How are tonoplast proteins degraded?

Abstract

Protein turnover is fundamental both for development and cellular homeostasis. The mechanisms responsible for the turnover of integral membrane proteins in plant cells are however still largely unknown. Recently, considerable attention has been devoted to the degradation of plasma membrane proteins. We have now studied the turnover of a tonoplast protein, the potassium channel TPK1, in fully differentiated Arabidopsis leaf cells and showed that its degradation occurs upon internalization into the vacuole. Here, we discuss the possible mechanisms and triggering events involved.

The plant cell vacuole undergoes functional and morphological changes during developmental programs such as the differentiation of meristematic cells, seed maturation and germination. These regulated events have been extensively studied and are reflected by changes in the protein content of the tonoplast and the vacuolar lumen. Thus, the full replacement of certain TIP aquaporins by other relatives in the course of germination^1-3 clearly indicates developmentally regulated tonoplast protein turnover. In eukaryotic cells, two major routes for protein degradation exist, involving either the cytosol-located proteasome or the proteolytic enzymes of the vacuolar lumen. Which of these routes is involved in tonoplast protein degradation? Upon soybean germination, a TIP isoform typical of the protein storage vacuoles (PSV) gradually disappears from the tonoplast of the PSVs of cotyledonary cells.⁴ During this process, vesicles that can be labeled with anti-TIP antiserum are detected by electron-microscopic immunochemistry within the vacuolar lumen, suggesting autophagic internalization of portions of the tonoplast followed by full degradation by vacuolar proteases.⁴ Similarly, in the growing root of germinating tobacco, TIP has recently been detected within condensed protein material in the lumen of PSVs that are being transformed into lytic vacuoles.⁵ Although these results do not provide definite proofs on the destiny of individual molecules over time, they strongly suggest tonoplast protein degradation within the vacuole during the transformation of PSVs into vegetative vacuoles.

Beside the replacement of the protein repertoire of a given compartment during programmed metabolic changes, proteins are constantly degraded in fully differentiated eukaryotic cells, to maintain proteostasis and to respond to environmental changes. We recently investigated the mechanism involved in the degradation of a model tonoplast protein in fully developed leaf tissue.⁶ When GFP was fused to the cytosolic C-terminus of the tonoplast Tandem-Pore Potassium channel AtTPK1 the resulting chimeric protein (TPK1-GFP) was correctly delivered to the tonoplast in transgenic Arabidopsis thaliana plants and still functional. By pulse-chase analysis, subcellular fractionation and protease protection experiments we demonstrated that intact TPK1-GFP progressively disappeared with a half-life time of 24 h and that a soluble degradation fragment, recognized by the GFP antiserum, accumulates in parallel in the vacuolar lumen. This degradation event is inhibited by treatment with brefeldin A, an inhibitor of Golgi-mediated protein traffic. We concluded that TPK1-GFP degradation requires the protein to reach the tonoplast and involves an internalization step (as the originally cytosol-exposed GFP tag accumulates within the vacuolar lumen), providing biochemical proof for degradation of a tonoplast protein within the vacuole.

Two possible models may account for this degradation. The first model (Fig. 1, panel A) is based on the mechanism involved in plasma membrane protein turnover,⁷ which is best characterized in yeast and animal cells but has been recently demonstrated to operate in plants as well.^8-12 Proteins are ubiquitinated and sorted from the plasma membrane via early endosomes to multivesicular bodies (MVB) where they are internalized. When MVBs fuse with the vacuole (lysosome in animal cells), the internalized vesicles end up in the vacuolar lumen where they are degraded. In an effort to detect ubiquitinated TPK1-GFP, we purified the fusion protein by immunoprecipitation with polyclonal or monoclonal anti-GFP antibody and performed detection with two different anti-ubiquitin antibodies. Only one combination resulted in signal detection. We also performed mass spectrometry analysis on purified TPK1-GFP, but we could not detect ubiquitinated peptides. Our results suggest that only a minor fraction of TPK1-GFP may be ubiquitinated, which is expected as the ubiquitinated molecules should be readily degraded.

Figure 1.

Two models for the pathway of tonoplast protein degradation within the vacuole. (A) multivesicular body-mediated pathway; (B) direct pathway. In both panels, numbers indicate the sequence of steps. The known endocytic pathway for the degradation of plasma membrane proteins is also illustrated. V, vacuole; PM, plasma membrane; MVB, multivesicular body; TGN/EE, Trans-Golgi Network/early endosome.

In the second model (Fig. 1, panel B), the internalization event would occur directly at the vacuolar membrane through invagination of a portion of the tonoplast into the vacuolar lumen. The resulting double-membrane vesicle is then degraded by the vacuolar lipases and proteases. Although the central lytic vacuole is often perceived as a rubber balloon swollen with the internal osmotic pressure, the tonoplast may produce invaginations or tubular structures encasing cytoplasmic strands across the vacuole lumen.¹³ Internal spherical structures similar to those observed in PSVs have also been reported in vegetative vacuoles. They were originally described in enlarging cells of cotyledons and hypocotyls of young Arabidopsis thaliana seedlings expressing a fusion between the tonoplast aquaporin γ-TIP and GFP.¹⁴ These so-called “bulbs” appear as double membrane invaginations originating from the tonoplast and protruding into the vacuolar lumen, where they encircle vacuolar content. Since their first description, vacuolar luminal structures were further studied and several functions were suggested.^15-17 Upon salt stress, TIP1;1-GFP, but not other TIP isoforms, relocalizes in part in bulb-like structures; the authors hypothesized that this may be related to specific degradation of certain TIP isoforms or to stress-induced reshaping of vacuoles.¹⁸ To date, however, a clear function of bulbs still needs to be demonstrated. We commonly observed the presence of luminal vesicles in our transgenic plants expressing TPK1-GFP.⁶ These structures are usually brighter than the tonoplast, indicating that they could be limited by a double membrane or contain a specific accumulation of our fusion protein (Fig. 2). We hypothesized that these membrane invaginations can represent a pathway for the direct internalization of tonoplast proteins. Compared with the classical “vesiculation” mechanism, a double membrane structure reduces the quantity of soluble material engulfed and degraded.

Figure 2.

Membrane structures within the vacuole labeled by TPK1-GFP. Leaf guard cells from transgenic Arabidopsis expressing TPK1-GFP under the CaMV 35S promoter were observed by epifluorescence microscopy. Notice that the recombinant protein appears more concentrated in the membranes within the vacuole than in the tonoplast. This may indicate either actual higher concentration in the lipid bilayer or the presence of multiple, stacked bilayers.

A mechanism of direct membrane budding from the tonoplast to the cytosol has not been identified, whereas invagination of the tonoplast has instead been observed in many studies, as described above. Therefore the latter model for the degradation of TPK1-GFP seems favored. However it should be noticed that membrane flow from the tonoplast to not well characterized “autolysosomes” has been detected in starved tobacco cells.¹⁹ It has been suggested that this may occur if the outer membrane of double membrane vesicles generated by invagination fuses to the tonoplast, thus releasing vesicles with a single membrane into the cytosol.²⁰ These could in theory be delivered to MVBs and enter the pathway for vacuolar degradation. This would be an intermediate mechanism between those shown in Figure 1.

But what could trigger degradation? This is still a highly debated general question in proteostasis research.²¹ Degron sequences have a role, but perhaps the major determinant is that most protein molecules inevitably acquire with time folding defects that make them recognized as malfunctional, triggering their disposal: in cells, individual proteins tend to be at the concentration limit of their aggregation²² thus making in vivo denaturation a quite frequent event, and those with faster turnover have higher propensity to aggregate.²³ The observation that brefeldin A inhibits the fragmentation of TPK1-GFP⁶ indicates that either misfolding is stimulated by the permanence of the protein at its compartment of action (the tonoplast) or the degradation machinery is unable to dispose of “aged,” misfolded TPK1-GFP if the protein is not at the final destination. This is consistent with the fact that a number of TPK1 domain exchange misfolded constructs defective in correct assembly and retained in the ER were not subjected to rapid degradation.⁶ Judging from fluorescence microscopy data, the same seems to hold true for Arabidopsis and rice TPK1 mutated constructs produced in other laboratories.^24,25 This would point to a different selectivity between “proximal” and “distal” protein quality control along the secretory pathway of plant cells, a feature that has been recently observed in mammalian cells.²⁶ The molecular recognition features of this difference are still unknown.

Which subcellular compartments are turned over as a unit, and which have different rates of turnover among their components is another important issue of proteostasis research. It is difficult to imagine that all proteins of the tonoplast have the same probability to acquire with time folding defects. Therefore, if misfolding is the determinant for degradation the process must be selective and the cargo must be specifically enriched in the invaginating or budding portion of the tonoplast. While γTIP-GFP is present at higher density in vacuolar bulbs than at the tonoplast, GFP-AtRab75c labels the tonoplast but is excluded from bulbs.¹⁴ Provided that bulbs are the structures leading to degradation, this supports a selective process. Comparisons of the degradation rates of tonoplast proteins present in the same cell, as well as biochemical approaches to follow the rate of acquisition of defects that could trigger and signal degradation will be needed to cast light on this issue.

Maîtrejean M, Wudick M M, Voelker C, Prinsi B, Mueller-Roeber B, Czempinski K, Pedrazzini E, Vitale A. Assembly and sorting of the tonoplast potassium channel AtTPK1 and its turnover by internalization into the vacuole. Plant Physiol. 2011;156 doi: 10.1104/pp.111.177816.