Plastoglobule Proteome

Originally identified in early electron microscopic studies as “osmiophilic globuli” in plastids (Greenwood et al., 1963; Bailey and Whyborn, 1963), plastoglobules are now recognized as distinct, lipoprotein structures. Although believed to play an active role in the plastid, their specific role is still unclear. To understand the function of plastoglobules, the identification of the proteins associated with them would be most helpful. This month's High Impact focuses on this question with the article by Ytterberg et al. (2006) titled “Protein profiling of plastoglobules in chloroplasts and chromoplasts. A surprising site for differential accumulation of metabolic enzymes.”

BACKGROUND

Present in all plastid types, plastoglobuli are lipoprotein particles, the shape and size of which change during development, plastid differentiation, and under stress conditions (for review, see Bréhélin et al., 2007). Etioplasts have large numbers of plastoglobules, and increase in plastoglobule size and numbers is also observed during of senescence (for review, see Hopkins et al., 2007). In chloroplasts, plastoglobules contain quinones, α-tocopherol, and lipids, and those in chromoplasts accumulate high levels of carotenoids (Deruere et al., 1994). Other than the structural proteins fibrillins (described below), little is known about the protein composition of plastoglobules.

A recent microscopy study demonstrated that plastoglobules arise from a “blistering” of the stroma-side leaflet of the thylakoid membrane, predominantly along highly curved margins (Austin et al., 2006), and are thus a single lipid layer studded with proteins. Unlike the globules that form on the endoplasmic reticulum, plastoglobules remain attached to the thylakoid membrane—even those at a distance are linked via other plastoglobules to the thylakoid (Austin et al., 2006).

Clustering of large groups of connected plastoglobules is also observed, particularly during senescence and under stress conditions. The trigger for the formation of the clusters, or indeed how plastoglobules arise, is unknown. One hint comes from a study in plants overexpressing fibrillin (plastoglobulin), a plastoglobule structural protein. An increase in clusters was observed in these plants (Rey et al., 2000; Simkin et al., 2007), suggesting that fibrillin (plastoglobulin) is involved in the formation of plastoglobule clusters. The significance of this clustering is unknown.

WHAT WAS SHOWN

Yield is a perennial problem for anyone studying proteins, and when identification of the proteins is the goal, this can be an even greater problem due to the potential loss of underrepresented proteins. Ytterberg et al. (2006) improved upon existing plastoglobule isolation protocols, increasing yield while shortening isolation time. To determine the composition of the plastoglobule proteome, plastoglobules were isolated from both wild-type Arabidopsis (Arabidopsis thaliana) and the clp2-1 mutants, which have reduced expression of the chloroplast protease ClpR2. Plastoglobules were also isolated from wild-type plants after exposure to 7 d of either high light or dark to investigate changes in plastoglobule protein composition under stress conditions. Thirty-two proteins were identified in the plastoglobules isolated from all of the conditions (wild type, stressed, and clpr2-1). As a developmental comparison, plastoglobules from red pepper (Capsicum annuum) chromoplasts were also isolated. Twelve of the 28 proteins identified from the pepper chromoplast plastoglobule were in common with the Arabidopsis chloroplast plastoglobule proteome.

Although some of the plastoglobule-associated proteins had been previously identified in thylakoid studies, most were unique to this structure and had not been found in either the stromal or envelope proteome. Considering the thylakoid origin of plastoglobules, the overlap with some thylakoid proteins is not unexpected. Also not unusual was the finding that only one out of 32 proteins had a predicted transmembrane domain since the plastoglobule lipid monolayer could not support such transmembrane domains. The plastoglobule proteome was also compared with proteins isolated from the prolamellar bodies from rice (Orzya sativa) etioplasts and little-to-no overlap was found between the two, demonstrating that the plastoglobule is distinct from prolamellar bodies.

The authors tentatively assigned the proteins to four functional classes: fibrillins (plastoglobulins; plastoglobule protein coat), lipid metabolism or fatty acid mobilization, quinone synthesis or regulation, and those with no obvious function. Fibrillins, found in almost all oxygenic photosynthetic organisms, were the most prominent proteins in the plastoglobule proteome. There are 11 or more fibrillin proteins in land plants associated with both plastoglobules and thylakoid membranes. They have one or more hydrophobic regions and are proposed to stud the surface of the plastoglobule to prevent them from coalescing (Deruere et al., 1994). Fibrillin expression increased following the high-light exposure, potentially corresponding to the observed increase in plastoglobule number during stress conditions.

Of the four enzymes possibly involved in lipid and hormone metabolism, one was identified as allene oxide synthase, the first enzyme of the lipoxygenase pathway leading to the formation of jasmonic acid. The function of the remaining three enzymes is unknown. Tocopherol cyclase, VTE1, orthologous to maize (Zea mays) SUCROSE EXPORT DEFECTIVE, was found in the plastoglobule proteome from both chloroplasts (Vidi et al., 2006; Ytterberg et al., 2006) and chromoplasts (Ytterberg et al., 2006). It is a key enzyme in vitamin E synthesis responsible for the production of γ- and δ-tocopherol. Plants defective in this enzyme are tocopherol deficient (for review, see Maeda and DellaPenna, 2007). Plastoglobules have been thought to be storage compartments for vitamin E, but the presence of VTE1 in both chloroplast and chromoplast plastoglobules suggests that plastoglobules might not only store vitamin E but also play an active role in its synthesis.

Previous studies have found that plastoquinones are a large component of plastoglobules and, to a lesser extent, so are phylloquinone (vitamin K1) and α-tocopherol quinone. Six proteins potentially involved in quinone synthesis—two members of the UbiE family and four ABC1 kinases—were identified in this study, suggesting also an active role for plastoglobules in quinone metabolism. The four ABC1 kinases have not been identified in Arabidopsis chloroplasts before but could possibly be involved in the regulation of quinone monoxygenases. As noted in Ytterberg et al. (2006), ABC1 proteins have been shown to be necessary for quinone synthesis in yeast mitochondria and Escherichia coli, and further study of these regulatory proteins should be exciting.

THE IMPACT

One of the “other proteins” identified in the plastoglobules proteome is At1g09340, an ortholog to the Chlamydomonas reinhardtii ribosomal associate protein RAP38 (Ytterberg et al., 2006). In Chlamydomonas, RAP38 copurifies with 70S chloroplast ribosomes (Yamaguchi et al., 2003). These proteins contain epimerase/dehydrase domains but their function is not clear. Recently, At1g09340 has been identified as CHLOROPLAST RNA BINDING (CRB) in Arabidopsis, a putative RNA binding protein (Hassidim et al., 2007). CRB is important for chloroplast function, as plants with mutations in this gene are smaller than the wild type and have paler leaves and abnormal chloroplast structure. The thylakoid membranes of CRB mutants have thicker grana stacks with decreased stroma lamella regions and a marked decrease in chlorophyll a. Interestingly, the circadian system is also altered with changes in the message levels of both oscillator and output genes, suggesting “the functional state of the chloroplast might be an important factor that affects the circadian system” (Hassidim et al., 2007, p. 551).

Another of the “other proteins,” At2g34460 with a predicted NAD-dependent epimerase/dehydrase domain, is a member of the Group II subfamily of Tic62 (Balsera et al., 2007). The best known member of this family, Tic62, is a member of protein translocon localized to the chloroplast inner membrane and acts as a redox sensor, possibly regulating protein import into the chloroplast (Stengel et al., 2008). In vascular plants, Tic62 is a “bimodular” protein with the N terminus of Tic62 binding to pyridine nucleotides, while the C-terminal module binds ferredoxin-NAD(P)-oxido-reductase. A comprehensive study of the Tic62-NAD(P)-related protein family by Balsera et al. (2007) compared all available sequences and found that although Tic62 related proteins are of ancient origin, the C-terminal ferredoxin-NAD(P)-oxido-reductase binding is present only in the Group I family of vascular plants.

CONCLUSION

Together, the studies by Ytterberg et al. (2006) as well as another on plastoglobule proteins by Vidi et al. (2006) demonstrate an active role for plastoglobules in synthesis and recycling of specific compounds in plastids. Additionally, plastoglobules have been suggested to be a new target for recombinant proteins in the chloroplast (Vidi et al., 2007).