The Schmeing Lab
Structure and Function of Macromolecular Machines
The general goal of the lab is to understand how some of the large enzymes in the cell act to perform their important functions. These enzymes often require supramolecular organization and complex architecture to function. For example, both the ribosome and some nonribosomal peptide synthetases use more than 100,000 atoms to make peptide bonds, while the proteases that break these bonds can be very small. Of course, these assemblies require regulation, processivity and fidelity, which contribute to their increased size. Our lab investigates both the manner by which cellular machines achieve these roles, and the mechanisms of their principal functions. To do this, we combine X-ray crystallography, electron microscopy and biochemical techniques.
Nonribosomal Peptide Synthetases
Nonribosomal peptide synthetases (NRPS) are large macromolecular machines that, like the ribosome, catalyze peptide bond formation. Instead of making proteins, these enzymes produce a large variety of small molecules with important and diverse biological activity. For example, NRPSs synthesize anti-fungals, anti-bacterials, anti-virals, anti-tumourigenics, siderophores, and immunosuppressants, including classic therapeutics such as penicillin and cyclosporin, and modern billion-dollar antibiotics like daptomycin.
NRPSs use an elegant synthetic strategy whereby individual domains work together in sets called modules to make a build a peptide product. In a basic elongation module, the adenylation (A) domain selects the cognate amino acid and adenylates it, then attaches it to a prosthetic phosphopantetheine group on the peptide carrier protein (PCP) domain. The PCP domain transports the amino acid to the condensation (C) domain, which catalyzes peptide bond formation between this amino acid and the peptide attached to the PCP domain of the preceding module, thus elongating the peptide chain. Next, the PCP domain brings the elongated peptide chain to the downstream module, where it is passed off and further elongated in the next peptidyl transferase reaction. Each elongation module will add a new amino acid until the growing chain reaches the termination module, where a thioesterase (TE), will typically release the peptide by oligomerization, hydrolysis or cyclization.
In addition to the core domains, NRPS modules also often include tailoring domains, such as epimerisation, oxidase, ketoreductase and methyltransferase domains, to co-synthetically modify the peptide. The wide range of tailoring domains, combined with the >500 monomer substrates used by NRPSs, including D-amino acids, aryl acids, hydroxy acids, and fatty acids, allows NRPS products to access an astounding area of chemical space.
The straightforward logic of the NRPS synthetic cycle make them attractive for bioengineering experiments: Swapping of specificity-determining residues, domains or modules should result in predictable outcome in the structure of the NRPS product, and lead to novel or improved chemical and therapeutic properties. However, to make these approaches broadly successful, a deeper understanding of NRPSs is required.
In the Schmeing lab, we use multipronged approaches to study the structures and mechanisms of NRPS, both to potentiate the bioengineering of NRPSs to synthesize novel therapeutics and green chemicals and to provide a fundamental understanding of these elegant natural nano-factories.
Three subjects of particular interest to the lab are amide bond forming domains, tailoring domains and large NRPS proteins.
Amide bond forming domains
In basic NRPS modules, the C domain catalyzes the key catalytic event of NRPS function, amide bond formation to link substrate monomers and grow the peptide chain. The C domain is a pseudodimer of ~450 amino acids, with N- and C-terminal subdomains. The active site includes a HHxxDG sequence motif that sits at the bottom of a covered “canyon” formed by the two subdomains. With our first crystal structure solved in the lab, we described important movements that occur between the subdomains that may facilitate C domain catalysis.
Co-complexes between substrate analogues and enzymes can provide excellent insight into substrate specificity and chemical mechanism. Such complexes with the C domain have been elusive because the native substrates rely on transient protein-protein interactions for delivery to the active site and small molecule analogues have low affinities. To overcome this, we developed and used a novel chemical biology approach to capture complexes of substrate analogues bound to the condensation domain. These substrate analogs become covalently tethered near the active site, to mimic covalent substrate delivery by carrier domains. They are competent for reaction and behave similarly to native substrates. This chemical biology approach has enabled determination of co-complexes of the condensation domain for the first time. The complexes strongly suggest that the catalytic histidine’s principal role is to properly position the substrates for catalysis by nucleophilic attack, and allowed insight into the determinants of substrate specificity.
In some NRPS systems, the C domain is replaced by other domains, like an amide-forming transglutaminase domain, or the heterocyclization (Cy) domain. The Cy domain is especially interesting as it not only catalyzes amide bond formation, but also then catalyzes a two-step cyclodehydration between a thiol or hydroxyl side chain and the carbonyl of the newly-formed amide bond, making thiazoline, oxazoline or methyloxazoline rings in the peptide backbone. These heterocyclic rings are important for the bioactivity of peptides such as bacitracin, bleomycin, argyrin, yersiniabactin and colibactin. Strikingly, Cy domains have an extremely conserved DxxxxD motif in the place of the HHxxxDG. Recently, we and Dowling et al. determined the first structures of Cy domains. The structure, biochemistry, bioinformatics and mutational analysis pinpointed new active site residues far in sequences but close in space to the DxxxxD motif that is responsible for the cyclodehydration in Cy domains, allowed proposal of the catalytic mechanisms. Efforts to visualize the Cy domain with substrate and intermediate analogues, which would provide more interesting insight into these multi-functional domains, are underway.
Relevant Schmeing lab papers: ; ; ;
Tailoring domains in NRPSs allow products to access far more chemical space and are commonly found in these synthetases. For example, cyclosporin synthetase contains methyltransferase domains; daptomycin (Cubicin) synthetase, epimerization domains; bactitracin (BACiiM) synthetase, a heterocyclization domain. These domains enable key functionalities of the product by providing protease resistance, enabling novel interactions, improving affinity or allowing product to assume its active conformation. We use structural biology, chemical biology, biochemistry and biophysics to show how tailoring domains are embedded into megaenzymes and act in their synthetic cycle. For example, our studies with the formylation (F) domain of linear gramicidin synthetase visualized how a horizontal gene transfer of a sugar formyltransferase gene produced the F domain, and how it interacts productively with the core NRPS domains. Other tailoring domains of particular interest to the lab include oxidase and reductase domains.
Relevant Schmeing lab papers: ; ; ;
NRPSs need specialized termination domains to release the final product. This is often the thioesterase (TE) domain. The TE catalytic mechanism proceeds through two half reactions: In the first step, the linear substrate made by the enzyme is transferred from the final PCP domain to the TE domain active site serine, forming a covalently attached acyl-enzyme intermediate. In the second step, the TE domain can catalyze one of three alternatives: 1). The acyl-enzyme intermediate can be attacked by a water nucleophile and hydrolyzed to release a linear product. 2) The acyl-enzyme intermediate can be attacked by nucleophile in the intermediate itself, for example a hydroxyl in a side chain or the free N-terminal amino group, to release a cyclic product. 3) The acyl-enzyme intermediate can wait for the upstream modules to produce a new peptide on the final PCP domain to oligomerize the peptide. The TE domain later catalyzes the release of the oligomerized peptide by cyclization or hydrolysis. The decision to hydrolyze, cyclize or oligomerize appears to depend on both the identity of the TE domain catalyzing the reaction and the identity of the substrate of the reaction, and has been difficult to understand or predict. The mode of release and thus the form of the product (linear, cyclic, multimer, cyclic multimer) has a huge impact on the activity of the product: A cyclic antibiotic will NOT have potency or antibiotic activity if it is released as a linear peptide. Thus, this is an extremely important step to characterize, for the fundamental understanding of megaenzymes.
Relevant Schmeing lab papers:: ;
Large NRPS proteins
In their assembly-line logic and complicated catalytic cycle, NRPS domains and modules must work in concert to synthesize the nonribosomal peptide product. To investigate the holistic workings of NRPSs, we use X-ray crystallography, chemical biology tools, biophysical techniques and biochemical methods to elucidate their structures and functions. Good examples of these studies include our solving of four independent structures of the initiation module of the NRPS that synthesizes the clinically important antibiotic gramicidin. It provides fundamental insight into the initial stages of gramicidin synthesis, including the crucial formylation event, and into nonribosomal synthesis in general. The movements required for synthesis are staggering, with both the PCP domain translocating 61 Å and rotating 75° and the A subdomain rotating 180° in just one of these required transitions, to transport substrate between two distal active sites. The structures also highlight the great versatility of NRPSs, as small domains repurpose and recycle their limited interfaces to interact with their various binding partners.
We combined electron microscopy and X-ray crystallography to get even wider views of NRPSs in action. Our recent EM studies on a dimodular bacillibactin synthetase protein produced the first 3D views of a multi-modular NRPS. Accompanying pan-module X-ray structures begin to show the higher-order architectural features NRPSs require for to perform their syntheses. Further studies of large NRPSs by EM and X-ray crystallography are a major initiative in the lab.
Relevant Schmeing lab papers: ; , ; ; ;
The ribosome (past research)
The ribosome is the cell’s protein factory. It translates the genetic information in mRNA into protein, rapidly and with high fidelity, using aminoacyl-tRNAs as substrates. A large number of accessory protein factors are necessary for in vivo protein synthesis, and the interplay between these factors and the ribosome is extremely complex. Deregulation of protein synthesis in humans in associated with cancers, and many important antibiotics target the bacterial ribosome.
Some NRPS products
daptomycin (Cubicin ®)
A simplified version of the assembly- line style cycle used by a basic elongation module of an NRPS to processively make peptide products
Calcium-dependent antibiotic synthetase
An overview of bacterial protein synthesis
The 70S ribosome, tRNAs and EF-Tu
These funding bodies are gratefully acknowledged for their support
Canada Research Chairs
Centre for Structural Biology
FRQS - GRASP
A condensation state of LgrA
EM and crystallography of bacillibactin synthetase
A tailoring initiation cycle
The oligomerizing valinomycin sythetase TE