Interchangeable domains in the Kdo transferases of Escherichia coli and Haemophilus influenzae

A conserved chaperone protein is required for the formation of a noncanonical type VI secretion system spike tip complex

Type VI secretion systems (T6SSs) are dynamic protein nanomachines found in Gram-negative bacteria that deliver toxic effector proteins into target cells in a contact-dependent manner. Prior to secretion, many T6SS effector proteins require chaperones and/or accessory proteins for proper loading onto the structural components of the T6SS apparatus. However, despite their established importance, the precise molecular function of several T6SS accessory protein families remains unclear. In this study, we set out to characterize the DUF2169 family of T6SS accessory proteins. Using gene co-occurrence analyses, we find that DUF2169-encoding genes strictly co-occur with genes encoding T6SS spike complexes formed by valine-glycine repeat protein G (VgrG) and DUF4150 domains. Although structurally similar to Pro-Ala-Ala-Arg (PAAR) domains, "PAAR-like" DUF4150 domains lack PAAR motifs and instead contain a conserved PIPY motif, leading us to designate them PIPY domains. Next, we present both genetic and biochemical evidence that PIPY domains require a cognate DUF2169 protein to form a functional T6SS spike complex with VgrG. This contrasts with canonical PAAR proteins, which bind VgrG on their own to form functional spike complexes. By solving the first crystal structure of a DUF2169 protein, we show that this T6SS accessory protein adopts a novel protein fold. Furthermore, biophysical and structural modeling data suggest that DUF2169 contains a dynamic loop that physically interacts with a hydrophobic patch on the surface of its cognate PIPY domain. Based on these findings, we propose a model whereby DUF2169 proteins function as molecular chaperones that maintain VgrG-PIPY spike complexes in a secretion-competent state prior to their export by the T6SS apparatus.

SGI1 excludes IncA and IncC plasmids

SGI1 and its many variant forms are integrative mobilizable elements that rely on IncA or IncC plasmids for transfer functions. However, the coexistence of SGI1 with the plasmid is unstable in the longer term. Here, we have investigated the effect of SGI1 type integrative elements on the initial entry of these plasmids. Using two transfer proficient IncC plasmids and the IncA plasmid RA1, exclusion indices were 40-100-fold for SGI1-I or SGI1-D which have a complete backbone. Using the SGI1-K and SGI1-LK1 variants that lack backbone segments, loss of a region of 5793 bp that includes the traHG transfer genes and the downstream open reading frame S010 was found to abolish exclusion. S010 was shown to be co-transcribed with traHG and hence also under the control of an AcaDC inducible promoter. However, complementation with a 5.2 kbp fragment that included the traHG-S010 operon did not restore exclusion activity to SGI1-LK1. Part of S013 that encodes a small polypeptide of unknown function, was also lost from SGI1-LK1. S013 and the adjacent S014 gene were also co-transcribed. However complementation with S013-S014 did not restore exclusion activity to SGI1-LK1. Hence, the precise cause of the SGI1-mediated plasmid exclusion remains elusive.

Gamma-Mobile-Trio systems are mobile elements rich in bacterial defensive and offensive tools

The evolutionary arms race between bacteria and phages led to the emergence of bacterial immune systems whose diversity and dynamics remain poorly understood. Here we use comparative genomics to describe a widespread genetic element, defined by the presence of the Gamma-Mobile-Trio (GMT) proteins, that serves as a reservoir of offensive and defensive tools. We demonstrate, using Vibrio parahaemolyticus as a model, that GMT-containing genomic islands are active mobile elements. Furthermore, we show that GMT islands' cargoes contain various anti-phage defence systems, antibacterial type VI secretion system (T6SS) effectors and antibiotic-resistance genes. We reveal four anti-phage defence systems encoded within GMT islands and further characterize one system, GAPS1, showing it is triggered by a phage capsid protein to induce cell dormancy. Our findings underscore the need to broaden the concept of 'defence islands' to include defensive and offensive tools, as both share the same mobile elements for dissemination.

The coral pathogen Vibrio coralliilyticus uses a T6SS to secrete a group of novel anti-eukaryotic effectors that contribute to virulence

Vibrio coralliilyticus is a pathogen of coral and shellfish, leading to devastating economic and ecological consequences worldwide. Although rising ocean temperatures correlate with increased V. coralliilyticus pathogenicity, the specific molecular mechanisms and determinants contributing to virulence remain poorly understood. Here, we systematically analyzed the type VI secretion system (T6SS), a contact-dependent toxin delivery apparatus, in V. coralliilyticus. We identified 2 omnipresent T6SSs that are activated at temperatures in which V. coralliilyticus becomes virulent; T6SS1 is an antibacterial system mediating interbacterial competition, whereas T6SS2 mediates anti-eukaryotic toxicity and contributes to mortality during infection of an aquatic model organism, Artemia salina. Using comparative proteomics, we identified the T6SS1 and T6SS2 toxin arsenals of 3 V. coralliilyticus strains with distinct disease etiologies. Remarkably, T6SS2 secretes at least 9 novel anti-eukaryotic toxins comprising core and accessory repertoires. We propose that T6SSs differently contribute to V. coralliilyticus's virulence: T6SS2 plays a direct role by targeting the host, while T6SS1 plays an indirect role by eliminating competitors.

Engineering Escherichia coli for constitutive production of monophosphoryl lipid A vaccine adjuvant

During the COVID-19 pandemic, expedient vaccine production has been slowed by the shortage of safe and effective raw materials, such as adjuvants, essential components to enhance the efficacy of vaccines. Monophosphoryl lipid A (MPLA) is a potent and safe adjuvant used in human vaccines, including the Shingles vaccine, Shingrix. 3-O-desacyl-4'-monophosphoryl lipid A (MPL), a representative MPLA adjuvant commercialized by GSK, was prepared via chemical conversion of precursors isolated from Salmonella typhimurium R595. However, the high price of these materials limits their use in premium vaccines. To combat the scarcity and high cost of safe raw materials for vaccines, we need to develop a feasible MPLA production method that is easily scaled up to meet industrial requirements. In this study, we engineered peptidoglycan and outer membrane biosynthetic pathways in Escherichia coli and developed a Escherichia coli strain, KHSC0055, that constitutively produces EcML (E. coli-produced monophosphoryl lipid A) without additives such as antibiotics or overexpression inducers. EcML production was optimized on an industrial scale via high-density fed-batch fermentation, and obtained 2.7 g of EcML (about 135,000 doses of vaccine) from a 30-L-scale fermentation. Using KHSC0055, we simplified the production process and decreased the production costs of MPLA. Then, we applied EcML purified from KHSC0055 as an adjuvant for a COVID-19 vaccine candidate (EuCorVac-19) currently in clinical trial stage III in the Philippines. By probing the efficacy and safety of EcML in humans, we established KHSC0055 as an efficient cell factory for MPLA adjuvant production.

The bacterial magnesium transporter MgtA reveals highly selective interaction with specific cardiolipin species

The bacterial magnesium transporter A (MgtA) is a specialized P-type ATPase important for Mg2+ import into the cytoplasm; disrupted magnesium homeostasis is linked to intrinsic ribosome instability and antibacterial resistance in Salmonella strains. Here, we show that MgtA has functional specificity for cardiolipin 18:1. Still, it reaches maximum activity only in combination with cardiolipin 16:0, equivalent to the major components of native cardiolipin found in E. coli membranes. Native mass spectrometry indicates the presence of two binding sites for cardiolipin, agreeing with the kinetic studies revealing that a cooperative relationship likely exists between the two cardiolipin variants. This is the first experimental evidence of cooperative effects between lipids of the same class, with only minor variations in their acyl chain composition, acting on a membrane protein. In summary, our results reveal that MgtA exhibits a highly complex interaction with one cardiolipin 18:1 and one cardiolipin 16:0, affecting protein activity and stability, contributing to our understanding of the particular interactions between lipid environment and membrane proteins. Further, a better understanding of Mg2+ homeostasis in bacteria, due to its role as a virulence regulator, will provide further insights into the regulation and mechanism of bacterial infections.

An Adenosylcobalamin Specific Whole-Cell Biosensor

Vitamin B12 (cobalamin) is essential for human health and its deficiency results in anemia and neurological damage. Vitamin B12 exists in different forms with various bioactivity but most sensors are unable to discriminate between them. Here, a whole-cell agglutination assay that is specific for adenosylcobalamin (AboB12), which is one of two bioactive forms, is reported. This biosensor consists of Escherichia coli that express the AdoB12 specific binding domain of CarH at their surface. In the presence of AdoB12, CarH forms tetramers, which leads to specific bacterial cell-cell adhesions and agglutination. These CarH tetramers disassemble upon green light illumination such that reversion of the bacterial aggregation can serve as internal quality control. The agglutination assay has a detection limit of 500 nм AdoB12, works in protein-poor biofluids such as urine, and has high specificity to AdoB12 over other forms of vitamin B12 as also demonstrated with commercially available supplements. This work is a proof of concept for a cheap and easy-to-readout AdoB12 sensor that can be implemented at the point-of-care to monitor high-dose vitamin B12 supplementation.

Engineering Escherichia coli MG1655 to Efficiently Produce 3-Deacyl-4'-monophosphoryl Lipid A

Monophosphoryl lipid A, derived from Salmonella minnesota R595, has been used in various adjuvant formulations. Escherichia coli can produce lipid A, but its structure is different. In this study, E. coli MG1655 has been engineered to efficiently produce the monophosphoryl lipid A. First, 126 genes relevant to the biosynthesis of the fimbriae, flagella, and ECA were deleted in MG1655, resulting in WQM027. Second, the genes pldA, mlaA, and mlaC related to the phospholipid transport system, the gene ptsG related to the carbohydrate phosphotransferase system, and the gene eptA encoding phosphoethanolamine transferase for lipid A modification were further deleted from WQM027, resulting in MW020. Third, lpxE from Francisella novicida and pagP and pagL from Salmonella were overexpressed in pFT24, resulting in pTEPL. pTEPL was transformed into MW020, resulting in MW020/pTEPL. Finally, fabI encoding an enoyl-ACP reductase was deleted from the genome of MW020/pTEPL, resulting in MW021/pTEPL. MW021/pTEPL could produce 85.31 mg/L of lipid A species after 26 h of fed-batch fermentation. Mainly two monophosphoryl lipid A species were produced in MW021/pTEPL, one is 3-deacyl-2-acyloxyacyl-4'-monophosphoryl lipid A and the other is 3-deacyl-4'-monophosphoryl lipid A. E. coli MW021/pTEPL constructed in this study could be an ideal host for the industrial production of monophosphoryl lipid A.

Effect of the S008-sgaCD operon on IncC plasmid stability in the presence of SGI1-K or absence of an SGI1 variant

An IncC or IncA plasmid is needed to enable transfer of SGI1 type integrative mobilisable elements but an IncC plasmid does not stably co-exist with SGI1. However, the plasmid is stably maintained with SGI1-K, a natural SGI1 deletion variant that lacks the sgaDC genes (S007 and S006) and the upstream open reading frame (S008) found in the SGI1 backbone. Here, the effect of the sgaDC genes and S008 on the stability of an IncC plasmid in an Escherichia coli strain with or without SGI1-K was examined. Co-transcription of the S008 open reading frame with the downstream sgaDC genes was established. When a strain containing SGI1-K complemented with a pK18 plasmid that included S008-sgaDC or sgaDC expressed from the constitutive pUC promoter was grown without antibiotic selection, the resident IncC plasmid was rapidly lost but loss was slower when S008 was present. In contrast, SGI1-K and the S008-sgaDC or sgaDC plasmid were quite stably maintained for >100 generations. However, the high copy number plasmids carrying the SGI1-derived S008-sgaDC or sgaDC genes constitutively expressed could not be introduced into an E. coli strain carrying the IncC plasmid but without SGI1-K. Using equivalent plasmids with S008-sgaDC or sgaDC genes controlled by an arabinose-inducible promoter, under inducing conditions the IncC plasmid was stable but the plasmid containing the SGI1-derived genes was rapidly lost. This unexpected observation indicates that there are multiple interactions between the IncC plasmid and SGI1 in which the transcriptional activator genes sgaDC play a role. These interactions will require further investigation.

Effect of Glycosylation on Self-Assembly of Lipid A Lipopolysaccharides in Aqueous Solutions

Lipopolysaccharides (LPSs) based on lipid A produced by bacteria are of interest due to their bioactivity in stimulating immune responses, as are simpler synthetic components or analogues. Here, the self-assembly in water of two monodisperse lipid A derivatives based on simplified bacterial LPS structures is examined and compared to that of a native Escherichia coli LPS using small-angle X-ray scattering and cryogenic transmission electron microscopy. The critical aggregation concentration is obtained from fluorescence probe experiments, and conformation is probed using circular dichroism spectroscopy. The E. coli LPS is found to form wormlike micelles, whereas the synthetic analogues bearing six lipid chains and with four or two saccharide head groups (Kdo₂-lipid A and monophosphoryl lipid A) self-assemble into nanosheets or vesicles, respectively. These observations are rationalized by considering the surfactant packing parameter.

Interplays of mutations in waaA, cmk, and ail contribute to phage resistance in Yersinia pestis

Plague caused by Yersinia pestis remains a public health threat worldwide. Because multidrug-resistant Y. pestis strains have been found in both humans and animals, phage therapy has attracted increasing attention as an alternative strategy against plague. However, phage resistance is a potential drawback of phage therapies, and the mechanism of phage resistance in Y. pestis is yet to be investigated. In this study, we obtained a bacteriophage-resistant strain of Y. pestis (S56) by continuously challenging Y. pestis 614F with the bacteriophage Yep-phi. Genome analysis identified three mutations in strain S56: waaA* (9-bp in-frame deletion ₂₄₉GTCATCGTG₂₅₇), cmk* (10-bp frameshift deletion ₁₅CCGGTGATAA₂₄), and ail* (1-bp frameshift deletion A₅₃₈). WaaA (3-deoxy-D-manno-octulosonic acid transferase) is a key enzyme in lipopolysaccharide biosynthesis. The waaA* mutation leads to decreased phage adsorption because of the failure to synthesize the lipopolysaccharide core. The mutation in cmk (encoding cytidine monophosphate kinase) increased phage resistance, independent of phage adsorption, and caused in vitro growth defects in Y. pestis. The mutation in ail inhibited phage adsorption while restoring the growth of the waaA null mutant and accelerating the growth of the cmk null mutant. Our results confirmed that mutations in the WaaA-Cmk-Ail cascade in Y. pestis contribute to resistance against bacteriophage. Our findings help in understanding the interactions between Y. pestis and its phages.

Structural basis for the toxic activity of MafB2 from maf genomic island 2 (MGI-2) in N. meningitidis B16B6

The Maf polymorphic toxin system is involved in conflict between strains found in pathogenic Neisseria species such as Neisseria meningitidis and Neisseria gonorrhoeae. The genes encoding the Maf polymorphic toxin system are found in specific genomic islands called maf genomic islands (MGIs). In the MGIs, the MafB and MafI encode toxin and immunity proteins, respectively. Although the C-terminal region of MafB (MafB-CT) is specific for toxic activity, the underlying enzymatic activity that renders MafB-CT toxic is unknown in many MafB proteins due to lack of homology with domain of known function. Here we present the crystal structure of the MafB2-CT_MGI-2B16B6/MafI2_MGI-2B16B6 complex from N. meningitidis B16B6. MafB2-CT_MGI-2B16B6 displays an RNase A fold similar to mouse RNase 1, although the sequence identity is only ~ 14.0%. MafB2-CT_MGI-2B16B6 forms a 1:1 complex with MafI2_MGI-2B16B6 with a Kd value of ~ 40 nM. The complementary charge interaction of MafI2_MGI-2B16B6 with the substrate binding surface of MafB2-CT_MGI-2B16B6 suggests that MafI2_MGI-2B16B6 inhibits MafB2-CT_MGI-2B16B6 by blocking access of RNA to the catalytic site. An in vitro enzymatic assay showed that MafB2-CT_MGI-2B16B6 has ribonuclease activity. Mutagenesis and cell toxicity assays demonstrated that His335, His402 and His409 are important for the toxic activity of MafB2-CT_MGI-2B16B6, suggesting that these residues are critical for its ribonuclease activity. These data provide structural and biochemical evidence that the origin of the toxic activity of MafB2_MGI-2B16B6 is the enzymatic activity degrading ribonucleotides.

An antibacterial T6SS in Pantoea agglomerans pv. betae delivers a lysozyme-like effector to antagonize competitors

The type VI secretion system (T6SS) is deployed by numerous Gram-negative bacteria to deliver toxic effectors into neighbouring cells. The genome of Pantoea agglomerans pv. betae (Pab) phytopathogenic bacteria contains a gene cluster (T6SS1) predicted to encode a complete T6SS. Using secretion and competition assays, we found that T6SS1 in Pab is a functional antibacterial system that allows this pathogen to outcompete rival plant-associated bacteria found in its natural environment. Computational analysis of the T6SS1 gene cluster revealed that antibacterial effector and immunity proteins are encoded within three genomic islands that also harbour arrays of orphan immunity genes or toxin and immunity cassettes. Functional analyses indicated that VgrG, a specialized antibacterial effector, contains a C-terminal catalytically active glucosaminidase domain that is used to degrade prey peptidoglycan. Moreover, we confirmed that a bicistronic unit at the end of the T6SS1 cluster encodes a novel antibacterial T6SS effector and immunity pair. Together, these results demonstrate that Pab T6SS1 is an antibacterial system delivering a lysozyme-like effector to eliminate competitors, and indicate that this bacterium contains additional novel T6SS effectors.

Remodelling of the Gram-negative bacterial Kdo2-lipid A and its functional implications

The lipopolysaccharide (LPS) is a characteristic molecule of the outer leaflet of the Gram-negative bacterial outer membrane, which consists of lipid A, core oligosaccharide, and O antigen. The lipid A is embedded in outer membrane and provides an efficient permeability barrier, which is particularly important to reduce the permeability of antibiotics, toxic cationic metals, and antimicrobial peptides. LPS, an important modulator of innate immune responses ranging from localized inflammation to disseminated sepsis, displays a high level of structural and functional heterogeneity, which arise due to regulated differences in the acylation of the lipid A and the incorporation of non-stoichiometric modifications in lipid A and the core oligosaccharide. This review focuses on the current mechanistic understanding of the synthesis and assembly of the lipid A molecule and its most salient non-stoichiometric modifications.

Molecular basis of glycyl-tRNAGly acetylation by TacT from Salmonella Typhimurium

Bacterial toxin-antitoxin modules contribute to the stress adaptation, persistence, and dormancy of bacteria for survival under environmental stresses and are involved in bacterial pathogenesis. In Salmonella Typhimurium, the Gcn5-related N-acetyltransferase toxin TacT reportedly acetylates the α-amino groups of the aminoacyl moieties of several aminoacyl-tRNAs, inhibits protein synthesis, and promotes persister formation during the infection of macrophages. Here, we show that TacT exclusively acetylates Gly-tRNA^Glyin vivo and in vitro. The crystal structure of the TacT:acetyl-Gly-tRNA^Gly complex and the biochemical analysis reveal that TacT specifically recognizes the discriminator U73 and G71 in tRNA^Gly, a combination that is only found in tRNA^Gly isoacceptors, and discriminates tRNA^Gly from other tRNA species. Thus, TacT is a Gly-tRNA^Gly-specific acetyltransferase toxin. The molecular basis of the specific aminoacyl-tRNA acetylation by TacT provides advanced information for the design of drugs targeting Salmonella.

Engineering Escherichia coli to produce Bordetella pertussis oligosaccharide with multiple trisaccharide units

The immunogenicity of the pertussis vaccine can be significantly improved by adding Bordetella pertussis oligosaccharide with multiple trisaccharide units. The more trisaccharide units there are, the better the efficiency of the immune response induction. However, natural B. pertussis oligosaccharides usually contain only a single terminal trisaccharide unit. In addition, B. pertussis is pathogenic, and there are potential safety hazards when preparing oligosaccharides from B. pertussis. In this study, Escherichia coli MG1655 was engineered to produce B. pertussis oligosaccharides containing multiple trisaccharide units. Fifty-nine genes relevant to the biosynthesis of the O-antigen and core oligosaccharide of lipopolysaccharide, enterobacterial common antigen, and colanic acid were deleted in MG1655, resulting in strain MDCO020. Then, 25 genes relevant to the biosynthesis of the oligosaccharide antigen in B. pertussis and 3 genes relevant to the repeating trisaccharide unit in Pseudomonas aeruginosa PAO1 were overexpressed in MDCO020, resulting in the recombinant E. coli MDCO020/pWpBpD5. The production of B. pertussis oligosaccharide with multiple trisaccharide units by MDCO020/pWpBpD5 was confirmed by SDS-PAGE and ¹H NMR analyses, and its immune response-stimulating activity was confirmed by using rabbit anti-pertussis serum.

Exploring a combined Escherichia coli-based glycosylation and in vitro transglycosylation approach for expression of glycosylated interferon alpha

The conventional use of E. coli system for protein expression is limited to non-glycosylated proteins. While yeast, insect and mammalian systems are available to produce heterologous glycoproteins, developing an engineered E. coli-based glycosylation platform will provide a faster, more economical, and more convenient alternative. In this work, we present a two-step approach for production of a homogeneously glycosylated eukaryotic protein using the E. coli expression system. Human interferon α-2b (IFNα) is used as a model protein to illustrate this glycosylation scheme. In the first step, the N-glycosyltransferase from Actinobacillus pleuropneumoniae (ApNGT) is co-expressed for in vivo transfer of a glucose residue to IFNα at an NX(S/T) N-glycosylation sequon. Several E. coli systems were examined to evaluate the efficiency of IFNα N-glucosylation. In the second step, the N-glucosylated protein is efficiently elaborated with biantennary sialylated complex-type N-glycan using an in vitro chemoenzymatic method. The N-glycosylated IFNα product was found to be biologically active and displayed significantly improved proteolytic stability. This work presents a feasible E. coli-based glycosylation machinery for producing therapeutic eukaryotic glycoproteins.

Metabolic engineering of Escherichia coli to produce a monophosphoryl lipid A adjuvant

Monophosphoryl lipid A (MPLA) species, including MPL (a trade name of GlaxoSmithKline) and GLA (a trade name of Immune Design, a subsidiary of Merck), are widely used as an adjuvant in vaccines, allergy drugs, and immunotherapy to boost the immune response. Even though MPLA is a derivative of lipopolysaccharide (LPS), a component of the outer membrane of Gram-negative bacteria, bacterial strains producing MPLA have not been found in nature nor engineered. In fact, MPLA generation involves expensive and laborious procedures based on synthetic routes or chemical transformation of precursors isolated from Gram-negative bacteria. Here, we report the engineering of an Escherichia coli strain for in situ production and accumulation of MPLA. Furthermore, we establish a succinct method for purifying MPLA from the engineered E. coli strain. We show that the purified MPLA (named EcML) stimulates the mouse immune system to generate antigen-specific IgG antibodies similarly to commercially available MPLA, but with a dramatically reduced manufacturing time and cost. Our system, employing the first engineered E. coli strain that directly produces the adjuvant EcML, could transform the current standard of industrial MPLA production.

Interplay between mitochondria and diet mediates pathogen and stress resistance in Caenorhabditis elegans

Diet is a crucial determinant of organismal biology; interactions between the host, its diet, and its microbiota are critical to determining the health of an organism. A variety of genetic and biochemical means were used to assay stress sensitivity in C. elegans reared on two standard laboratory diets: E. coli OP50, the most commonly used food for C. elegans, or E. coli HT115, which is typically used for RNAi-mediated gene knockdown. We demonstrated that the relatively subtle shift to a diet of E. coli HT115 had a dramatic impact on C. elegans's survival after exposure to pathogenic or abiotic stresses. Interestingly, this was independent of canonical host defense pathways. Instead the change arises from improvements in mitochondrial health, likely due to alleviation of a vitamin B12 deficiency exhibited by worms reared on an E. coli OP50 diet. Increasing B12 availability, by feeding on E. coli HT115, supplementing E. coli OP50 with exogenous vitamin B12, or overexpression of the B12 transporter, improved mitochondrial homeostasis and increased resistance. Loss of the methylmalonyl-CoA mutase gene mmcm-1/MUT, which requires vitamin B12 as a cofactor, abolished these improvements, establishing a genetic basis for the E. coli OP50-incurred sensitivity. Our study forges a mechanistic link between a dietary deficiency (nutrition/microbiota) and a physiological consequence (host sensitivity), using the host-microbiota-diet framework.

Biochemical and Structural Insights into an Fe(II)/α-Ketoglutarate/O2-Dependent Dioxygenase, Kdo 3-Hydroxylase (KdoO)

During lipopolysaccharide biosynthesis in several pathogens, including Burkholderia and Yersinia, 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) 3-hydroxylase, otherwise referred to as KdoO, converts Kdo to d-glycero-d-talo-oct-2-ulosonic acid (Ko) in an Fe(II)/α-ketoglutarate (α-KG)/O₂-dependent manner. This conversion renders the bacterial outer membrane more stable and resistant to stresses such as an acidic environment. KdoO is a membrane-associated, deoxy-sugar hydroxylase that does not show significant sequence identity with any known enzymes, and its structural information has not been previously reported. Here, we report the biochemical and structural characterization of KdoO, Minf_1012 (Kdo_MI), from Methylacidiphilum infernorum V4. The de novo structure of Kdo_MI apoprotein indicates that KdoO_MI consists of 13 α helices and 11 β strands, and has the jelly roll fold containing a metal binding motif, HXDX₁₁₁H. Structures of Kdo_MI bound to Co(II), Kdo_MI bound to α-KG and Fe(III), and Kdo_MI bound to succinate and Fe(III), in addition to mutagenesis analysis, indicate that His146, His260, and Asp148 play critical roles in Fe(II) binding, while Arg127, Arg162, Arg174, and Trp176 stabilize α-KG. It was also observed that His225 is adjacent to the active site and plays an important role in the catalysis of KdoO_MI without affecting substrate binding, possibly being involved in oxygen activation. The crystal structure of KdoO_MI is the first completed structure of a deoxy-sugar hydroxylase, and the data presented here have provided mechanistic insights into deoxy-sugar hydroxylase, KdoO, and lipopolysaccharide biosynthesis.

Crystal structure and activity of Francisella novicida UDP-N-acetylglucosamine acyltransferase

The first step of lipid A biosynthesis in Escherichia coli (E. coli) is catalyzed by LpxA (EcLpxA), an acyltransferase selective for UDP-N-acetylglucosamine (UDP-GlcNAc) and R-3-hydroxymyristoyl-acyl carrier protein (3-OH-C14-ACP), and is an essential step in majority of Gram-negative bacteria. Since the majority of lipid A species isolated from F. novicida contains 3-OH-C16 or 3-OH-C18 at its C3 and C3' positions, FnLpxA was thought to be selective for longer acyl chain (3-OH-C16 and 3-OH-C18) over short acyl chain (3-OH-C14, 3-OH-C12, and 3-OH-C10). Here we demonstrate that Francisella novicida (F. novicida) lpxA functionally complements an E. coli lpxA knockout mutant and efficiently transfers 3-OH-C14 as well as 3-OH-C16 in E. coli. Our results implicate that the acyl chain length of lipid A is determined by several factors including acyl chain selectivity of LpxA and downstream enzymes, as well as the composition of the acyl-ACP pool in vivo. We also report the crystal structure of F. novicida LpxA (FnLpxA) at 2.06 Å. The N-terminal parallel beta-helix (LβH) and C-terminal alpha-helical domain are similar to other reported structures of LpxAs. However, our structure indicates that the supposed ruler residues for hydrocarbon length, 171L in one monomer and 168H in the adjacent monomer in a functional trimer of FnLpxA, are located just 3.8 Å apart that renders not enough space for binding of 3-OH-C12 or longer acyl chains. This implicates that FnLpxA may have an alternative hydrophobic pocket, or the acyl chain may bend while binding to FnLpxA. In addition, the FnLpxA structure suggests a potential inhibitor binding site for development of antibiotics.

Kdo hydroxylase is an inner core assembly enzyme in the Ko-containing lipopolysaccharide biosynthesis

The lipopolysaccharide (LPS) isolated from certain important Gram-negative pathogens including a human pathogen Yersinia pestis and opportunistic pathogens Burkholderia mallei and Burkholderia pseudomallei contains d-glycero-d-talo-oct-2-ulosonic acid (Ko), an isosteric analog of 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo). Kdo 3-hydroxylase (KdoO), a Fe(2+)/α-KG/O2 dependent dioxygenase from Burkholderia ambifaria and Yersinia pestis is responsible for Ko formation with Kdo2-lipid A as a substrate, but in which stage KdoO functions during the LPS biosynthesis has not been established. Here we purify KdoO from B. ambifaria (BaKdoO) to homogeneity for the first time and characterize its substrates. BaKdoO utilizes Kdo2-lipid IVA or Kdo2-lipid A as a substrate, but not Kdo-lipid IVAin vivo as well as in vitro and Kdo-(Hep)kdo-lipid A in vitro. These data suggest that KdoO is an inner core assembly enzyme that functions after the Kdo-transferase KdtA but before the heptosyl-transferase WaaC enzyme during the Ko-containing LPS biosynthesis.

Structural basis of lipid binding for the membrane-embedded tetraacyldisaccharide-1-phosphate 4'-kinase LpxK

The membrane-bound tetraacyldisaccharide-1-phosphate 4'-kinase, LpxK, catalyzes the sixth step of the lipid A (Raetz) biosynthetic pathway and is a viable antibiotic target against emerging Gram-negative pathogens. We report the crystal structure of lipid IVA, the LpxK product, bound to the enzyme, providing a rare glimpse into interfacial catalysis and the surface scanning strategy by which many poorly understood lipid modification enzymes operate. Unlike the few previously structurally characterized proteins that bind lipid A or its precursors, LpxK binds almost exclusively to the glucosamine/phosphate moieties of the lipid molecule. Steady-state kinetic analysis of multiple point mutants of the lipid-binding pocket pinpoints critical residues involved in substrate binding, and characterization of N-terminal helix truncation mutants uncovers the role of this substructure as a hydrophobic membrane anchor. These studies make critical contributions to the limited knowledge surrounding membrane-bound enzymes that act upon lipid substrates and provide a structural template for designing small molecule inhibitors targeting this essential kinase.

Mutants resistant to LpxC inhibitors by rebalancing cellular homeostasis

LpxC, the deacetylase that catalyzes the second and committed step of lipid A biosynthesis in Escherichia coli, is an essential enzyme in virtually all gram-negative bacteria and is one of the most promising antibiotic targets for treatment of multidrug-resistant gram-negative infections. Despite the rapid development of LpxC-targeting antibiotics, the potential mechanisms of bacterial resistance to LpxC inhibitors remain poorly understood. Here, we report the isolation and biochemical characterization of spontaneously arising E. coli mutants that are over 200-fold more resistant to LpxC inhibitors than the wild-type strain. These mutants have two chromosomal point mutations that account for resistance additively and independently; one is in fabZ, a dehydratase in fatty acid biosynthesis; the other is in thrS, the Thr-tRNA ligase. For both enzymes, the isolated mutations result in reduced enzymatic activities in vitro. Unexpectedly, we observed a decreased level of LpxC in bacterial cells harboring fabZ mutations in the absence of LpxC inhibitors, suggesting that the biosyntheses of fatty acids and lipid A are tightly regulated to maintain a balance between phospholipids and lipid A. Additionally, we show that the mutation in thrS slows protein production and cellular growth, indicating that reduced protein biosynthesis can confer a suppressive effect on inhibition of membrane biosynthesis. Altogether, our studies reveal a previously unrecognized mechanism of antibiotic resistance by rebalancing cellular homeostasis.

Structural and mechanistic analysis of the membrane-embedded glycosyltransferase WaaA required for lipopolysaccharide synthesis

WaaA is a key enzyme in the biosynthesis of LPS, a critical component of the outer envelope of Gram-negative bacteria. Embedded in the cytoplasmic face of the inner membrane, WaaA catalyzes the transfer of 3-deoxy-d-manno-oct-2-ulosonic acid (Kdo) to the lipid A precursor of LPS. Here we present crystal structures of the free and CMP-bound forms of WaaA from Aquifex aeolicus, an ancient Gram-negative hyperthermophile. These structures reveal details of the CMP-binding site and implicate a unique sequence motif (GGS/TX(5)GXNXLE) in Kdo binding. In addition, a cluster of highly conserved amino acid residues was identified which represents the potential membrane-attachment and acceptor-substrate binding site of WaaA. A series of site-directed mutagenesis experiments revealed critical roles for glycine 30 and glutamate 31 in Kdo transfer. Our results provide the structural basis of a critical reaction in LPS biosynthesis and allowed the development of a detailed model of the catalytic mechanism of WaaA.

A two-component Kdo hydrolase in the inner membrane of Francisella novicida

Lipid A coats the outer surface of the outer membrane of Gram-negative bacteria. In Francisella tularensis subspecies novicida lipid A is present either as the covalently attached anchor of lipopolysaccharide (LPS) or as free lipid A. The lipid A moiety of Francisella LPS is linked to the core domain by a single 2-keto-3-deoxy-D-manno-octulosonic acid (Kdo) residue. F. novicida KdtA is bi-functional, but F. novicida contains a membrane-bound Kdo hydrolase that removes the outer Kdo unit. The hydrolase consists of two proteins (KdoH1 and KdoH2), which are expressed from adjacent, co-transcribed genes. KdoH1 (related to sialidases) has a single predicted N-terminal transmembrane segment. KdoH2 contains 7 putative transmembrane sequences. Neither protein alone catalyses Kdo cleavage when expressed in E. coli. Activity requires simultaneous expression of both proteins or mixing of membranes from strains expressing the individual proteins under in vitro assay conditions in the presence of non-ionic detergent. In E. coli expressing KdoH1 and KdoH2, hydrolase activity is localized in the inner membrane. WBB06, a heptose-deficient E. coli mutant that makes Kdo(2) -lipid A as its sole LPS, accumulates Kdo-lipid A when expressing the both hydrolase components, and 1-dephospho-Kdo-lipid A when expressing both the hydrolase and the Francisella lipid A 1-phosphatase (LpxE).

Three phosphatidylglycerol-phosphate phosphatases in the inner membrane of Escherichia coli

The phospholipids of Escherichia coli consist mainly of phosphatidylethanolamine, phosphatidylglycerol (PG), and cardiolipin. PG makes up ∼25% of the cellular phospholipid and is essential for growth in wild-type cells. PG is synthesized on the inner surface of the inner membrane from cytidine diphosphate-diacylglycerol and glycerol 3-phosphate, generating the precursor phosphatidylglycerol-phosphate (PGP). This compound is present at low levels (∼0.1% of the total lipid). Dephosphorylation of PGP to PG is catalyzed by several PGP-phosphatases. The pgpA and pgpB genes, which encode structurally distinct PGP-phosphatases, were identified previously. Double deletion mutants lacking pgpA and pgpB are viable and still make PG, suggesting the presence of additional phosphatase(s). We have identified a third PGP-phosphatase gene (previously annotated as yfhB but renamed pgpC) using an expression cloning strategy. A mutant with deletions in all three phosphatase genes is not viable unless covered by a plasmid expressing either pgpA, pgpB, or pgpC. When the triple mutant is covered with the temperature-sensitive plasmid pMAK705 expressing any one of the three pgp genes, the cells grow at 30 but not 42 °C. As growth slows at 42 °C, PGP accumulates to high levels, and the PG content declines. PgpC orthologs are present in many other bacteria.