Introduction
Bacterial lipoproteins comprise a family of proteins that are posttranslationally modified with acyl chains following their transport across the plasma membrane (1,2). These proteins play a key role in a variety of cellular processes, including nutrient uptake, sporulation, cell-wall synthesis and protein transport. Moreover, lipoproteins are clinically important as they are involved in conferring antibiotic resistance and are required for virulence in some pathogenic bacteria such as Neisseria meningitidis and Streptococcus pneumoniae (3). Lipoproteins are produced by both Gram-positive and Gram-negative organisms. In Gram-negative bacteria, the vast majority of lipoproteins is localized to the inner leaflet of the outer membrane (facing the periplasm), while in Gram-positive species they are anchored to the outer leaflet of the plasma membrane (1,2) and are thus exposed to the extracellular surface (figure 1). Lipoproteins are typically synthesized in the cytosol as precursors (preprolipoproteins) equipped with a short signal peptide of about 20 residues. This contains a highly conserved motif, the lipobox, that is modified with covalent addition of acyl chains to the thiol group of a crucial cysteine residue and includes the cleavage site for signal peptidase II (1,2). Lipoproteins are easily identified in silico through their lipobox motif and this enables genome-wide predictions of the number of lipoproteins.
For example, it has been estimated that the genome of the Gram-negative bacterium Escherichia coli encodes about 100 potential lipoproteins, whereas the proteome of the Gram-positive pathogen Staphylococcus aureus encompasses 70 putative lipoproteins (4,5). Most lipoprotein precursors are translocated across the cytoplasmic membrane via the SecYEG translocon. However, evidence has been presented showing that a subset of lipoproteins is secreted through the TAT system (1,2). Maturation and processing of lipoproteins occurs after translocation across the plasma membrane at the extracytoplasmic side and requires three membrane-embedded enzymes, namely: diacylglyceryl transferase (Lgt), signal peptidase II (Lsp/SPaseII) and N-acyl transferase (Lnt). Lgt attaches a diacyl moiety to the lipobox cysteine of prolipoproteins. Subsequently, the signal peptide is proteolytically removed from the diacylated lipoprotein precursor by Lsp, yielding apolipoprotein with the lipobox cysteine as new N-terminal residue. This cysteine is in Gram-negative and some Gram-positive bacteria subjected to acetylation by Lnt, resulting in the final triacylated, mature lipoprotein (1,2). Many Gram-negative lipoproteins are localized to the outer membrane, which implies that, following their maturation in the plasma membrane, these proteins are transported to their final destination. In Gram-positive bacteria, however, lipoprotein maturation represents their final biosynthesis step. Lipoprotein maturation and processing is essential for Gram-negative cell viability, while their biogenesis is not vital in Gram-positive bacteria. The enzymes that catalyze the posttranslational modification steps in the biogenesis of lipoproteins are not present in eukaryotes and therefore the lipoprotein maturation pathway represents an attractive target for the development of novel antibiotics. With regards to this it should be noted that detailed structural information is available for all lipoprotein maturation enzymes (6,7,8), while recently high-resolution structures of SPaseII from the human pathogen methicillin-resistant S. aureus complexed to different antibiotics were reported (9), providing profound insight into their inhibitory mechanism. It can therefore be expected that these novel structures will contribute to the design of novel small molecule drugs that interfere with lipoprotein maturation. Here, I will discuss these recent structures of SPaseII as well as their current functional understanding.
Overview of lipoprotein biosynthesis
Bacterial lipoproteins are a large group of secreted proteins with many different functions, ranging from transport to cell-wall biosynthesis (1,2). These proteins are produced by virtually all bacteria, including important human pathogens such as Yersinia pestis and Mycobacterium tuberculosis. Not surprisingly, lipoproteins are required for virulence in many pathogens (3). About 2-3% of all the genes that make up a typical bacterial genome are predicted to encode lipoproteins. The defining feature of bacterial lipoproteins is the presence of a N-terminal signal peptide, followed by a canonical cysteine of the lipobox that is posttranslationally modified through covalent addition of acyl chains. Figure 2 (adapted from 7) shows the posttranslational modification steps of Gram-negative lipoproteins. These are initially synthesized as cytosolic preprolipoprotein precursors that mainly utilize the SecYEG machinery for translocation across the cytoplasmic membrane, although a subset of lipoproteins uses the TAT system for this purpose. Following membrane translocation, the preprolipoprotein precursor is anchored to the cytoplasmic membrane through its N-terminal signal peptide (green) that is inserted into the lipid bilayer. The processing and maturation of lipoproteins occurs at the periplasmic face of the cytoplasmic membrane and requires three membrane-embedded enzymes. Diacyl glycerol transferase (Lgt) represents the first enzyme of the lipoprotein maturation pathway and catalyzes the modification of the lipobox cysteine with two acyl chains, yielding diacylglycerol-prolipoprotein. Next, SPaseII, the second maturation enzyme, cleaves the signal peptide of diacylated lipoprotein precursor. In fact, this enzyme only accepts the diacylated product of the first maturation step.
As a result of the SPaseII catalyzed reaction, the diacylated lipobox cysteine becomes the first residue of apolipoprotein. Finally, a third acyl group is covalently added to the N-terminal cysteine of apolipoprotein by N-acyl transferase (Lnt), the third maturation enzyme, generating a triacylated mature lipoprotein. Thus, Gram-negative lipoproteins are triacylated, while Gram-positive ones display a greater variety in acylation that include diacylated as well as lyso species (1,2). This is due to the lack of the Lnt enzyme in most Gram-positive bacteria, although it is not well understood how this greater variety in acylation is achieved. Phospholipids from the plasma membrane serve as donor for the lipid moieties of lipoproteins. Specifically, Lgt transfers the diacylglyceryl group from phosphatidylglycerol, while Lnt uses phosphatidylethanolamine as acyl donor (1,2). Maturation represents the final step in the biosynthesis of Gram-positive lipoproteins. In contrast, many Gram-negative lipoproteins (up to 90% in E. coli) are localized to the inner leaflet of the outer membrane. This means that these lipoproteins are transported from the periplasmic face of the cytoplasmic membrane to their final destination in the outer membrane, following maturation and processing. This logistical feat is accomplished by the Lol system, which represents the main machinery by which lipoproteins are shipped from the plasma membrane to the outer membrane (1,2). The Lol system (figure 2) is made up of five proteins (LolABCDE) with components in all parts of the Gram-negative cell envelope, namely: the plasma membrane-embedded ABC transporter LolCDE, a soluble periplasmic chaperone known as LolA and the outer membrane lipoprotein LolB. Work on E. coli lipoproteins has shown that transport of mature lipoproteins is initiated by their extraction from the plasma membrane, which is achieved by the LolCDE complex upon ATP hydrolysis. This ABC transporter comprises two copies of LolD, the ATPase subunit, and one copy each of the membrane-embedded subunits LolC and LolE. LolD hydrolyzes ATP at the cytoplasmic side of the plasma membrane, while LolC and LolE recognize and release mature lipoproteins from the periplasmic leaflet of the plasma membrane. Following membrane extraction, lipoproteins are handed over to LolA for passage across the periplasm and delivery at the outer membrane. Here, LolB, the outer membrane acceptor, receives incoming lipoproteins from LolA and anchors them into the inner leaflet of the outer membrane (figure 2). It has long been accepted that all the components of the Lol pathway were essential for cell viability. However, physiological conditions were reported recently that allow E. coli cells to grow in the absence of both LolA and LolB (10), thereby pointing towards alternative sorting mechanisms for lipoproteins. All Gram-negative lipoproteins mature at the periplasmic leaflet of the plasma membrane and although the majority is transferred to the outer membrane, a subset of lipoproteins is retained in the plasma membrane. This raises the question what molecular features of a lipoprotein determine whether it is transferred to the outer membrane or not. Studies on E. coli lipoproteins identified a simple targeting signal in which aspartate at the second position of the mature lipoprotein (the lipidated cysteine represents the first residue) retains it in the plasma membrane, whereas other residues at this position ensure transfer to the outer membrane. This targeting signal was therefore termed the” +2 rule” and the presence of aspartate at the second position is known as a Lol avoidance signal (1,2).
SPaseII as drug target
The enzymes of the lipoprotein maturation pathway are ubiquitously present in bacteria and are absent from eukaryotic cells, while being essential for Gram-negative cell viability (1,2). Therefore, the lipoprotein maturation enzymes represent an attractive target for the development of novel antibiotics. Of these, SPaseII is of particular pharmaceutical interest owing to its crucial role in lipoprotein maturation in both Gram-negative and Gram-positive bacteria (11). In the latter organisms, SpaseII is also involved in pathogenesis. Inhibitors of this enzyme prevent the proteolytic removal of the lipoprotein signal peptide and cause toxic accumulation of lipoprotein precursors in the plasma membrane. Thus, inhibiting SPaseII is a promising strategy that could be used for the design of novel broad-spectrum antibiotics. However, only two compounds are currently known that block SPaseII, namely: globomycin and myxovirescin (figure 3). Globomycin is a 19-membered hydrophobic cyclic peptide that includes five amino acids (Ser, Thr, Gly, Leu and Ile) with a molecular formula of C32H57N5O2 and a mass of 655 (12). It was first identified in 1978 as an antibiotic produced by different Streptomyces strains and exhibits bactericidal activity towards different Gram-negative and Gram-positive bacteria (11). Myxovirescin is a 28-membered macrolactam lactone with a molecular formula of C35H61NO8 and a mass of 624. This macrocyclic compound is produced by Myxococcus xanthus and displays bactericidal activity against many Gram-negative and some Gram-positive bacteria (13). Despite their broad-spectrum bactericidal activity, globomycin and myxovirescin will probably never make it directly to the clinic owing to their poor stability and cumbersome production (11). Globomycin and myxovirescin, however, emphasize that SPaseII is an effective target for the development of broad-spectrum antibiotics. To further the discovery of robust SPaseII inhibitors, a compound library of 646,275 molecules was screened recently employing a novel FRET-based high-throughput screen for the detection of SPaseII activity (14). This yielded several compounds that effectively inhibited SPaseII activity and displayed a potent inhibitory effect on E. coli cell growth after chemical optimization.
Biochemical and structural features of SPaseII
To fully understand how SPaseII is inhibited and to aid the design of novel inhibitors, detailed structural information of SPaseII in complex with these molecules is required. It is therefore important to note that recently high-resolution structures of SPaseII from the Gram-positive pathogen S. aureus bound to globomycin and myxovirescin were determined at 1.9 and 2.3Å, respectively (9). This enzyme is a small membrane protein of about 18 kDa with four predicted transmembrane domains (TMDs) and is not essential for cell viability in S. aureus unlike in Gram-negative bacteria. Figure 4 shows the structures of SPaseII in ribbon representation and colored according to secondary structure complexed with globomycin (left panel) or myxovirescin (right panel). This reveals that the enzyme comprises two domains, namely: a membrane-embedded part with 4 TMDs and both termini in the cytoplasm as well as an extracytoplasmic domain with the active site. Aspartate 118 and aspartate 136 represent the catalytic dyad located on extracytoplasmic face of the plasma membrane. Additionally, SPaseII contains a subdomain, the β-cradle, made up of several β-sheets located on top of the plasma membrane adjacent to the transmembrane part. The β-cradle is thought to accommodate the residues of the diacylated lipoprotein precursor C-terminally of the cleavage site. Globomycin (in yellow ball and sticks) is located in the substrate-binding pocket and is associated with the enzyme through hydrogen bonds and hydrophobic interactions.
The signal peptide of the lipoprotein precursor is probably helical and is located in the lipid bilayer, whereas the lipobox is thought to form an extended peptide with its cleavage site next to the catalytic aspartates. It is therefore conceivable that globomycin mimics this lipobox sequence. As noted above, myxovirescin is another inhibitor of SPaseII with antibiotic properties. Like globomycin, it is a cyclic compound (figure 3) with hydroxyl and amide groups but they are otherwise chemically unrelated. The overall structure of SPaseII bound to myxovirescin is similar to the structure with globomycin. Interestingly, globomycin and myxovirescin interact with SPaseII through associating with opposite sides of the enzyme. A structural alignment of the globomycin and myxovirescin-bound structures is presented in figure 5, confirming an overall similar structure and revealing a marked difference at the EL loop, spanning asparagine 53 to lysine 63 and includes the conserved tryptophan 57. In the globomycin-containing structure, this loop comprises a half-form helix, while in the complex with myxovirescin this loop is unfolded. This difference points towards considerable flexibility of the EL loop, which ensures effective binding of the antibiotics in the substrate-binding pocket with tryptophan 57 locking globomycin and myxovirescin in place.
Mechanism of SPaseII inhibition
Despite different chemical structures and interacting with opposite sites of the enzyme, globomycin and myxovirescin inhibit SPaseII through a similar mechanism. Figure 6 shows a structural overlay of both antibiotics at the active site (left panel), revealing that globomycin and myxovirescin are bound here in the same way. Globomycin has the OH group of its serine residue located between the catalytic residues (right panel). Likewise, the 6-OH group of myxovirescin is also placed between the catalytic aspartates, although its macrocycle is bound on the other side of the substrate-binding pocket. The OH group of globomycin’s serine and the 6-OH group of myxovirescin represent blocking hydroxyls that prevent Asp118 and Asp136 mediated hydrolysis. Thus, both antibiotics inhibit SPaseII by acting as a non-cleavable catalytic intermediate. Although globomycin and myxovirescin are chemically unrelated, a stretch of 19 C-atoms structurally aligns (indicated by a yellow circle in the right panel) that extends from the active site. This stretch comprises the so-called spine atoms and these probably correspond to a part of the acylated lipoprotein precursor in its bound form. The macrocycle of globomycin and myxovirescin contain a polar and apolar side (figure 3). The apolar part extends into the membrane away from the spine atoms to the left and right of the catalytic center. However, the polar side faces the active site and displays significant structural overlap through the spine atoms.
Conclusion
The global threat of increasing antibiotic resistance emphasizes the need for novel antimicrobials as well as new leads for potential drug targets (11). The inhibition of SPaseII, a key enzyme in the maturation of bacterial lipoproteins, by globomycin and myxovirescin demonstrates that it is an effective target for the development of broad-spectrum antibiotics. Lipoproteins are group of proteins that are posttranslationally modified with acyl chains (1,2). These proteins are produced by nearly all bacteria and fulfill a variety of functions, ranging from transport to cell-wall biosynthesis. Moreover, lipoproteins are required for virulence in many pathogens (3). These are initially synthesized as precursors in the cytosol with an N-terminal signal peptide, which directs it to either the Sec-translocon or Tat machinery for membrane translocation. After passage of the plasma membrane, lipoproteins are subjected to a series of posttranslational maturation and processing steps that are catalyzed by three membrane-embedded enzymes (1,2). These are: diacyl glycerol transferase (Lgt), the first enzyme of the lipoprotein maturation pathway and catalyzes the modification of the lipoprotein precursor with two acyl chain. Next, SPaseII, the second maturation enzyme, cleaves the signal peptide of diacylated lipoprotein precursor. Finally, a third acyl group is covalently added to the N-terminal cysteine of apolipoprotein by N-acyl transferase (Lnt). To fully understand how SPaseII is inhibited and to aid the design of novel inhibitors, detailed structural information of SPaseII in complex with these molecules is required. The recent high-resolution structures of SPaseII from the Gram-positive pathogen S. aureus complexed with globomycin and myxovirescin (9) showed that both antibiotics inhibit SPaseII by acting as a non-cleavable catalytic intermediate. It can therefore be expected that these structures will be valuable tool in the design of novel antimicrobials
References
1. Grabowicz M. 2019. Lipoproteins and Their Trafficking to the Outer Membrane. EcoSal Plus. doi: 10.1128.
2. Buddelmeijer N. The molecular mechanism of bacterial lipoprotein modification--how, when and why? 2015. FEMS Microbiol Rev. 39: 246-261.
3. Kovacs-Simon A, Titball RW, Michell SL. 2011. Lipoproteins of bacterial pathogens. Infect Immun. 79: 548-561.
4. Juncker AS, Willenbrock H, Von Heijne G. et al. 2003. Prediction of lipoprotein signal peptides in Gram-negative bacteria. Protein Sci. 12: 1652-1662.
5. Shahmirzadi SV, Nguyen MT, Götz F. 2016. Evaluation of Staphylococcus aureus Lipoproteins: Role in Nutritional Acquisition and Pathogenicity. Front Microbiol. 7:1404.
6. Mao G, Zhao Y, Kang X. 2016. Crystal structure of E. coli lipoprotein diacylglyceryl transferase. Nat Commun. 7: 10198.
7. Vogeley L, El Arnaout T, Bailey J. 2016. Structural basis of lipoprotein signal peptidase II action and inhibition by the antibiotic globomycin. Science. 351: 876-880.
8. Wiktor M, Weichert D, Howe N. et al. 2017. Structural insights into the mechanism of the membrane integral N-acyltransferase step in bacterial lipoprotein synthesis. Nat Commun. 8: 15952.
9. Olatunji S, Yu X, Bailey J. et al. 2020. Structures of lipoprotein signal peptidase II from Staphylococcus aureus complexed with antibiotics globomycin and myxovirescin. Nat Commun. 11:140.
10. Grabowicz M, Silhavy TJ. 2017. Redefining the essential trafficking pathway for outer membrane lipoproteins. Proc Natl Acad Sci U S A. 114: 4769-4774.
11. Lehman KM, Grabowicz M. 2019. Countering Gram-Negative Antibiotic Resistance: Recent Progress in Disrupting the Outer Membrane with Novel Therapeutics. Antibiotics. 8: pii: E163.
12. Kiho T, Nakayama M, Kogen H. 2003. Total synthesis and NMR conformational study of signal peptidase II inhibitors, globomycin and SF-1902 A5. Tetrahedron. 59: 1685-1697.
13. Xiao Y, Gerth K, Müller R. et al. 2012. Myxobacterium-produced antibiotic TA (myxovirescin) inhibits type II signal peptidase. Antimicrob Agents Chemother. 56: 2014-2021.
14. Kitamura S, Owensby A, Wall D. et al. 2018. Lipoprotein Signal Peptidase Inhibitors with Antibiotic Properties Identified through Design of a Robust In Vitro HT Platform. Cell Chem Biol. 25: 301-308.e12.
Reactie schrijven