Outer membrane vesicle production by Escherichia coli is independent of membrane instability

It has been long noted that gram-negative bacteria produce outer membrane vesicles, and recent data demonstrate that vesicles released by pathogenic strains can transmit virulence factors to host cells. However, the mechanism of vesicle release has remained undetermined. This genetic study addresses whether these structures are merely a result of membrane instability or are formed by a more directed process. To elucidate the regulatory mechanisms and physiological basis of vesiculation, we conducted a screen in Escherichia coli to identify gene disruptions that caused vesicle over- or underproduction. Only a few low-vesiculation mutants and no null mutants were recovered, suggesting that vesiculation may be a fundamental characteristic of gram-negative bacterial growth. Gene disruptions were identified that caused differences in vesicle production ranging from a 5-fold decrease to a 200-fold increase relative to wild-type levels. These disruptions included loci governing outer membrane components and peptidoglycan synthesis as well as the sigma(E) cell envelope stress response. Mutations causing vesicle overproduction did not result in upregulation of the ompC gene encoding a major outer membrane protein. Detergent sensitivity, leakiness, and growth characteristics of the novel vesiculation mutant strains did not correlate with vesiculation levels, demonstrating that vesicle production is not predictive of envelope instability.     

Proteomic analysis of outer membrane vesicles from the probiotic strain Escherichia coli Nissle 1917

Escherichia coli Nissle 1917 (EcN) is a probiotic used for the treatment of intestinal disorders. EcN improves gastrointestinal homeostasis and microbiota balance; however, little is known about how this probiotic delivers effector molecules to the host. Outer membrane vesicles (OMVs) are constitutively produced by Gram‐negative bacteria and have a relevant role in bacteria–host interactions. Using 1D SDS–PAGE and highly sensitive LC–MS/MS analysis we identified in this study 192 EcN vesicular proteins with high confidence in three independent biological replicates. Of these proteins, 18 were encoded by strain‐linked genes and 57 were common to pathogen‐derived OMVs. These proteins may contribute to the ability of this probiotic to colonize the human gut as they fulfil functions related to adhesion, immune modulation or bacterial survival in host niches. This study describes the first global OMV proteome of a probiotic strain and provides evidence that probiotic‐derived OMVs contain proteins that can target these vesicles to the host and mediate their beneficial effects on intestinal function. All MS data have been deposited in the ProteomeXchange with identifier PXD000367 ( http://proteomecentral.proteomexchange.org/dataset/PXD000367 ).     

Antigenic architecture of membrane vesicles from Escherichia coli.

The antigenic architecture of membrane vesicles prepared from Escherichia coli ML 308--225 has been studied using crossed immunoelectrophoresis. Progressive immunoadsorption experiments conducted with control vesicles and with physically disrupted vesicles were used to monitor and quantitate the expression of 14 different immunogens. Eleven immunogens, including NADH dehydrogenase (EC 1.6.33.3), D-lactate dehydrogenase (EC 1.1.1.27), dihydro-orotate dehydrogenase (EC 1.3.3.1), 6-phosphogluconate dehydrogenase (EC 1.1.1.43), polynucleotide phosphorylase (EC 2.3.7.8), and beta-galactosidase (EC 3.2.1.23), exhibit minimal expression (10% or less) unless the vesicles are disrupted. Three unidentified antigens are expressed to a similar extent in untreated and disrupted vesicles. Consideration of these and other results [Owen, P., & Kaback, H. R. (1978) Proc. Natl. Acad. Sci. U.S.A. 75, 3148] in terms of membrane polarity, dislocation of antigens, and possible transmembrane orientation of some immunogens reveals that over 95% of the membrane in the vesicle preparations is in the form of sealed sacculi with the same orientation as the intact cell. Furthermore, antigens are distributed across the membrane in a highly asymmetric manner, indicating that dislocation of components from the inner to the outer surface of the membrane during vesicle preparation does not occur to an extent exceeding 10%.     

Immunochemical analysis of membrane vesicles from Escherichia coli.

Membrane vesicles isolated from Escherichia coli ML 308--225 have been analyzed by crossed immunoelectrophoresis, and immunoprecipitates corresponding to the following cellular components have been identified: ATPase (EC 3.6.1,3), two or three NADH dehydrogenases (EC 1.6.99.3), D-lactate dehydrogenase (EC 1.1.1.27), glutamate dehydrogenase (EC 1.4.1.4), dihydro-orotate dehydrogenase (EC 1.3.3.1), 6-phosphogluconate dehydrogenase (EC 1.1.1.43), polynucleotide phosphorylase (EC 2.3.7.8), beta-galactosidase (EC 3.2.1.23), lipopolysaccharide, and Braun's lipoprotein. The cellular origin of many of the vesicle immunogens is determined, and Braun's lipoprotein is used as a marker to quantitate the extent of outer membrane contamination (less than 3%). Membrane antigens are also characterized with regard to their amphiphilic or hydrophilic properties by charge-shift crossed immunoelectrophoresis. Furthermore, the following immunogens cross-react with components in membrane vesicles prepared from Salmonella typhimurium: one of the three NADH dehydrogenases, ATPase, polynucleotide phosphorylase, 6-phosphogluconate dehydrogenase, Braun's lipoprotein, and three unidentified antigens. In the accompanying paper [Owen, P., & Kaback, H. R. (1979) Biochemistry 18 (following paper in this issue)] quantitative immunoadsorption is utilized to establish the topology of the vesicles with respect to the distribution of antigens on the inner and outer faces of the membrane.     

Engineering Escherichia coli for biodiesel production utilizing a bacterial fatty acid methyltransferase

The production of low-cost biofuels in engineered microorganisms is of great interest due to the continual increase in the world's energy demands. Biodiesel is a renewable fuel that can potentially be produced in microbes cost-effectively. Fatty acid methyl esters (FAMEs) are a common component of biodiesel and can be synthesized from either triacylglycerol or free fatty acids (FFAs). Here we report the identification of a novel bacterial fatty acid methyltransferase (FAMT) that catalyzes the formation of FAMEs and 3-hydroxyl fatty acid methyl esters (3-OH-FAMEs) from the respective free acids and S-adenosylmethionine (AdoMet). FAMT exhibits a higher specificity toward 3-hydroxy free fatty acids (3-OH-FFAs) than FFAs, synthesizing 3-hydroxy fatty acid methyl esters (3-OH-FAMEs) in vivo. We have also identified bacterial members of the fatty acyl-acyl carrier protein (ACP) thioesterase (FAT) enzyme family with distinct acyl chain specificities. These bacterial FATs exhibit increased specificity toward 3-hydroxyacyl-ACP, generating 3-OH-FFAs, which can subsequently be utilized by FAMTs to produce 3-OH-FAMEs. PhaG (3-hydroxyacyl ACP:coenzyme A [CoA] transacylase) constitutes an alternative route to 3-OH-FFA synthesis; the coexpression of PhaG with FAMT led to the highest level of accumulation of 3-OH-FAMEs and FAMEs. The availability of AdoMet, the second substrate for FAMT, is an important factor regulating the amount of methyl esters produced by bacterial cells. Our results indicate that the deletion of the global methionine regulator metJ and the overexpression of methionine adenosyltransferase result in increased methyl ester synthesis.     

The membrane potential in everted vesicles of Escherichia coli.

Ion-selective electrodes were used to measure the equilibration of thiocyanate across the membrane of everted (“inside-out”) vesicles of Escherichia coli W1485. Membrane potentials, vesicle interior positive, generated by the oxidation of NADH, succinate, and d-lactate, or by the hydrolysis of ATP, fell in the range of 100–150 mV depending on the carbon source for cell growth and the substrate used to energize the membranes. There was no relationship between the rate of oxidation of different substrates and the membrane potential they generated. The membrane potential generated by oxidation of NADH was relatively constant between pH 7.0 and 8.5. Somewhat lower values obtained at pH 5.5 to 6.5 were attributed to the effect of pH on substrate oxidation.     

Reconstitution of Escherichia coli membrane vesicles with D-amino acid dehydrogenase

When purified D-amino acid dehydrogenase [Olsiewski, P. J., Kaczorowski, G. J., & Walsh, C. T. (1980) J. Biol. Chem. 255, 4487] is incubated with right-side-out membrane vesicles from Escherichia coli, the enzyme binds to the membrane in a time- and concentration-dependent manner. As a result, the vesicles acquire the ability to oxidize D-alanine and catalyze D-alanine-dependent active transport. Similarly, incubation of D-amino acid dehydrogenase with inside-out vesicles results in binding of enzyme and D-alanine oxidase activity. Antibody inhibition studies indicate that the enzyme is bound exclusively to the inner cytoplasmic surface of the membrane in native vesicles (i.e., membrane vesicles prepared from cells induced for D-amino acid dehydrogenase). In contrast, similar studies with reconstituted vesicles demonstrate that enzyme binds to the surface exposed to the medium regardless of the orientation of the membrane. Thus, enzyme bound to right-side-out vesicles is located on the opposite side of the membrane from where it is normally found. Remarkably, in the presence of D-alanine, reconstituted right-side-out and inside-out vesicles generate electrochemical proton gradients of similar magnitude but opposite polarity, indicating that enzyme bound to either surface of the membrane is physiologically functional. The results suggest that vectorial proton translocation via the respiratory chain occurs at a point distal to the site where electrons enter the respiratory chain from the primary dehydrogenase, a conclusion that is inconsistent with the notion that the dehydrogenase forms part of a proton-translocating loop.     

High performance anion exchange chromatography purification of probiotic bacterial extracellular vesicles enhances purity and anti-inflammatory efficacy

Bacterial extracellular vesicles (BEVs), including outer membrane vesicles, have emerged as a promising new class of vaccines and therapeutics to treat cancer and inflammatory diseases, among other applications. However, clinical translation of BEVs is hindered by a current lack of scalable and efficient purification methods. Here, we address downstream BEV biomanufacturing limitations by developing a method for orthogonal size- and charge-based BEV enrichment using tangential flow filtration (TFF) in tandem with high performance anion exchange chromatography (HPAEC). The data show that size-based separation coisolated protein contaminants, whereas size-based TFF with charged-based HPAEC dramatically improved purity of BEVs produced by probiotic Gram-negative Escherichia coli and Gram-positive lactic acid bacteria (LAB). Escherichia coli BEV purity was quantified using established biochemical markers while improved LAB BEV purity was assessed via observed potentiation of anti-inflammatory bioactivity. Overall, this work establishes orthogonal TFF + HPAEC as a scalable and efficient method for BEV purification that holds promise for future large-scale biomanufacturing of therapeutic BEV products.     

Some growth peculiarities and membrane characteristics of probiotic strains of Escherichia coli

It is found that the probiotic strains of Escherichia coli, G35 N 59 isolated from “Okarin” and from “ASAP” drug formulas, and the commensal strain 5–1 isolated from healthy human intestine, manifest higher specific growth rates, lower acidifying capacity during glucose fermentation in the external medium, and lower rates of decrease in redox potential in comparison with the wild-type strain (MC4100). At the same time, despite similar values of their membrane potentials, these bacteria differ essentially in total and N,N′-dicyclohexylcarbodiimide-sensitive rates of energy-dependent transmembrane H⁺ and K⁺ transport and display a low level of H₂ production. It is suggested that the difference between their membrane characteristics reflects changes in the activity of proton-translocating F₀F₁-ATPase and is crucial for bacterial growth and probiotic activity of E. coli.     

Mechanism of the melibiose porter in membrane vesicles of Escherichia coli

The melibiose transport system of Escherichia coli catalyzes sodium--methyl 1-thio-beta-D-galactopyranoside (TMG) symport, and the cation is required not only for respiration-driven active transport but also for binding of substrate to the carrier in the absence of energy and for carrier-mediated TMG efflux. As opposed to the proton--beta-galactoside symport system [Kaczorowski, G. J., & Kaback, H. R. (1979) Biochemistry 18, 3691], efflux and exchange of TMG occur at the same rate, implying that the rates of the two processes are limited by a common step, most likely the translocation of substrate across the membrane. Furthermore, the rate of exchange, as well as efflux, is influenced by imposition of a membrane potential (delta psi; interior negative), suggesting that the ternary complex between sodium, TMG, and the porter may bear a net positive charge. Consistently, energization of the vesicles leads to a large increase in the Vmax for TMG influx, with little or no change in the apparent Km of the process. It is proposed that the sodium gradient (Na+out < Na+in) and the delta psi (interior negative) may affect different steps in the overall mechanism of active TMG accumulation in the following manner: the sodium gradient causes an increased affinity for TMG on the outer surface of the membrane relative to the inside and the delta psi facilitates a reaction involved with the translocation of the positively charged ternary complex to the inner surface of the membrane.     

Uncoupling action of amytal in membrane vesicles from Escherichia coli.

The barbiturate amytal (5-ethyl-5-isopentylbarbituric acid) has been shown to inhibit amino acid transport in membrane vesicles from anaerobically grown Escherichia coli. Amytal has no effect on the activity of the enzymes of the nitrate respiration system, nor on electron transfer in this system. However, addition of amytal to the membrane vesicles results in a decrease of the membrane potential from -90 mV to -72 mV, and to a decrease of the pH-gradient of -61 mV to undetectable values. Furthermore, amytal causes an increase in the rate of ferricyanide reduction in liposomes, indicating that amytal increases the proton permeability of phospholipid membranes. These results demonstrate that amytal acts as an uncoupler in membrane vesicles from anaerobically grown E. coli.     

Functional mosaicism of membrane proteins in vesicles of Escherichia coli.

Membrane vesicles of Escherichia coli prepared by osmotic lysis of lysozyme ethylenediaminetetracetate (EDTA) spheroplasts have approximately 60% of the total membrane-bound reduced nicotinamide adenine dinucleotide (NADH) dehydrogenase (ED 1.6.99.3) and Mg2+-adenosine triphosphatase (ATPase) (EC 3.6.1.3) activities exposed on the outer surface of the inner membrane. Absorption of these vesicles with antiserum prepared against the purified soluble Mg2+-ATPase resulted in agglutination of approximately 95% of the inner membrane vesicles, as determined by dehydrogenase activity, and about 50% of the total membrane protein. The unagglutinated vesicles lacked all dehydrogenase activity and may consist of outer membrane. Lysozyme-EDTA vesicles actively transported calcium ion, using either NADH or adenosine 5'-triphosphate (ATP) as energy source. However, neither D-lactate nor reduced phenazine methosulfate energized calcium uptake, suggesting that the observed calcium uptake was not due to a small population of everted vesicles. Transport of calcium driven by either NADH or ATP was inhibited by simultaneous addition of D-lactate or reduced phenazine methosulfate. Proline transport driven by D-lactate oxidation was inhibited by either NADH oxidation or ATP hydrolysis. These results suggest that the portion of the total population of vesicles capable of active transport, i.e., the inner membrane vesicles, are functionally a homogeneous population but cannot be categorized as either right-side-out or everted, since activities normally associated with only one side of the inner membrane can be found on both sides of the membrane of these vesicles. Moreover, the data indicate that oxidation of NADH or hydrolysis of ATP by externally localized NADH dehydrogenase or Mg2+-ATPase establishes a protonmotive force of the opposite polarity from that established through D-lactate oxidation.     

A powerful approach for effective finding of significantly differentially expressed genes

The problem of identifying significantly differentially expressed genes for replicated microarray experiments is accepted as significant and has been tackled by several researchers. Patterns from Gene Expression (PaGE) and q-values are two of the well-known approaches developed to handle this problem. This paper proposes a powerful approach to handle this problem. We first propose a method for estimating the prior probabilities used in the first version of the PaGE algorithm. This way, the problem definition of PaGE stays intact and we just estimate the needed prior probabilities. Our estimation method is similar to Storey's estimator without being its direct extension. Then, we modify the problem formulation to find significantly differentially expressed genes and present an efficient method for finding them. This formulation increases the power by directly incorporating Storey's estimator. We report the preliminary results on the BRCA data set to demonstrate the applicability and effectiveness of our approach.     

Energetic studies of lactose active transport in Escherichia coli membrane vesicles

The energetics of D-lactate-driven active transport of lactose in right-side-out Escherichia coli membrane vesicles has been investigated with a microcalorimetric method. Changes of enthalpy (delta Hox), free energy (delta Gox), and entropy (delta Sox) during the D-lactate oxidation reaction in the presence of membrane vesicles are -39.9 kcal, -46.4 kcal, and 22 cal/deg per mole of D-lactate, respectively. The free energy released by this reaction is utilized to form a proton electrochemical potential (delta-microH+) across the membrane. The higher observed heat in the D-lactate oxidation reaction in the presence of carbonylcyanide m-chlorophenylhydrazone (a proton ionophore) supports the postulate that delta-microH+ is formed across the membrane vesicles. Thermodynamic quantities for the formation of delta-microH+ are delta Hm = 14.1 kcal, delta Gm = 0.6 kcal, and delta Sm = 45 cal/deg per mole of D-lactate. The efficiency in the free energy transfer from the oxidation reaction to the formation of delta-microH+ (defined by delta Gm/delta Gox) was 2%, as compared to that in the heat transfer (defined by delta Hm/delta Hox) of 35%. The energetics of the movement of lactose in symport with proton across the membrane as a consequence of the formation of delta-microH+ are delta H1 = -19 kcal, delta G1 = -0.5 kcal, and delta S1 = -62 cal/deg per mole of lactose. No heat of reaction is contributed by lactose movement across the membrane without symport with H+.     

Incorporation of heterologous outer membrane and periplasmic proteins into Escherichia coli outer membrane vesicles

Gram-negative bacteria shed outer membrane vesicles composed of outer membrane and periplasmic components. Since vesicles from pathogenic bacteria contain virulence factors and have been shown to interact with eukaryotic cells, it has been proposed that vesicles behave as delivery vehicles. We wanted to determine whether heterologously expressed proteins would be incorporated into the membrane and lumen of vesicles and whether these altered vesicles would associate with host cells. Ail, an outer membrane adhesin/invasin from Yersinia enterocolitica, was detected in purified outer membrane and in vesicles from Escherichia coli strains DH5alpha, HB101, and MC4100 transformed with plasmid-encoded Ail. In vesicle-host cell co-incubation assays we found that vesicles containing Ail were internalized by eukaryotic cells, unlike vesicles without Ail. To determine whether lumenal vesicle contents could be modified and delivered to host cells, we used periplasmically expressed green fluorescent protein (GFP). GFP fused with the Tat signal sequence was secreted into the periplasm via the twin arginine transporter (Tat) in both the laboratory E. coli strain DH5alpha and the pathogenic enterotoxigenic E. coli ATCC strain 43886. Pronase-resistant fluorescence was detectable in vesicles from Tat-GFP-transformed strains, demonstrating that GFP was inside intact vesicles. Inclusion of GFP cargo increased vesicle density but did not result in morphological changes in vesicles. These studies are the first to demonstrate the incorporation of heterologously expressed outer membrane and periplasmic proteins into bacterial vesicles.     

Stapled NRPS enhances the production of valinomycin in Escherichia coli

Nonribosomal peptides (NRPs) are a large family of secondary metabolites with notable bioactivities, which distribute widely in natural resources across microbes and plants. To obtain these molecules, heterologous production of NRPs in robust surrogate hosts like Escherichia coli represent a feasible approach. However, reconstitution of the full biosynthetic pathway in a host often leads to low productivity, which is at least in part due to the low efficiency of enzyme interaction in vivo except for the well-known reasons of metabolic burden (e.g., expression of large NRP synthetases—NRPSs with molecular weights of >100 kDa) and cellular toxicity on host cells. To enhance the catalytic efficiency of large NRPSs in vivo, here we propose to staple NRPS enzymes by using short peptide/protein pairs (e.g., SpyTag/SpyCatcher) for enhanced NRP production. We achieve this goal by introducing a stapled NRPS system for the biosynthesis of the antibiotic NRP valinomycin in E. coli. The results indicate that stapled valinomycin synthetase (Vlm1 and Vlm2) enables higher product accumulation than those two free enzymes (e.g., the maximum improvement is nearly fourfold). After further optimization by strain and bioprocess engineering, the final valinomycin titer maximally reaches about 2800 µg/L, which is 73 times higher than the initial titer of 38 µg/L. We expect that stapling NRPS enzymes will be a promising catalytic strategy for high-level biosynthesis of NRP natural products.     

Mechanisms of active transport in isolated bacterial membrane vesicles. XII. Active transport by a mutant of Escherichia coli uncoupled for oxidative phosphorylation

Amino acid and β-galactoside transport activity catalyzed by whole cells and membrane vesicles prepared from an Escherichia coli mutant uncoupled for oxidative phosphorylation is comparable to the activity of analogous preparations from the parent strain. Valinomycin-induced rubidium uptake is also similar in membrane vesicles prepared from wild-type and mutant cells. The properties of the transport systems in mutant vesicles are the same as those of wild-type vesicles with respect to electron donors which stimulate transport, and with respect to inhibition by anoxia, cyanide, and 2,4-dinitrophenol.Magnesium ion markedly stimulates the ATPase activity of wild-type membrane vesicles and ethylenediaminetetraacetate markedly inhibits. However, these compounds have relatively slight effects on either the initial rate or extent of transport. Dicyclohexylcarbodiimide does not inhibit respiration-dependent transport despite inhibition of the calcium, magnesium-activated ATPase activity of wild-type vesicles.These results confirm earlier observations indicating that oxidative phosphorylation is not involved in respiration-linked active transport.