Polyethylene glycol fractionation improved detection of low-abundant proteins by two-dimensional electrophoresis analysis of plant proteome

Abstract

Poor detection of low-abundant proteins is a common problem in two-dimensional electrophoresis (2-DE) for separation of proteins in a proteome analysis. This is attributed partially, at least, to the existence of high-abundant proteins, e.g. ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) in plants. They engage a large proportion of the whole-cell proteins and thus prevent low-abundant proteins from being up-taken by immobilized pH gradient (IPG) strip, consequently making the latter poorly detectable by 2-DE. In this work, we report a straightforward protocol for preparation of whole-cell proteins through differential polyethylene glycol (PEG) precipitation aiming at elimination of Rubisco from plant protein samples. In comparison with 2-DE analysis of protein samples prepared using a conventional TCA/acetone method, a relatively high reproducibility of proteins was achieved using a PEG fractionation protocol in terms of protein yield and protein species. As expected, the large subunit of Rubisco was precipitated predominantly in the 16% PEG fraction. This allowed proteins of the Rubisco-containing fraction to be analyzed separately from those of other PEG fractions. After taking into account the overlapping protein spots among 2-DE gels of all fractions through image and statistical analyses, we detected with this protocol a total 5077 protein spots, among which ca. 80% are proteins undetectable with the TCA/acetone method, while the rest of proteins exhibited a significant increase in their abundance. This protocol was developed using Arabidopsis as a source of protein and thus may also be applicable to protein preparations of other plants.

Introduction

Proteomics is developing rapidly in the post-genomic era and has become increasingly desirable for exploration of gene function (Phizicky et al., 2003). Preparation of proteins at the proteome scale is however essential for proteomic analysis. Although many techniques have been used successfully for proteomic analysis, such as chromatography, isotope coded affinity tag (ICAT) and protein micro-arrays (Adam et al., 2002, Templin et al., 2003), two-dimensional electrophoresis (2-DE) is currently the only technique that can be routinely applied for parallel quantitative profiling of large sets of complex protein mixtures. To perform proteome analysis using 2-DE, sample preparation is crucial to achieve efficient protein resolution, which in turn affects 2-DE results in terms of image quality and protein species distribution (Lilley et al., 2002). It is noteworthy that the majority of proteins revealed by 2-DE are abundant “housekeeping” proteins that are present in copy number per cell from 10⁵ to 10⁶ (Patlerson and Aebersold, 2003, Görg et al., 2004). However, many low-abundant proteins, in particular those that are regulatory factors and receptor molecules are present in cells at relatively low concentrations (typically, 100 molecules per cell), and are thus usually not detectable (Gygi et al., 2000, Tirumalai et al., 2003). Development of efficient methods or strategies for protein sample preparation has therefore become an important goal in applications of 2-DE for proteome analysis.

Gene expression of a large copy number of high-abundant proteins is essential for living organisms to maintain their fundamental physiological processes. In plants, for example, ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is the most abundant protein and is involved in CO₂ fixation of photosynthesis (Gutteridge and Gatenby, 1995); indeed, it makes up more than half of the total leaf proteins in some species (Ellis, 1979). Since these high-abundant proteins usually do not participate in gene regulatory events, they become interfering factors in 2-DE proteomic analysis of regulatory proteins in two ways: (1) Rubisco co-migrates with the low-abundant proteins during 2-DE causing the latter to be undetectable (Corthals et al., 2000); (2) since Rubisco accounts for a large proportion of total protein sample and protein loading capacity for a given immobilized pH gradient (IPG) strip is limited, low-abundant proteins may only be up-taken by the IPG strip in a small quantity and hence hardly be visualized by 2-DE analysis. It is apparent that one of the considerations to improve the detection of low-abundant proteins by 2-DE is to minimize the high-abundant proteins.

Unlike nucleic acids that can be amplified by the polymerase chain reaction (PCR), low-abundant proteins from living organisms are hardly amplified due to the high dynamic range and diversity of their gene expression. Many strategies have been developed to promote detection of low-abundant proteins by 2-DE. For example, enrichment of proteins can be achieved by SDS–PAGE-based size fractionation (Sun et al., 2003) and a combination of complementary multidimensional technologies (Issaq et al., 2002) for proteome analysis. In addition, a three-phase partitioning (TPP) fractionation method was used on a new group of low abundant proteins in the proteome of the stripped thylakoid membrane (Peltier et al., 2004a). Another way to visualize as many proteins as possible in 2-DE is to perform subcellular fractionation for sub-proteome studies (Millar et al., 2001, Ferro et al., 2003, Pendle et al., 2005). As for diminishing the interference of Rubisco, the high-abundant proteins in plants and other organisms, selective precipitation (Sun et al., 2001), sucrose density gradient centrifugation (Peltier et al., 2004b) and FPLC anion-exchange chromatography (Wienkoop et al., 2004), etc. were also reported to deplete Rubisco successfully from different plant materials. However, many of these protocols are either laborious and time-consuming or require expensive equipments (Giavalisco et al., 2003).

In this work, we report a differential precipitation protocol using polyethylene glycol (PEG) to facilitate proteomic analysis by 2-DE. Using this protocol, Rubisco could be enriched into a specific PEG concentration fraction and thus detection of low-abundant proteins was improved significantly. PEG is a nontoxic water soluble synthetic polymer. It is understood that protein solubility (S) is depended on PEG concentration (C) following equation: log S = βC + constant (Atha and Ingham, 1981 and references therein). It can be seen from the equation that the linearity of C extends over a wide range of S. This advantage has made PEG an excellent differential precipitation reagent for purification of proteins from a variety of sources (Juckes, 1971). This protocol was characterized through detailed studies on protein precipitation efficiency using 2-DE image comparison and computational statistic analysis. The results showed that differential PEG fractionation is a straightforward method that can be used to visualize more proteins in 2-DE analysis of plant proteome.