Research in the van Wijk lab

  1. Comparative leaf development and cell-specific differentiation in C3 and C4 leaves of maize and rice; a systems analysis

  2. Proteolysis in plastid biogenesis and homeostasis with focus on the Clp protease system

  3. Macromolecular networks of plastid gene expression and biogenesis

  4. The role of lipoprotein particles (plastoglobules) in chloroplast biogenesis, metabolism and stress response

  5. Characterization of the leaf and chloroplast proteomes and specialized structures

  6. Tools for plant proteomics, mass spectrometry and the Plant Proteome Data Base (PPDB)

1. Comparative leaf development and cell-specific differentiation in C3 and C4 leaves of maize and rice; a systems analysis
Plants are classified as C3 or C4 species based on the primary product of carbon fixation in photosynthesis. C4 plants account for 20-30% of terrestrial biomass production, even if they represent only few percent of all plant species. C4-type plants, such as maize, have traits that greatly increase their efficiency of carbon-fixation especially when water or nitrogen are limiting. Therefore C4 species are attractive for biofuel production while introducing C4 traits into rice may help to raise productivity and alleviate food shortages. The key C4 traits are (i) specialization and cooperation of two leaf photosynthetic cell types (mesophyll and bundle sheath) for photosynthesis, (ii) enhanced movement of metabolites between cooperating cells, and (iii) high density of leaf venation.

We are studying C3 and C4 leaf development and cellular differentiation mainly through large scale comparative proteome analysis (Majeran et al (2005) Plant Cell; Majeran et al 2008 MCP; Friso et al 2010, Plant Physiology). Discoveries and putative key regulators and functions will be pursued using various functional essays and research strategies. Complementary analyses are carried out by collaborating groups. Supported by the National Science Foundation.

Morphology and anatomy of a developing maize leaf. Insert: Electron microscopy image of BS and M chloroplasts.

The distribution of major primary and secondary pathways, as well as of metabolic transporters, between bundle sheath and mesophyll cells of maize leaves. From Majeran and van Wijk (2009) Trends in Plant Science

2. Proteolysis in plastid biogenesis and homeostasis with focus on the Clp protease system
The plastid-localized Clp protease system is essential for plant growth and development. The plastid Clp system is of prokaryotic origin, but has greatly expanded in higher plants and consists of a 325 kDa barrel-shaped ClpPR complex and ATP-dependent chaperones that interact with this barrel as hexameric rings. During evolution, the plastid Clp system has acquired at least two specific proteins, assigned ClpT1,T2 and has maintained a homologue of the bacterial substrate 'adaptor' protein, ClpS.

Using reverse genetics, protein modeling, biochemistry and proteomics, we are characterizing structures, functions and substrates of this proteolytic Clp system and we take a 'system' view of the chloroplast and surrounding cell to understand functional and regulatory networks (see Peltier et al 2004 JBC; Rudella et al 2006 Plant Cell; Kim et al 2009 Plant Cell; Zybailov et al 2009 MCP). Supported by the National Science Foundation.

3-D homology models of plastid Arabidopsis proteins. Left and middle: The ClpPR core and its interactions with ClpT. Two out of the 14 ClpPR protein are shown in space filling models. Right. Interaction between the ClpPR core and a hexameric (from Pleture ring of ClpC1. Six ATP molecules bound to the ATP binding site in each of the ClpC proteins is shown as a space-filling white molecule. (From Peltier et al 2004, JBC)

Examples of Clp mutants in Arabidopsis and comparison to wild-type plants (From Kim et al 2009 Plant Cell)

3. Macromolecular networks of plastid gene expression and biogenesis
Chloroplast biogenesis requires a series of functionally and physically-connected gene expression and assembly processes. The goal of this project is to advance understanding of the network of DNA/RNA/protein interactions that underlie the biogenesis of the photosynthetic apparatus in chloroplasts. Macromolecular assemblies selected for study are anticipated to be rich in proteins of unknown function, and will be dissected through concerted genetic, proteome, and ribosome analyses. Maize is our primary experimental organism because its attributes make it especially well suited for these methods. Moreover it is an important crop species and it is a prime model for understanding C4 photosynthesis. Selective comparisons with Arabidopsis will facilitate extrapolation of results to both monocot and dicot species and will accelerate the determination of basic protein functions.

The project involves characterization of: i) new chloroplast biogenesis genes through a forward-genetic strategy that combines "next generation sequencing" with a deep collection of transposon-induced non-photosynthetic maize mutants, ii) Identifying protein components of immunopurified macromolecular assemblies involved in plastid biogenesis through high sensitivity mass spectrometry and iii) Use of genome-wide RNA/DNA coimmunoprecipitation assays to identify the RNA or DNA sequences associated with nucleic-acid binding proteins. This a collaborative project with Dr Alice Barkan at the University of Oregon, sponsored by NSF.

This figure shows the series of steps needed for chloroplast biogenesis (Blue arrows). Note that chloroplast biogenesis requires expression of both nuclear- and chloroplast-encoded genes. The involvement of nuclear-encoded proteins is indicated with black arrows.

4. The role of lipoprotein particles (plastoglobules) in chloroplast biogenesis, metabolism and stress response
Plastoglobules (PGs) are dynamic lipoprotein particles attached to the thylakoid membranes of chloroplasts, but they are also present in non-photosynthetic plastids, such as red-colored chromoplasts in fruits (e.g. pepper, tomato). These PGs are visible in transmission electron micrographs as electron-dense round or oval shaped particles or as rod-like, fibrillous structures. Although they were first thought of as passive lipid storage particles, a more dynamic and active role is suggested by their varying size in different conditions and recent proteome and metabolite analyses.

PGs in chloroplasts are formed by the swelling of the outer lipid leaflet of the thylakoid membrane, creating a hydrophobic interior comprised of phylloquinone (vitamin K1), plastoquinone, α-tocopherol (vitamin E), and neutral lipids. Proteome analyses identified some 35 structural proteins and metabolic enzymes (Ytterberg et al (2006) Plant Physiology; Brehelin, Kessler and van Wijk (2007) TIPS). Using different approaches (including molecular genetics, proteomics and metabolite analysis), we aim to understand the role of PGs in plastid biogenesis, membrane (re)modeling and stress responses.

The figure shows Electron microscopy images of a chloroplast from unstressed Arabidopsis plants and of a chloroplast of leaves exposed to higher light intensities for 3 days. This shows the increases number and size of the PGs upon stress. The right hand panel showed how we purify PGs and shows that these purified PGs are yellow in color.

5. Characterization of leaf subcellular proteomes and protein complexes
To understand the role of specialized chloroplast structures and organization of chloroplast protein complexes, we take various experimental and bioinformatics approaches. For instance, we develop and employed a three-phase partitioning system with organic solvents to characterize the hydrophobic thylakoid membrane proteome of Arabidopsis (Peltier et al (2004) JBC). A native gel approach was used to determine the oligomeric state of the soluble stromal proteome of Arabidopsis (Peltier et al (2006) MCP) (Fig. 1). This experimental data, combined with other information, was also use to describe to determine the actual Calvin cycle enzymes in Arabidopsis and determine their relative abundance and oligomeric state (Fig. 2). Currently, we are investigating the chromatin of the plastid chromosome, as well as the organization of Megadalton complexes, including chloroplast ribosomes.

Many proteins undergo post-translational modifications; these modifications can affect function in several different ways. We are employing Mass spectrometry, combined with retention time analysis, aided by various bioinformatics approaches to characterize these important protein modifications in plants (Zybailov et al (2009) Analytical chem.)

Fig. 1 shows a native gel of the soluble stromal proteome of Arabidopsis chloroplasts, with a annotion for a few of the identified proteins. Fig. 2 shows a reconstruction of the Calvin cycle enzymes with details about protein abundance. Only the underlined proteins are found at significant levels in chloroplasts. (see Peltier et al 2006 MCP).

6. Tools for plant proteomics, mass spectrometry and the Plant Proteome DataBase (PPDB)
Most cellular functions are carried out by proteins. Therefore knowing the set of expressed proteins in a cell, their subcellular localization and protein interactions and modifications is important. Proteomics is the systematic analysis of large sets of proteins and relies on the modern mass spectrometers, combined with the availability of sequenced genomes and bioinformatics tools. We develop and implement proteomics tools to solve question in plant biology. To that end we maintain and use several mass spectrometers (LTQ-Orbitrap (below), MALDI-TOF DE-STR) and other proteomics infrastructure (HPLC, FPLC, Spot picker, Automatic protein digester). We developed a bioinformatics workflow to process our mass spectrometry data (see workflow schema below). We employ our expertise also in various external collaborations.

Together with our Cornell colleague Dr Qi Sun, we also developed the Plant Proteome Database (PPDB; to provide an integrated resource for experimentally identified proteins in the key species Arabidopsis, maize (Zea mays) and rice (Sun et al. (2009) NAR).

The left hand figure show our LTQ-Orbitrap mass spectrometer interfaced with HPLC. The right hand figure shows our current workflow for processing LTQ-Orbitrap peptide data (from Zybailov et al 2009 Anal. Chem.)

Klaas Jan van Wijk
Department of Plant Biology, Emerson Hall
Cornell University
Ithaca NY 14853