehensive table. Clin Chem 2002;48: Thongboonkerd V, Chutipongtanate S, Kanlaya R. Systematic evaluation of sample preparation methods for gel-based human urinary proteomics: quantity, quality, and variability. J Proteome Res 2006;5: Traum AZ, Wells MP, Aivado M, Libermann TA, Ramoni MF, Schachter AD. SELDI-TOF MS of quadruplicate urine and serum samples to evaluate changes related to storage conditions. Proteomics 2006;6: Boguski MS, McIntosh MW. Biomedical informatics for proteomics. Nature 2003;422: White CN, Chan DW, Zhang Z. Bioinformatics strategies for proteomic profiling. Clin Biochem 2004;37: Baumann S, Ceglarek U, Fiedler GM, Lembcke J, Leichtle A, Thiery J. 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: Findeisen P, Sismanidis D, Riedl M, Costina V, Neumaier M. Preanalytical impact of sample handling on proteome profiling experiments with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Clin Chem 2005;51: Mengel-Jorgensen J, Sanchez JJ, Borsting C, Kirpekar F, Morling N. MALDI-TOF mass spectrometric detection of multiplex single base extended primers. A study of 17 y-chromosome single-nucleotide polymorphisms. Anal Chem 2004;76: