Structural Studies of Biopolymer Membrane Transport

The selective transport of specific substrates across the membrane is an essential part of biology. Organisms have evolved a basic mechanism for the transport of ions and small molecule such as sugars across the membrane. This mechanism is termed ‘alternating access’ and involves the alternating exposure of a substrate-binding site to either side of the membrane. While the molecular details of ‘alternating access’ have been revealed for a number of different transporters, this scheme seems infeasible for the transport of long biopolymers such as nucleic acids, proteins, and polysaccharides because ‘alternating access’ requires that the transporter forms a binding site that is large enough to bind the entire substrate. Here, I present novel data addressing the transport mechanism for two bacterial biopolymer transporter proteins: Bacterial Cellulose Synthase which synthesizes and secretes the polysaccharide, cellulose; and PrtD, an ABC transporter which is involved in the secretion of an extracellular protease.

Cellulose is a linear polymer of glucose units that forms a major component of plant cell walls as well as many bacterial biofilms. In bacteria, the components required for cellulose biosynthesis are encoded in a single operon minimally made up of the genes bcsA, bcsB, bcsC, and bcsZ. BcsA is an integral inner-membrane (IM) glycosyltransferase enzyme that is responsible for coupling the synthesis of cellulose with its transport across the IM. BcsB is a periplasmic protein with C-terminal IM anchor, and BcsB forms a complex with BcsA (BcsA–B) that is sufficient for in vitro cellulose synthesis. BcsZ is a periplasmic cellulase enzyme, and BcsC is an outer-membrane β-barrel that presumably forms a pore for the cellulose polymer to cross the outer membrane.

BcsA–B activity is stimulated by the bacterial signaling molecule, cyclic-di-GMP (c-di-GMP), which is a key regulator of biofilm formation. I present crystal structures of the c-di-GMP-bound BcsA–B complex in the presence and absence of UDP, a competitive inhibitor and substrate mimic. The structures reveal that c-di-GMP releases an auto-inhibited state of the enzyme. A salt bridge stabilizes one of the signature c-di-GMP-binding Arg residues in a position to tether a conserved ‘gating loop’ in front of the active site. The binding of c-di-GMP releases the tether and allows substrate to access the active site. Additionally, the UDP-bound structure reveals an additional role for the ‘gating loop’ in coordinating substrate at the active site. Functional experiments confirm the structural interpretation by revealing that disruption of the auto-inhibitory salt bridge by mutagenesis generates a constitutively-active cellulose synthase. The mechanistic insights presented here represent the first examples of how c-di-GMP allosterically modulates enzymatic functions.

To address the mechanism by which BcsA–B transports cellulose across the IM, I use in crystallo enzymology with the c-di-GMP-bound BcsA–B crystals. Because crystallized BcsA–B is catalytically active, these experiments provide a detailed molecular movie of the complete cellulose biosynthesis cycle, from substrate binding to polymer translocation. A substrate-bound structure of BcsA reveals the basis for substrate recognition, and structural snapshots show that BcsA translocates cellulose via a ratcheting mechanism, which involves the upward and downward movement of a ‘finger helix’ that contacts the cellulose polymer's terminal glucose. The movement of this finger helix is coupled to the insertion and retraction of the ‘gating loop’ in response to substrate binding and polymer extension, respectively. Thus, insertion of the gating loop pushes the ‘finger helix’ upwards, which then pushes the elongated polymer into BcsA’s TM channel. This mechanism is validated experimentally by tethering BcsA's finger helix, which inhibits polymer translocation but not elongation.

While the BcsA–B enzyme couples the synthesis of the biopolymer with its transport across the IM, polypeptide secretion requires that these two processes are carried out by separate machinery. Type 1 secretion systems (T1SSs) represent a widespread mode of protein secretion across the cell envelope in Gram-negatives. The T1SS is composed of an inner-membrane ABC transporter, a periplasmic membrane fusion protein (MFP), and an outer-membrane TolC-like barrel. These three components assemble into a complex spanning both membranes and providing a conduit for the translocation of unfolded polypeptides. Utilizing the Dickeya dadantii PrtDEF (DdPrtDEF) system, I show that ATP hydrolysis and assembly of a complete PrtDEF T1SS complex is necessary for protein translocation. Further, I present a 3.15Å crystal structure of Aquifex aeolicus PrtD (AaPrtD), a homologue of DdPrtD. The structure suggests a substrate entry window just above the transporter's nucleotide binding domains (NBDs). In addition, highly kinked transmembrane helices frame a narrow channel not observed in canonical peptide transporters and are likely implicated in substrate translocation. Combined, the AaPrtD structure suggests a polypeptide transport mechanism distinct from alternating access.