Yungkang Lee (Presenter)
Berkshire Healthcare Systems
Bio: Dr. Yungkang Lee is a board certified clinical chemist at the Berkshire Medical Center in Pittsfield, MA. He completed his B.S. degree in chemical engineering in Taiwan and subsequently came to the US for his graduate study. He received his PhD in pharmacology and toxicology from the University of Southern California. Dr. Lee completed his fellowship training in clinical chemistry by a joint effort from Cedar-Sinai Medical Center, USC Department of Pathology, and LAC+USC Medical Center. Dr. Lee specializes in various forms of liquid chromatography-mass spectrometry for quantitative clinical test development as well as translational researches. He has authored many scientific publications in peer-review journals. In addition to his clinical services and researches, he also devoted his time in teaching selective topics in laboratory sciences to medical residents and medical technologists.
Authorship: Yungkang Lee (1,2,3), Keane KY Lai (2), SM Hossein Sadrzadeh (4)
(1) Berkshire Healthcare Systems, Pittsfield, MA (2) USC Department of Pathology, Los Angeles, CA (3) Precision Beacon, Austin, TX (4) Pathology and Lab Medicine, University of Calgary, Calgary, AB
| Short Abstract Successful translation during biomarker development with complex human serum proteome requires ultra-high resolution protein identification technologies. Here we devised a triple component serum proteome profiling method that integrated serum abundant protein depletion, MudPIT, and segment survey scan mass screening, by which its effectiveness to increasing peptide and protein coverages as well as run-to-run reproducibility at very low false discovery rate was studied, comparing to one-dimension shotgun proteomic approach. We further applied this method to biomarker development for diagnosis and prognosis of chronic myelogenous leukemia and chronic lymphocytic leukemia, due to the facts that clinical symptoms of chronic leukemia are often non-specific where the American Cancer Society estimates that at least one-fifth of the people with leukemia have been underdiagnosed. |
Long Abstract
Background: Successful translation from biomarker discovery with complex human serum proteome by global proteome profiling, to biomarker verification by accurate inclusion mass screening (AIMS) requires ultra-high resolution protein identification technologies. Here we devised a triple component (TC) serum proteome profiling method that integrated serum abundant protein depletion, multi-dimensional protein identification technology (MudPIT), and segment survey scan mass screening by which its effectiveness to increasing peptide and protein coverages as well as run-to-run reproducibility at very low false discovery rate (FDR) was studied, comparing to one-dimension shotgun proteomic approach. We further applied this method to biomarker development for diagnosis and prognosis of chronic myelogenous leukemia (CML) and chronic lymphocytic leukemia (CLL), due to the facts that clinical symptoms of chronic leukemia are often non-specific where the American Cancer Society estimates that at least one-fifth of the people with leukemia have been underdiagnosed.
Methods: Serum abundant protein depletion was achieved by immunoprecipitation with Agilent MARS Hu-7 or Hu-14 phase under stringent wash conditions to minimize sample loss. Peptides for shotgun proteome profiling were generated by treating sera with TCEP and iodoacetamide, followed by digestion with modified trypsin. Our MudPIT platform consists of online two-dimension capillary LC (partisphere SCX 5um, Aqua C18 3um, packed in 100um ID capillary with pulled emitter tip) coupled with LTQ Velos Pro. Peptide separation was achieved by 33 steps of increasing salt concentrations from 0 to 2M in 5% acetonitrile/95% water, and followed by 5%-80% acetonitrile gradient in 0.1% formic acid/water for two hours to each salt step. Peptide mass spectra obtained from segment survey scan mass screening, by which survey scans was divided into three m/z ranges of 300-800, 800-1200, 1200-2000 and the most abundant 5 ions from each survey scan was fragmented by CID, was fed to Agilent Spectrum Mill for library matching against human IPI database, where false discovery rate at protein level was kept at 1%. AIMS was executed with the capillary LC setup described above despite mass filters derived from discovery experiments were applied to biomarker verification with LTQ Orbitrap XL.
Results: Our triple component method identified 755 unique proteins among which only 67 proteins were identified by single peptides, in contrast to 163 proteins identified by one-dimension shotgun method (RP only, no serum abundant protein depletion and segment survey scan screening), among which 71 proteins were identified by single peptides. 696 out of 755 proteins identified by the TC method were reproducible from repeated analyses and relative quantitation from label-free repeated analyses demonstrated excellent correlation (R2=0.9964, y=0.9971x). In subsequent AIMS analyses, 645 of 755 originally identified proteins by TC method were verified, compared to 91 out of 163 proteins by the one-dimension shotgun method.
Conclusions: From this study we concluded that by removing serum most abundant proteins while minimizing sample loss, our triple component method not only effectively increases peptide and protein coverages, but also greatly enhances reproducibility on both protein identifications and relative quantitation from repeated analyses. This platform stability provides the essential basis for detecting sensitive protein fold changes between patient and control groups, which in turn ups the chance of successful translation of discovery data to validated clinical assays. It also reduces the biomarker attrition as well as contains the cost during biomarker development. A prospective study that aims to validate biomarker panels as well as to assess their diagnostic and prognostic values using isobaric labeling technique on CML and CLL is underway. At the conclusion of this project this biomarker development study will provide a non-invasive screening tool with quantitative definition on those subtypes of leukemia.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |