| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Proteomics and Protein Markers |
-Trypsin Inhibitor Heavy Chain 4 by a Surface-Enhanced Laser Desorption/Ionization-Based Immunoassay
1 Center for Biomarker Discovery, Department of Pathology, and4
Department of Pathology, Johns Hopkins Medical Institutions, Baltimore, MD.
2 Ciphergen Biosystems, Fremont, CA.
3 Department of Physiological Chemistry, School of Pharmaceutical Sciences, Showa University, Shinagawa, Tokyo, Japan.
aAddress correspondence to this author at: Department of Pathology, Johns Hopkins Medical Institutions, 419 N. Caroline St., Baltimore, MD 21231. Fax 410-502-7882; e-mail zzhang7{at}jhmi.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-trypsin inhibitor heavy chain 4 (ITIH4) have been identified by surface-enhanced or matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS or MALDI-TOF-MS) as potential disease markers. Methods: Previously, we developed a SELDI-based immunoassay that can simultaneously distinguish and quantify multiple isoforms/variants of a protein/peptide of interest. In this study, we used this high-throughput approach to quantify and characterize the extensive fragmentation within the PRR of human serum ITIH4 and determined its association with different disease conditions. The ITIH4-related fragments were first immunocaptured by use of beads coupled with peptide-specific antibodies. The eluates were then studied by SELDI-TOF-MS. In addition, freshly collected and immediately processed serum and plasma samples were used to analyze the ex vivo stability of these ITIH4 fragments.
Results: Human serum ITIH4 was shown to be extensively proteolytically processed within the PRR, and its fragmentation patterns were closely associated with different disease conditions. Fragmentation patterns were generally consistent with cleavages by endoprotease followed by exoprotease actions. Observed fragments changed little under different assay conditions or blood collection and processing procedures.
Conclusions: The fragmentation patterns within the PRR of human serum ITIH4 are associated with different disease conditions and may hold important diagnostic information. These fragmentation patterns could be useful as potential biomarkers for detection and classification of cancer.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-trypsin inhibitor heavy chain 4 (ITIH4)
1
is a plasma glycoprotein with a relative molecular mass of 120 000 that is expressed mainly in liver and that acts as an acute-phase protein in several species (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11). Unlike other members of the inter--trypsin inhibitor family (12)(13), ITIH4 is present in plasma as a single-chain protein because its COOH terminus lacks the consensus sequence (DPHFII) for bikunin assembly through the glycosaminoglycan bridges (2)(3)(12)(13)(14). The COOH terminus of human ITIH4 contains a proline-rich region (PRR; Gly611–Gln730; GenBank accession no. Q14624) that is 
120 residues in length and harbors species- and tissue-specific domains (2)(4)(15). ITIH4 is highly sensitive to plasma kallikrein and has been proposed to be a potential precursor for plasma kallikrein-induced bioactive peptides (1)(2). ITIH4 is readily cleaved to the N-terminal Mr 85 000 and the C-terminal Mr 35 000 fragments by plasma kallikrein. The N-terminal Mr 85 000 fragment is further cleaved to the N-terminal Mr 57 000 fragment and a putative Mr 28 000 fragment (1)(3). However, the putative Mr 28 000 fragment that covers 65% of the PRR has hitherto not been detected, suggesting that it in turn is rapidly cleaved into unidentified smaller fragments (Fig. 1A
). The exact biological function of ITIH4 in vivo remains unclear.
|
A cleavage fragment of ITIH4 from serum proteomic analysis is a candidate biomarker for detecting early-stage ovarian cancer (16). The fragment m/z 3272 (Met658–Phe687) falls in the same domain as the proposed kallikrein-released, bradykinin-like fragment (Pro662–Arg688) within the PRR. Preliminary analysis using a surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS)–based immunoassay showed that this fragment was processed differently in the sera of patients with certain types of cancer (17). The identified fragments might indicate different proteolytic processing of ITIH4 in disease.
Conventional immunoassays often cannot distinguish a parent protein from its cleaved fragments, although the latter could possess the greater diagnostic potential. Previously, we developed a SELDI-based immunoassay to examine the significance of the presence and quantitative changes of specific fragments of human serum ITIH4 (17). This method consists of 2 steps: immunocapture followed by ProteinChip array assay. In this study, we used this high-throughput approach to quantify and characterize the extensive fragmentation of human serum ITIH4 within the PRR and to further explore its association with different disease conditions.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
antibodies to human itih4
The 3 peptide antibodies (designated H4NT, H4CT, and H4CT-S) against human ITIH4 fragments were prepared by SynPep. For H4NT, a peptide was chemically synthesized as CKIPKPEASFSPR (see Fig. 1B
). For H4CT, a peptide was synthesized as CMNFRPGVLSSRQLGLPGPPDVPDHAAYHPF (17). For H4CT-S, a shorter peptide than that of H4CT was synthesized as CLGLPGPPDVPDH. All synthesized peptides had cysteine at the NH2 terminus, were conjugated to keyhole limpet hemocyanin, and were injected into rabbits according to a 69-day standard protocol used by SynPep. All peptide antibodies above were purified by both protein-A and affinity chromatography (Pierce Biotechnology). Both anti-ITIH4 rabbit antiserum (antiserum) and anti-ITIH4 mouse monoclonal antibody (1A4) were prepared by Dr. N.H. Choi-Miura (3)(5). Antibody 1A4 recognizes an epitope located on the C-terminal Mr 35 000 fragment of ITIH4.
sodium dodecyl sulfate–polyacrylamide gel electrophoresis and immunoblot analysis
Human serum samples from cancer or noncancer patients were diluted 40-fold in 1x phosphate-buffered saline (pH 7.4, 5.60 mmol/L Na2HOP4, 1.06 mmol/L KH2PO4, 154.0 mmol/L NaCl in culture-grade water) with 1x protease inhibitor cocktail (Roche). Each sample was incubated with U9 buffer (9 mol/L urea–20 g/L CHAPS–50 mmol/L Tris-HCl; 2 volumes of sample per 3 volumes of U9 buffer) for 30 min at 4 °C. The mixture was then heated for 5 min at 95 °C. The denatured samples were electrophoresed on 4%–15% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) gels, electroblotted on nitrocellulose membranes, and probed with the H4NT, H4CT, H4CT-S, antiserum, and 1A4 antibodies, respectively. The bound antibodies were visualized with horseradish peroxidase–conjugated secondary antibodies and enhanced chemiluminescence (Amersham Biosciences).
seldi-based immunoassays
SELDI-based immunoassays were performed as described previously with modification (17). This method consists of 2 steps: immunocapture followed by ProteinChip array assay (see the Methods in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol52/issue6). All immunocaptures were carried out in 96-well filter plates (Nalge Nunc International), and all ProteinChip array incubations were performed in 96-well bioprocessors (Ciphergen). All procedures, including sample dispensing, washing, and matrix application except for bead distribution, were performed by Biomek 2000 robots (Beckman Coulter Inc.). Both immunocaptures and ProteinChip array incubations were run in duplicate for each patient. Eluates of each individual patient sample were pooled before splitting in the next step. Placement of samples in the 96-well filter plates and 96-well bioprocessors was randomized. A mixture of serum samples containing known peptides or proteins was used as calibrator and quality-control samples throughout the experiment and accounted for 7%–8% of the total samples processed.
itih4 fragmentation stability assay
To test the stability of fragmentation under different assay conditions, we examined 3 factors during the immunocaptures: the effect of protease inhibitors (with vs without protease inhibitor), incubation temperature (room temperature vs 4 °C), and duration of incubation (2 vs 4 h). To assess the stability of fragmentation subjected to different sample preprocessing procedures, we repeatedly froze/thawed the freshly collected serum samples 10 times and incubated them at room temperature for different durations (1 h, 3 h, and overnight) before performing the experiment. To determine the effect of clotting procedure on the fragmentation, we checked 3 pairs of freshly collected and processed plasma and serum samples. All procedures of the SELDI-based immunoassays were performed as described above.
data analysis
Mass spectral data preprocessing was performed with Ciphergen ProteinChip software, Ver. 3.2.1. Spectra were internally calibrated based on a few known peaks, including m/z 3272.6, and baseline-subtracted by use of a fitting window of 8 times the expected peak width. Univariate analysis was performed with the Mann–Whitney U-test for each pairwise comparison. Differences with P <0.01 were considered statistically significant. ROC curve analyses were performed to evaluate the diagnostic potential of the observed fragments. All statistical analyses were performed with Statistica 6.1 (Statsoft) and Analyze-it 1.71 (Analyze-It Software).
computational analyses of sequences
Bioinformatics tools at the Expert Protein Analysis System (ExPASy) proteomics server of the Swiss Institute of Bioinformatics (SIB) were used for protein sequence and structure analyses (http://us.expasy.org/). More specifically, the PeptideMass tool was used to calculate the theoretical masses of peptides generated by the chemical or enzymatic cleavage of human ITIH4 protein and to predict the putative cleavage sites with different enzymes or reagents; the ProtParam tool was used to compute the amino acid composition and instability indexes of generated peptides.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). The Mr 28 000 fragment was still undetectable even with more specific peptide antibodies, which agreed with previous observations (1)(3) and further underscored that this fragment may be rapidly cleaved into smaller fragments. However, the differences in ITIH4 cleavage patterns among different samples observed here need to be explored further in a much larger number of experiments.
|
fragmentation within the prr of itih4
Because of the unique characteristic of the PRR of human ITIH4 and because all previous SELDI-TOF-MS or matrix-assisted laser desorption/ionization (MALDI)-TOF-MS–identified disease-associated fragments were derived from this region (16)(17)(18)(19), we were interested in further characterizing the fragmentation of ITIH4 within this region. Using a SELDI-based immunoassay, we quantitatively measured the relative abundance of the specific fragments of ITIH4 within the PRR. The SELDI-based immunoassay showed a linear quantitative response spanning a more than 16-fold difference in concentrations for a human serum specimen serially diluted in the binding buffer (see Fig. 1
in the online Data Supplement). Although the epitopes recognized by H4NT and H4CT are very close (only 13 amino acid residues apart; see Fig. 1B
), the SELDI-based immunoassay using beads coupled with respective antibodies revealed completely different spectra within the preferred mass range from the same serum sample (see top panel of Fig. 3A
). When we used H4CT, we detected not only the antigenic peptide (m/z 3273) but also a series of larger and smaller peptides with a potential C-terminal cleavage and a "sequential" N-terminal truncation (see Table 1
and the top panel of Fig. 3A
). This cleavage site suggested by our data matched the proposed first cleavage site of ITIH4 (Arg688–Arg689), which splits the intact protein into Mr 85 000 and 35 000 fragments (2)(14). All of these identified antigenic peptide-related fragments were located in the same domain of the kallikrein-released, bradykinin-like fragment (Pro662–Arg688). In contrast, when we used H4NT, we detected only 1 major peptide (m/z 3157) without any obvious further truncation as above, indicating that this fragment is more stable than fragments recognized by H4CT (see Table 1
and the top panel of Fig. 3A
). This fragment was located in a domain with potential alternative splicing (missing Gly621–Gln650), derived from the longer variant form (isoform 1) of human ITIH4 (4). These heterogeneous spectra indicated that Asn615–Arg616 and Arg644–Arg645 are preferential cleavage sites within the PRR of ITIH4, which is consistent with cleavage sites predicted by computational analyses (trypsin/chymotrypsin cleavages; data not shown). In addition, either with or without serum denaturation, H4CT-S recognized fragments different from those detected by H4CT (Fig. 3B
), which suggested that there might be a conformational folding within this range that partially blocks the epitope recognized by H4CT-S.
|
|
Using the SELDI-based immunoassay, we also tested the stability of the observed fragmentation within the PRR of ITIH4. We first examined the fragmentation pattern of ITIH4 under different assay conditions (Fig. 4A
). The experimental data revealed that the addition of protease inhibitors had no obvious effect on the intensity and pattern of fragmentation. A longer incubation did decrease the overall intensity but did not alter the pattern of fragmentation. Incubation at room temperature gave slightly lower intensity than incubation at 4 °C but had no effect on the overall pattern of fragmentation. We next evaluated the fragmentation pattern of ITIH4 subjected to different sample preprocessing procedures (Fig. 4B
). Repeated freezing/thawing of the freshly collected serum samples (10 times) and incubation of the same sera at room temperature for different durations (1 h, 3 h, and overnight) before experiments had no obvious effect on the intensity and pattern of fragmentation. Notably, neither immunoblot analysis nor SELDI-based immunoassay detected any cleavage or truncation from the intact Mr 120 000 protein in the freshly collected serum from a healthy control (serum 1) subjected to these procedures (data not shown). Finally, we compared the overall fragmentation patterns of ITIH4, using 3 pairs of freshly collected and processed plasma and serum samples (data not shown). The experimental results showed that there were no obvious alterations of intensity and pattern of fragmentation related to the clotting procedure. These data suggest that the observed fragmentation within the PRR of ITIH4 occurred mainly in vivo with negligible influence, if any, from the assay process, different sample preprocessing procedures, or the clotting procedure.
|
association between fragmentation within the prr of itih4 and disease conditions
To determine the association of fragmentation within the PRR of ITIH4 with different disease conditions, we used the SELDI-based immunoassay to analyze 178 human serum specimens. The representative overall spectrum acquired by use of pooled beads coupled with H4CT or H4NT is given in the bottom panel of Fig. 3A
. Scatter plots of intensity values obtained with the SELDI-based immunoassay for each of identified ITIH4 fragments across samples from patients with ovarian, breast, colon, prostate, or pancreatic cancer and diabetes as well as from healthy controls are shown in Fig. 2
of the online Data Supplement. The data presented in Table 1
further demonstrate that the fragmentation within the PRR of ITIH4 is indeed associated with different disease conditions. Some fragments were significantly increased or decreased in patients with different diseases, but not in colon cancer. Different diseases also showed different fragmentation patterns. For example, one or all of the fragments recognized by H4CT were significantly decreased in ovarian and pancreatic cancer, whereas several fragments recognized by H4CT were significantly increased in breast and prostate cancer and diabetes. The m/z 3156.6 fragment recognized by H4NT was significantly decreased in pancreatic cancer, whereas it was significantly increased in prostate cancer. ROC analysis results (Table 1
) indicated that most of these fragments demonstrated a noticeable power for disease detection even within this relatively small set of specimens.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because of the extreme susceptibility of ITIH4 to plasma kallikrein and the partial homology of the amino acid sequence to that of bradykinin released from high–molecular-mass kininogen by plasma kallikrein, human ITIH4 had been proposed to be a potential precursor of bioactive peptides (1)(2). However, how these fragments are further cleaved, which protease(s) are directly responsible for their cleavages, and the putative biological activity of bioactive peptide(s) remain to be determined experimentally. The identification of disease-associated fragments within the PRR of human ITIH4 by SELDI- or MALDI-based platforms is undoubtedly providing some important insights into these interesting issues. Our study revealed frequent cleavages/truncations in the "bradykinin-like domain", but few events in the "alternative splicing domain", consistent with the primary structure analyses obtained with the ProtParam tool, which classified the former as an unstable form and the latter as a stable form (data not shown). Similar truncation frequencies of fragments derived from these 2 domains of ITIH4 in human serum (20) and plasma (21) have been described. Some of these fragments may be bioactive peptide(s)/propeptide(s), and others are seemingly random internal fragments. However, the cleavage/truncation patterns were generally consistent with those initial cleavages by trypsin/chymotrypsin-like serine proteases (e.g., kallikreins) followed by subsequent terminal truncations resulting from aminopeptidase and carboxypeptidase activities. Similar 2-step proteolytic processes (endoprotease followed by exoprotease) that generate the most abundant layer of the serum/plasma peptidome have also been described (17)(20)(21). In addition, exoproteases play an important role in the regulation of bioactive peptides (22)(23)(24)(25)(26)(27). For example, some regulatory peptides are metabolized/modulated by membrane-bound peptidases such as dipeptidyl peptidase IV (DPPIV or CD26) and angiotensin-converting enzyme (22)(23)(24)(25). However, a thorough assessment of the biological activity of the peptides identified in this study demands extensive research in vitro and in vivo.
A key question is whether these identified fragments or fragmentation patterns represent valuable biomarkers or nonspecific epiphenomena (21)(28). Related questions include (a) Is serum suitable for this kind of experiment, with the possibility that the proteases activated during the clotting process alter the existing biologically relevant fragmentation patterns? (b) How should samples be collected and processed? (c) Is it necessary to block the inherent enzymatic activity at the time of sample collection and during the course of assays? The identified disease-associated fragments of human ITIH4 from serum were also observed in plasma, which suggested that the primary fragmentation of ITIH4 was not related solely to clotting (18)(19)(21). Exoprotease activities superimposed on the ex vivo coagulation and complement-degradation pathways contribute to generation of not only cancer-specific but also cancer-type–specific serum peptides (20). In this study, through the pairwise comparisons of freshly collected and immediately processed serum and plasma samples from individual patients, we saw no obvious alterations of fragmentation patterns of ITIH4 related to clotting. Moreover, once fragments were formed, they changed little under different assay conditions or when subjected to different sample preprocessing procedures. These results indicated that the observed fragmentations within the PRR of ITIH4 occurred mainly in vivo, not ex vivo. It is generally thought that use of protease inhibitors during sample collection will be necessary to asses whether ex vivo proteolytic activity contributes to the observed alterations (17)(21)(29), but not all agree (20). In our experiments, protease inhibitor cocktail was used to block the inherent enzymatic activity and minimize background experimental variation from ex vivo proteolytic activity.
Proteases have been extensively implicated in the development and progression of cancer (30)(31). The proteolytic processing of high-abundance host-response proteins actually amplifies the signal of potentially low-abundance biologically active disease markers such as proteases. Therefore, one may expect to find the more convenient and reliable blood protein(s)/peptide(s) that simply serve as an endogenous substrate pool for proteases as surrogate markers for detection and classification of cancer. The fragmentation within the PRR of human ITIH4 might be such a case. However, how the proteases may contribute mechanistically to the observed differences in the fragmentation patterns of human serum ITIH4 among the different disease conditions remains unanswered. In this study, the overall significant decrease in ITIH4 fragments in pancreatic cancer suggested that the fragmentation pattern of human ITIH4 might be a good candidate biomarker for this disease. Consistent with our result, one cleavage fragment within the PRR, derived from the shorter variant form (isoform 2) of human ITIH4, was also found to be decreased in pancreatic cancer (19). The m/z 2358 fragment was significantly increased in breast cancer and the m/z 3272, 2724, 2358, and 3156 fragments were significantly increased in prostate cancer, which are in good agreement with the results reported by Villanueva et al. (20) and our preliminary analysis (17). Previously, the m/z 3272 fragment was identified as a candidate biomarker (up-regulated) for detecting early-stage ovarian cancer (16). In this study, this up-regulation was not observed in late-stage ovarian cancers. This could be an indication that the fragmentation within the PRR of ITIH4 is related to the stage of this disease. We confirmed the down-regulation of the m/z 2627 fragment in late-stage ovarian cancers (17). Two fragments, m/z 3141 and 2358, were significantly increased in diabetes. Some identified disease-associated fragments in specific disease groups (17)(20) were not validated in this experiment. ROC analyses demonstrated the potential diagnostic value of these fragments. Assessment of diagnostic accuracy will require large-scale studies. SELDI and MALDI have limitations for biomarker discovery (21)(28). Discovery of truly disease-associated biomarkers also requires standardized sample collection and handling protocols, statistically sound study designs, and appropriate study populations selected based on clearly defined clinical utility. Furthermore, technology development for high-throughput sample preparation with low analytical variability is also critically needed to allow for the analysis of subproteomes of complex specimens (32).
In conclusion, in this study, we further quantified and characterized the fragmentation within the PRR of human ITIH4 and demonstrated the associations of the fragmentation of ITIH4 with different disease conditions. Our experimental results suggested that proteolytic processing patterns of human serum ITIH4 hold important diagnostic information that could be used as biomarkers for detection and classification of cancer.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
-trypsin inhibitor heavy chain 4; PRR, proline-rich region; SELDI-TOF-MS, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry; SDS-PAGE, sodium dodecyl sulfate–polyacrylamide gel electrophoresis; and MALDI, matrix-assisted laser desorption/ionization. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-trypsin inhibitor family from human plasma. J Biochem (Tokyo) 1995;117:400-407.-inhibitor family largely differs in the structure and synthesis of its prolin-rich region from rat to human. Biochem Biophys Res Commun 1998;243:522-530.[CrossRef][ISI][Medline]
[Order article via Infotrieve]-inhibitor family genes under normal or acute inflammatory conditions in rat. Arch Biochem Biophys 1998;350:315-323.[CrossRef][ISI][Medline]
[Order article via Infotrieve]-trypsin inhibitor heavy chain 4) is a new acute-phase protein isolated from cattle during experimental infection. Infect Immun 2004;72:3777-3782.-inhibitor family: from structure to regulation. Biochem J 1996;315(Pt 1):1-9.-trypsin inhibitor proteoglycan family—a group of proteins binding and stabilizing the extracellular matrix. Eur J Biochem 1998;252:339-346.[ISI][Medline]
[Order article via Infotrieve]-trypsin inhibitor family heavy chain-related protein (IHRP), a novel human plasma glycoprotein. J Biochem (Tokyo) 1995;117:14-18.-trypsin inhibitor family heavy chain-related protein. J Biochem (Tokyo) 1996;119:577-584.The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  J. M. Koomen, E. B. Haura, G. Bepler, R. Sutphen, E. R. Remily-Wood, K. Benson, M. Hussein, L. A. Hazlehurst, T. J. Yeatman, L. T. Hildreth, et al. Proteomic Contributions to Personalized Cancer Care Mol. Cell. Proteomics, October 1, 2008; 7(10): 1780 - 1794. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Gammerman, V. Vovk, B. Burford, I. Nouretdinov, Z. Luo, A. Chervonenkis, M. Waterfield, R. Cramer, P. Tempst, J. Villanueva, et al. Serum Proteomic Abnormality Predating Screen Detection of Ovarian Cancer The Computer Journal, April 15, 2008; (2008) bxn021v3. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Tang, K. Kitisin, W. Jogunoori, C. Li, C.-X. Deng, S. C. Mueller, H. W. Ressom, A. Rashid, A. R. He, J. S. Mendelson, et al. Progenitor/stem cells give rise to liver cancer due to aberrant TGF-{beta} and IL-6 signaling PNAS, February 19, 2008; 105(7): 2445 - 2450. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. A. Liotta and E. F. Petricoin Putting the "Bio" back into Biomarkers: Orienting Proteomic Discovery toward Biology and away from the Measurement Platform Clin. Chem., January 1, 2008; 54(1): 3 - 5. [Full Text] [PDF]
|
|
|
|
|
|
M. J. Bennett Untargeted Metabolomic Analysis Hits the Target Clin. Chem., December 1, 2007; 53(12): 2037 - 2039. [Full Text] [PDF]
|
|
|
|
|
|
J. F. Timms, E. Arslan-Low, A. Gentry-Maharaj, Z. Luo, D. T'Jampens, V. N. Podust, J. Ford, E. T. Fung, A. Gammerman, I. Jacobs, et al. Preanalytic Influence of Sample Handling on SELDI-TOF Serum Protein Profiles Clin. Chem., April 1, 2007; 53(4): 645 - 656. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
