The role of the cytoskeleton in signal transduction and gene expression in plants

Shunnosuke Abe

Laboratory of Molecular Cell Biology, Department of Biological Resources, College of Agriculture, Ehime University, 3-5-7 Tarumi, Matsuyama 790, Japan.

It is becoming indispensable to understand the structure and function of the cytoarchitecture, which is responsible for controlling cell cycles, differentiation, morphogenesis, the movement of cytoplasm and organelles, signaltransduction and gene expression, and many other important cellular processes. The cytoskeleton plays a central role in organization of such cytoarchitecture. In plant cells, in contrast to animal cells, fully-enlarged cells consist of a plasma membrane surrounding a thin smear of cytoplasm, which, in turn, encloses a massive vacuole, while the plasma membrane is almost always enclosed in a tough cell wall. Thus the part of the plant cell of interest to us, the cytoplasm and its organelles including the cytoskeleton, comprises only about10% of the total cell volume. The wall occupies another 10% and the vacuole as much as 80%. This cytoarchitecture has severely hampered studies on the plant cytoskeleton.

In plant cells, as in other eukaryotes, the cytoskeleton exists as an elaborate and highly dynamic network of filamentous and tubular structures, and the major cytoskeleton elements include microfilaments, which are composed of a double stranded helical array of the protein, actin; and microtubules of a helical array of a heterodimeric protein, tubulin, forming a hollow fiber. The third major component of the cytoskeleton, the intermediate filaments, are far less well understood in plants. These filaments may be assembled into larger aggregate structures, with or without associated factors, to exert their functions. In addition to these major cytoskeleton proteins, numerous other proteins are found associated with the cytoskeleton in higher plants.

Microtubules and microfilaments exist as extensive, interacting cytoskeletal networks that are clearly involved in dynamic aspects of cell behavior, and microfilaments often play a scaffolding role in this interaction. Disruption of microfilaments, but not microtubules, ruins the recovery of the spatial organization of cell division plane and cytoplasmic streaming, and therefore microfilaments act as "memory" of the spatial organization. Not only do close associations exist between microfilaments and microtubules, but also between both of these structures and endo-membranes, and these cytoskeleton-membrane complexes may be involved in cytoarchitectural coherence of organelles such as protein bodies, mitochondria, organelle trafficking or other functions. The cytoskeleton is also closely associated with the plasma membrane and with the membrane skeleton in higher plants. Use of various cytoskeleton and calcium inhibitors showed that the dynamics of the cytoskeleton association with membranes is at least in part calcium-dependent. Microfilaments attach to the plasma membrane indirectly. A spectrin-like polypeptide, which is a component of the membrane skeleton originally found in erythrocytes, is present in rice root plasma membrane preparations.

Additional evidence shows that cytoplasmic strands are connected to an integral protein in the plasma membrane and this junction appears to be dependent on arrays of microtubules. These connections may comprise a major component of the cytoskeleton-plasma membrane-cell wall (extra-cellular matrix) interface that has been the focus of much attention recently. This interface appears to be very important for signal transduction from the environment to the cell interior. The F-actin seems to be involved in at least some aspects of the gravity perception, hormone reception, pressure sensing, and infection. Firm evidence does exist for the localization of three kinases from the phosphatidylinositol (PI) pathway on the plant cytoskeleton. Furthermore, a protein identified through its PI kinase activity also acts as an actin-bundling protein and an elongation factor. A synthetic auxin (NPA)-binding proteins is now known to be cytoskeleton-bound, and may also play a role in the signal transduction through the cytoskeleton.

Wounding a plant in one region causes a 75% inhibition of protein synthesis in tissue 10 cm distant. In this, agents that disrupt cytoplasmic streaming (a process known to be microfilament-dependent) inhibit protein synthesis with similar kinetics, and both events (inhibition of protein synthesis and cytoplasmic streaming) did not take place in the presence of calcium chelators, thus indicating that Ca++influx was a necessary part of both responses. These findings, along with those from studies of cytoplasmic streaming, have been gathered together to form a working hypothesis on signal transduction through the cytoskeleton. There seem to be a calcium mechanism: Ca++ entering the cell via voltage-gated, mechanosensitive, or hormone-dependent channels binds to a cytoskeleton-associated Calcium-dependent protein kinase (CDPK). The CDPK then initiates a wave of phosphorylation through the cytoskeleton causing the phosphorylation of myosin, which inhibits cytoplasmic streaming, and of elongation factor, which inhibits translation. The slow recovery from the inhibition may arise from dephosphorylation caused by protein phosphatases activated by calmodulin which is also a cytoskeleton-associated enzyme in plants. Nuclei are also often found to be encircled by the cytoskeleton, and this may play a role in the regulation of gene expression in response to environmental cues, especially if there is continuity between the extra-cellular matrix (cell wall), the plasma membrane, the cytoskeleton and the nuclear matrix.

The evidence for an association between polysomes and the cytoskeleton association in plants is growing. The first biochemical evidence for cytoskeleton-bound polysomes (CBP) in plants was published 11 years ago in attempts to explain aggregation of membrane-bound polysomes by me. Further biochemical evidences for the existence of CBP in plants rested on the development of techniques which would permit the isolation of large fragments of the plant cytoskeleton (Abe and Davies, 1991) and these methods led to the finding that CBP are present in pea stems, pea roots and corn endosperm. Unlikely in animal cells, most evidence points to a role for the ribosome, but not via the mRNA and perhaps the nascent proteins.

Not only do plants have polysomes attached to the cytoskeleton, they also have polysomes attached to both the cytoskeleton and membranes -the cytoskeleton-membrane-bound polysomes (CMBP). These CMBP were characterized first in pea stems in my laboratory. In pea stems, CMBP constitute a major population of the cytoplasmic polysomes, and very few free polysomes are found. Protein bodies are unique to plants as organelles involved in the accumulation of proteins in the seed storage zones (endosperm or cotyledons). Both biochemical and fluorescence microscopic evidences indicate that polysomes involved in the synthesis of storage proteins which will be sequestered in the protein bodies of corn endosperm are attached primarily to the cytoskeleton and secondarily to the protein body membrane. In rice endosperm, protein bodies which store prolamin seem to attach to an endoplasmic reticulum with which the cytoskeleton is associated and the prolamine mRNA is also found in such membranes. These recent findings lead us to the conclusion that the classical description of the sub-cellular localization of polysomes (membrane-bound and free polysomes) is nolonger valid and that the role of the CBP and CMBP in gene expression should be seriously considered.

*NOTE: Also see "A Look beyond Transcription (Plant Physiology)"