A novel method to capture methylated human DNA from stool: implications for colorectal cancer screening.
Several groups have detected methylated DNA markers in stool, including markers located on p16, MGMT, MLH1, SFRP2, HIC1, and vimentin genes (3-9). Methylated markers are attractive for screening because they occur with high frequency in early-stage neoplasia (10) and are predictable assay targets on gene promoter regions. DNA marker assay in stool is compromised, however, when crude stool DNA is used directly as a PCR template, because stool DNA composition is overwhelmingly bacterial and dietary, with human DNA making up <0.1% (2). Analyte enrichment would logically enhance detection of methylated DNA markers from stool.
Methyl-binding domain (MBD) is a functional region of McCP2 protein that specifically binds symmetrically methylated CpGs in any sequence context and is involved in mediating methylation-dependent repression (11). Cross et al. (12) described use of MBD protein bound to a nickel-agarose resin column to purify methylated CpG islands, and such a column has been used to enrich methylated CpG islands in lung cancer tissues (13). However, this method has not been applied to capture the extremely low amounts of methylated human DNA found within a complicated medium such as stool. Furthermore, the affinity of MBD binding to bacterial DNA, normally methylated at adenine and cytosine residues through Dam and Dcm transmethylases (14), is untested. Because of the large excess of bacterial DNA in stool compared with the minute amounts of exfoliated human DNA, even loose binding of MBD to methylated DNA sequences in bacteria could hamper this approach.
We describe a method using MBD to universally capture methylated human DNA from stool and thereby increase the sensitivity of this marker class for the detection of CRC.
Materials and Methods
This study included 4 related experiments: The 1st experiment was designed to test whether the MBD column could separate methylated DNA from a high background of bacterial DNA and unmethylated DNA. The 2nd experiment optimized a buffer panel for enriching methylated DNA with the MBD column. The 3rd experiment tested the effect of MBD enrichment on assay sensitivity in detecting trace amounts of methylated cancer cell DNA added to stool DNA. The 4th experiment explored whether MBD enrichment could increase assay sensitivity for detecting tumor-specific methylated markers in patient stools, which contained low amounts of human DNA. Approval of this study was obtained from the Institutional Review Board of Mayo Foundation.
Four cell lines derived from human digestive cancer were used, including 2 colon cancer cell lines (RKO and SW480), 1 gastric cancer cell line (RF-1), and 1 pancreatic cancer cell line (Capan2). RKO was grown in RPMI 1640 medium, SW480 and RF-1 in Leibovitz's L-15 medium, and Capan2 in DMEM. Media were supplemented with 10% fetal bovine serum, 100 000 U/L of penicillin, 100 g/L of streptomycin, and 2 mmol/L of L-glutamine. Cells were incubated at 37[degrees]C in the presence of 5% [CO.sub.2].
TISSUE AND STOOL SAMPLES
Fourteen paraffin-embedded colon cancer tissues were used to check the methylation status of vimentin in tumor. Stools from 8 patients with a corresponding methylated tumor were selected to test the MBD column. Stools from 6 individuals with normal colonoscopy findings were used as controls. All stools were collected before colonoscopy or surgery. None of the CRC patients had undergone chemotherapy or radiotherapy before stool collection. Any previous instrumentation or cathartic preparation had occurred more than 2 weeks before stool collection. A plastic bucket device was used to collect whole stool. Stools in sealed buckets were immediately transported to our laboratory and stored at -80 [degrees]C.
Tissue sections were examined by a pathologist who circled histologically distinct lesions to direct careful microdissection. Genomic DNA from both microdissected tissues and cell lines was extracted by use of a Qiagen DNA Mini Kit (Qiagen). Stool was homogenized in ASL buffer (1 g stool:10 mL buffer) and extracted with a QIAamp[R] DNA Stool Mini Kit (Qiagen).
Sodium bisulfite converts unmethylated, but not methylated, cytosine residues to uracil. DNA from tissue, cell line, and total stool was bisulfite modified using the EZ DNA Methylation Kit (Zymo Research). Because stool DNA samples after capture with MBD column typically contain <1 [micro]g DNA, whole DNA purified from each MBD elute was bisulfite modified (see below). We used 30 [micro]L of buffer to elute bisulfite-modified tissue and cell-line DNA and 10 [micro]L for stool DNA.
Bisulfite-modified DNA (1 [micro]L for tissue and cell DNA and 4 [micro]L for stool DNA) was amplified in a total volume of 25 [micro]L containing 1x PCR buffer, 1.5 mmol/L [MgCl.sub.2], 200 [micro]mol/L of each dNTP, 400 nmol/L of each primer, and 1.25 unit of AmpliTaq Gold polymerase (Applied Biosystems). Amplification included hot start at 95 [degrees]C for 12 min, 35 cycles for tissue and cell DNA or 40 cycles for stool DNA at 95 [degrees]C for 45 s, annealing temperatures for 45 s, 72 [degrees]C for 45 s, and a final 10-min extension step at 72 [degrees]C. The methylation-specific primers for vimentin were 5'-TCG TTT CGA GGT TTT CGC GTT AGA GAC-3' (sense) and 5'-CGA CTA AAA CTC GAC CGA CTC GCG A-3' (antisense), and the annealing temperature was 68 [degrees]C (4). The unmethylation-specific primers for vimentin were 5'-TTG GTG GAT TTT TTG TTG GTT GAT G-3' (sense) and 5'-CAC AAC TTA CCT TAA CCC TTA AAC TAC TCA-3' (antisense), and the annealing temperature was 60 [degrees]C. The methylation-specific primers for TPEF were 5'-CGG TAA AGA TTC GAG TAA GGA ACG T-3' (sense) and 5'-AAA ACA TCG ACC GAA CAA CGA CGT C-3' (antisense), and the annealing temperature was 65 [degrees]C. The unmethylation-specific primers for TPEF were 5'-GTT ATT TGG TAA AGA TTT GAG TAA GGA ATG-3' (sense) and 5'-AAA ACA TCA ACC AAA CAA CAA CAT C-3' (antisense), and the annealing temperature was 60 [degrees]C. Bisulfite-treated human genomic DNA and CpGenome[TM] Universal Methylated DNA (Chemicon) were used as positive controls for unmethylation and methylation, respectively. The methylation-specific PCR (MSP) amplicons for both vimentin and TPEF were located in CpG island regions without MseI cutting sites (TTAA).
PREPARATION OF MBD COLUMN
A previously reported method of column preparation was followed (12). MBD protein tagged with 6 histidines was expressed from a pET6HMBD plasmid-containing a MBD encoding DNA (a gift from Dr. Adrian Bird, Wellcome Trust Centre for Cell Biology, University of Edinburgh), which was cloned from rat McCP2 gene (12). Briefly, MBD was produced in BL21 star (DE3) pLysS (Invitrogen), and partially purified with a cation exchange resin, Fractogel[R] EMD [SO.sub.3] (M; EMD Chemicals), in a EconoPac Column (Bio-Rad). MBD protein was then coupled to an Ni-NTA Superflow (Qiagen), a nickel-agarose gel that specifically binds to protein with proteins tagged with 6 histidines, in a 10-mL Poly-Prep Chromatography Column (Bio-Rad) to generate the MBD column. Approximately 10 mg MBD was coupled per mL Ni-NTA Superflow.
Stool DNA and cancer cell DNA to be loaded on the MBD column for separation were first digested overnight with MseI (4 units/[micro]g), which recognizes the sequence TTAA, so the majority of CpG islands were kept intact, but other genomic regions were cut into short fragments. Digested DNA was then extracted with phenol/ chloroform/ isoamyl alcohol (25:24:1), precipitated in ethanol, and eluted in nuclease-free water.
TESTING SPECIFICITY OF MBD COLUMN
To test whether the MBD column separates methylated DNA from a high background of bacterial DNA, 5 [micro]g of cancer cell DNA and 50 /[micro]g of stool DNA (all MseI cut) were loaded on the MBD column in MBD buffer/0.1 mol/L NaCl (20 mmol/L HEPES, pH 7.9, 10% glycerol, 0.1% Triton X-100, 0.1 mol/L NaCl). The cancer cell DNA, which was used to simulate DNA exfoliated from cancers in digestive tract, was from a methylated cell line and an unmethylated cell line (2.5 [micro]g each; RKO with SW480, or RF-1 with Capan2). Vimentin is methylated in RKO, but not SW480; TPEF is methylated in RF-1, but not Capan2. DNA fragments bound on MBD protein were eluted using MBD buffers with gradient concentrations (0.2-1.0 mol/L) of NaCl.
Each eluate sample (5 mL) from the MBD column was concentrated with Amicon Ultra Centrifugal Filter Devices (30 000 MWCO, Millipore) to 200 [micro]L and then extracted with phenol/ chloroform/ isoamyl alcohol (25: 24:1), precipitated in ethanol, and eluted in nuclease-free water. DNA from each eluate was amplified with primers specific to Escherichia coli DNA, or bisulfite-treated for methylation analysis.
E. coli, a common bacterium in human stool, was used to represent fecal bacteria. Two sets of primers were designed to amplify E. coli DNA, 1 set targeting the dnaK gene and the other targeting a randomly selected undefined region. The primers specific for dnaK gene were 5'-GTG CCG GAT TAG CCA ACT TA-3' (sense) and 5'-GTG ACG ATT CCA GCC GTA CT-3' (antisense), and the primers for the undefined E. coli DNA region were 5'-ACT CCT GCG AAA CAT CAT CC-3' (sense) and 5'-CGG CAC CTT GCT AAG TCT TC-3' (antisense). We amplified 1 [micro]L of stool DNA in a total volume of 25 [micro]L containing 1X iQ[TM] Supermix (Bio-Rad), 200 nmol/L of each primer under the following conditions: 95 [degrees]C for 3 min, followed by 28 cycles of 95 [degrees]C for 30 s, 60 [degrees]C for 30 s, and 72 [degrees]C for 40 s, and a final 10-min extension step at 72 [degrees]C.
CAPTURE OF CANCER CELL DNA ADDED TO STOOLS
To test the sensitivity of the MBD column to capture methylated DNA in a stool model, trace amounts of cancer cell DNA (0, 2, 10, and 50 ng; RKO with SW480) were added to stool aliquots (1 g each) from a homogenized normal stool. Stool DNA was extracted and digested, as described above. Whole DNA from each stool aliquot was loaded on the MBD column in MBD buffer/ 0.52 mol/L NaCl. Loosely bound DNA was washed away in MBD buffer/0.6 mol/L NaCl. Tightly bound DNA was eluted in MBD buffer/1.0 mol/L NaCl, concentrated, extracted, and bisulfite-treated for methylation analysis targeting tumor-specific methylated vimentin gene (4).
CAPTURE OF METHYLATED HUMAN DNA FROM PATIENT STOOLS
The MBD column was used to capture methylated human DNA in clinical stool samples from 8 CRC patients and 6 control individuals described above. Stool DNA was extracted, digested, and loaded on the MBD column in MBD buffer/0.52 mol/L NaCl, and then washed with MBD buffer/0.6 mol/L NaCl. Methylated DNA bound to the column was retrieved with MBD buffer/1.0 mol/L NaCl and prepared as above for amplifying methylated vimentin.
REAL-TIME Alu PCR
Human DNA in patient stools was quantified using a real-time Alu PCR method, as we first reported (15). Primers specific for the human Alu sequences (sense, 5'-ACG CCT GTA ATC CCA GCA CTT-3'; and antisense, 5'-TCG CCC AGG CTG GAG TGC A-3') were used to amplify sequences approximately 245 by inside Alu repeats (15,16). Stool DNA was diluted 1:5 with nuclease-free water for PCR amplification. We amplified 1 [micro]L water-diluted stool DNA in a total volume of 25 [micro]L containing 1x iQ[TM] SYBR[R] Green Supermix (Bio-Rad) and 200 nmol/L of each primer under the following conditions: 95 [degrees]C for 3 min, followed by 23 cycles of 95 [degrees]C for 30 s, 60 [degrees]C for 30 s, and 72 [degrees]C for 40 s in a real-time iCycler (Bio-Rad). A calibration curve was created for each plate by amplifying 10-fold serially diluted human genomic DNA samples (Novagen). A melting curve was made after each PCR to confirm that only 1 product was amplified for all samples. Amplification was carried out in 96-well plates in an iCycler (Bio-Rad). Each plate consisted of stool DNA samples and multiple positive and negative controls. Each assay was performed in duplicate.
SEPARATION OF METHYLATED CANCER CELL DNA FROM HIGH BACKGROUND BACTERIAL DNA
This study was performed to determine whether methylated human DNA derived from a cancer cell line could be separated from a background of highly abundant bacterial DNA as would be found in stool. Stool DNA with added cancer cell DNA was loaded on the MBD column and then eluted using MBD buffers with stepwise increases in the concentration of NaCl. More tightly bound DNA requires elution with buffers at higher concentrations of NaCl. By either set of primers for the E. coli genome, E. coli DNA was amplified mostly in gradient MBD eluates with 0.4-0.6 mol/L NaCl, but less in eluates with >0.6 mol/L NaCl. Methylated vimentin and TPEF were found mostly in MBD eluates with >0.6 mol/L NaCl, and less in eluates with [less than or equal to]0.6 mol/L NaCl. Unmethylated vimentin was mainly in eluates with 0.4-0.72 mol/L NaCl, and unmethylated TPEF was detected in almost all eluates with 0.4-1.0 mol/L NaCl (Fig. 1).
[FIGURE 1 OMITTED]
OPTIMIZATION OF BUFFER PANEL FOR MBD ENRICHMENT
This study was performed to optimize a buffer panel for the application of MBD column in enriching methylated DNA in stool. When stool DNA without added cell DNA was loaded to the column in MBD buffer/0.1 mol/L NaCl and then sequentially eluted with MBD buffers with 0.2, 0.52, 0.6, and 1.0 mol/L NaCl, most E. coli DNA was eluted into buffers with [less than or equal to]0.6 NaCl (Fig. 2). Quantification with a photospectrometer revealed that <1% of total stool DNA was left in the 1.0 mol/L NaCl eluate.
Because a buffer cutoff at 0.6 mol/L NaCl separated methylated DNA from background bacterial DNA, we were able to use a selected sequence of MBD buffers for the enrichment process. Stool DNA was loaded to the MBD column in MBD buffer/0.52 mol/L NaCl to allow the binding of most methylated human DNA, but little bacterial DNA; an additional MBD buffer/0.6 mol/L NaCl further washed off loosely bound bacterial DNA and part of unmethylated human DNA; and a last MBD buffer/1.0 mol/L NaCl retrieved most methylated human DNA and a portion of unmethylated human DNA.
INCREASED SENSITIVITY OF DETECTING METHYLATED MARKER IN STOOL WITH ADDED DNA
This experiment was designed to test the effect of MBD enrichment on detection of trace amounts of methylated cancer cell DNA added to stool DNA. Trace amounts of cancer cell DNA were captured from stools using the MBD column and then tested with MSP by detecting methylated vimentin. With MBD enrichment, methylated vimentin was detectable in stool aliquots to which 10 and 50 ng RKO and SW480 cancer cell DNA were added, but not in those with 0 and 2 ng cancer cell DNA. Without MBD enrichment, methylated vimentin was not detectable in any stool aliquot with added DNA (Fig. 3).
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
ENHANCED DETECTION OF METHYLATED MARKER IN PATIENT STOOLS
This study was designed to test whether MBD enrichment increases the sensitivity of detecting methylated markers in patient stools. The effect of MBD enrichment on the clinical detection of CRC detection was evaluated on stools from patients whose matched CRC tissue demonstrated methylated vimentin. Vimentin was methylated in 8 of the 14 paraffin-embedded CRC tissues. The 8 stools with methylated tissues were captured with the MBD column and then tested with MSP by detecting methylated vimentin. With MBD enrichment, methylated vimentin was detected in 4 CRC stool samples with 4, 27, 408, and 832 ng human DNA, but not in the other 4 samples with 0.5, 1, 2, and 10 ng of human DNA; without MBD enrichment, methylated vimentin was detectable in only 1 CRC stool sample with 832 ng human DNA (Fig. 4). Methylated vimentin was not detected in 6 normal stools with or without MBD enrichment (data not shown).
This report describes a new method to capture methylated human DNA from stool samples using MBD protein chelated into a nickel-agarose matrix in a chromatography column. With the enrichment of the MBD column, methylated vimentin could be detected by MSP in a stool model with as little as 10 ng of added human cancer cell DNA and in patient stool with only 4 ng total human DNA. However, when MSP was directly applied to crude stool DNA, methylated vimentin was detectable only in a stool sample that contained a large amount of human DNA (832 ng). These results demonstrate that capturing methylated CpG islands with MBD can increase the detection sensitivity of stool-based methylated marker assays, even in stools with low concentrations of template human DNA.
[FIGURE 4 OMITTED]
Recently, we reported that the median concentration of human DNA in fresh CRC stools was 309 ng/g stool, with a range of 5-21 115 ng/g (15). Stools in the present study were selected to include multiple samples with low concentrations of human DNA; the median concentration was 24 ng/g and the range was 3-6027 ng/g. These concentrations corresponded to a range of 0.5-832 ng of human DNA loaded onto the MBD column. Thus, it would appear that the large majority of CRC stools in practice contain human DNA amplifiable with MBD capture. However, considering that human DNA concentrations are <100 ng/g stool in approximately 40% of CRC stools (15), the detection of methylation markers without enrichment may be limited in a substantial proportion of stools. Analyte enrichment by MBD capture or some other approach will likely be necessary to achieve adequately sensitive stool-based methylated marker assays intended for the early detection of colorectal neoplasia.
Another method of stool analyte enrichment has been developed in which specific DNA sequences are captured with complementary oligonucleotide probes (17-19). This method requires separate capture probes for each marker assayed. In contrast, because the methyl-CpG binding domain is a shared functional domain of MBD proteins (MeCP2, MBD1-4) (20) and binds to the large majority of methylated CpG islands (21), the MBD column method could theoretically capture most methylated CpG islands and so enrich multiple methylated genes in a single capture. But additional experiments using tumor-specific methylated genes are needed to fully test the universal capture potential of this method.
Of note, our results show that MBD protein has very low affinity to bacterial DNA, which has a higher CG:AT ratio than the human genome and is densely methylated at cytosine and adenine residues by Dcm and Dam methyltransferase (14). Bacterial DNA fragments could be eluted into buffers at concentrations <0.6 mol/L NaCl. Methylated human DNA was enriched approximately 100-fold by the MBD column without interference by the abundant bacterial DNA in stool.
Although the current study has demonstrated the feasibility of the MBD capture method to increase stool assay sensitivity of methylated vimentin for CRC detection, additional studies are needed to assess the broad use of this method with other methylated genes and its application in clinical practice. Results must be corroborated in large patient studies, and technological refinements to simplify this method would enhance its pragmatic use.
Grant/funding support: A generous grant from the Charles Oswald Foundation.
Financial disclosures: None declared.
Acknowledgments: We thank Dr. Adrian Bird for providing pET6HMBD plasmid, Ross Aleff for technical support, Dr. Sally Cross for valuable suggestions, Ann Kolb for colleting stool samples, and Jaci McCormick for clerical support.
Received January 25, 2007; accepted July 3, 2007. Previously published online at DOI: 10.1373/clinchem.2007.086223
(1.) Jemal A, Siegel R, Ward E, Murray T, Xu J, Smigal C, et al. Cancer statistics, 2006. CA Cancer J Clin 2006;56:106-30.
(2.) Osborn NK, Ahlquist DA. Stool screening for colorectal cancer: molecular approaches. Gastroenterology 2005;128:192-206.
(3.) Ahlquist DA, Klatt KK, Harrington JJ, Cunningham JM. Novel use of hypermethylated DNA markers in stool for detection of colorectal cancer: a feasibility study. Gastroenterology 2002;122(Suppl): A40.
(4.) Chen WD, Han ZJ, Skoletsky J, Olson J, Sah J, Myeroff L, et al. Detection in fecal DNA of colon cancer-specific methylation of the nonexpressed vimentin gene. J Natl Cancer Inst 2005;97:1124-32.
(5.) Petko Z, Ghiassi M, Shuber A, Gorham J, Smalley W, Washington MK, et al. Aberrantly methylated CDKN2A, MGMT, and MLH1 in colon polyps and in fecal DNA from patients with colorectal polyps. Clin Cancer Res 2005;11:1203-9.
(6.) Lenhard K, Bommer GT, Asutay S, Schauer R, Brabletz T, Goke B, et al. Analysis of promoter methylation in stool: a novel method for the detection of colorectal cancer. Clin Gastroenterol Hepatol 2005;3:142-9.
(7.) Leung WK, To KF, Man EP, Chan MW, Bai AH, Hui AJ, et al. Detection of epigenetic changes in fecal DNA as a molecular screening test for colorectal cancer: a feasibility study. Clin Chem 2004;50:2179-82.
(8.) Belshaw NJ, Elliott GO, Williams EA, Bradburn DM, Mills SJ, Mathers JC, et al. Use of DNA from human stools to detect aberrant CpG island methylation of genes implicated in colorectal cancer. Cancer Epidemiol Biomarkers Prev 2004;13:1495-501.
(9.) Muller HM, Oberwalder M, Fiegl H, Morandell M, Goebel G, Zitt M, et al. Methylation changes in faecal DNA: a marker for colorectal cancer screening? Lancet 2004;363:1283-5.
(10.) Zou H, Osborn NK, Harrington JJ, Klatt KK, Molina JR, Burgart LJ, et al. Frequent methylation of eyes absent 4 gene in Barrett's esophagus and esophageal adenocarcinoma. Cancer Epidemiol Biomarkers Prev 2005;14:830-4.
(11.) Nan X, Meehan RR, Bird A. Dissection of the methyl-CpG binding domain from the chromosomal protein McCP2. Nucleic Acids Res 1993;21:4886-92.
(12.) Cross SH, Chariton JA, Nan X, Bird AP. Purification of CpG islands using a methylated DNA binding column. Nat Genet 1994;6:236-44.
(13.) Shiraishi M, Chuu YH, Sekiya T. Isolation of DNA fragments associated with methylated CpG islands in human adenocarcinomas of the lung using a methylated DNA binding column and denaturing gradient gel electrophoresis. Proc Natl Acad Sci U S A 1999;96:2913-8.
(14.) Palmer BR, Marinus MG. The dam and dcm strains of Escherichia coli: a review. Gene 1994;143:1-12.
(15.) Zou H, Harrington JJ, Klatt KK, Ahlquist DA. A sensitive method to quantify human long DNA in stool: relevance to colorectal cancer screening. Cancer Epidemiol Biomarkers Prev 2006;15:1115-9.
(16.) Zijlstra A, Mellor R, Panzarella G, Aimes RT, Hooper JD, Marchenko ND, et al. A quantitative analysis of rate-limiting steps in the metastatic cascade using human-specific real-time polymerase chain reaction. Cancer Res 2002;62:7083-92.
(17.) Ahlquist DA, Skoletsky JE, Boynton KA, Harrington JJ, Mahoney DW, Pierceall WE, et al. Colorectal cancer screening by detection of altered human DNA in stool: feasibility of a multitarget assay panel. Gastroenterology 2000;119:1219-27.
(18.) Whitney D, Skoletsky J, Moore K, Boynton K, Kann L, Brand R, et al. Enhanced retrieval of DNA from human fecal samples results in improved performance of colorectal cancer screening test. J Mol Diagn 2004;6:386-95.
(19.) Traverso G, Shuber A, Levin B, Johnson C, Olsson L, Schoetz DJ, et al. Detection of APC mutations in fecal DNA from patients with colorectal tumors. N Engl J Med 2002;346:311-20.
(20.) Hendrich B, Bird A. Identification and characterization of a family of mammalian methyl-CpG binding proteins. Mol Cell Biol 1998; 18:6538-47.
(21.) Lopez-Serra L, Ballestar E, Fraga MF, Alaminos M, Setien F, Esteller M. A profile of methyl-CpG binding domain protein occupancy of hypermethylated promoter CpG islands of tumor suppressor genes in human cancer. Cancer Res 2006;66:8342-6.
 Nonstandard abbreviations: CRC, colorectal cancer; MBD, methyl-binding domain; MSP, methylation-specific PCR.
HONGZHI ZOU, JONATHAN HARRINGTON, RAFAELA L. REGO, and DAVID A. AHLQUIST *
Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN.
* Address correspondence to this author at: Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN 55905. Fax 507-266-0350; e-mail email@example.com.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Cancer Diagnostics|
|Author:||Zou, Hongzhi; Harrington, Jonathan; Rego, Rafaela L.; Ahlquist, David A.|
|Article Type:||Clinical report|
|Date:||Sep 1, 2007|
|Previous Article:||Characterization of diabetic nephropathy by urinary proteomic analysis: identification of a processed ubiquitin form as a differentially excreted...|
|Next Article:||Oxidative stress markers are associated with persistent atrial fibrillation.|