Kidney International, Vol. 62 (2002), pp. 1461–1469 TECHNICAL NOTE Proteomic analysis of normal human urinary proteins isolated by acetone precipitation or ultracentrifugation VISITH THONGBOONKERD,KENNETH R. MCLEISH,JOHN M. ARTHUR, and JON B. KLEIN Core Proteomics Laboratory and Molecular Signaling Group, Kidney Disease Program, Department of Medicine, and Department of Biochemistry and Molecular Biology, University of Louisville, and Veterans Affairs Medical Center, Louisville, Kentucky, and Nephrology, Department of Medicine, Medical University of South Carolina, Charleston, South Carolina, USA; and Nephrology, Department of Medicine, Chiang Mai University, Thailand Proteomic analysis of normal human urinary proteins isolated The ability to examine the expression of a majority of by acetone precipitation or ultracentrifugation. urinary proteins simultaneously would represent a sig- Background. Proteomic techniques have recently become available for large-scale protein analysis. The utility of these nificant technical advance. Identification of urinary pro- techniques in identification of urinary proteins is poorly defined. teins may lead to an enhanced understanding of renal We constructed a proteome map of normal human urine as a physiology and pathophysiology, and thus lead to the reference protein database by using two differential fraction- discovery of novel “biomarkers” for diseases. ated techniques to isolate the proteins. In the post-genomic era, proteomic analysis has been Methods. Proteins were isolated from urine obtained from normal human volunteers by acetone precipitation or ultracen- developed for the large-scale study of proteins in micro- trifugation, separated by two-dimensional polyacrylamide gel organisms, cells and tissues [3]. Commonly, two-dimen- electrophoresis (2D-PAGE) and identified by matrix-assisted sional polyacrylamide gel electrophoresis (2D-PAGE) laser desorption ionization-time-of-flight (MALDI-TOF) mass is used for protein separation, and mass spectrometry spectrometry followed by peptide mass fingerprinting. Results. A total of 67 protein forms of 47 unique proteins (MS) is used for protein identification. Using these tech- were identified, including transporters, adhesion molecules, niques, up to 10,000 protein spots can be studied simulta- complement, chaperones, receptors, enzymes, serpins, cell sig- neously [4]. There are, however, limited proteomic data naling proteins and matrix proteins. Acetone precipitated more on normal human urine due to the low abundance of acidic and hydrophilic proteins, whereas ultracentrifugation fractionated more basic, hydrophobic, and membrane proteins. proteins. Concentrating methods, such as dye-precipita- Bioinformatic analysis predicted glycosylation to be the most tion and lyophilization, have been applied to urine [5, 6]. common explanation for multiple forms of the same protein. Although a number of proteins were identified, the ab- Conclusions. Combining two differential isolation tech- sence of transporters and cotransporters suggests a sig- niques magnified protein identification from human urine. Pro- nificant limitation to previous approaches [6]. teomic analysis of urinary proteins is a promising tool to study renal physiology and pathophysiology and to determine bio- The present study was designed to establish optimal markers of renal disease. techniques for the creation of a proteome map of normal human urinary proteins. Because of variability in physi- cal and chemical properties of the proteins, it is unlikely Quantitative and qualitative analyses of urinary pro- that a single protein isolation method will identify all of teins have been used to study renal physiology and as a the protein components [7]. We used two differential diagnostic tool in renal and systemic diseases. Western techniques, acetone precipitation and ultracentrifugation, blotting and other immunological methods are the most to isolate proteins. We determined that acetone precipi- successful techniques previously employed to identify uri- tated more acidic and hydrophilic proteins, whereas ul- nary proteins [1, 2]. These techniques are limited, how- tracentrifugation fractionated more basic, hydrophobic, ever, by availability of specific antibodies and by the and membrane proteins. These data indicate a usefulness ability to examine only few proteins in each experiment. of different isolation techniques to identify the complete urine proteome. Key words: protein analysis, transporter, adhesion molecule, chaper- one, receptor, post-translational modifications, biomarker, large-scale analysis. METHODS Received for publication March 13, 2002 The study was conducted in accordance with the ethi- and in revised form April 19, 2002 cal principles described by the Declaration of Helsinki Accepted for publication May 8, 2002 and was approved by the University of Louisville Human 2002 by the International Society of Nephrology Studies and Biosafety Committees. 1461 1462 Thongboonkerd et al: Proteomic analysis of human urine Fig. 1. Proteome map of human urine. The urine samples were fractionated by two different methods, acetone precipitation (A) and ultracentrifuga- tion (B). The proteins were separated by 2D-PAGE based on their differential pH value for the isoelectric point (pI; x-axis) and molecular weights (y-axis). The protein spots were excised and underwent in-gel tryptic digestion followed by MALDI-TOF MS. Peptide mass fingerprinting were performed to identify the proteins using the Profound search engine query to the entire NCBI protein database. Only the identities with significant Z scores (greater than 1.65) were included. Number labeling in the figure corresponds to the number in Tables 1 and 2. Urine collection the supernatants underwent microscopic examination Young healthy donors who had a history of normal with a hemacytometry counting chamber and there was renal function were recruited into this study. The donors no remaining cell or particle. The supernatants were had no acute or chronic medical illness and were not taking fractionated either by 50% acetone precipitation for 10 any prescription or over the counter medicines. Urine sam- min followed by centrifugation at 12,000 ϫ g for five ples (10 mL) were collected in 1 mL of protease-inhibitors minutes or by ultracentrifugation at 200,000 ϫ g for 120 cocktail [0.1 mg/mL leupeptin, 0.1 mg/mL phenylmethyl- minutes. The pellets were resuspended in 250 mmol/L sulfonyl fluoride (PMSF) and 1 mmol/L sodium azide in sucrose with 10 mmol/L triethanolamine. Protein con- 1 mol/L Tris, pH 6.8) and were fractionated immediately. centration of each sample was measured by spectropho- tometry using a Biorad protein microassay based on the Sample preparation method of Bradford [8]. All fractionation procedures were performed at 4ЊC. The samples were passed through 0.34 mm Whatman First dimensional 2D-PAGE chromatography paper and then centrifuged at 1000 ϫ An immobilized pH gradient (IPG) strip (pH 3 to 10; g for five minutes. After removal of cell debris and nuclei, Amersham Pharmacia Biotech Inc., Piscataway, NJ, USA) Thongboonkerd et al: Proteomic analysis of human urine 1463 Fig. 1. (Continued). was used for isoelectric focusing of acetone-precipitated buffer was 0.625 mol/L Tris/acetate. Electrophoresis was samples, and a mobile ampholyte tube gel (pH 3 to 10; performed with a maximum of 500 V and 20,000 mW Genomic Solutions Inc., Ann Arbor, MI, USA) was used per gel. for ultracentrifuged samples. IPG strips were loaded with 150 ␮g protein and tube gels were loaded with 75 ␮g Staining protein. The samples were focused for 24 hours for a The protocol was modified from the EMBL silver stain total of 100,000 volt-hours for IPG strips and for 17.5 described by Shevchenko et al [9]. The gel slabs were hours to reach 18,000 volt-hours for the tube gels. fixed in 50% methanol and 5% acetic acid for 20 minutes and washed with 50% methanol for 10 minutes and with Second dimensional 2D-PAGE deionized (18 meg Ohm) water for two hours. The gels The first dimensional gels were incubated in equilibra- were treated with 0.02% sodium thiosulphate for one tion buffer containing 112 mmol/L Tris base, 6 mol/L minute and washed with deionized water for one minute urea, 130 mmol/L dithiothreitol (DTT), and 4% sodium twice. Pre-chilled 0.1% silver nitrate was used for stain- dodecyl sulfate (SDS) before loading onto 10% homoge- ing at 4ЊC for 20 minutes. The gels were washed with neous, 22 ϫ 22 cm, duracryl gels (Genomic Solutions deionized water for one minute twice before developing Inc.). The upper running buffer contained 0.2 mol/L Tris, with 0.04% formalin in 2% sodium carbonate solution 0.2 mol/L Tricine and 0.4% SDS, and the lower running three times (30 sec, 90 sec and 90 sec). The developing 1464 Thongboonkerd et al: Proteomic analysis of human urine Table 1. The identified proteins from acetone precipitated urine Protein No. Identifier pI Molecular weight Adhesion molecules E-cadherin 1gi|1617084 4.50 91.19 E-cadherin 2gi|1617084 4.50 91.19 Activated leukocyte cell adhesion molecule 18 gi|16157538 6.57 56.93 Chaperones Chaperonin containing TCP1, subunit 6A (zeta 1) 27 gi|4502643 6.24 58.02 Enzymes Amylase, alpha 1A; salivary 8 gi|14722513 6.47 57.77 Amylase, alpha 2A; pancreatic 9 gi|4502085 6.60 57.71 Acid phosphatase, prostate 21 gi|16740983 5.89 44.54 Prostatic acid phosphatase precursor 22 gi|6382064 5.83 44.57 Hypothetical protein XP_015332a 30 gi|15302930 6.68 44.70 Hypothetical protein XP_015332a 31 gi|15302930 6.68 44.70 Alpha, alpha-trehalase