ヒト血中タンパク質のプロテオーム(Proteome)解析において、
前処理としての多量に存在する各種ヒト血漿タンパク質
[HAP(Highly Abundant Proteins of Plasma]除去方法の比較

近年におけるバイオマーカーなどの探索において、ヒト血中タンパク質のプロテオーム解析が重要になってきました。しかし非常に微量でしか存在しないタンパク質(LAP: Low Abundant Proteins)を検出するのは、アルブミンなどの多量で存在するタンパク質(HAP: Highly Abundant Proteins)などに覆われていたりして、とても困難でした。このことからHAPを除去する前処理方法として、今回ここで紹介いたします米国Dr. Hua Linのグループが、抗原／抗体反応に基づく新しい三種類の市販品を、従来から販売されている製品と比較しました。

Evaluation of Methods for Removal of Highly Abundant Human Plasma Proteins

Hua Lin, Thomas A.Shaler, Jing Wang, Melissa Chen, Erika Price, Jaya Kothule, Sophia Chen, and Christopher H. Becker, PPD Biomarker Discovery Sciences, LLC, Menlo Park, CA; ASMS(American Society for Mass Spectrometry)2007, Poster

Overview

Three new commercially available antibody-based depletion approaches have been compared to the earlier Agilent 7-antibody method for removal of high-abundance proteins in human plasma. The results showed good depletion efficiency and reproducibility for all three new methods investigated. Signals from low-abundance plasma proteins were significantly enriched with the extended level of antibody removal, especially for Genway two-antibody column, two?step depletion approach.

Introduction

Plasma protein concentrations have extremely wide dynamic range. The top 12 to 14 proteins make up 〜 96% of plasma by mass. The sensitivity for plasma proteomic analysis and biomarker discovery is often limited by interferences and charge suppression from high-abundance proteins. One way to overcome this dynamic range issue is to use affinity columns to specifically remove sets of high-abundance proteins. Agilent and GenWay Biotech/Beckman Coulter are lead companies developing commercial products for this application.

Agilent Technologies offers an antibody column designed to remove 7 proteins from human plasma in a single step. This column has been used in our biomarker discovery platform routinely and has proven to be beneficial and reliable. More recently, Agilent has expanded its antibody panel to include 14 antibodies. GenWay Biotec offers not only the onestep but also two-step removal products. The first IgY-12 or IgY-14 column removes 12 or 14 abundant proteins, mostly in common to the Agilent column. Then, a second antibody “SuperMix” column with antibodies produced against the flow-through proteins of IgY-12, is applied to the flow-through of the first IgY-12 or IgY-14 column, resulting in a further depletion. Here we present comparative evaluation of these depletion methods for detecting low-abundance human plasma proteins.

Table 1: Summary of the evaluated depletion methods

* removed protein

Methods

Experiment #1. For comparison of one-step removal methods, high-abundance proteins were depleted from a pooled EDTA human plasma sample (5 healthy donors from the Stanford blood bank) with a) the 7-antibody column (Agilent, Hu-7), b) the 14-antibody column (Agilent, Hu-14, prototype), and c) another 14-antibody column (GenWay, IgY-14).

Experiment #2. For comparison of the two-step removal method, three fractionated samples were purchased from GenWay Biotec, also prepared from an EDTA human plasma control sample. In this experiment, the 12-antibody column (Beckman/GenWay, IgY-12) was used for the first step of protein removal. The elution from the IgY-12 column bound protein represents the first sample investigated. For the second protein removal step, a second antibody column (GenWay, SuperMix) was used to further deplete the already IgY-12 depleted sample, with the elution of the “SuperMix” bound and the flow-through material representing the second and third samples investigated, respectively.

A small fraction of each sample investigated was used to determine total protein amount using BCA assay and the rest was concentrated, reduced, alkylated and digested by trypsin prior to a one-dimensional on-line capillary C-18 LC-ESI-MS or LC-ESI-MS/MS analysis. Approximately 10µg fractionated plasma proteins were injected into the mass spectrometer.

Label free MS profiling and MS/MS peptide/protein identification: Relative fractionation and depletion efficiencies were evaluated based on the MS signal of LC-ESI-TOF-MS (Waters, LCT) runs of the depleted samples. Peptide identification data were acquired on a LC-ESI-MS/MS system with a linear ion trap (ThermoFisher, LTQ). Each sample investigated was subjected to 4 separate MS/MS library runs;1) m/z range 300-700, ions 1-10; 2) m/z range 700-2000, ions 1-10; 3) m/z 300-700 range, ions 11-20; and 4) m/z 700-2000, ions 11-20.
HPLC gradient: 0 to 40% ACN with 0.1% formic in 46 min. Proteins/peptides were identified using database searching with Mascot software against a human database containing reverse sequences. The Mascot score cutoff for identification at individual peptide level was determined based reaching a local 30% false discovery rate using reverse sequence “hits.” A typical cutoff score is 38.

Results

Experiment #1: Comparison of one-step removal methods

Reproducibility

Figure 1: Overlay of UV traces @280 nm of antibody depletion of multiple injections with two difference size of new 14-antibody columns.

Depletion efficiencies for high-abundance proteins

Table 2: Comparison of one-step antibody removal efficiency for depleted proteins

LCT count from Hu-14 or IgY-14 column was scaled to the loading of Hu-7 column with equivalent plasma 1.25 µL pre-depletion loaded on LCT.

Signal enrichment for low-abundance proteins

Figure 2: Comparison of signal enrichment for non-depleted proteins with one-step depletion methods

For Hu-14 and IgY-14 columns, data were obtained with the LCT and LTQ loading of sample equivalent to 5 µL of non-depleted plasma, whereas for Hu-7 column, equivalent 1.25 µL pre-depletion plasma was loaded for comparison.

Table 3: Comparison of one-step antibody depletion methods for nondepleted low-abundance protein identifications, each sample investigated was subjected to 4 separate MS/MS runs with different m/z ranges.

Since this experiment, Agilent has improved the loading protocol for its 14-antibody column, new results are not included.

Experiment #2: Comparison of two-step removal fractions

Table 4: Summary of protein/peptide identification from three fractions of the IgY-12 plus “SuperMix” depletion method, starting with 250 µL plasma.
Each fraction was subjected to 4 separate MS/MS runs with different m/z ranges.

Conclusion

Several new antibody depletion methods have been compared with the earlier available Agilent 7-antibody removal method using pooled normal human plasma samples. The GenWay technology with two antibody column depletion (IgY-12 plus “SuperMix”) results in detection of human plasma proteins at lower concentration than the either of the Agilent or GenWay one-step 14-antibody methods. For example, several lower abundance proteins, such as insulin-like growth factor-binding proteins, Lselectin, fibrillin-1, galectin-3-binding protein, cartilage oligomeric matrix protein, and mannose-binding protein C were enriched significantly and detected via a one-dimensional LC-MS/MS analysis. However, the detection of low abundance proteins using the “SuperMix” depletion strategy could come at the expense of not monitoring some intermediate or higher abundance proteins such as orosmucoid and HDL which can serve as biomarkers in certain disease/drug situations. Thus if resources permit, more than one method or one fraction will result in a more complete coverage. One possible scenario for this complementary approach is to use a one-dimensional chromatographic approach for the first column’s depleted (eluted) material, and a one- or two-dimensional chromatographic approach for the “SuperMix” flow-through fraction.

The authors would like to acknowledge Agilent and GenWay for their valuable discussions.