Piecing together nonribosomal peptide synthesis

https://doi.org/10.1016/j.sbi.2018.01.011Get rights and content

Highlights

  • Nonribosomal peptide synthetase (NRPSs) are modular biosynthetic factories.

  • Most functional states of NRPS modules have now been visualized at high resolution.

  • Massive conformational changes occur within a module at different stages of catalysis.

  • X-ray and EM studies hint at a lack of regular higher order multimodular architecture.

Nonribosomal peptide synthetases (NRPSs) produce peptide products with wide-ranging biological activities. NRPSs are macromolecular machines with modular assembly-line logic, a complex catalytic cycle, moving parts and multiple active sites. They are organized into repeating sets of domains, called modules. Each module contains all functionality to introduce a building block into the growing peptide, many also perform cosynthetic tailoring. Structures of individual domains have provided insights into their catalytic mechanisms, but with one exception, larger NRPS proteins were refractory to structure determination. Recently, structure determination succeeded for four multi-domain NRPS proteins: an alternative formylating initiation and two termination modules as well as a large cross-module construct. This review highlights how these data, together with novel didomain structures, contribute to a holistic view of the architecture, domain–domain interactions and conformational changes in NRPS megaenzymes.

Introduction

Nonribosomal peptide synthetases (NRPSs) are a family of microbial megaenzymes that produce natural products that are useful to society as therapeutics (antibiotics, antivirals, antitumours, and immunosuppressants) and green chemicals (agricultural agents, emulsifiers, siderophores, and research tools) (Figure 1a) [3]. NRPSs typically synthesize their products through amide bond formation between aminoacyl (or other acyl) monomers. Their architecture is dissimilar to the more famous peptide maker. Whereas ribosomes use the same active sites for each amino acid added to the ribosomal peptide, NRPSs typically employ a dedicated set of enzyme domains for each amino acid added to the nonribosomal peptide. This set of domains is termed a module, and the synthetic strategy dictates that, normally, the number and specificity of the modules correspond to the length and sequence of amino acids in the peptide product. NRPSs can consist of a single polypeptide of between 2 and 18 modules, with a mass of ∼220 kDa to 2.2 MDa, or be split over multiple proteins that assemble non-covalently (Figure 1) [4, 5, 6, 7].

Within a module, the domains work together to incorporate the incoming amino acid into the growing peptide (Figure 1b) [5, 8]. A basic elongation module contains a condensation (C) domain, an adenylation (A) domain, and a peptidyl carrier protein (PCP) domain. The A domain selects and adenylates the cognate amino acid, then attaches it by a thioester link to a prosthetic phosphopantetheinyl (PPE) group on the PCP domain. The PCP domain then transports the amino acid to the C domain, which catalyzes amide bond formation between this amino acid and the peptide attached to the PCP domain of the preceding module, elongating the peptide by a single residue. Next, the PCP domain brings the elongated peptide to the downstream module, where it is passed off and further elongated in the next condensation reaction. Once a PCP domain has donated its peptide, it can accept a new amino acid from the A domain and participate in the next cycle of assembly-line synthesis. Initiation modules lack the C domain, and termination modules usually contain a thioesterase (Te) domain, which releases the peptide by cyclization or hydrolysis. A canonical organization of a basic NRPS is A–PCP-(C–A–PCP)n–Te (Figure 1c). Additionally, NRPS modules very often have tailoring domains, including oxidase, reductase, epimerization, ketoreductase, aminotransferase and methyltransferase domains, and the action of these domains is incorporated into the catalytic cycle of the module [9]. NPRSs can alternatively end in a reductase [10] or terminal C domain [11•, 12]. This wide range of tailoring domains, combined with the over five hundred monomers that can be used as substrates, including D-amino acids, aryl acids, hydroxy acids, and fatty acids, allows nonribosomal peptides to occupy a diverse area of chemical space [13].

Starting with Conti's determination of an A domain in 1997 [14], each core NRPS domain has been structurally characterized [5, 15]. The A domain has a large Acore (∼450 amino acids) portion with binding sites for ATP and substrate amino acid, and a small Asub (∼100 amino acids) portion that changes position depending on functional state [14, 16, 17, 18, 19, 20]. (Acore and Asub are also called the large/N-terminal and small/C-terminal subdomains [21].) The C domain is a ∼450 amino acid V-shaped pseudodimer of chloramphenicol acetyltransferase folds, with an active site at the middle of a tunnel connecting binding sites for donor and acceptor PCP domains. The PCP domain is an oblong 4-helix bundle of ∼80 amino acids, homologous to fatty acid synthase and polyketide synthase acyl carrier protein domains, with the PPE attached to a conserved serine at one end. The Te domain is a ∼275 amino acid α/β hydrolase domain with an active site topped by a variable ‘lid’ region. Studies investigating the catalytic mechanisms of each individual reaction in the NRPS cycle are reviewed elsewhere [5, 15, 16, 22, 23]. Furthermore, some tailoring domains have been structurally characterized, several didomain structures have been determined, and a landmark study in 2008 from Marahiel's lab produced a snapshot of an entire termination module (reviewed in [5]). However, a most intriguing aspect of NRPS function is how these domains transiently and productively interact during the catalytic cycle and how domains are built up into modules and full NRPS megaenzymes [8]. Multiple structures of large constructs of NRPSs in recognizable functional states are required to understand these megaenzymes. Over the last ∼2 years, several studies, including on full modules and module-sized constructs of NRPSs, have started to reveal the catalytic cycle in the context of the large macromolecular machine. In this review, we follow a hypothetical nascent peptide from initiation to termination, with special emphasis on insight gained from these recently-determined structures of large NRPS proteins.

Section snippets

Start it off

Minimal initiation modules contain only A and PCP domains. However, it is not uncommon for NRPSs to start with more complex initiation modules. Lipopeptides synthetases for peptide like daptomycin contain starter C domains that use acyl-PCPs as donor substrates; depsipeptide synthetases for valinomycin and cereulide contain ketoreductase domains; and kolossin A (Figure 1) and linear gramicidin synthetase contain formylation (F) domains. Recently, we determined a series of crystal structures of

Pass it on

The (formyl-)aminoacyl-PCP moves from the initiation module to the donor site of the C domain. This donor site was identified by apo structures of the C domain and biochemically [31, 32]. The structure of an excised PCP–C didomain from TycC visualized the two domains required for donation, but was in a non-productive conformation [33]. However, recently two didomain structures of PCP and C domain homologues, a terminal cyclizing C (CT) [11] and an epimerization (E) domain [34] produced the

Keep it going

The first view of any module was that of the termination module of surfactin synthetase, SrfA-C, solved in the peptide-accepting state almost a decade ago [35]. With a domain architecture of C–A–PCP–Te, it represents both a minimal C–A–PCP elongation domain and the most common type of bacterial termination module. The SrfA-C structure showed very large distances between active sites in NRPSs indicating that substantial conformations changes would occur in the catalytic cycle (Figure 3a-i).

Finish him!

This elongation cycle (or an expanded cycle with a tailoring step) occurs in every canonical elongation module along the NRPS assembly line, until the termination module. Here, after one more elongation cycle, the PCP domain delivers the elongated peptide to the chain-terminating domain. In fungal termination modules, that is often a CT domain, and the PCP–CT structure is as discussed above, but for bacterial NRPSs, it is most commonly a Te domain and second most commonly a terminal reductase

Bring it all together

The above-described structures, in context of excellent existing functional studies, provide a wealth of insight into each step of the NRPS cycle. They are less informative about how modules form intact NRPS megaenzymes. High resolution structures of multimodular NRPSs are an outstanding goal in the field, and multiple such structures are required to answer questions of NRPS architecture, but a view of higher order structure is beginning to form.

We recently solved the structure of a

Onward

Recent work has illuminated many aspects of NRPSs structure. The next several years will likely be equally insightful. It will be interesting to see whether the coming structures of multimodular NRPSs will be determined by crystallography or cryo-EM. Cryo-EM is now the technique of choice for large macromolecules but the moderate size of NRPS domains and especially the massive array of domain–domain and module–module conformations which are absolutely required for NRPS function (many of which

Conflict of interest statement

Nothing declared.

References and recommended reading

Papers of particular interest, published within the period of review, have been highlighted as:

  • • of special interest

  • •• of outstanding interest

Acknowledgments

We thank the Schmeing lab for helpful discussions. Funding to TMS from the Canadian Institutes of Health Research (FDN-148472), Canada Research Chairs, and the Natural Sciences and Engineering Research Council (Discovery Grant 418420-12) is gratefully acknowledged.

References (52)

  • D.A. Herbst et al.

    Structural basis of the interaction of MbtH-like proteins, putative regulators of nonribosomal peptide biosynthesis, with adenylating enzymes

    J Biol Chem

    (2013)
  • K. Bloudoff et al.

    Chemical probes allow structural insight into the condensation reaction of nonribosomal peptide synthetases

    Cell Chem Biol

    (2016)
  • J. Alfermann et al.

    FRET monitoring of a nonribosomal peptide synthetase

    Nat Chem Biol

    (2017)
  • M.J. Jaremko et al.

    Structure and substrate sequestration in the pyoluteorin type ii peptidyl carrier protein PltL

    J Am Chem Soc

    (2015)
  • A.C. Goodrich et al.

    Solution structure of a nonribosomal peptide synthetase carrier protein loaded with its substrate reveals transient, well-defined contacts

    J Am Chem Soc

    (2015)
  • U. Linne et al.

    Control of directionality in nonribosomal peptide synthesis: role of the condensation domain in preventing misinitiation and timing of epimerization

    Biochemistry

    (2000)
  • E.A. Felnagle et al.

    Nonribosomal peptide synthetases involved in the production of medically relevant natural products

    Mol Pharm

    (2008)
  • C.T. Walsh

    Polyketide and nonribosomal peptide antibiotics: modularity and versatility

    Science

    (2004)
  • K.J. Weissman

    The structural biology of biosynthetic megaenzymes

    Nat Chem Biol

    (2015)
  • D. Schwarzer et al.

    Nonribosomal peptides: from genes to products

    Nat Prod Rep

    (2003)
  • D. Gahloth et al.

    Structures of carboxylic acid reductase reveal domain dynamics underlying catalysis

    Nat Chem Biol

    (2017)
  • J. Zhang et al.

    Structural basis of nonribosomal peptide macrocyclization in fungi

    Nat Chem Biol

    (2016)
  • K. Bloudoff et al.

    Structural and mutational analysis of the nonribosomal peptide synthetase heterocyclization domain provides insight into catalysis

    Proc Natl Acad Sci U S A

    (2017)
  • S. Caboche et al.

    NORINE: a database of nonribosomal peptides

    Nucleic Acids Res

    (2008)
  • E. Conti et al.

    Structural basis for the activation of phenylalanine in the non-ribosomal biosynthesis of gramicidin S

    EMBO J

    (1997)
  • B.R. Miller et al.

    Structural biology of nonribosomal peptide synthetases

    Methods Mol Biol

    (2016)
  • Cited by (69)

    • Proteins

      2023, Pharmacognosy: Fundamentals, Applications, and Strategies, Second Edition
    • Utilization of agroindustrial and food by-products for the production of amphiphilic proteins and peptides

      2023, Value-Addition in Agri-Food Industry Waste through Enzyme Technology
    • Antifungal evaluation and mechanistic investigations of membrane active short synthetic peptides-based amphiphiles

      2022, Bioorganic Chemistry
      Citation Excerpt :

      Around 18 peptide-based antimicrobials have been approved by the Food and Drug Administration (FDA), which account for about 17% of all peptide drugs approved till date whereas more than 100 are being currently investigated in pre-clinical and clinical trials for a diverse range of microbial infections [3]. Owing to the modern developments in new, sustainable, simplified and faster synthetic techniques and tools for peptide synthesis, the number of peptide-based therapeutics generated and being considered in the preclinical and clinical investigations has increased tremendously in the past few years [4–7]. Identification of side effects associated with conventional antibiotics such as nephrotoxicity with polyene antifungals, neurotoxicity related to β-lactam and quinolone class of antibacterials along with azole antifungals, and ototoxicity associated with use of aminoglycosides indicated the necessity of superior classes of antimicrobials [8,9].

    View all citing articles on Scopus
    1

    Equal contribution.

    View full text