| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 North Carolina State University, W.M. Keck FT-ICR Mass Spectrometry Laboratory, Department of Chemistry, Raleigh, NC
2 Mayo Clinic College of Medicine, Department of Internal Medicine, Division of Cardiovascular Diseases, Rochester, MN
aAddress correspondence to this author at: North Carolina State University, W. M. Keck FT-ICR Mass Spectrometry Laboratory, 2620 Yarbrough Drive, Box 8204, Raleigh, NC 27695, Fax 919.513.7993, e-mail adam_hawkridge{at}ncsu.edu
To the Editor:
Depletion of abundant proteins from plasma and serum is an important initial step in many biomarker discovery platforms(1). Decreasing the concentrations of highly abundant proteins (e.g., albumin, IgG, and antitrypsin) facilitates the use of contemporary proteomics technologies, such as gel electrophoresis and mass spectrometry, for detection and identification of low-abundant proteins. Furthermore, decreasing abundant protein concentrations may also improve immunoprecipitation recovery efficiencies for targeted low-abundant species by decreasing nonspecific binding (i.e., shielding the antigen-binding domain) to the antibody and/or solid supports. A notable pitfall to depletion strategies is the potential for unintentionally removing low-abundant plasma or serum proteins. These low-abundant species may be bound specifically or nonspecifically to the depletion ligand, depletion target protein (e.g., carrier proteins), or the solid support(s). Thus, it is important to critically evaluate the effectiveness of abundant plasma protein depletion for enhancing the study of low-abundant protein biomarker(s).
We have been actively developing a targeted biomarker discovery platform for characterizing the circulating forms of B-type natriuretic peptide (BNP) that includes protein depletion strategies, immunoprecipitation, gel electrophoresis, isotope dilution (absolute quantification), and nanoflow liquid chromatography coupled to high-performance hybrid Fourier transform mass spectrometry(2). BNP has emerged as a very important biomarker for the clinical diagnosis and management of heart failure(3). Despite the increasing clinical use of BNP, very little is known about the secretion, posttranslational processing, and receptor binding of BNP at the molecular level(4). This gap in knowledge between the diagnostic and prognostic utility of BNP and the basic understanding of the molecular form(s) and function(s) of BNP is presently hindering the advancement of heart failure research. Increasing evidence in our laboratory(2) and others [reviewed by Goetze(4) and Liang et al.(5)] has shown that BNP circulates in heterogeneous forms that interfere with the commercially available BNP tests and potentially affect the pathophysiology of heart failure. These heterogeneous forms could have significant diagnostic, prognostic, and therapeutic value for the management of heart failure. The present obstacle to identifying circulating forms of BNP is the detection-limit gap that exists between the circulating concentrations of BNP, recovery efficiencies from plasma, and the high-performance hybrid Fourier transform mass spectrometry method. We sought to establish whether BNP-32 bound specifically or nonspecifically to the 6 most abundant proteins in plasma.
The general experimental set-up in this study involved tracking radiolabeled BNP-32 (125I-[Tyr]-BNP-32; specific activity = 1322.66 Ci/mmol: Phoenix Pharmaceuticals) in human plasma (single individual samples collected from donors who gave informed consent and were deidentified in accordance with the Mayo Clinic’s Institutional Review Board) through the Multiple Affinity Removal System (MARS) (4.6 x 100 mm MARS 6, Agilent Technologies). Unbound (%B = 0) and bound (%B = 100) fractions were collected from the MARS 6 eluent and the 125I-[Tyr]-BNP-32 in each fraction was measured (1480 Wizard 3 Gamma Counter, Spectrofuge) as shown in Fig. 1
. In the first experiment, 123 µL of plasma was combined with 123 µL of an 8 µg/L stock solution of 125I-[Tyr]-BNP-32 [in 110 mmol/L PBS (0.22 g NaH2PO4 + 1.280 g Na2HPO4 + 0.40 g NaCl + 0.01 NaN3 in 100 mL deionized H2O; pH adjusted to 7.4 with NaOH) and 154 mmol/L NaCl] at 37 °C, then 25 µL aliquots of the 125I-[Tyr]-BNP-32/plasma sample were taken at 0, 1, and 2 h, diluted with 75 µL buffer A (Agilent Technologies; more information on buffer A can be found at http://www.chem.agilent.com/en-US/Support/FAQs/Proteomics/MARS/Pages/KB000818.aspx), filtered, and injected. The purpose of this experiment was to determine the effects of 125I-[Tyr]-BNP-32 incubation time in plasma on its binding to high-abundant proteins. Results for the 3 time intervals sampled showed that 98% of the 125I-[Tyr]-BNP-32 was eluted in each of the 3 unbound fractions (Fig. 1
). The data clearly indicate that 125I-[Tyr]-BNP-32 does not bind to any of the top 6 most abundant proteins in plasma (albumin, IgG, antitrypsin, IgA, transferrin, and haptoglobin). A second set of experiments was performed to determine the reproducibility of the 125I-[Tyr]-BNP-32 recoveries shown in Fig. 1
. The results show that 125I-[Tyr]-BNP-32 does not bind to these proteins with mean (SD) 98% (2%) (n = 3) eluting in the unbound fraction.
|
High-abundant plasma protein depletion platforms have become increasingly popular for preparing plasma before mass spectral analysis(1). By measuring the radioactivity of the bound and unbound fractions from a MARS 6 column, we determined that 125I-[Tyr]-BNP-32 does not bind to the 6 most abundant proteins in plasma. Although we are cautious in assuming that the binding properties of BNP-32 and 125I-[Tyr]-BNP-32 are the same with regard to the 6 most abundant proteins, we can make reasonable inferences that the MARS 6 platform will not remove BNP-32. Another important issue that must also be determined is whether plasma protein depletion removes alternative molecular forms of BNP, such as glycosylated proBNP(5).
Acknowledgments
Grant/Funding Support: We thank the W.M. Keck Foundation and NIH-NHLBI (5RO1HL036634) for generous funding.
Financial Disclosures: None declared.
Disclosures: These data were presented at the 55th American Society for Mass Spectrometry Conference in Indianapolis, IN, June 3–7, 2007. Abstracts are published online (ASMS website) but are not refereed and cannot therefore be referenced in primary literature.
Acknowledgments: We thank Professor Jon Horowitz and Margaret Goralska (NCSU Veterinary Medicine) for access and assistance with the 
counter used in this study.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
