Susan M. Lea

In chickens and humans, classical class I molecules of the major histocompatibility complex (MHC) have a spectrum of correlated properties, including cell surface expression and peptide repertoire. Understanding the impact of these promiscuous generalists and fastidious specialists is of considerable interest. The promiscuous BF2 molecule from the chicken B21 haplotype, BF2*21:01, binds a wide range of peptides by remodelling the peptide-binding site, allowing co-variation of the anchor residues at peptide positions P ₂ and P _c-2 , and binding of an anchor residue at Pc. Using in vitro refolding assays with peptides and peptide libraries, determining thermostability and crystal structures, and analysing a chicken B21 cell line by immunopeptidomics, we found that BF2*21:01 will accommodate many possible combinations at P ₂ and P _c-2 , as well as several hydrophobic amino acids at P _c . However, marked preferences for particular peptide lengths, particular amino acids at the three anchor residues, and particular combinations of amino acids at P ₂ and P _c-2 as well as at P ₃ and P _c-3 together lead to high frequencies of major peptides while still allowing the possibility of presenting a wide peptide repertoire. These simplifying principles may eventually allow predictions of pathogen peptides with stable binding for this iconic promiscuous class I molecule, as well as providing the data for more sophisticated peptide prediction methods.

Cell surface lipoproteins (SLPs) play crucial roles in the pathogenesis and cell biology of gram-negative bacteria, yet the route by which they reach the cell surface has remained unknown except in special cases ^1–3 . Here we show that in Bacteroidota bacteria, a phylum that includes prominent human commensals and anaerobic pathogens, SLP export is carried out by the outer membrane lipopolysaccharide (LPS) translocon. Isolation and structural characterisation of LPS translocon complexes from two Bacteroidota species reveals that they are massively enlarged and elaborated relative to the Escherichia coli paradigm with up to eight additional subunits located within, and on either side of, the outer membrane. We establish a role for the universally conserved core LptD subunit and four of the additional Bacteroidota-specific subunits in SLP export. Expansion of the LPS translocon in the Bacteroidota may allow export of the unusually complex SLPs found in this phylum. Our discoveries raise the possibility that SLP export is a conserved second function of the LPS transport pathway across gram-negative bacteria.

Protein trafficking and organelle identity depends on precise localisation of membrane proteins, anchored by small GTPases that bind to specific cellular compartments. Here we show that the rhomboid-like pseudoprotease TMEM115 defines an unexpected two-factor system with Rab6A to target the golgin TMF1. In the absence of TMEM115, TMF1 is delocalised from the rim of the Golgi apparatus; instead, it is retained in Golgi-targeting vesicles. TMEM115, TMF1, and Rab6A assemble into a ternary complex which, in striking contrast to canonical rhomboid-like mechanisms, is independent of transmembrane domain interactions. Disruption of the TMEM115-Rab6A-TMF1 complex affects the efficiency of trafficking of proteins from the Golgi; moreover, human disease-associated mutations target the interfaces of this ternary complex, thereby delocalising TMF1 from the Golgi. Finally, loss of TMEM115 in Drosophila causes abnormal Golgi morphology, and mouse knockouts undergo perinatal death. We propose that two-factor effector recruitment defines a mechanism for precise Rab GTPase-mediated membrane targeting.

In chickens and humans, classical class I molecules of the major histocompatibility complex (MHC) have a spectrum of correlated properties, including cell surface expression and peptide repertoire. Understanding the impact of these promiscuous generalists and fastidious specialists is of considerable interest. The promiscuous BF2 molecule from the chicken B21 haplotype, BF2*21:01, binds a wide range of peptides by remodelling the peptide-binding site, allowing co-variation of the anchor residues at peptide positions P2 and Pc-2, and binding of an anchor residue at Pc. Using in vitro refolding assays with peptides and peptide libraries, determining thermostability and crystal structures, and analysing a chicken B21 cell line by immunopeptidomics, we found that BF2*21:01 will accommodate many possible combinations at P2 and Pc-2, as well as several hydrophobic amino acids at Pc. However, marked preferences for particular peptide lengths, particular amino acids at the three anchor residues, and particular combinations of amino acids at P2 and Pc-2 as well as at P3 and Pc-3 together lead to high frequencies of major peptides while still allowing the possibility of presenting a wide peptide repertoire. These simplifying principles may eventually allow predictions of pathogen peptides with stable binding for this iconic promiscuous class I molecule, as well as providing the data for more sophisticated peptide prediction methods.

Mitotic centrosome maturation requires Polo/PLK1-dependent expansion of the pericentriolar material (PCM). In Drosophila , Centrosomin (Cnn) assembles a scaffold around mitotic centrioles through interactions between its PReM and CM2 domains. Here, we show that PReM adopts an autoinhibited helical hairpin conformation that prevents CM2 binding. Polo/PLK1 phosphorylation relieves this autoinhibition, enabling scaffold assembly, whereas phospho-blocking mutations disrupt PReM–CM2 binding in vitro and Cnn scaffold assembly in vivo. Potential functionally analogous domains have been identified in the human and C. elegans Cnn homologues CDK5RAP2 and SPD-5. We find that the human protein appears to share a structurally similar mechanism for scaffold assembly, but the worm protein does not. Consistent with this, deletion of these domains alters the dynamics of Cnn condensates in vitro, but has little effect on SPD-5 condensate dynamics. We conclude that Polo/PLK1 promotes mitotic centrosome assembly, at least in part, by relieving autoinhibitory intramolecular interactions.

Lipid droplets (LDs) are key organelles in cellular lipid homeostasis that form at the endoplasmic reticulum (ER) through a sequence of membrane rearrangements. While seipin emerged as an essential protein complex for LD biogenesis, how seipin-mediated LD formation proceeds beyond the initial step of neutral lipid nucleation remains unknown. Using a combination of in vitro and in-cell cryogenic electron microscopy (cryo-EM), ultrastructural expansion microscopy, molecular simulations and tailored genetic perturbations, we show that the seipin transmembrane domains undergo large-scale conformational rearrangements that define the architecture of the ER-LD interface and enable LD growth. Cryo-EM of purified Xenopus seipin revealed coexistence of two states: a compact closed conformation, consistent with early LD biogenesis, and an open conformation in which the transmembrane helices splay out laterally. Molecular dynamics simulations indicate that this open state induces local membrane curvature and promotes triacylglycerol accumulation. We identify conserved flexible linkers between the seipin luminal and transmembrane regions that act as mechanical hinges, enabling this conformational transition. We demonstrate that mutations in these hinge regions hinder seipin opening and affect LD formation in yeast and human cells. Analysis of native ER-LD contacts in human cells using light microscopy and cryo-electron tomography confirms that the seipin complex opens to establish stereotypical ~21-nm necks connecting the ER bilayer and LD monolayer. Moreover, we identify the liver-enriched microprotein SMLR1 as an inhibitor of this seipin conformational transition, providing a regulatory mechanism for seipin-dependent lipid storage in a tissue-specific manner. Together, these data establish seipin opening as a key structural rearrangement at the ER-LD interface that is essential for LD biogenesis and growth.

The twin-arginine translocation (Tat) system is a mechanistically unique protein transport pathway moving folded proteins across membranes. It is found in all domains of life and is essential for bacterial virulence and plant photosynthesis. The membrane proteins, TatA, TatB and TatC form a core complex to which substrate proteins bind, triggering the recruitment of additional TatA protomers to form the transport site. Here we present cryo-electron microscopy structures of the prototypical TatBC complex from Escherichia coli and the atypical complexes from Nitratifactor salsuginis and Myxococcus xanthus in a resting state, alongside TatAC substrate-bound TatBC and TatABC complexes from E. coli in the early stages of transport. These structures demonstrate that substrate proteins associate with the core complex solely through their N-terminal signal peptides. The Tat targeting sequences of the signal peptides make specific contacts with TatC, and the peptide body is clamped by TatB. The core complex contains highly tilted transmembrane helices that drive extreme local membrane thinning. On the basis of our structures and biochemical and functional analyses, we propose a model for the early steps in Tat transport.

Bacterial type III secretion systems export proteins across the inner membrane using the proton-motive force (PMF), but how proton flow is coupled to opening of the export channel is unknown ^XX,XX . Here we present single-particle cryo-EM structures at 2.5-4.2 Å resolution of the intact flagellar Export Apparatus from Salmonella enterica , obtained using optimized extraction conditions of the endogenous assembly that retain the complete transmembrane complex. The transmembrane domain of FlhA forms a nonameric funnel-shaped basket beneath the Export Gate, with each subunit harboring a buried pathway of conserved hydrophilic residues spanning the hydrophobic core of the membrane. FlhB is resolved for the first time within the intact gate, revealing helices that thread through the FlhA channel, completely sealing it at rest. A symmetry-free reconstruction captures one FlhA protomer hinged outward at the water-filled cavity, breaking the rotational symmetry. Together, these structures define the architecture of the proton-transducing element of the type III secretion system and suggest a rotary gating mechanism, with parallels to the F ₀ motor of ATP synthase, in which PMF-driven conformational cycling of FlhA drives regulated opening of the export channel.

Signal peptide peptidase (SPP) is an intramembrane aspartyl protease that cleaves transmembrane segments of diverse origin, such as remnant signal peptides of secretory proteins or transmembrane helices of tail-anchored proteins. Consistent with its ability to cleave multiple substrates, SPP is essential in mammals, and its activity is linked to a wide range of processes from immune regulation to protein quality control and cancer. Here we determine cryo-EM structures of human SPP in its apo form and bound to a signal peptide substrate. Together with molecular dynamics simulations and functional assays, our data show that SPP forms a constitutive homodimer that locally curves and thins the membrane, placing the conserved active site within the bilayer. Substrate engagement drives major conformational changes, including movement of a “latch” helix that positions the signal peptide for cleavage. The substrate adopts a tilted helix–unwound–β-strand architecture that defines the cleavage register independently of sequence. Steric exclusion by folded luminal domains or additional transmembrane helices explains selective processing of certain protein termini.