Structure of the Plant-Conserved Region of Cellulose Synthase and Its Interactions with the Catalytic Core

The processive plant cellulose synthase (CESA) synthesizes (1→4)-β-D-glucans. CESAs assemble into a six-fold symmetrical cellulose synthase complex (CSC), with an unknown symmetry and number of CESA isomers. The CSC synthesizes a cellulose microfibril as the fundamental scaffolding unit of the plant cell wall. CESAs are approximately 110 kDa glycosyltransferases with an N-terminal RING-type zinc finger domain (ZnF), seven transmembrane α-helices (TMHs) and a cytoplasmic catalytic domain (CatD). In the CatD, the uridine diphosphate glucose (UDP-Glc) substrate is synthesized into (1→4)-β-D-glucans. The ZnF is likely to facilitate dimers in the CSC. Recombinant class-specific region (CSR), a plant specific insertion to the C-terminal end of the CatD is also known to form dimers in vitro. The CSR sequence is the primary source of distinction between CESA isoforms and class structure. Also within the CESA CatD is a 125-amino acid insertion known as the plant-conserved region (P-CR), whose molecular structure was unknown. The function of the P-CR is still unclear, especially in the context of complete CESA and CSC structures. Thus, one major knowledge gap is understanding how multimeric CSCs synthesize multiple chains of (1→4)-β-D-glucans that coalesce to form microfibrils. The specific number of CESAs in a CSC and how interactions of individual CESA isoforms contribute to the CSC are not known. Elucidating the structure-function relationships of the P-CR domain, and with the consideration of the ability of CSR and ZnF domains to dimerize, it is possible to more completely model the structure of the CSC.

Recombinantly expressed rice (Oryza sativa) secondary cell wall OsCESA8 P-CR domain purifies as a monomer and shows distinct α-helical secondary structure by circular dichroism analysis. A molecular envelope of the P-CR was derived by small angle X-ray scattering (SAXS). The P-CR was crystallized and structure solved to 2.4 Å resolution revealing an anti-parallel coiled-coiled domain. Connecting the coiled-coil α-helices is an ordered loop that bends back towards the coiled-coils. The P-CR crystal structure fits the molecular envelope derived by SAXS, which in turn fits into the CatD molecular envelope. The best fit places the P-CR between the membrane and substrate entry portal. In depth analysis of structural similarity to other proteins, and 3D-surface structure of the P-CR, leads to hypotheses that it could function in protein-protein interactions as a dimer, trimer or tetramer in the CSC, that it could form protein-protein interactions with CESA-interacting proteins, and/or modulate substrate entry through its N- and/or C-terminus. From modeling, hypothetically important residues within the P-CR or related to the P-CR through potential protein contacts were mutated in Arabidopsis thaliana AtCESA1 constructs. These constructs were expressed in the temperature-sensitive radial swelling (rsw) rsw1-1 mutant of AtCESA1 to test for complementation of growth phenotypes at restrictive temperatures. Preliminary experiments indicate that some mutated CESA1 sequences fail to complement the rsw1-1 phenotype, suggesting that specific functions of individual amino can be tested using this system.