Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways

Available online at www.sciencedirect.com

ScienceDirect Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Nathanael A Caveney1, Franco KK Li1 and Natalie CJ Strynadka The bacterial cell wall is a complex polymeric structure with essential roles in defence, survival and pathogenesis. Common to both Gram-positive and Gram-negative bacteria is the meshlike peptidoglycan sacculus that surrounds the outer leaflet of the cytoplasmic membrane. Recent crystallographic studies of enzymes that comprise the peptidoglycan biosynthetic pathway have led to significant new understanding of all stages. These include initial multi-step cytosolic formation of sugar-pentapeptide precursors, transfer of the precursors to activated polyprenyl lipids at the membrane inner leaflet and flippase mediated relocalization of the resulting lipid II precursors to the outer leaflet where glycopolymerization and subsequent peptide crosslinking are finalized. Additional, species-specific enzymes allow customized peptidoglycan modifications and biosynthetic regulation that are important to bacterial virulence and survival. These studies have reinforced the unique and specific catalytic mechanisms at play in cell wall biogenesis and expanded the atomic foundation to develop novel, structure guided, antibacterial agents. Address University of British Columbia, Biochemistry and Molecular Biology and the Center for Blood Research, Rm 4350 Life Sciences Center, 2350 Health Sciences Mall, Vancouver V6T 1Z3 Canada Corresponding author: Strynadka, Natalie CJ ([email protected]) These authors contributed equally to the preparation of this manuscript.

1

Current Opinion in Physiology 2018, 53:45–58 This review comes from a themed issue on Catalysis and regulation Edited by Hazel Holden and Alice Vrielink

https://doi.org/10.1016/j.sbi.2018.05.002 2468-8673/ã 2018 Elsevier Ltd. All rights reserved.

Introduction The bacterial cell wall, and the unique enzymes and chemistries responsible for its biogenesis remain as attractive targets for the development of novel antibiotics. The effectiveness of disrupting peptidoglycan (PG) synthesis for treating bacterial infections is long proven and best exemplified by b-lactam antibiotics, which inhibit penicillin-binding proteins (PBPs) from catalyzing the crosslinking transpeptidation of PG [1]. PG is composed of polymerized glycan strands of alternating b-1-4 linked www.sciencedirect.com

N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (MurNAc), with the latter covalently bound via the D-lactoyl moiety to characteristic pentapeptide ‘stems’ (commonly L-Ala-g-D-Glu-meso-DAP (or L-Lys)D-Ala-D-Ala where DAP is diaminopimelic acid) (Figure 1a). Crosslinking of adjacent glycan strands generally occurs between the carboxyl group of the penultimate D-Ala at position 4 and the amino group of the diamino acid at position 3, either directly or through a short peptide bridge, with the terminal D-Ala lost during the reaction. PG crosslinking is essential for the strength and viability of a bacterium and disruption of any step of PG biosynthesis, up to and including the late stage transpeptidation, typically results in cell death of bacteria in their physiological environs [2,3]. Despite several decades of successful efforts to effectively target PG biosynthesis by the pharmaceutical industry, specifically PBP mediated transpeptidation, bacteria have inevitably developed a variety of mechanisms of resistance to these antibiotics. Examples include expression of b-lactam hydrolyzing enzymes known as b-lactamases, production of PBPs with low affinity for b-lactams, and particularly in Gram-negative bacteria, utilization of drug efflux pumps [2]. While research into developing novel b-lactams, new classes of transpeptidase inhibitors and combinatorial strategies for combating the resistance crisis are ongoing, this represents only a molecular tip of the potential multienzyme PG pathway to be capitalized on therapeutically (Figure 1b). A major hurdle in this regard has been the lack of atomic level detail on many of the critical enzymes in the pathway, particularly those spanning or associated with the cytosolic membrane. Several recent structures however have now helped pave the way for future drug discovery efforts. An additional potential avenue for addressing resistance is inhibition of the known chemical modifications of PG, several of which have been shown to be important for downstream virulence. Although often more species-specific, a substantial common PG modification amongst Gram-positive bacteria is the anionic polymer known as wall teichoic acid (WTA). WTAs are phosphate-rich polymers covalently attached to the C6 hydroxyl of the MurNAc sugar component of PG (Figure 1a), and its significant molecular mass can be attributed for up to 60% of the total cell wall weight [4]. WTA modification of PG has been shown to have important advantages for bacteria including guiding the spatiotemporal localization of PBPs and autolysins for PG crosslinking [5] and cell division [6], respectively. In addition, it contributes to pathogenicity Current Opinion in Structural Biology 2018, 53:45–58

46 Catalysis and regulation

Figure 1

(a)

(b)

Current Opinion in Structural Biology

Current Opinion in Structural Biology 2018, 53:45–58

www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 47

with WTA-deficient mutants displaying diminished ability in biofilm formation [7], host cell adherence [8] and colonization [9]. Some commonalities are found amongst the two, likely coordinated biosynthetic pathways of PG and WTA. Notably, both polymers are initially synthesized in the cytoplasm on a lipid carrier, commonly undecaprenyl phosphate (C55-P), with lipid-bound precursors then necessarily flipped across the membrane for covalent attachment of the polar “head” groups onto the existing cell wall in the periplasm. Subsequent recycling of the released lipid carrier back to the cytoplasmic leaflet for PG/WTA biogenesis to continue efficiently is also an essential and catalyzed end point of the process. The lipid carrier is an energetically costly molecule to synthesize and potentially membrane perturbing at high concentrations. The importance of recycling the lipid carrier is demonstrated by the lethality of accumulating non-functional lipid-bound WTA intermediates in the absence of certain WTA biosynthetic enzymes [10]. Targeting both PG synthesis and WTA induced virulence is a promising combinatorial strategy for combating antibiotic resistant species such as Methicillin Resistant Staphylococcus aureus (MRSA). This approach is supported by the observed resensitization of resistant strains to b-lactams in the absence of WTAs, presumably due to additional disruption of PG synthesis incurred through mislocalization of PBPs [11]. Importantly, characterization of WTA biosynthetic enzymes has also expanded our understanding in the biogenesis of other bacterial cell wall polymers, such as arabinogalactan of mycobacteria and lipopolysaccharide of Gram-negative bacteria, as the pathways share some common enzymes which are also of interest to drug discovery. This review features a structural overview of recently characterized enzymes with key roles in PG and WTA biogenesis.

Assembly of the lipid II building block of peptidoglycan PG biosynthesis is initiated in the bacterial cytoplasm and catalyzed by several well characterized enzymes (the Mur pathway) which act successively to form a uridine diphosphate (UDP)-MurNAc pentapeptide (Supplementary Figure 2). In the first membrane-associated step of PG synthesis, this product is transferred to a C55-P lipid carrier. De novo synthesis of C55-P occurs through successive condensation catalyzed by the synthase UppS to form C55-PP [12]. This product is then dephosphorylated by the polytopic membrane spanning phosphatase UppP to form the membrane localized C55-P lipid carrier [13] (Figure 2a). The structure of E. coli UppP has recently been solved, providing the first structural insight into lipid carrier recycling [14

,15
].

UppP is largely composed of ten transmembrane (TM) a-helices, six of which are full-span TM helices and four which, unexpectedly, form two antiparallel re-entrant helix-loop-helix regions. UppP is a dimer both in solution and in the crystal packing with a twofold axis of symmetry bisecting the membrane. The UppP active site is a membrane embedded hydrophobic cleft, as opposed to the surface localized active sites of other characterized phosphatases. A monoolein lipid used in the lipidic cubic phase crystallization lies within the active site, acting as a substrate mimic of C55-PP and allowing for insight into the phosphatase mechanism [14

]. The catalytic serine carries out nucleophilic attack on the terminal phosphate of the C55-P, with the pentavalent transition state stabilized by a coordinated metal ion and arginine. The resultant phosphoenzyme intermediate is then hydrolyzed by a glutamate-activated water, resulting in the release of C55-P and inorganic phosphate. Interestingly, UppP has structural similarities to various proteins involved in cross-membrane transport suggesting the possibility for a phosphatase activated transport mechanism of C55-P back to the cytoplasmic leaflet, a recycling process that ostensibly plays an important role in the generation of dephosphorylated lipid carrier for ongoing biosynthesis of both WTA and PG. The subsequent phospho-MurNAc pentapeptide transfer to C55-P and membrane is catalyzed by a second polytopic membrane protein MraY, a member of the polyisoprenylphosphate N-acetylglucosaminosugar-1-phosphate-transferase (PNPT) superfamily of prokaryotic and eukaryotic prenyl sugar transferases (Figure 2b). Recent structures of MraY from both Aquifex aeolicus and Clostridium bolteae in the apo and inhibitor bound forms have provided important first insights into this critical step of PG biosynthesis [16

,17
,18

]. MraY is comprised of 10TM a-helices per monomer. A homodimeric form is observed in all structures with 2 monomers arranged in a parallel fashion relative to the membrane axis and the predicted active site regions of both monomers lying at the cytosolic face. Although structures with donor (UDP-MurNAc pentapeptide) or acceptor (C55-P) substrates were not captured in these studies, an observed extended hydrophobic groove is proposed to accommodate the acyl lipid tail of the substrate. In structures of MraY in complex with the natural product inhibitors muraymycin D2 and tunicamycin there is seen to be rearrangement of TM helices 1, 5 and cytosolic loops 5, 7 and 9 to accommodate the compounds and allowed for prediction of potential nucleotide binding and active site candidates near the cytoplasmic face of the enzyme [17
,18

]. The conformational plasticity of MraY observed in comparisons of the

(Figure 1 Legend) Overview of peptidoglycan and wall teichoic acid biogenesis pathways. (a) Chemical structures of PG and PG modifications. Modifications at C6 hydroxyl of PG MurNAc include O-acetylation and addition of a large species-dependent glycopolymer such as WTA, teichuronic acid, capsular polysaccharide and arabinogalactan. The chemical structure of S. aureus WTA is depicted. (b) Schematic representation of PG and S. aureus WTA biogenesis pathways. A further annotated version of Figure 1b is available as Supplementary Figure 1. PDB accession codes are available in Supplementary Table 1.

www.sciencedirect.com

Current Opinion in Structural Biology 2018, 53:45–58

48 Catalysis and regulation

Figure 2

i

UppP

C

ii

N

C55PP

cytosolic membrane

(a)

C55P

iv

iii

C55PP P P

P P

cytoplasmic leaflet

outer leaflet

P

P

UppP

i

ii

cytosolic membrane

(b)

C55P

MraY +

UMP

lipid I

MraY

UDP-MurNAc pentapeptide

iii

apo (c)

C55P

muraymycin D2

i

tunicamycin

ii UDP-GlcNAc

lipid II

MurG

MurG lipid I

UDP

cytosolic membrane

i

MurJ ii

cytosolic membrane

(d)

TM 13-14

lipid II M

G

MurJ

cytoplasmic leaflet

P P G

P P

M

G P P

outer leaflet

M

P M

P G

lipid II C lobe

N lobe Current Opinion in Structural Biology

Current Opinion in Structural Biology 2018, 53:45–58

www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 49

apo form and these complexes may be important for enzyme catalysis, as well as presumably enable the accommodation of the wide array of diverse natural product inhibitors identified for this enzyme class. The natural product inhibitor tunicamycin is of note, as it is seen to effectively inhibit both PG and WTA biosynthesis through inhibition of MraY and paralog TarO respectively, and indeed is commonly used as an inhibitor in other members of the PNPT family including those involved in N-linked glycosylation [19
]. The structure of one such N-linked glycosylase, human GlcNAc-1-Ptransferase (hGPT), was recently solved in complex with tunicamycin [19
]. Modification of the GlcNAc portion of tunicamycin is seen to reduce the inhibition of hGPT while maintaining inhibition of MraY [19
] and gives promise that downstream structure-guided drug design can lead to the development of further novel and specific MraY inhibitors. The chemical mechanism of MraY has been debated with latest evidence lending support to a one-step, single-displacement over a two-step variant [20–22]. Regardless of whether a covalent intermediate is formed, the mechanism is proposed to involve deprotonation of the phosphate moiety of C55-P by a catalytic aspartate. Also of note is an essential magnesium which presumably coordinates the phosphate groups of the nucleotide substrate. Transfer of phospho-MurNAc pentapeptide to C55-P results in the first lipid precursor of PG biosynthesis termed lipid I. Lipid I is subsequently converted into lipid II with the transfer of a GlcNAc from a UDP-GlcNAc donor by the membrane-tethered glycosyltransferase (GT) MurG (Figure 2c). This conversion is presumably closely coupled to MraY activity, as little lipid I is observed to exist in the bilayer [23,24]. Structures of MurG have a common GT-B fold comprised of two a/b domains with an extended active site cleft at their interface [25]. MurG catalysis occurs via a sequential Bi Bi ordered mechanism, whereby the UDP-GlcNAc donor binds first, inducing the requisite rearrangements in a loop containing a conserved GGS motif to facilitate subsequent lipid I acceptor binding and conversion to lipid II product [25].

Lipid II transfer across the cytoplasmic membrane For PG assembly to proceed, the large, electronegative head group of lipid II must be efficiently flipped to the outer leaflet to be acted upon by PBPs in the final stages of the biosynthetic pathway. It has long been proposed that the necessary flipping would be specifically catalyzed by a component of the biosynthetic machinery, as transfer of lipid II across the bilayer is non-spontaneous [26]. Two candidates for the lipid II flippase role have been previously suggested, MurJ and FtsW/RodA. Bioinformatic approaches first identified MurJ as a potential flippase of lipid II, and recent structural studies have supported this claim [27,28

] (Figure 2d). MurJ belongs to the mouse virulence factor (MVF) subfamily of the multidrug/oligosaccharidyl-lipid/polysaccharide (MOP) superfamily. Of note in MurJ are two a-helices in addition to the 12 core TM helices common to most members of the MOP superfamily. The structure of MurJ from Thermosipho africanus recently solved was observed to be in an inward facing ‘N-shaped’ conformation, distinct from the outward facing ‘V-shaped’ conformations of all other MOP superfamily structures solved to date. At the intramembranal interface of the bi-lobed structure lies an internalized polar cavity, a commonly localized binding site for cargo in the MOP superfamily. In the inward facing conformation of MurJ, a hydrophobic groove, formed by the additional two helices, extends toward the central portal suggesting a unique modification that can accommodate the hydrophobic tail of lipid II with the negatively charged head group occupying a cationic portion of the central portal. It is likely that MurJ acts to flip lipid II in an alternating-access transport mechanism [28

], with MurJ transitioning between inward and outward facing conformations to export lipid II to the outer leaflet. Despite the evidence for MurJ as a lipid II flippase, there is a compelling argument that FtsW also has potential to play a role in this critical step of PG synthesis. FtsW, a member of the SEDS protein family (shape, elongation,

(Figure 2 Legend) Initial membrane associated steps of peptidoglycan biosynthesis. Lipid carrier synthesis and recycling, membrane associated Lipid II synthesis, and transfer to the outer leaflet of the cytosolic membrane. (a) (i) Structure of UppP (green) in complex with structural lipid and substrate mimic, monoolein (pink) (E. coli, 6CB2). (ii) Proposed reaction mechanism for UppP mediated dephosphorylation of C55PP. (iii) Schematic representation of proposed UppP phosphatase-couple lipid flippase activity. (iv) Two-fold symmetry axis of the UppP dimer (black) and twofold pseudosymmetry axes of each monomer (grey). (b) (i) Apo MraY dimer (purple) (Aquifex aeolicus, 4J72) with placement in the cytosolic membrane (grey). The catalytically important aspartate residues and magnesium ion are shown in green. (ii) The proposed one-step mechanism for MraY. (iii) Conformational changes seen upon binding of either muraymycin D2 (Aquifex aeolicus, 5CKR) or tunicamycin (Clostridium bolteae, 5JNQ) inhibitors. Catalytic aspartate residues and the two inhibitors are shown in green. Changes are seen to the end of TM helix 1 (blue), the end of TM helix 5 and loop 5 (yellow), loop 7 (cyan) and loop 9 (red). It is proposed that the conformational flexibility seen could play a role in MraY binding and function with its natural substrate. (c) (i) The bi-domain architecture of MurG (green) in complex with UDP-GlcNAc substrate (magenta) (E. coli, 1NLM), shown in relation to the cytosolic membrane. MurG is seen to associate with the cytosolic membrane and thought to receive lipid I substrate directly from MraY. (ii) Chemical structures of the MurG catalyzed transition from lipid I to lipid II with the addition of GlcNAc. (d) (i) Structure of inward facing MurJ (Thermosipho africanus, 5T77). With the N lobe (blue), C lobe (purple) and TMs 13-14 (green) forming a unique N-shaped architecture. (ii) The proposed alternating-access mechanism for lipid II flipping by MurJ. Upon lipid II binding, the inward N-shaped conformation (Thermosipho africanus, 5T77) switching to a proposed V-shaped conformation, as modelled previously [28

], allowing transfer of the lipid II to the outer leaflet of the cytosolic membrane.

www.sciencedirect.com

Current Opinion in Structural Biology 2018, 53:45–58

50 Catalysis and regulation

division and sporulation), has been shown to directly translocate 7-nitro-2,1,3-benzoxadiazol-4-yl lipid II across membranes [29]. In the same work, no appreciable activity from MurJ could be detected by the assay. As well, it is known that FtsW and other closely related SEDS proteins, RodA and SpoVE, interact with PBPs [30,31]. This aligns with the notion that lipid transfer is coupled to the GT/TP activities of PBPs [26]. Remarkably, SEDS proteins have recently been proposed to have in of themselves inherent lipid II polymerizing GT activity [32
,33], albeit with as yet little clarity toward their ability to also or instead act as lipid II flippases. One appealing proposal is that these SEDS proteins act as coupled lipid II flippaseglycosyltransferases, perhaps the latter driving the flipping energetically. Using evolutionary guided de novo structural prediction, a model of FtsW was developed which supports this dual-functionality hypothesis [34]. Due to the current lack of structural information for these SEDS family of proteins, and the apparent importance of their function, they remain an exciting area of interest in the field of PG biosynthesis.

Peptidoglycan glycopolymerization and transpeptidation by penicillin binding proteins The final steps in the biosynthesis of PG involve catalysis by membrane anchored PBPs on the outer leaflet of the cytoplasmic membrane. These PBPs first act to polymerize lipid II disaccharide units into PG strands, followed by (regulated) transpeptidase-mediated crosslinking to the existing sacculus to form mature PG [2] (Figure 3). In most bacterial species the bulk of the GT activity has historically been proposed to involve specific PBPs, with various species having several acting in different capacities during growth, division, environmental stress and antibiotic resistance. These PBPs exist within complex regulatory networks, involving numerous protein-protein interactions, to ensure proper spatial and temporal regulation of PBP activity (reviewed in [35]). Of note are the crystallographic and NMR structures of two regulatory outer-membrane anchored lipoproteins LpoA [36] and LpoB [37,38], which have been observed to interact with select PBP partners to stimulate late stage PG synthesis [35]. The high molecular weight Class A PBPs, bifunctional enzymes which encapsulate both GT and transpeptidase (TP) activities in distinct active sites, as well as monofunctional GT variants, act to catalyze the attachment of the non-reducing end of lipid II disaccharide to the reducing end of the donor PG strand via a b1,4glycosidic linkage [[39
]]. The first structures of fulllength class A PBPs revealed an unprecedented GT domain with an active site cleft at the interface between a solvent exposed a-helical ‘head’ subdomain somewhat reminiscent of the PG hydrolase lysozyme and a distinct membrane embedded ‘jaw’ subdomain common only to these lipid activated family of PG glycosyltransferases [39
]. As with lysozyme, the structure indicated six sugar Current Opinion in Structural Biology 2018, 53:45–58

moieties occupy the active site cleft and orient the growing PG donor strand away from the identified catalytic glutamate. A hydrophobic channel in the ‘jaw’ subdomain acts as a binding site for the extended acyl chain of the donor lipid II. Co-crystallization with a natural product inhibitor, Moenomycin A (Moe A) [39
,40], and lipid II analogs [41] validated these predictions and has provided further insight into PG polymerization. The Moe A complex structures indicate this potent, sub nM inhibitor acts as a substrate mimic of the tetrasaccharideC55PP product of the first round of polymerization, lipid IV [39
,41,42]. The GT mechanism is initiated with action of the catalytic glutamate acting as a Brønsted base in deprotonation of the acceptor C4 hydroxyl of the lipid II, followed by a nucleophilic attack on the donor MurNAc anomeric C1 of the acyl phosphate in an SN2 like reaction with inversion of stereochemistry to form the b1,4-glycosidic linkage [39
,41]. The final step in PG biosynthesis is the periodic crosslinking (transpeptidation) of glycan strands into the existing PG sacculus. TP activity provides rigidity and structure to the PG layer necessary for its protective function [2] and occurs through a two-step mechanism that begins with a serine-mediated acylation of the penultimate residue, D-alanine, of the pentapeptide on the growing PG strand. The covalent acyl-enzyme intermediate is subsequently deacylated via a nucleophilic attack of a side chain or terminal amino group of the third residue, meso-DAP, L-lysine, or L-lysine-pentaglycine, on an adjacent PG strand. This reaction is famously inhibited by b-lactam antibiotics, such as penicillin, which act as a substrate mimic of donor peptide and inhibit through long lived acylation of the catalytic serine. Alternate PBPs, such as PBP2a in MRSA, which are poorly acylated by b-lactam antibiotics, can be upregulated in the presence of b-lactam stress to provide resistance [43]. Further, structural and computational analysis of PBP2a indicates the potential for allosteric effects of bound PG fragments in its resistance phenotype [44,45]. For class A bifunctional PBPs, TP activity is thought to be coupled to intramolecular GT activity, with the glycan strand snaking from the membrane embedded GT catalytic domain directly to the extracellular TP domain [46]. For class B monofunctional PBPs such as the aforementioned PBP2a, TP activity is likely coupled to the GT activity of a class A PBP or a monofunctional glycosyltransferase (including potentially the newly proposed SEDS GT/flippase family). Indeed, the major class A and B PBPs of cell division in E. coli are thought to exist only in tight complex with FtsW at the divisome [31]. Interestingly, the 4-3, D,D PG crosslinks formed by traditional D,D-transpeptidases such as class A and B PBPs can be bypassed under certain stress including b-lactam antibiotics with crosslinking instead accomplished by 3-3, L,D-transpeptidases, which undergo slow www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 51

Figure 3

(a)

(c)

(b)

ly po

TP

PG

Moe A, doner mimetic

m er iza

ti o

Glu

n

UB2H

Ser acceptor mimetic

GT TM

aztreonam, peptide mimetic

PBP1b cytosolic membrane

acceptor Lipid II

(d)

GTase

+ C55P

doner PG

(e)

catalytic Ser D-Ala

3-4, D-D crosslink m-DAP

TP

TP

Current Opinion in Structural Biology

Final stages of PG biosynthesis by PBPs. (a) Acyl-aztreonam and moenomycin A (Moe A) inhibited PBP1b, a bi-functional PBP (E. coli, 5HLB). GT domain seen in green, TP domain in blue, TM helix in purple and UB2H interaction domain in orange. (b) GT domain of PBP1b inhibited by Moe A (magenta) (E. coli, 5HLB), which acts as a donor PG strand mimetic to inhibit the GT activity of PBPs. Lipid II acceptor mimetic (cyan) from a monofunctional GTase co-structure (S. aureus, 3VMT) is shown aligned with its potential position in the GT domain of PBP1b. The catalytic glutamate is shown (green) in its position between the donor and acceptor. (c) Active site of PBP1b with acyl-aztreonam (E. coli, 5HLB). Aztreonam (magenta) is a monobactam which acts as a peptide mimetic and suicide inhibitor of the TP activity of PBPs. (d) The catalytic mechanism of PBP GT activity, showing the polymerization of an acceptor lipid II into the donor, growing strand. (e) The catalytic mechanism of the 3-4, D-D, transpeptidase activity of PBPs.

acylation by b-lactam antibiotics. In Mycobacterium, L,Dtranspeptidation makes up the large majority of PG crosslinking even in the absence of stressors [47]. Importantly, L,D-transpeptidases often act on PG glycan strands with www.sciencedirect.com

tetrapeptide as opposed to pentapeptide extensions. L,Dtranspeptidation is therefore likely coupled to the carboxypeptidase activity of class C PBPs, which are observed to be quite resistant to b-lactam antibiotics. Current Opinion in Structural Biology 2018, 53:45–58

52 Catalysis and regulation

Despite decades since the first detection of L,D-transpeptidation, specifics of this resistance pathway have been slow to come to light. Recently an L,D-transpeptidase, YcbB, and a carboxypeptidase, PBP5, have been shown to play a role in b-lactam resistance in E. coli [48]. Despite mechanistic similarities in D,D and L,D-transpeptidases, the latter use a reactive cysteine thiol rather than serine hydroxyl for the initial attack of substrate or b-lactam. The differences in nucleophile and nature of the covalent thiol ester intermediate generated result in the fast deacylation of most b-lactam antibiotics and a lack of inhibition, leading to the resistance phenotype. It is clear that these final stages of PG synthesis are far more complex than previously thought, with recent work suggesting that many of the enzymes involved interact as a highly regulated macromolecular machine [35].

Peptidoglycan modifications PG modifications, with critical functions in virulence, immune evasion and autolysin regulation, are another vital aspect of cell wall assembly (reviewed in [49]). These modifications are primarily localized at the C2 acetyl group and C6 hydroxyl moiety (Figure 1a). Ndeacetylation of both MurNAc and GlcNAc, and N-glycolylation of MurNAc occur at the C2 position. At the C6 hydroxyl of MurNAc, the modifications include O-acetylation and the addition of a large species-dependent glycopolymer such as WTA, teichuronic acid, capsular polysaccharide and arabinogalactan. In addition, MurNAc can be modified into a spore-specific muramic acid d-lactam that allows the spore PG to be recognized by germination associated lytic enzymes. Lastly, the terminal PG MurNAc residues of Gram-negative bacteria and some Gram-positive bacteria can be modified into 1,6-anhydroMurNAc. One of the most well characterized secondary glycopolymers is WTA of Gram-positive pathogens. The role of WTA in virulence and b-lactam resistance has garnered much attention and has led to recent structural characterization of enzymes involved in its synthesis and modification.

Modification of peptidoglycan by wall teichoic acid Although prominent outliers have been characterized (e. g. Streptococcal WTA [50,51]), the composition of WTA polymers centers on two common features: the sugarbased linkage unit (N-acetylglucosamine-N-acetyl-DGlcNAc-Manmannosamine-[glycerol-phosphate]1-2; NAc-GroP1-2) that is transferred to PG, and the polyol nature of the main chain (typically poly-ribitol-5-phosphate or poly-glycerol-3-phosphate; poly-RboP or polyGroP) extending from the linkage unit (Figure 1a) [52]. Differences in WTA synthesis amongst various species generally center on the precise chemical composition, formation and modification of the polyol main chain. However, despite these differences, the WTA polymers are functionally similar and are often built by a Current Opinion in Structural Biology 2018, 53:45–58

homologous set of enzymes. This review emphasizes Staphylococcal variants in WTA biogenesis for which several enzyme structures have recently been described. The synthesis of S. aureus WTA can be divided into 6 stages: initiation via synthesis of an intracellular linkage unit on C55-P, polymerization, glycosylation, export, Dalanylation and attachment to PG (Figure 1b).

Initiation of wall teichoic acid biosynthesis TarO initiates creation of a membrane anchored ‘linkage unit’ by catalyzing transfer of GlcNAc-1-phosphate from a UDP-GlcNAc donor to C55-P [53]. TarO is a paralog of MraY and an ortholog in other bacterial species, WecA, is involved in arabinogalactan, enterobacterial common antigen and O-antigen lipopolysaccharide synthesis; only MraY thus far has been characterized at the atomic level [16

,17
,18

]. Selective inhibition of TarO has been demonstrated by compounds such as the tarocins [54] and further structural understanding of subfamily specific features will serve to further guide the modification of existing PNPT inhibitors to bypass hGPT inhibition and instead target selected bacterial variants. Interestingly, the PNPT inhibitor, tunicamycin, was recently discovered to influence not only the activity of TarO but also that of MnaA and Cap5P, which are UDP-GlcNAc 2epimerases responsible for conversion of UDP-GlcNAc to UDP-ManNAc [55]. The latter is the requisite donor for incorporation of the second sugar, ManNAc, to the linkage unit by TarA [56]. Subsequently, synthesis of the linkage unit is completed by TarB and TarF through consecutive addition of two GroP molecules from cytidine 50 -diphosphate (CDP)-glycerol to ManNAc [57].

Polymerization of wall teichoic acid TarL/TarK catalyze the polymerization of the S. aureus RboP main chain, and are members of the TagF-like family of monotopic membrane proteins. Structural understanding of this family is provided by a series of Staphylococcus epidermidis TagF (residues 312-721) crystal structures [58

], an ortholog which catalyzes formation of a GroP rather than RboP main chain. The crystallized region of TagF encompasses a two-helix domain (amphipathic helices a1 and a2) involved in monotopic membrane binding and a GT-B domain involved in catalysis (Figure 4a). Insertions of basic residues between the b-strands of the typical GT-B fold localize adjacent to the active site and are proposed to be involved in dimerization, stabilization of the membrane binding helices and expansion of the active site cleft, presumably to bind the growing polymer. An SN2-like reaction involving the deprotonation of the terminal GroP moiety of the polyol chain by His444 and the subsequent nucleophilic attack on the b-phosphate of the donor, CDP-glycerol, is supported by a catalytic mutant (His444Asn) structure of S. epidermidis TagF in complex with CDP-glycerol. www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 53

Figure 4

(a)

GT-B domain

i

Terminal GroP of WTA

ii

His444

TagF His584

CDP-glycerol WTA GroP chain

C

α2 α1

(b)

i

CMP

kcal/mol

N

cytosolic membrane

-10

0

ii

TarM

+

10

α-glyceryl-GlcNAc

kcal/mol

10

0

trimerization domain

-10

GT-B NTD

TarS

(c)

UDP

GT-B CTD

C2 domain

linker

fondaparinux (WTA analog)

(d)

kcal/mol

10

0

-10

C1 domain GT-A domain

LCP domain monomer

trimer

Cps2A accessory domain Current Opinion in Structural Biology

WTA biogenesis: polymerization, GlcNAcylation and ligation. (a) (i) Structure of S. epidermidis TagF His444Asn (residues 312-721) in complex with CDP-glycerol (3L7K). Helix a1, helix a2, GT-B domain and the lipid bilayer are coloured orange, purple, green and grey, respectively. The second monomer of the homodimeric assembly is depicted in a transparent electrostatic potential surface. CDP-glycerol is shown in light grey and heteroatoms are coloured by type (N, blue; O, red; P, orange; S; yellow). (ii) Mechanistic details of WTA GroP main chain elongation by TagF. (b) (i) S. aureus TarM homotrimer (4X6L). Ribbon structures of GT-B NTD, GT-B CTD and trimerization domain are coloured green, blue and purple,

www.sciencedirect.com

Current Opinion in Structural Biology 2018, 53:45–58

54 Catalysis and regulation

Glycosylation of wall teichoic acid The decoration of S. aureus WTA with GlcNAc residues has been directly linked to human antibody response [59], phage binding [60,61], genetic exchange [62] and antibiotic resistance [60]. Either or both a-linked and b-linked GlcNAc are transferred to the C4 hydroxyls of the polyRboP WTA main chain (Supplementary Figure 3a) [60,61]. b-O-GlcNAcylation is of particular interest because it is linked to b-lactam resistance of S. aureus [60]. The addition of a-O-GlcNAc and b-O-GlcNAc are catalyzed by TarM and TarS, respectively. Both enzymes form homotrimeric assemblies, but possess different oligomerization domains and interfaces [63
,64,65
]. TarM is a retaining enzyme that catalyzes attachment of a-O-GlcNAc from UDP-GlcNAc to the polymerized chain of WTA [61]. The enzyme encompasses an Nterminal trimerization domain formed by 10 antiparallel b-strands and a C-terminal region that adopts a GT-B fold (Figure 4b) [63
,64]. TarM assembles into a propellershaped complex with the trimerization domains as the central hub and the GT-B domains as blades [64]. Electropositive surface grooves extend from the acceptor binding site of the GT-B domain to the trimerization domain of the neighbouring protomer. These grooves were occupied by a pentameric heparin analog (fondaparinux) that appeared to bind as an acceptor substrate mimic in the structure of a trimerization mutant [63
]. This structure also included UDP and a-glycerylGlcNAc, which presents the first enzyme-catalyzed product complex for retaining GT-B-type glycosyltransferases and provides support of an internal front-face nucleophilic substitution (SNi)-like mechanism [66]. The proposed mechanism involves the base activation of the nucleophile (ribitol hydroxyl) by the leaving group (phosphate oxygen of UDP) on the same face of the nucleophilic attack [63
]. The stereospecific counterpart, TarS, is an inverting GTA-type enzyme that transfers b-O-GlcNAc from UDPGlcNAc to WTA [60,65
]. The catalytic GT-A domain is linked to a C-terminal trimerization domain that encompasses tandem carbohydrate binding modules (C1/C2) proposed to bind WTA (Figure 4c) [65
]. The trimeric assembly adopts a hanging basket like structure with the GT-A domains pointed outwards in opposing orientations. Electropositive grooves occupied by sulfate ions suggest the extended acceptor binding site for WTA. Two loops, designated as the catalytic (CS) loop and

the substrate access (SA) loop, play crucial roles in substrate binding and catalysis, and were found to adopt different conformations in the UDP- and UDP-GlcNAcbound structures. TarS is proposed to utilize a concerted SN2-like catalytic mechanism that has been assigned to other inverting GT-A glycosyltransferases. The mechanism involves the deprotonation of the poly-RboP C4 hydroxyl by the general base on the CS loop (Asp178) in concert with nucleophilic attack at the b-face of the UDPGlcNAc C1 anomeric centre; the UDP leaving group is stabilized by a manganese/magnesium ion.

Export, D-alanylation and ligation of wall teichoic acid Flipping of the lipid-linked WTA intermediate is facilitated by TarG and TarH, a two-component ATP-binding cassette (ABC) transporter [67]. Three distinct mechanisms of lipid flipping by ABC transporters have been proposed based on structures of MsbA, a lipopolysaccharide flippase, and PglK, a polyprenyl-linked oligosaccharide flippase involved in protein N-glycosylation [68–70]. The shared commonalities in PglK and TarGH substrates may hint at analogy in their general mechanisms of flipflop, but nonetheless underlines the importance of determining future TarGH structures to understand the molecular details governing specific passage of WTA polymers to the outer face of the cytosolic membrane. Once transported, the main chain of WTA is further modified by the addition of D-alanine residues to the C2 hydroxyl groups of the ribitol moieties through a process that has not been fully characterized. Attachment of WTA and other secondary polymers to PG is mediated by the LytR-CpsA-Psr (LCP) family of enzymes [71

,72,73,74]. Typically, these enzymes have an LCP domain, an extracellular catalytic region comprised of a b-sheet sandwiched between a-helices, anchored to the outer leaflet of the cytosolic membrane by a single TM helix (Figure 4d) [71

,75] Certain members possess additional extracellular regions of unknown function such as the ‘accessory’ domain of Streptococcus pneumoniae Cps2A, a mixed a/b fold with similarities to globular domains of periplasmic binding proteins such as phosphonate binding protein PhnD. Structures of various lipidated donor analogs have been captured in the hydrophobic active site pocket of LCP enzymes showing conserved arginine residues position the pyrophosphate moiety which in turn coordinates, along with two conserved aspartates, a magnesium ion shown to be essential

(Figure 4 Legend Continued) respectively. The electrostatic potential surface of two monomers is depicted and the electropositive WTA binding groove is outlined in yellow. (ii) Close-up of fondaparinux (light blue), UDP (purple) and a-glyceryl-GlcNAc (cyan) in the structure of TarM G117R (4 X7R). Heteroatoms are colored by type and ribbon structures are coloured as in b i. (c) S. aureus TarS homotrimer and monomer (5TZ8) with UDP (light grey), manganese ion (black) and sulphate ions superimposed from a TarS structure encompassing residues 1-349 (5TZK). Ribbon structures of the GT-A domain, linker, C1 domain and C2 domain are coloured green, orange, cyan and purple, respectively. Heteroatoms are colored by type. The electrostatic potential surface and WTA binding groove are depicted as in b i (d) S. pneumoniae Cps2A (residues 98–481; 3TEP) in complex with octaprenyl pyrophosphate (green). Ribbon structures of the accessory domain and LCP domain are coloured in purple and blue, respectively. Heteroatoms are colored by type.

Current Opinion in Structural Biology 2018, 53:45–58

www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 55

for activity [71

,76,77]. Interestingly, recent structures of Bacillus subtilis TagT revealed limited interactions between the enzyme and the saccharides of the donor substrate [77]. LCP enzymes likely utilize a general acid/ base mechanism, with catalytic candidates suggested based on various structures [71

,77]. The localization of the LCP enzymes makes them attractive anti-virulence targets, potentially in combination with b-lactam drugs; however, their redundancy and as yet unestablished specificity at the atomic level, still leaves questions as to the potential efficacy of such a strategy.

Scholar program (to N.C.J.S.) and the Canadian Foundation of Innovation and British Columbia Knowledge Development Fund (to N.C.J.S.). N.C.J. S. is a Tier I Canada Research Chair in Antibiotic Discovery.

Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10. 1016/j.sbi.2018.05.002.

References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as
of special interest

of outstanding interest

Conclusions Recent structural milestones in the PG biosynthesis field have provided significant new information for functional understanding of PG biosynthesis, modification and guided design of novel antibiotic strategies. Notably, the first atomic information of previously uncharacterized polytopic membrane enzymes, such as UppP, MraY and MurJ, have now been elucidated. Although substrate bound complexes are required to further illuminate the precise mechanism and specificity in several cases, the newly captured data has already allowed for new avenues of enquiry into structure guided antibacterial development. Also enabled are new anti-virulence strategies targeting structures such as those involved in PG and lipid carrier recycling, or WTA biosynthesis. In recent years, structural studies on WTA biosynthetic enzymes provided novel mechanistic insights into the formation and glycosylation of the polyol chain, and the attachment of the polymer to PG. These findings have potential application to homologous enzymes involved in synthesizing other bacterial cell wall polymers of therapeutic interest. The characterization of even the classic PG enzymes, such as the PBPs has also advanced with focus on critical pathogens known to be associated with antimicrobial resistance. Although our understanding of individual enzymes in these complex pathways is greatly expanding, it is clear that there is still much to be learned regarding the interfacial recognition and interplay of the PG biosynthetic pathway members as a dynamic nanomachine during growth, division or under antibiotic stress. Perhaps, facilitated by the new resolution revolution in single particle cryoEM and cryo-tomography, atomic visualization of this nanomachinery will be the next step in our molecular understanding of this important antibiotic target.

Conflict of interest statement The authors declare no conflict of interest.

Acknowledgements This work was supported by the Natural Sciences and Engineering Research Council (to F.K.K.L.), the Canadian Institutes of Health Research (to N.C.J.S.), the Howard Hughes Medical Institute International Senior www.sciencedirect.com

1.

King DT, Sobhanifar S, Strynadka NCJ: One ring to rule them all: current trends in combating bacterial resistance to the b-lactams. Protein Sci 2016, 25:787-803.

2.

Sobhanifar S, King DT, Strynadka NCJ: Fortifying the wall: synthesis, regulation and degradation of bacterial peptidoglycan. Curr Opin Struct Biol 2013, 23:695-703.

3.

Lovering AL, Safadi SS, Strynadka NCJ: Structural perspective of peptidoglycan biosynthesis and assembly. Annu Rev Biochem 2012, 81:451-478.

4.

Hancock IC: Bacterial cell surface carbohydrates: structure and assembly. Biochem Soc Trans 1997, 25:183-187.

5.

Atilano ML, Pereira PM, Yates J, Reed P, Veiga H, Pinho MG, Filipe SR: Teichoic acids are temporal and spatial regulators of peptidoglycan cross-linking in Staphylococcus aureus. Proc Natl Acad Sci U S A 2010, 107:18991-18996.

6.

Schlag M, Biswas R, Krismer B, Kohler T, Zoll S, Yu W, Schwarz H, Peschel A, Go¨tz F: Role of staphylococcal wall teichoic acid in targeting the major autolysin Atl. Mol Microbiol 2010, 75:864873.

7.

Holland LM, Conlon B, O’Gara JP: Mutation of tagO reveals an essential role for wall teichoic acids in Staphylococcus epidermidis biofilm development. Microbiology 2011, 157:408418.

8.

Weidenmaier C, Peschel A, Xiong Y-Q, a Kristian S, Dietz K, Yeaman MR, Bayer AS: Lack of wall teichoic acids in Staphylococcus aureus leads to reduced interactions with endothelial cells and to attenuated virulence in a rabbit model of endocarditis. J Infect Dis 2005, 191:1771-1777.

9.

Misawa Y, Kelley Ka, Wang X, Wang L, Park WB, Birtel J, Saslowsky D, Lee JC: Staphylococcus aureus colonization of the mouse gastrointestinal tract is modulated by wall teichoic acid, capsule, and surface proteins. PLoS Pathog 2015, 11: e1005061.

10. D’Elia Ma, Millar KE, Bhavsar AP, Tomljenovic AM, Hutter B, Schaab C, Moreno-Hagelsieb G, Brown ED: Probing teichoic acid genetics with bioactive molecules reveals new interactions among diverse processes in bacterial cell wall biogenesis. Chem Biol 2009, 16:548-556. 11. Farha Ma, Leung A, Sewell EW, D’Elia Ma, Allison SE, Ejim L, Pereira PM, Pinho MG, Wright GD, Brown ED: Inhibition of WTA synthesis blocks the cooperative action of PBPs and sensitizes MRSA to b-lactams. ACS Chem Biol 2013, 8:226-233. 12. Fujihashi M, Zhang YW, Higuchi Y, Li XY, Koyama T, Miki K: Crystal structure of cis-prenyl chain elongating enzyme, undecaprenyl diphosphate synthase. Proc Natl Acad Sci U S A 2001, 98:4337-4342. 13. Manat G, Roure S, Auger R, Bouhss A, Barreteau H, MenginLecreulx D, Touze´ T: Deciphering the metabolism of undecaprenyl-phosphate: the bacterial cell-wall unit carrier at the membrane frontier. Microb Drug Resist 2014, 20:199-214. 14. Workman SD, Worrall LJ, Strynadka NCJ: Crystal structure of an

intramembranal phosphatase central to bacterial cell-wall Current Opinion in Structural Biology 2018, 53:45–58

56 Catalysis and regulation

peptidoglycan biosynthesis and lipid recycling. Nat Commun 2018, 9:1159. This paper provides the first structure, at 2.0 A˚, of UppP, a phosphatase for conversion of C55-PP to C55-P for use in both PG and WTA biogenesis. The structure provides insight into the unique phosphatase activity deep within the membrane bilayer including proposed substrate and metal binding and further suggest the transporter like fold of UppP may act in catalytically coupled recycling of C55-P precursor back to the cytosol. 15. El Ghachi M, Howe N, Huang C-Y, Olieric V, Warshamanage R, Touze´ T, Weichert D, Stansfeld PJ, Wang M, Kerff F et al.: Crystal
structure of undecaprenyl-pyrophosphate phosphatase and its role in peptidoglycan biosynthesis. Nat Commun 2018, 9:1078. This paper, alongside the Workman et al. paper, provides the first structure, at 2.6 A˚, of UppP, a phosphatase for conversion of C55-PP to C55-P for use in both PG and WTA biogenesis. 16. Chung BC, Zhao J, Gillespie RA, Kwon D-Y, Guan Z, Hong J,

Zhou P, Lee S-Y: Crystal structure of MraY, an essential membrane enzyme for bacterial cell wall synthesis. Science 2013, 341:1012-1016. This paper provides the first structure of MraY, an essential N-glycosylase in lipid II synthesis. MraY is an entising antibacterial drug target, due to its essential nature for bacterial growth and division. 17. Chung BC, Mashalidis EH, Tanino T, Kim M, Matsuda A, Hong J, Ichikawa S, Lee S-Y: Structural insights into inhibition of lipid I
production in bacterial cell wall synthesis. Nature 2016, 533:557-560. This paper provides the first structural insight into inhibition of MraY. A structure was solved of MraY in complex with the naturally occurring inhibitor muraymycin D2 highlighting the plasticity of the enzyme’s active site. 18. Hakulinen JK, Hering J, Bra¨nde´n G, Chen H, Snijder A, Ek M,

Johansson P: MraY-antibiotic complex reveals details of tunicamycin mode of action. Nat Chem Biol 2017, 13:265-267. This paper provides the structure of MraY-tunicamycin complex. Derivatization of tunicamycin poses an enticing avenue for the development of novel and effective antibacterial agents. 19. Yoo J, Mashalidis EH, Kuk ACY, Yamamoto K, Kaeser B, Ichikawa S, Lee S-Y: GlcNAc-1-P-transferase-tunicamycin
complex structure reveals basis for inhibition of Nglycosylation. Nat Struct Mol Biol 2018, 25:217-224. This paper provides the first structure of the human GlcNAc-1-P-transferase (GPT). This enzyme is of interest, as the potent MraY/TarO inhibitor, tunicamycin, is seen to inhibit GPT as well. Allows for structure guided development of MraY/TarO inhibitors that are not cross inhibitory for GPT. 20. Al-Dabbagh B, Olatunji S, Crouvoisier M, El Ghachi M, Blanot D, Mengin-Lecreulx D, Bouhss A: Catalytic mechanism of MraY and WecA, two paralogues of the polyprenyl-phosphate Nacetylhexosamine 1-phosphate transferase superfamily. Biochimie 2016, 127:249-257. 21. Stickgold RA, Neuhaus FC: On the initial stage in peptidoglycan synthesis. Effect of 5-fluorouracil substitution on phospho-Nacetylmuramyl-pentapeptide translocase (uridine 50 phosphate). J Biol Chem 1967, 242:1331-1337. 22. Al-Dabbagh B, Henry X, El Ghachi M, Auger G, Blanot D, Parquet C, Mengin-Lecreulx D, Bouhss A: Active site mapping of MraY, a member of the polyprenyl-phosphate Nacetylhexosamine 1-phosphate transferase superfamily, catalyzing the first membrane step of peptidoglycan biosynthesis. Biochemistry 2008, 47:8919-8928. 23. van Heijenoort Y, Go´mez M, Derrien M, Ayala J, van Heijenoort J: Membrane intermediates in the peptidoglycan metabolism of Escherichia coli: possible roles of PBP 1b and PBP 3. J Bacteriol 1992, 174:3549-3557. 24. Bouhss A, Crouvoisier M, Blanot D, Mengin-Lecreulx D: Purification and characterization of the bacterial MraY translocase catalyzing the first membrane step of peptidoglycan biosynthesis. J Biol Chem 2004, 279:2997429980. 25. Hu Y, Chen L, Ha S, Gross B, Falcone B, Walker D, Mokhtarzadeh M, Walker S: Crystal structure of the MurG:UDPGlcNAc complex reveals common structural principles of a Current Opinion in Structural Biology 2018, 53:45–58

superfamily of glycosyltransferases. Proc Natl Acad Sci U S A 2003, 100:845-849. 26. van Dam V, Sijbrandi R, Kol M, Swiezewska E, de Kruijff B, Breukink E: Transmembrane transport of peptidoglycan precursors across model and bacterial membranes. Mol Microbiol 2007, 64:1105-1114. 27. Ruiz N: Bioinformatics identification of MurJ (MviN) as the peptidoglycan lipid II flippase in Escherichia coli. Proc Natl Acad Sci U S A 2008, 105:15553-15557. 28. Kuk ACY, Mashalidis EH, Lee S: Crystal structure of the MOP

flippase MurJ in an inward-facing conformation. Nat Struct Mol Biol 2017, 24:171-176. This paper provides the first structure of MurJ, the likely flippase for the transition of lipid II across the cytosolic membrane. MurJ is the first MATE family protein characterized in an inward-facing conformation. This structure provides insight into the conformational rearrangement of not only MurJ, but numerous other MATE transporters. 29. Mohammadi T, van Dam V, Sijbrandi R, Vernet T, Zapun A, Bouhss A, Diepeveen-de Bruin M, Nguyen-Diste`che M, de Kruijff B, Breukink E: Identification of FtsW as a transporter of lipid-linked cell wall precursors across the membrane. EMBO J 2011, 30:1425-1432. 30. Fraipont C, Alexeeva S, Wolf B, van der Ploeg R, Schloesser M, den Blaauwen T, Nguyen-Diste`che M: The integral membrane FtsW protein and peptidoglycan synthase PBP3 form a subcomplex in Escherichia coli. Microbiology 2011, 157:251259. 31. Leclercq S, Derouaux A, Olatunji S, Fraipont C, Egan AJF, Vollmer W, Breukink E, Terrak M: Interplay between Penicillinbinding proteins and SEDS proteins promotes bacterial cell wall synthesis. Sci Rep 2017, 7:43306. 32. Meeske AJ, Riley EP, Robins WP, Uehara T, Mekalanos JJ,
Kahne D, Walker S, Kruse AC, Bernhardt TG, Rudner DZ: SEDS proteins are a widespread family of bacterial cell wall polymerases. Nature 2016, 537:634-638. This paper provided the first significant evidence that SEDs proteins may function as PG GTases. Overexpression of RodA was shown to compensate for deletion of aPBPsin vivo. Preliminary in vitro evidence that RodA is able to polymerise PG. 33. Emami K, Guyet A, Kawai Y, Devi J, Wu LJ, Allenby N, Daniel RA, Errington J: RodA as the missing glycosyltransferase in Bacillus subtilis and antibiotic discovery for the peptidoglycan polymerase pathway. Nat Microbiol 2017, 2:16253. 34. Ovchinnikov S, Kinch L, Park H, Liao Y, Pei J, Kim DE, Kamisetty H, Grishin NV, Baker D: Large-scale determination of previously unsolved protein structures using evolutionary information. Elife 2015, 4:e09248. 35. Egan AJF, Biboy J, van’t Veer I, Breukink E, Vollmer W: Activities and regulation of peptidoglycan synthases. Philos Trans R Soc B Biol Sci 2015, 370:20150031. 36. Sathiyamoorthy K, Vijayalakshmi J, Tirupati B, Fan L, Saper MA: Structural analyses of the Haemophilus influenzae peptidoglycan synthase activator LpoA suggest multiple conformations in solution. J Biol Chem 2017, 292:17626-17642. 37. Egan AJF, Jean NL, Koumoutsi A, Bougault CM, Biboy J, Sassine J, Solovyova AS, Breukink E, Typas A, Vollmer W et al.: Outer-membrane lipoprotein LpoB spans the periplasm to stimulate the peptidoglycan synthase PBP1B. Proc Natl Acad Sci U S A 2014, 111:8197-8202. 38. King DT, Lameignere E, Strynadka NCJ: Structural insights into the lipoprotein outer membrane regulator of penicillin-binding protein 1B. J Biol Chem 2014, 289:19245-19253. 39. Lovering AL, de Castro LH, Lim D, Strynadka NCJ: Structural
insight into the transglycosylation step of bacterial cell-wall biosynthesis. Science 2007, 315:1402-1405. This paper provides the first structure of a bi-functional, class A, penicillin binding protein. Class A PBPs have long been considered enticing antibiotic targets. b-lactam antibiotics are seen to inhibit the TP domain of these aPBPs. This was the first structure to provide insight into the GTase activity of these proteins. www.sciencedirect.com

Enzyme structures of the bacterial peptidoglycan and wall teichoic acid biogenesis pathways Caveney, Li and Strynadka 57

40. King DT, Wasney GA, Nosella M, Fong A, Strynadka NCJ: Structural insights into inhibition of Escherichia coli penicillinbinding protein 1B. J Biol Chem 2017, 292:979-993. 41. Huang C-Y, Shih H-W, Lin L-Y, Tien Y-W, Cheng T-JR, Cheng WC, Wong C-H, Ma C: Crystal structure of Staphylococcus aureus transglycosylase in complex with a lipid II analog and elucidation of peptidoglycan synthesis mechanism. Proc Natl Acad Sci U S A 2012, 109:6496-6501. 42. King DT, Wasney GA, Nosella M, Fong A, Strynadka NCJ: Structural insights into inhibition of Escherichia coli penicillinbinding protein 1B. J Biol Chem 2017, 292:979-993. 43. Lim D, Strynadka NCJ: Structural basis for the beta lactam resistance of PBP2a from methicillin-resistant Staphylococcus aureus. Nat Struct Biol 2002, 9:870-876. 44. Otero LH, Rojas-Altuve A, Llarrull LI, Carrasco-Lopez C, Kumarasiri M, Lastochkin E, Fishovitz J, Dawley M, Hesek D, Lee M et al.: How allosteric control of Staphylococcus aureus penicillin binding protein 2a enables methicillin resistance and physiological function. Proc Natl Acad Sci U S A 2013, 110:16808-16813. 45. Mahasenan KV, Molina R, Bouley R, Batuecas MT, Fisher JF, Hermoso JA, Chang M, Mobashery S: Conformational dynamics in penicillin-binding protein 2a of methicillin-resistant Staphylococcus aureus, allosteric communication network and enablement of catalysis. J Am Chem Soc 2017, 139:21022110. 46. Sung M, Lai Y, Huang C-Y, Chou L, Shih H-W, Cheng W-C, Wong C-H, Ma C: Crystal structure of the membrane-bound bifunctional transglycosylase PBP1b from Escherichia coli. Proc Natl Acad Sci U S A 2009, 106:8824-8829. 47. Lavollay M, Arthur M, Fourgeaud M, Dubost L, Marie A, Veziris N, Blanot D, Gutmann L, J-L Mainardi: The peptidoglycan of stationary-phase Mycobacterium tuberculosis predominantly contains cross-links generated by L,D-transpeptidation. J Bacteriol 2008, 190:4360-4366. 48. Hugonnet J-E, Mengin-Lecreulx D, Monton A, den Blaauwen T, Carbonnelle E, Veckerle´ C, Brun YV, van Nieuwenhze M, Bouchier C, Tu K et al.: Factors essential for L,Dtranspeptidase-mediated peptidoglycan cross-linking and b-lactam resistance in Escherichia coli. Elife 2016, 5:1-22. 49. Vollmer W: Structural variation in the glycan strands of bacterial peptidoglycan. FEMS Microbiol Rev 2008, 32:287-306. 50. Denapaite D, Bru¨ckner R, Hakenbeck R, Vollmer W: Biosynthesis of teichoic acids in Streptococcus pneumoniae and closely related species: lessons from genomes. Microb Drug Resist 2012, 18:344-358. 51. Gisch N, Kohler T, Ulmer AJ, Mut¨hing J, Pribyl T, Fischer K, Lindner B, Hammerschmidt S, Zah¨ringer U: Structural reevaluation of Streptococcus pneumoniae lipoteichoic acid and new insights into its immunostimulatory potency. J Biol Chem 2013, 288:15654-15667. 52. Pereira MP, Brown ED: Chapter 19 – biosynthesis of cell wall teichoic acid polymers. Microb Glycobiol 2010 http://dx.doi.org/ 10.1016/B978-0-12-374546-0.00019-5. 53. Soldo B, Lazarevic V, Karamata D: tagO is involved in the synthesis of all anionic cell-wall polymers in Bacillus subtilis 168. Microbiology 2002, 148:2079-2087. 54. Lee SH, Wang H, Labroli M, Koseoglu S, Zuck P, Mayhood T, Gill C, Mann P, Sher X, Ha S et al.: TarO-specific inhibitors of wall teichoic acid biosynthesis restore b-lactam efficacy against methicillin-resistant staphylococci. Sci Transl Med 2016, 8:329ra. 55. Mann PA, Mu¨ller A, Wolff KA, Fischmann T, Wang H, Reed P, Hou Y, Li W, Mu¨ller CE, Xiao J et al.: Chemical genetic analysis and functional characterization of Staphylococcal wall teichoic acid 2-epimerases reveals unconventional antibiotic drug targets. PLOS Pathog 2016, 12:e1005585. 56. D’Elia Ma, Henderson Ja, Beveridge TJ, Heinrichs DE, Brown ED: The N-acetylmannosamine transferase catalyzes the first www.sciencedirect.com

committed step of teichoic acid assembly in Bacillus subtilis and Staphylococcus aureus. J Bacteriol 2009, 191:4030-4034. 57. Brown S, Zhang Y, Walker S: A revised pathway proposed for Staphylococcus aureus wall teichoic acid biosynthesis based on in vitro reconstitution of the intracellular steps. Chem Biol 2008, 15:12-21. 58. Lovering AL, Lin LY-C, Sewell EW, Spreter T, Brown ED,

Strynadka NCJ: Structure of the bacterial teichoic acid polymerase TagF provides insights into membrane association and catalysis. Nat Struct Mol Biol 2010, 17:582-589. This study presents a set ofS. epidermidis TagF crystal structures encompassing the membrane-binding domain and catalytic domain. These structures provide insight into membrane-binding and substrate-binding features, as well as oligomerization and catalysis. 59. Nathenson SG, Ishimoto N, Anderson JS, Strominger JL: Enzymatic synthesis and immunochemistry of alpha- and beta-N-acetylglucosaminylribitol linkages in teichoic acids from several strains of Staphylococcus aureus. J Biol Chem 1966, 241:651-658. 60. Brown S, Xia G, Luhachack LG, Campbell J, Meredith TC, Chen C, Winstel V, Gekeler C, Irazoqui JE, Peschel A et al.: Methicillin resistance in Staphylococcus aureus requires glycosylated wall teichoic acids. Proc Natl Acad Sci U S A 2012, 109:1890918914. 61. Xia G, Maier L, Sanchez-Carballo P, Li M, Otto M, Holst O, Peschel A: Glycosylation of wall teichoic acid in Staphylococcus aureus by TarM. J Biol Chem 2010, 285:13405-13415. 62. Winstel V, Liang C, Sanchez-Carballo P, Steglich M, Munar M, Bro¨ker BM, Penade´s JR, Nu¨bel U, Holst O, Dandekar T et al.: Wall teichoic acid structure governs horizontal gene transfer between major bacterial pathogens. Nat Commun 2013, 4:2345. 63. Sobhanifar S, Worrall LJ, Gruninger RJ, Wasney Ga, Blaukopf M,
Baumann L, Lameignere E, Solomonson M, Brown ED, Withers SG et al.: Structure and mechanism of Staphylococcus aureus TarM, the wall teichoic acid a-glycosyltransferase. Proc Natl Acad Sci U S A 2015, 112:E576-E585. A series ofS. aureus TarM crystal structures are presented in this paper. Notably, a structure of an enzyme-catalyzed product complex provided support for an internal nucleophilic substitution-like mechanism of anomeric retention proposed for retaining GT-B family members. 64. Koc¸ C, Gerlach D, Beck S, Peschel A, Xia G, Stehle T: Structural and enzymatic analysis of TarM glycosyltransferase from Staphylococcus aureus reveals an oligomeric protein specific for the glycosylation of wall teichoic acid. J Biol Chem 2015, 290:9874-9885. 65. Sobhanifar S, Worrall LJ, King DT, Wasney GA, Baumann L,
Gale RT, Nosella M, Brown ED, Withers SG, Strynadka NCJ: Structure and mechanism of Staphylococcus aureus TarS, the wall teichoic acid b-glycosyltransferase involved in methicillin resistance. PLoS Pathog 2016, 12:e1006067. This study presents apo and substrate/product-bound structures ofS. aureus TarS. Loops with crucial roles in substrate binding and catalysis were identified, and stacked carbohydrate binding modules were found to mediate trimerization. 66. Lee SS, Hong SY, Errey JC, Izumi A, Davies GJ, Davis BG: Mechanistic evidence for a front-side, SNi-type reaction in a retaining glycosyltransferase. Nat Chem Biol 2011, 7:631-638. 67. Lazarevic V, Karamata D: The tagGH operon of Bacillus subtilis 168 encodes a two-component ABC transporter involved in the metabolism of two wall teichoic acids. Mol Microbiol 1995, 16:345-355. 68. Ward A, Reyes CL, Yu J, Roth CB, Chang G: Flexibility in the ABC transporter MsbA: alternating access with a twist. Proc Natl Acad Sci U S A 2007, 104:19005-19010. 69. Mi W, Li Y, Yoon SH, Ernst RK, Walz T, Liao M: Structural basis of MsbA-mediated lipopolysaccharide transport. Nature 2017, 549:233-237. 70. Perez C, Gerber S, Boilevin J, Bucher M, Darbre T, Aebi M, Reymond J, Locher KP: Structure and mechanism of an active lipid-linked oligosaccharide flippase. Nature 2015, 524:433438. Current Opinion in Structural Biology 2018, 53:45–58

58 Catalysis and regulation

71. Kawai Y, Marles-Wright J, Cleverley RM, Emmins R, Ishikawa S,

Kuwano M, Heinz N, Bui NK, Hoyland CN, Ogasawara N et al.: A widespread family of bacterial cell wall assembly proteins. EMBO J 2011, 30:4931-4941. This study reports the involvement of LCP enzymes in catalyzing the transfer of anionic bacterial cell wall polymers to PG. The LCP enzymes came into focus when they were found to associate with the MreB cytoskeleton. Pyrophosphatase activity was demonstrated and crystal structures of an LCP enzyme in complex with different polyprenyl lipids were reported. 72. Harrison J, Lloyd G, Joe M, Lowary TL, Reynolds E, Walters-Morgan H, Bhatt A, Lovering A, Besra GS, Alderwick LJ: Lcp1 is a phosphotransferase responsible for ligating arabinogalactan to peptidoglycan in Mycobacterium tuberculosis. MBio 2016, 7:1-12. 73. Schaefer K, Matano LM, Qiao Y, Kahne D, Walker S: In vitro reconstitution demonstrates the cell wall ligase activity of LCP proteins. Nat Chem Biol 2017, 13:396-401.

Current Opinion in Structural Biology 2018, 53:45–58

74. Gale RT, Li FKK, Sun T, Strynadka NCJ, Brown ED: B. subtilis LytR-CpsA-Psr enzymes transfer wall teichoic acids from authentic lipid-linked substrates to mature peptidoglycan in vitro. Cell Chem Biol 2017, 24:1537-1546 e4. 75. Hu¨bscher J, Lu¨thy L, Berger-Ba¨chi B, Stutzmann Meier P: Phylogenetic distribution and membrane topology of the LytRCpsA-Psr protein family. BMC Genom 2008, 9:617. 76. Eberhardt A, Hoyland CN, Vollmer D, Bisle S, Cleverley RM, Johnsborg O, Ha˚varstein LS, Lewis RJ, Vollmer W: Attachment of capsular polysaccharide to the cell wall in Streptococcus pneumoniae. Microb Drug Resist 2012, 18:240-255. 77. Schaefer K, Owens TW, Kahne D, Walker S: Substrate preferences establish the order of cell wall assembly in Staphylococcus aureus. J Am Chem Soc 2018, 140:2442-2445.

www.sciencedirect.com