Escherichia coli

Escherichia coli (commonly abbreviated E. coli) is a rod-shaped, Gram-negative bacterium of the family Enterobacteriaceae. Theodor Escherich first isolated it from infant feces in 1885 and called it Bacterium coli commune. It colonizes the intestines of humans and other warm-blooded animals, where most strains are harmless. Some strains, however, cause diarrhea, urinary tract infections, sepsis, and other illnesses, and E. coli is a standard indicator of fecal contamination in water.

Its significance goes well beyond medicine. Because it grows quickly, is inexpensive to culture, and can be manipulated genetically with precision, E. coli has become one of the principal model organisms of molecular biology. Foundational work on gene regulation, DNA replication, and bacterial genetic exchange was carried out in this organism, and it is the workhorse of recombinant DNA technology, including the production of human insulin.

This entry covers the taxonomy and discovery of E. coli, its cell structure and genome, its metabolism and growth, the regulation of its genes, its genetic exchange and laboratory strains, its role in research and biotechnology, its place in the human gut, its pathogenic variants, its evolution in the laboratory and the environment, and the spread of antibiotic resistance.

Taxonomy, discovery and naming

Classification

Escherichia coli belongs to the domain Bacteria. In the standard classification it sits in the family Enterobacteriaceae, within the order Enterobacterales and the class Gammaproteobacteria. Strains are Gram-negative, facultatively anaerobic, non-spore-forming, rod-shaped organisms, so they can respire oxygen or switch to other metabolic routes when oxygen is scarce, and they produce no endospores. The species is the type species of the genus Escherichia, and that genus in turn is the type genus of the family, which gives the organism a formal anchoring role in the taxonomy of its group (E. coli is the type species of the genus Escherichia, which, in turn, is the type genus of the family Enterobacteriaceae.).

Discovery and the original name

The organism was first described in 1885 by the pediatrician Theodor Escherich (1857–1911), who was studying the intestinal bacteria of newborns and infants. In 1884, while living in Munich, Escherich began research on the bacteria of infant intestinal tracts and how they change after birth. He presented the work on July 14, 1885, to the Society for Morphology and Physiology under the name Bacterium coli commune (Escherich officially presented his work on Bacterium coli commune to the Society for Morphology and Physiology on July 14, 1885.). The epithet refers to the site of isolation: "Coli" refers to the fact that Escherich found the organism in the colon. The original isolate, designated NCTC 86, was isolated in 1885 from a child with no overt signs of diarrhoeal disease, and it was deposited in a national culture collection in 1919 (The strain was deposited in the National Collection of Type Cultures (NCTC) in 1919 by the Lister Institute). A separate type strain was later fixed for the species: in 1963, the strain U5/41, isolated in 1941 from the urine of a Danish patient with cystitis, was selected as the neotype.

The name changed several times. Following a revision of Bacteria it was reclassified as Bacillus coli by Migula in 1895. The current binomial is credited to Castellani and Chalmers in 1919, with formal recognition in the 1980 Approved Lists of bacterial names (Escherichia coli (Migula 1895) Castellani and Chalmers 1919 (Approved Lists 1980).). Sources differ on when the renaming became official. One reference states that the name was adopted in 1919 but was not officially endorsed until 1958.

Internal diversity and the Shigella boundary

The species is heterogeneous. There are E. coli strains that are harmless commensals of the intestinal tract and others that are major pathogens of humans and animals. Lineages are grouped into phylogroups, which are defined mainly by genomic and molecular markers; the historical strains mentioned above fall into subgroup A and subgroup B2 respectively (It belongs to E. coli subgroup A.; It carries a incFII-type plasmid and belongs to subgroup B2.).

The genus Shigella, which causes bacillary dysentery, is the main taxonomic complication. Shigella species were named separately from E. coli, and classification systems for each genus have been developed independently, with somatic (O) antigen serotyping established to identify Shigella species, as they lack surface (K) and flagellar (H) antigens. Genomic data have since undermined that separation. Escherichia coli and Shigella species are closely related; the aggregate biochemical reactions of members of these two genera are similar, and in phylogenomic analyses E. coli and Shigella species are not differentiated into distinct, separate phylogenetic groups. Instead, they form a single, interspersed clade. Some Shigella species carry O antigens that match those of E. coli. For this reason, many microbiologists regard the two as one species complex, while clinical and regulatory practice continues to use the two names.

Serotyping and designations such as O157:H7

Serotyping classifies strains by surface antigens rather than by genome. The O (somatic) antigen is a component of the lipopolysaccharide in the outer membrane, and the H antigen is the flagellin protein of the flagellum. Serotyping was based on the somatic lipopolysaccharide O and flagellar H antigens as detected in agglutination assays with specific rabbit antibodies. The diversity is large: more than 200 different O antigens and more than 50 H antigens have been identified for E. coli. The O-group defines the serogroup, and the combination of the O-antigen and the H-flagellar antigen defines the E. coli serotype.

Designations such as O157:H7 follow this rule. The number 157 refers to the O-group, and the 7 refers to the flagellar type, so the name identifies a specific O/H antigen combination rather than a single genetic lineage. This matters in outbreak work, because serogroup classifications based upon the O-somatic or H-flagellar antigens provide significant epidemiological information on clinical isolates, and this measure can provide the first indication of relatedness between strains during outbreak investigations. The method has limits: cross-reactions, the existence of "O rough" strains that have lost their ability to express the O antigen, and non-motile strains that do not express a flagellar antigen, all hamper the efficiency of the method.

Cell structure and genome

Cell envelope

Escherichia coli is a Gram-negative rod with three main envelope layers. The innermost is the inner membrane, a phospholipid bilayer that carries the respiratory chain and many transport systems. Between it and the outer membrane lies the periplasm, which contains a thin peptidoglycan layer only one or a few sheets thick. The outer membrane is an asymmetric bilayer whose outer leaflet is largely lipopolysaccharide (LPS). LPS has three parts: lipid A, which anchors it in the membrane; a core oligosaccharide; and often an O-antigen chain that extends into the environment and underlies the O serotypes.

This architecture determines the Gram stain. In the procedure, crystal violet and iodine form a complex inside the cell. The thick peptidoglycan mesh of Gram-positive bacteria retains this complex, whereas the decolorizing solvent dissolves the lipids of the Gram-negative outer membrane, so the complex is lost and the safranin counterstain colours the cells pink.

The outer membrane also restricts antibiotic entry. Small hydrophilic molecules pass mainly through porins such as OmpF and OmpC, which form channels that exclude larger solutes, while many hydrophobic drugs cross the lipid layer slowly. Agents that target peptidoglycan synthesis, such as penicillin-class drugs, must therefore reach their enzyme targets in the periplasm. Reduced porin permeability combined with efflux is a common route to resistance.

Surface structures

E. coli is peritrichous, with several flagella distributed over the cell surface. Each flagellum is a helical filament built from the protein flagellin and turned by a rotary motor in the envelope that runs on the proton motive force. Switching the rotation direction produces the run-and-tumble swimming used in chemotaxis.

Pili and fimbriae are thinner, hair-like appendages. Type 1 fimbriae are numerous, and their tip adhesin FimH binds mannosylated host glycoproteins, which supports attachment to the urinary tract and the intestinal wall. Sex pili, encoded by the F plasmid, are different: they contact recipient cells and are needed for conjugative DNA transfer, a process explained in the genetic exchange section.

Some strains also produce a capsule, a polysaccharide layer outside the outer membrane that is named for its K antigens. Capsules help cells resist desiccation and phagocytosis and contribute to biofilm formation. Certain capsule types are associated with neonatal meningitis.

Chromosome and genome

The best-studied laboratory strain, K-12 MG1655, has a single circular double-stranded DNA chromosome of about 4.6 million base pairs that encodes roughly 4,300 protein-coding genes. The chromosome is not enclosed by a membrane. It is compacted into a nucleoid by DNA supercoiling and by nucleoid-associated proteins such as H-NS and HU. Replication starts at one origin, oriC, and proceeds in both directions.

Comparisons among isolates separate the genome into a core and an accessory part. The core genome is shared by nearly all strains and encodes housekeeping functions such as central metabolism, transcription, translation and envelope assembly. The accessory genome varies between strains. It is enriched in prophages, genomic islands, plasmid-derived sequences, and genes for virulence, resistance and specialized metabolism. Because the accessory component differs so widely, the species pan-genome is far larger than any single genome, and one strain may carry many hundreds of genes that another lacks.

Plasmids and mobile elements

Many strains also carry plasmids, circular DNA molecules that replicate independently at characteristic copy numbers. Conjugative plasmids such as the F plasmid encode the machinery to transfer themselves to other cells, while many others are non-conjugative. Plasmids often carry resistance genes and virulence genes, and they are a major vehicle of horizontal gene transfer.

Insertion sequences are short elements, typically about one to two kilobases long, that encode only the enzyme needed for their own movement and are flanked by inverted repeats. They can land within genes and disrupt them, or they can mediate deletions and rearrangements. Transposons are larger and carry extra cargo genes, such as resistance genes, that can move between the chromosome and plasmids.

Integrated prophages are temperate phage genomes that sit in the chromosome at specific sites. The K-12 chromosome contains several such prophages, some of which are cryptic. In enterohemorrhagic strains, Shiga toxin genes are carried on prophages, and their induction releases the toxin.

Metabolism and growth

Escherichia coli is a facultative anaerobe with a flexible, versatile metabolism. It grows well on glucose and many other sugars, organic acids, and amino acids. Glucose is typically broken down to pyruvate by the Embden–Meyerhof–Parnas pathway of glycolysis. Under aerobic conditions, pyruvate is converted to acetyl-CoA and oxidized in the citric acid cycle, and the reduced carriers feed the electron transport chain, with oxygen as the terminal electron acceptor. When oxygen is scarce, the same central pathways continue with alternative acceptors. Nitrate, nitrite, fumarate, and dimethyl sulfoxide can all serve as terminal acceptors in anaerobic respiration, and the choice depends on which enzymes the cell has induced in response to the available electron acceptors. Because of this flexibility, E. coli can exploit the gut, where oxygen is limited, as well as the open environment, where it is abundant.

When oxygen is absent and no suitable respiratory acceptor is available, E. coli relies on mixed-acid fermentation. Pyruvate formate-lyase splits pyruvate into acetyl-CoA and formate, and the cell then reduces its intermediates into a mixture of end products: lactate, acetate, ethanol, succinate, and formate. Formate can be further cleaved by formate hydrogenlyase into carbon dioxide and hydrogen gas. The balance of these products, and especially the strong acid production, is the basis for several classical biochemical tests used to identify Enterobacteriaceae. In the IMViC series (indole, methyl red, Voges–Proskauer, citrate), E. coli is typically indole-positive and methyl red-positive, because its acid products drive the pH low enough to turn the indicator red. It is typically Voges–Proskauer-negative, because it does not channel pyruvate into acetoin, and citrate-negative, because it cannot use citrate as its sole carbon source. This pattern distinguishes it from many closely related species.

Growth in a closed batch culture follows a characteristic sequence, usually described by a bacterial growth curve. After inoculation into fresh medium, cells enter a lag phase in which they synthesize enzymes and ribosomes needed for the new environment. They then enter exponential (logarithmic) growth, during which each cell divides by binary fission at a constant rate. Under optimal conditions in rich medium at 37 °C, the doubling time is about 20 minutes. As nutrients become depleted and waste products accumulate, growth slows and the culture enters stationary phase, in which the number of viable cells is roughly constant and the cells alter their gene expression to survive. Eventually, the population declines in a death phase. The growth rate also depends strongly on temperature and pH. E. coli grows over a wide temperature range, but its growth is fastest near 37 °C, which matches the mammalian gut. It tolerates a broad pH range, with optimal growth close to neutral, and its internal pH is actively maintained near neutrality even when the external pH changes.

Energy capture in E. coli is coupled through the proton motive force. Respiratory complexes and the electron transport chain pump protons out of the cytoplasm across the inner membrane, creating an electrical potential and a pH gradient. The combined potential is typically on the order of 150–180 mV. The protons flow back into the cell through the membrane sector of ATP synthase, and this flow rotates the c-ring of the enzyme, which drives the synthesis of ATP in the catalytic F1 subunit. The E. coli c-ring contains ten subunits, so roughly three to four protons are needed per ATP made. The same proton gradient also powers flagellar rotation and active transport of nutrients, which means that the energy budget of the cell is shared among several processes.

Gene regulation

The lac operon

The lactose system of E. coli is the classic model of gene control. In the 1961 model proposed by François Jacob and Jacques Monod, three structural genes, lacZ (β-galactosidase), lacY (lactose permease) and lacA (thiogalactoside transacetylase), are transcribed together from one promoter as a single unit, which they called an operon. Adjacent to the promoter lies the operator, a DNA sequence to which the LacI repressor binds, physically blocking transcription by RNA polymerase. Lactose itself does not act directly on the repressor. A small amount of β-galactosidase converts lactose into allolactose, which binds LacI, alters its shape and releases it from the operator. Allolactose is therefore the natural inducer, and the non-metabolizable analogue IPTG is widely used in laboratories for the same purpose. Jacob and Monod distinguished constitutive mutants by where the defect lay: mutations in the operator affected only the adjacent genes (acting in cis), whereas mutations in the repressor gene affected all of them (acting in trans). This logic, with cis-acting DNA sites and diffusible trans-acting proteins, became a template for thinking about gene expression in all organisms.

Catabolite repression and the CRP–cAMP system

The lac operon is also subject to positive control that ensures glucose is used first. When glucose is abundant, the phosphotransferase system lowers the activity of adenylate cyclase, so intracellular cyclic AMP (cAMP) stays low. When glucose falls, cAMP rises and binds the catabolite activator protein CRP (also called CAP), a dimer that binds a site upstream of the lac promoter and helps recruit RNA polymerase. The result is a two-condition logic: lac genes are expressed strongly only when lactose is present and glucose is scarce. In batch cultures containing both sugars, this produces diauxic growth, in which cells consume glucose first, pause briefly, and then switch to lactose. The same CRP–cAMP mechanism controls many other catabolic operons, so it is a central link between carbohydrate availability and the overall metabolic program.

Sigma factors and stress-specific promoter sets

Bacterial RNA polymerase forms a holoenzyme by associating with a sigma factor, which recognizes promoter sequences. The housekeeping factor σ70 (encoded by rpoD) drives most genes in growing cells, recognizing consensus -10 (TATAAT) and -35 (TTGACA) elements. Alternative sigma factors redirect the same core enzyme to different sets of genes. σ32 (RpoH) controls the heat-shock response; it is unstable under normal conditions, but its levels rise when temperature increases or misfolded proteins accumulate, and the resulting induction of chaperones such as DnaK and GroEL helps restore protein folding. σS (RpoS) governs genes for the stationary phase and general stress, σE (RpoE) responds to damage in the cell envelope, σ54 (RpoN) mediates responses to nitrogen limitation, and σ28 (FliA) activates flagellar genes. Anti-sigma factors can sequester these proteins until a signal releases them. Because one sigma factor can change the transcription of dozens to hundreds of genes at once, switching sigma factors is an efficient way to reprogram the cell.

Two-component systems and global regulators

Many environmental signals are sensed by two-component systems. A membrane-bound sensor histidine kinase autophosphorylates on a histidine residue in response to a stimulus, then transfers the phosphate to an aspartate on a cognate response regulator, which usually acts as a transcription factor. The EnvZ–OmpR pair responds to osmolarity and adjusts the ratio of the porins OmpF and OmpC, which is linked to osmotic balance. The PhoQ–PhoP pair senses low magnesium and other conditions, and it controls genes involved in resistance to antimicrobial peptides. The E. coli genome encodes dozens of such systems, and cross-talk between them is possible. Other regulators integrate broader inputs. FNR and the ArcA/ArcB system respond to oxygen availability, Fur controls iron uptake, and the nucleoid protein H-NS silences many horizontally acquired, AT-rich genes.

Cell-to-cell variability

Gene expression is noisy at the level of single cells. In 2002, Elowitz and colleagues placed two reporter genes encoding different fluorescent proteins under identical promoters in the same cell and used the degree of correlation between them to separate intrinsic noise (from the random timing of individual transcription events) from extrinsic noise (from differences in shared cellular conditions). In 2004, Ozbudak and colleagues showed that the lac system can display bistability: at intermediate inducer concentrations, some cells become fully induced while others remain uninduced. Because lactose permease imports the inducer, a positive feedback loop amplifies small differences between cells. Such heterogeneity means that a clonal population can contain subpopulations with different phenotypes, a form of bet-hedging that can be advantageous when conditions fluctuate.

Genetic exchange and laboratory strains

Much of classical bacterial genetics was worked out in E. coli, and the organism's value as a model depends partly on the ways its DNA can move between cells. Three routes were established or studied in detail in this species: conjugation, interrupted mating, and transformation. The lineages that became standard laboratory strains also carry a documented history that affects how experimental results should be read.

Conjugation, the F plasmid and Hfr strains

In 1946 Joshua Lederberg and Edward Tatum mixed two auxotrophic strains of K-12, each unable to make certain nutrients. Neither strain grew on minimal medium when tested alone. But when Lederberg and Tatum mixed the two strains together first, then plated the mixture on that same bare minimal medium, colonies appeared. The prototrophic colonies showed that the parental cells had exchanged genetic information, a result reported under the title "Gene recombination in E. coli" in Nature. Later work by Davis showed in a series of experiments that direct cell contact was needed for transfer, which rules out free DNA as the route in this process.

The genetic agent is the F (fertility) plasmid, a circular element of about 100 kilobase pairs. Donor cells carry it and are designated F+, while recipients lacking it are F−. The F plasmid involves the formation of a sex pilus that allows the transfer of DNA from a donor strain (F+) to a recipient strain (F-). A recipient that receives F usually becomes F+ itself. Some F+ cells carry F integrated into the chromosome; these are called Hfr (high frequency of recombination) strains. During conjugation, portions of the donor's chromosomal DNA adjacent to the integrated plasmid are transferred to the recipient. The transfer is often incomplete, so the recipient remains F- but acquires new chromosomal genes that can be incorporated into its genome. Because the integration point is fixed, an Hfr strain will transfer genetic markers from the host in a fixed order from that point. Hfr strains were therefore more useful than F+ cells for studying chromosomal genes.

Interrupted mating and the circular map

Elie Wollman and François Jacob exploited this ordered transfer. Their interrupted mating technique mixed Hfr donors with F− recipients and then forcibly separated the mating pairs at different times, usually with a kitchen-type blender. By experimentally disrupting mating pairs of cells at different times after mixing, and assaying which genes had by then entered the female, genetic maps of the E. coli chromosome were established. The time at which a marker first appeared in recipients served as its map position, with distances expressed in minutes. The closer a gene marker is to the origin, the greater the chance it will be transferred to the recipient during conjugation. Complete transfer of the chromosome takes roughly 100 minutes, so the map is a time-based linear scale of the genome.

The circularity of the chromosome emerged from comparing several Hfr strains in which F had inserted at different sites. Each insertion point gave a different linear order, and the overlapping maps collectively confirmed the circular topology of the E. coli chromosome. Wollman and Jacob's approach also established gene order before DNA sequencing was available, and it became a basic tool of bacterial genetics; the Wollman and Jacob work laid the foundation for Francois Jacob's Nobel Prize-winning work.

Transformation

Transformation differs from conjugation because the recipient cell takes up free DNA from its surroundings without cell-to-cell contact. Most E. coli strains are not naturally competent for this, so laboratory protocols were developed to make them take up plasmid DNA. In standard methods, cells are chilled in a calcium chloride solution and then given a brief heat pulse, which allows plasmids to enter. Transformation is the basis of routine plasmid cloning, and it has been applied to introduce foreign DNA that the cell then propagates. Its efficiency is far lower than that of natural exchange in some other bacteria, and linear DNA is often degraded by the host's restriction systems, which is one reason circular plasmids are used.

Origins of standard laboratory strains

The two main laboratory lineages are K-12 and B. The origin of K-12 is straightforward; it was isolated from the stool of a convalescent diphtheria patient in Palo Alto, CA, in 1922. In the 1940s Charles Clifton and Edward Tatum adopted it as a model organism. Early mutants derived in the 1940s, such as strains 58 and 679 described by Gray and Tatum, are considered the most ancient documented derivatives, and all currently used E. coli K-12 strains are descendants from strain 58 or 679 or a very close relative of them. Over decades, researchers introduced further mutations by treating E. coli K-12 with agents such as nitrogen mustard, ultra-violet radiation, X-ray etc. The widely used MG1655 strain was derived and named by Mark Guyer from strain W1485, which was derived in Joshua Lederberg's lab from a stab-culture descendant of the original K-12 isolate.

The genetic background of these strains matters for experiments. The strain was maintained as a laboratory strain with minimal genetic manipulation, having only been cured of the temperate bacteriophage lambda and F plasmid, yet ancestral K-12 strains still carry F and phage λ, and they have undergone extensive laboratory-based evolution. K-12 also lacks O-antigen synthesis, a feature that results in the absence of O-antigen synthesis and separates it from many natural isolates. K-12 strains, like MG1655 and W3110, are very different from commensal or environmental isolates, so conclusions drawn from K-12 cannot be assumed to hold for other strains. The B lineage has a murkier history: the first naming of the strain as E. coli B was by Delbrück and Luria in 1942 in their study of bacteriophages T1 and T7. Because researchers shared and renamed samples, the history of B strain is a bit more convoluted, and the BL21 strain and its derivatives are now the most common examples.

Research history and model-organism status

Several foundational experiments in molecular biology used Escherichia coli either directly or through the viruses that infect it. The bacterium grows quickly on simple media, its genetics are easy to manipulate, and its viruses, the bacteriophages, made it possible to ask basic questions about heredity.

Phage studies and the nature of mutation and genetic material

In 1943, Salvador Luria, then at Indiana University, and Max Delbrück, then at Vanderbilt, published a test of whether E. coli resistance to the virus T1 arose before exposure or was induced by the virus. Their fluctuation test grew many small parallel cultures and then plated each on phage. If resistance was induced by the virus, the counts of resistant cells should vary little between cultures; instead, the counts varied widely, with many cultures containing no resistant cells at all. The result supported random, pre-existing mutations that the virus selected, and it is usually cited as an early demonstration of selection acting on random variation (Luria–Delbrück experiment).

In 1952, Alfred Hershey and Martha Chase used bacteriophage T2 to infect E. coli. They labeled the phage DNA with radioactive phosphorus and the protein coat with radioactive sulfur, because DNA contains phosphorus but not sulfur, and proteins contain sulfur but generally not phosphorus. After infection, a blender stripped empty phage coats from the bacterial surface, and centrifugation separated the cells from the debris. Most of the phosphorus label ended up inside the bacteria, while most of the sulfur label stayed outside. The authors concluded that DNA, not protein, enters the cell and directs the production of new phages, which established DNA as the carrier of genetic information (Hershey–Chase experiment).

Semiconservative DNA replication

In 1958, Matthew Meselson and Franklin Stahl reported in the Proceedings of the National Academy of Sciences how DNA replicates in E. coli. They grew the bacteria in medium containing the heavy nitrogen isotope ^15N, then transferred them to medium with the common ^14N isotope. Using density-gradient centrifugation in cesium chloride, which they had developed together in 1957, they separated DNA by density. After one round of replication, all DNA was a hybrid of heavy and light strands. After a second round, both hybrid and fully light DNA appeared. This pattern excluded conservative replication and supported the semiconservative model that Watson and Crick had proposed in 1953, in which each daughter molecule keeps one parental strand and gains one new strand (Meselson–Stahl experiment; see also DNA replication).

mRNA, ribosomes and the genetic code

Much of the early work on gene expression was carried out in E. coli. In 1961, Sydney Brenner, François Jacob, Matthew Meselson and colleagues identified an unstable RNA intermediate that carries information from genes to the ribosomes, which is now called messenger RNA. Cell-free extracts of E. coli, which contain ribosomes and the other components needed for translation, were then used by Marshall Nirenberg, Heinrich Matthaei and others to show that synthetic RNA could direct protein synthesis, and to assign amino acids to specific triplet codons. These results helped complete the genetic code.

Genome sequencing and reference status

In 1997, a team led by Frederick Blattner published the complete genome sequence of the E. coli K-12 strain MG1655, a circular chromosome of about 4.6 million base pairs encoding roughly 4,300 proteins. The paper is often described as the first complete sequence of a free-living organism, but that claim is inaccurate: the genome of Haemophilus influenzae had been published in 1995. The E. coli sequence is better regarded as the first complete genome of a widely used laboratory model, and it set a template for annotating other bacteria. The organism remains a standard reference because its strains, growth conditions, and genetic tools are thoroughly documented, so protocols developed with it can be reproduced across laboratories and taught to students as a model of bacterial genetics.

Biotechnology and industrial applications

Escherichia coli became the central host of molecular biotechnology because its genetics are well characterized, it grows quickly on inexpensive media, and it takes up foreign DNA readily. Cloning work typically relies on plasmids that carry an origin of replication, an antibiotic-resistance marker and a multiple cloning site, so that an insert can be propagated in the cell. Expression vectors add a promoter that drives strong transcription of the inserted gene, and host strains such as BL21 derivatives are chosen for high expression and reduced protease activity. These tools are the basis of most recombinant DNA work.

Recombinant protein production

Human insulin was the landmark product. In 1978, the first recombinant DNA (rDNA) human insulin was synthetized by David Goeddel and colleagues at Genentech by combining A- and B-insulin chains through gene expression in Escherichia coli. The first licensed drug produced using recombinant DNA technology was human insulin, which was developed by Genentech and licensed as well as marketed by Eli Lilly in 1982.

Overexpression often causes a practical problem. Heterologous proteins generally accumulate in E. coli as inclusion bodies, which comprise of insoluble misfolded aggregates of proteins. In the early study of insulin-producing cells, the aggregates were so abundant that at peak production, the inclusion bodies may occupy as much as 20 percent of the Escherichia coli cellular volume. Aggregation has advantages: the insoluble form of the protein product protects proinsulin from degradation by proteolysis within the microbial cells. Because the aggregates are dense, they can be separated from other cell debris by centrifugation. Recovery is then laborious. Downstream steps include solubilizing the aggregates, washing them, chemical cleavage, chromatographic purification and refolding into an active form, and all the critical aspects of downstream processing must be balanced against cost. Even so, the majority of recombinant insulin therapeutics are produced from E. coli inclusion bodies.

A second practical hazard is endotoxin. Lipopolysaccharide from the outer membrane is released when cells lyse during harvesting, and it is a potent pyrogen in humans. Preparations intended for injection must therefore be depleted of endotoxin, usually by chromatography, and tested with standard assays. Strains are often engineered to reduce this contamination.

Metabolic engineering and chassis strains

E. coli is also a platform for making chemicals. Escherichia coli has been chosen as a workhouse for the production of many valuable chemicals due to its clear genetic background, convenient to be genetically modified and good growth properties with low nutrient requirements. Engineered strains have been developed for biofuels, organic acids, amino acids, sugar alcohols and biopolymers. The amino acid market alone is large: more than two million tons of amino acids are produced annually, and the figure is cited for applications in feed, pharmaceuticals and food additives. Typical strategies raise the flux through a target pathway by increasing copies of its enzyme genes, and they remove competing pathways by disruption or knockout. Aromatic amino acids, for example, are built from chorismate in the metabolic shikimate route, which metabolic engineers often redirect. Reported yields show the scale of these efforts: one strain achieved 878 mM acetate with a yield of 0.5 g/g glucose, and in another case pyruvate yields exceeding 1.78 mol per mol glucose were reached in fed-batch fermentation.

Synthetic biology has pushed the idea of a standard host further. Designers want a predictable background, so they build "chassis" strains from which non-essential DNA has been removed. To behave predictably, these chassis are designed with reduced genomes that minimize biological complexity. The MDS42 strain, for instance, was created by systematically deleting non-essential genes, mobile DNA elements and cryptic prophages from the MG1655 strain. Another reduced strain, DGF-298, has a single circular chromosome of 2,991,126 bp and 2,831 genes. Reduction does not guarantee stability, however. In a chemostat experiment under starvation, genome reductions in the MDS42 did not stabilize the chassis under metabolic stress. The mutation rate rose in both the reduced and the parent strain, so the benefit is conditional.

Biosensors, bioremediation and structural biology

Engineered cells can also act as sensors. A common design links a gene-expression reporter, such as a fluorescent protein, to a regulated promoter that responds to a target compound; this approach has been applied to detect pollutants such as heavy metals and arsenic, and such devices are generally called biosensors. Related strains have been proposed for bioremediation, in which cells degrade or immobilize contaminants. Field use raises containment and biosafety questions, and these limit deployment outside controlled settings.

In structural biology, E. coli is the usual expression host for proteins destined for X-ray crystallography and cryo-electron microscopy. Its rapid growth and simple protein folding machinery make it practical to produce milligram quantities of a target, which may be fused to affinity tags for purification. Some proteins are labelled with isotopes or selenomethionine to aid structure determination. These uses depend on the same cloning and expression tools described above.

Commensal role in the human gut

Escherichia coli is one of the first microorganisms to settle in the human intestine after birth, and it persists into adulthood as a minor but stable part of the gut microbiota. A study of healthy newborns found that culturable E. coli was absent from the first-pass meconium and was not detected in babies delivered by cesarean section until about the third day. It then colonized them strongly by the fourth day, and the authors concluded that colonization occurs after birth regardless of delivery mode or sex. Because these early colonists are facultative anaerobes, they consume the oxygen present in the gut lumen and create the anaerobic conditions that allow obligate anaerobes such as Bifidobacterium, Clostridium and Bacteroides to establish themselves. In adults, E. coli usually makes up a small fraction of the community, with one review estimating less than 1 percent of the human gut microbiome. It is most often found in the mucus layer of the colon.

The species also contributes to host nutrition through its metabolism. E. coli synthesizes the electron carrier menaquinone (vitamin K2) from chorismate, which is supplied by the shikimate pathway. Six enzymes convert chorismate into a naphthoquinone ring, which is then joined to an isoprenoid side chain and modified by prenylation and methylation. Under anaerobic conditions the bacterium mainly produces MK-8, whereas Bacteroides species make longer forms, MK-10 and MK-11. Bacterial menaquinone is poorly absorbed, however, and diet remains the main functional source of vitamin K2 for humans. Conversion of dietary phylloquinone (vitamin K1) to MK-7 by colonic E. coli has been described, but its contribution to overall vitamin K status is considered minimal. The gut's menaquinone production is therefore better understood as a bacterial metabolic output than as a major nutritional supply.

E. coli also contributes to colonization resistance, the protection that a resident microbiota offers against invaders, which is sometimes called the barrier effect. Proposed mechanisms include direct competition for nutrients, occupation of adhesion sites, production of inhibitory substances, and stimulation of host defenses. Early human evidence came from volunteers who were fed E. coli strains isolated from their own feces; those strains failed to colonize. In mice, the human commensals HS and Nissle 1917 occupied the same sugar-defined niches used by the O157:H7 strain EDL933 and prevented its colonization, which led the authors to argue that the outcome depends on how much of the required nutrients the resident community consumes. Commensal E. coli biofilms also respond specifically when pathogenic E. coli or Klebsiella pneumoniae enter them, and deletion of the genes yiaF and bssS changed pathogen colonization in pre-colonized mice. One of the main tools of competition is bacteriocins such as colicins, which kill closely related bacteria. In a 2020 study, however, colicin production was not required for sustained colonization of healthy mice, and carrying both a colicin and its immunity gene appeared to carry a slight fitness cost. The role of colicins in pathogen exclusion is therefore context-dependent. Competition is also shaped by host tissue state: in a mouse model, streptomycin-induced inflammation increased E. coli colonization through nitrate respiration, showing that antibiotic disturbance can shift the balance in favor of the bacterium.

Commensal E. coli also interacts with the host immune system. In gnotobiotic mice, the commensal strain CEC15 was reported to reinforce intestinal defenses and to be protective in a chronic colitis model. Reviews of commensal E. coli describe roles in protecting the intestinal epithelium from injury, influencing host fat storage, and stimulating intestinal angiogenesis. Cross-feeding is another important link. Besides consuming oxygen and supplying menaquinone, the species produces fermentation products such as acetate, formate and lactate, which other gut anaerobes can use as substrates. The same commensal pool is also the source of pathogenic lineages, because some strains acquire virulence and resistance genes through horizontal gene transfer of plasmids, pathogenicity islands, transposons and phages.

Pathogenic strains and disease

Most E. coli cells living in the gut are harmless commensals, but several lineages have acquired virulence genes that allow them to cause disease. Some virulence genes sit on mobile genetic elements, so some strains carry virulence genes from more than one pathotype because of the transfer of mobile genetic elements. This is why pathotype boundaries are useful but not always sharp (see horizontal-gene-transfer).

Diarrheagenic pathovars

The diarrheagenic group is responsible for causing diarrhea, particularly in children and travelers. Five pathotypes are most often recognized in clinical practice, and they differ mainly in how they cause disease:

• Enterotoxigenic E. coli (ETEC) produces the plasmid-encoded heat-labile enterotoxin (LT) and heat-stable enterotoxin (ST) or both. The toxins drive fluid into the intestinal lumen, producing watery diarrhea. ETEC is the leading cause of adult traveler's diarrhea visiting endemic areas.
• Enteropathogenic E. coli (EPEC) is defined by attaching-and-effacing lesions mediated by the locus of enterocyte effacement (LEE) pathogenicity island. These lesions destroy the microvilli of small-intestinal cells. EPEC is an important cause of infant and pediatric diarrheal disease worldwide.
• Enterohemorrhagic E. coli (EHEC), also called STEC, is the group best known from the O157:H7 serotype. Its bacteriophage-encoded stx1 and/or stx2 genes determine the production of Shiga toxin. The O157:H7 serotype is associated with a Shiga-like toxin responsible for systemic clinical symptoms such as hemorrhagic colitis and hemolytic uremic syndrome. Because of this, STEC infection is the most common cause of hemolytic uremic syndrome (HUS).
• Enteroinvasive E. coli (EIEC) invades the cells of the colonic lining. It does not rely on enterotoxins and resembles Shigella clinically, causing dysentery-like illness. Its invasion genes lie on a large virulence plasmid that encodes a type III secretion system.
• Enteroaggregative E. coli (EAEC) adheres to the intestinal lining in a stacked, brick-like arrangement and forms a biofilm, which underlies its association with prolonged watery diarrhea.

EPEC, EHEC and EIEC all depend on a type III secretion system to inject effector proteins into host cells, whereas ETEC and EAEC use simpler, non-T3SS mechanisms.

Extraintestinal pathogenic E. coli

Extraintestinal pathogenic E. coli (ExPEC) form a separate category. ExPEC isolates are facultative pathogens: they reside as commensals in the gut and belong to the normal microbiota, while outside the intestine are mainly related to neonatal meningitis and urinary tract infections. Uropathogenic E. coli (UPEC) is the main etiological agent of urinary tract infection (UTI), and it is mainly responsible for recurrent urinary tract infections (RUTIs).

In newborns, E. coli is a major cause of early-onset sepsis and meningitis. The maternal colonization rate is approximately 20–30%, and vertical transmission during delivery occurs frequently. It is the second most common pathogen in term infants and the leading pathogen in preterm infants; together with GBS, it accounts for approximately 60% of all cases. Persistence in the gut can also drive relapse: in one reported infant case, a recurrent infection was linked to a strain that remained in the faeces. Its genome matched the original isolate, suggesting that NMEC could persist in the gut and cause repeat infection.

Virulence mechanisms

Pathogenic strains use several overlapping strategies. Adhesins, such as fimbrial proteins, mediate attachment to host surfaces, and the type III secretion system delivers effectors into epithelial cells. Toxins act directly on host tissue: the heat-labile and heat-stable enterotoxins disturb fluid transport, and Shiga toxin damages the endothelium of blood vessels. Iron acquisition is also important for extraintestinal disease. In a genomic study of neonatal meningitis isolates, the most prevalent genes were those involved in iron uptake, including the enterobactin (98%), yersiniabactin (98%), aerobactin (62%), and salmochelin (55%) siderophore systems.

Transmission and surveillance

Diarrheagenic strains spread by the fecal-oral route. Contaminated food and water, and close contact with infected people, are the main routes, and the same principles apply to outbreaks. Raw dairy and produce are recurring sources. In a 2026 outbreak linked to raw dairy, CDC reports a total of nine illnesses in three states. Three people were hospitalized and one person developed hemolytic uremic syndrome (HUS), a serious condition that can cause kidney failure. A larger outbreak in 2022 linked to a fast-food chain detected 109 cases from six states.

Outbreak detection depends on laboratory surveillance. DNA fingerprinting is performed on bacteria using a method called whole genome sequencing (WGS). WGS showed that bacteria from sick people's samples are closely related genetically. This allows public-health authorities to connect cases that are otherwise scattered. Treatment also requires care: antibiotics are not recommended for patients with Shiga toxin-producing E. coli infections, because they may increase the risk of HUS. Current outbreak reporting also shows that case counts are incomplete. In a September 2026 multistate outbreak of O26 infections, the true number of sick people in this outbreak is likely much higher than the number reported, because many mild infections are never tested.

Evolution and ecology

Long-term experimental evolution

Richard Lenski's Long-Term Evolution Experiment (LTEE) began on 24 February 1988 with twelve populations of asexual E. coli cells, all descended from a single ancestral clone and grown in a glucose-limited minimal medium that is transferred daily. The populations passed 80,000 generations by August 2024. Oversight passed from Lenski's lab to Jeffrey Barrick at Michigan State University after the stocks were frozen during the COVID-19 pandemic and later revived.

The experiment's best-known result is the evolution of citrate use. The medium contained abundant citrate, which serves as an iron chelator, but no population could use it as a carbon source under oxygen. After roughly 31,000 generations, one of the twelve populations gave rise to a lineage that could grow on citrate. A genetic duplication activated a previously unexpressed transporter for di- and tricarboxylates, and the trait arose in one of three coexisting lineages within that population. The generation count varies between reports, with some putting the event at about 33,000 generations. The original team treated the event as a possible speciation. A 2015 reanalysis argued that its rarity reflected the experimental conditions rather than a unique evolutionary event: 46 independent citrate-using mutants appeared within 12 to 100 generations in similar media. Across the twelve populations, the genetic changes have generally followed similar trajectories, which makes the citrate lineage a notable exception.

Population genetics

E. coli combines several features that shape its diversity. Its population is very large, both within a single host and across the hosts it colonizes, so selection can act efficiently on even small fitness differences. Its mutation supply is continuous, and mutations accumulate in the core genome over time, producing lineages that diverge through mutation, natural-selection and genetic-drift. The species is also highly open to horizontal-gene-transfer. plasmids, bacteriophages and transposable-elements move genes between lineages, so distantly related strains can acquire large blocks of accessory DNA. As a result, the species' diversity is better described as a shared core plus a flexible, shifting accessory pool than as a set of cleanly separated clades. Population-level analyses therefore often treat recombination and gene flow as important as point mutation.

Survival outside the host

E. coli is adapted to the gut, and most cells that reach soil, sediment or water die quickly. One reason may be that cells cannot slow their metabolism enough to cope with the scarce, usable carbon found outside the host. Persistence is nonetheless possible. A strain applied at high loading to a Podzol field plot survived for 13 years, and a soil isolate from Ireland was recovered after more than nine years. Such long-term residents appear mostly in wet or submerged, cool environments, and the ability of an isolate to persist does not reliably follow from its genotype or phenotype.

Temperature is the dominant environmental control. A meta-analysis of 42 manure-amended soil studies found that temperature was the strongest factor for both E. coli and Salmonella, and that field die-off was markedly slower at 0–10 °C than above 20 °C. Growth itself requires warmth: in laboratory soils, cells multiplied at 30–37 °C but did not grow at 15 °C, and fluctuating temperatures reduce survival more than constant ones. Available carbon is considered the main limiting resource, while higher moisture, lower oxygen and lower pH generally favor persistence. Virulence determinants such as Shiga toxin and intimin do not appear to be essential for survival in soil, so the pathogenic and commensal lineages behave similarly in this respect.

Fecal indicator in water testing

Because E. coli lives in the intestines of warm-blooded animals and is shed in feces, its presence in water is taken as evidence of fecal contamination. The World Health Organization prefers it as the indicator for 100 mL samples and recommends zero detectable cells in drinking water, and many regulators apply a similar non-detection standard. Culture methods such as membrane filtration and most-probable-number counts remain the standard, although faster molecular and enzyme-based tests are being developed.

The indicator has clear limits. A negative result does not prove that pathogens are absent, and coliform tests in particular do not reliably reflect viruses such as norovirus or protozoa such as Cryptosporidium and Giardia. Not all E. coli in the environment comes from humans, since strains from birds, cattle and other animals occur too, and pinpointing the source requires specialized laboratory differentiation. The logic of the indicator also assumes that the organism does not multiply once it leaves the host, yet E. coli can grow in warm, nutrient-rich soils and sediments, which complicates the interpretation of counts. These issues are discussed further in public-health and epidemiology practice.

Antibiotic resistance

E. coli resists antibiotics through two broad routes: genes it acquires from other bacteria, and chromosomal traits that limit drug access or disable drugs without any gene transfer. The acquired route is the one that spreads quickly between lineages and hosts, so it dominates clinical concern.

Acquired resistance on mobile genetic elements

Most clinically important resistance genes in E. coli sit on plasmids, transposons and integrons. Plasmids are the main vehicle. Conjugative plasmids can carry several resistance genes at once, so a single transfer event may confer resistance to multiple drug classes. Integrons capture and express gene cassettes, and transposons move DNA segments between replicons, which helps resistance genes shift onto new plasmids. Horizontal gene transfer is widely regarded as the main reason resistance spreads across bacterial populations.

The most common acquired resistance to penicillin-type and cephalosporin drugs comes from β-lactamases. Extended-spectrum β-lactamases (ESBLs) inactivate third-generation cephalosporins, and CTX-M enzymes now account for most ESBLs worldwide. In a Korean study of community-onset bacteremia, CTX-M-15 was the most frequent variant, followed by CTX-M-14. Carbapenemases hydrolyse carbapenems, which are often reserved for severe infections. Those enzymes include KPC, NDM, OXA-48-like and VIM types. Their genes often travel on plasmids, although not always. In a California survey of carbapenem-resistant isolates, just over half of the CTX-M genes were predicted to be plasmid-borne. Some carbapenemase genes are chromosomally integrated, and under meropenem pressure they can excise as plasmids and disseminate, which shows that the chromosome can serve as a reservoir for mobile resistance.

Multidrug-resistant lineages

Certain lineages carry resistance far more successfully than others. The best-known is sequence type ST131, usually of serotype O25b:H4, which is a leading cause of community-acquired urinary tract infections and bacteremia. Its ESBL prevalence is tied to CTX-M-15, which has been carried on IncF-type plasmids. One analysis argues that the high prevalence of CTX-M-15 among ST131 mainly reflects the expansion of a single highly virulent subclone, H30-Rx, rather than repeated plasmid acquisition alone. Other workers emphasise plasmid transfer, so both clonal expansion and plasmid spread likely contribute. Regional subclones also matter: a Southeast Asian ST131 subclone accounts for much of the excess beta-lactam resistance in that region. ST131 has also acquired carbapenemase genes, most frequently OXA-244 and OXA-48, according to European surveillance. Because lineages and plasmids move together, resistance tracking must look at both.

Chromosomal and non-enzymatic mechanisms

Several mechanisms do not require acquiring a new enzyme gene. The outer membrane acts as a permeability barrier, and reduced expression of porins such as OmpF lowers entry of some β-lactams and fluoroquinolones. Efflux pumps are a second layer. The AcrAB-TolC system is a tripartite assembly in which the inner-membrane transporter AcrB works with the periplasmic adaptor AcrA and the outer-membrane channel TolC to expel drugs from the cell. Cryo-electron microscopy has shown how ligand binding triggers channel opening. Its expression is controlled by the repressor AcrR, and meta-analyses have found elevated acrAB expression in many multidrug-resistant isolates. AcrAB-TolC also helps cells retain plasmid-encoded tetracycline resistance by lowering intracellular drug concentrations. Enzymatic inactivation, through aminoglycoside-modifying or chloramphenicol-acetylating enzymes, and target modification, such as mutations in gyrA and parC that reduce fluoroquinolone binding, complete the picture. Many isolates combine several of these mechanisms, which is why multidrug resistance often arises stepwise.

Surveillance and stewardship

Resistance is tracked by whole-genome sequencing alongside standard susceptibility testing, because phenotypic tests can miss some genetic threats. A 2019 European survey of 548 E. coli isolates from 156 hospitals in 32 countries used sequencing to build a reference dataset, while a Japanese study of 4,195 isolates identified carbapenemase-positive strains that tested susceptible to carbapenems. ECDC has recommended genomic surveillance of carbapenemase-carrying E. coli to detect community spread early. E. coli serves as a sentinel organism because it is abundant in human and animal guts, is easy to sample, and reflects selection pressure from antibiotic use across the community; its resistance profile therefore gives an early signal for wider resistance trends. Stewardship programmes aim to reduce unnecessary antibiotic prescribing, and their effectiveness is judged partly by changes in resistance rates in these indicator populations.

References

1. Escherichia - an overviewsciencedirect.com
2. E.coliwhatsinaname.hmnh.harvard.edu
3. Sequencing a piece of history: complete genome sequence of the original Escherichia coli strain - PMCpmc.ncbi.nlm.nih.gov
4. Escherichia coli in molecular biologyen.wikipedia.org
5. Escherichia coligbif.org
6. The Genus Escherichialink.springer.com
7. Comparative Genomics of Escherichia coli Serogroups 64474, O179, O188 and Shigella boydii O16doi.org
8. Identification of Escherichia coli and Shigella Species from Whole-Genome Sequencesjournals.asm.org
9. Escherichia coli O-Antigen Gene Clusters of Serogroups O62, O68, O131, O140, O142, and O163: DNA Sequences and Similarity between O62 and O68, and PCR-Based Serogroupingncbi.nlm.nih.gov
10. Sequence-based typing of genetic targets encoded outside of the O-antigen gene cluster is indicative of Shiga toxin-producing Escherichia coli serogroup lineagesncbi.nlm.nih.gov
11. The Escherichia coli Serogroup O1 and O2 Lipopolysaccharides Are Encoded by Multiple O-antigen Gene Clustersncbi.nlm.nih.gov
12. Escherichia colien.wikipedia.org
13. Comparative Genomics of Escherichia coli Serogroups 64474, O179, O188 and Shigella boydii O16 - PMCpmc.ncbi.nlm.nih.gov
14. Bacterial Conjugation and F Plasmid Transfer: Mechanism, Terminology, and Role in Antibiotic Resistance — Microbe Onlinemicrobeonline.com
15. The F-plasmid, a paradigm for bacterial conjugation (Chapter 5) - The Dynamic Bacterial Genomecambridge.org
16. Bacterial Conjugationtechnologynetworks.com
17. 1 Chapter 1 Bacterial Gene Transfer John C. Mordacq and Roberta W. Ellingtonableweb.org
18. Jacob and Monod: From Operons to EvoDevo: Current Biologycell.com
19. Interrupted Mating Technique (Bacterial Gene Mapping by Conjugation Time)studyguides.com
20. %C3%89lie Wollmanen.wikipedia.org
21. Tracing Ancestors and Relatives of Escherichia coli B, and the Derivation of B Strains REL606 and BL21(DE3) - ScienceDirectsciencedirect.com
22. Laboratory strains of Escherichia coli K-12: things are seldom what they seemncbi.nlm.nih.gov
23. APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 1995, p. 4135–4139 Vol. 61, No. 11journals.asm.org
24. Strain Information of E. coli K-12 MG1655genome.wisc.edu
25. Laboratory strains of *Escherichia coli* K-12: not such perfect role models after allbiorxiv.org
26. Plasmids 101: Common Lab E. coli Strainsblog.addgene.org
27. Bacterial Conjugation - an overviewsciencedirect.com
28. F Factor - an overviewsciencedirect.com
29. Bacterial Interrupted mating Experimentcharon.perso.libertysurf.fr
30. Interrupted Mating Technique (for Bacterial Gene Order)studyguides.com
31. Historical Highlight: The Luria-Delbrück Fluctuation Test – A Study of the Nature of Bacterial Mutations Conferring Resistance to Infection by Bacteriophagencbi.nlm.nih.gov
32. The Luria-Delbrück Fluctuation Testpaijournal.com
33. The River Divides into Thousands of Branches: Chapter 2: Luria and Delbrűck show that bacteria are subject to Natural Selectionscalar.fas.harvard.edu
34. Hershey and Chase Experiment - DNA as Genetic Material - %%sitename%% - GeeksforGeeksgeeksforgeeks.org
35. The Hershey–Chase Experimentmacmillanlearning.com
36. Martha Chaseen.wikipedia.org
37. Introducing the Matthew Meselson Collectioncshl.edu
38. The Replication of DNA in Escherichia Colimun.ca
39. Semiconservative replicationen.wikipedia.org
40. Meselson-Stahl Experiment Explained: Definition, Examples, Practice & Video Lessonspearson.com
41. Recombinant Human Insulin - an overviewsciencedirect.com
42. Cell factories for insulin productionmicrobialcellfactories.biomedcentral.com
43. Cytoplasmic Inclusion Bodies in Escherichia coli Producing Biosynthetic Human Insulin Proteinsscience.org
44. Insulin Production -Dr. Ekta Khare Department of Microbiologygyansanchay.csjmu.ac.in
45. Downstream processing of recombinant human insulin and its analogues production from E. coli inclusion bodiesncbi.nlm.nih.gov
46. Downstream processing of recombinant human insulin and ...d-nb.info
47. Metabolic engineering of Escherichia coli for biotechnological production of high-value organic acids and alcoholslink.springer.com
48. Metabolic engineering of Escherichia coli: A sustainable industrial platform for bio-based chemical production - ScienceDirectsciencedirect.com
49. (PDF) Metabolic engineering of Escherichia coli and Corynebacterium glutamicum for biotechnological production of organic acids and amino acidsacademia.edu
50. The effect of metabolic stress on genome stability of a synthetic biology chassis Escherichia coli K12 strainmicrobialcellfactories.biomedcentral.com
51. The effect of metabolic stress on genome stability of a synthetic biology chassis Escherichia coli K12 strain - PMCpmc.ncbi.nlm.nih.gov
52. Complete sequence of the genome-reduced Escherichia coli DGF-298journals.asm.org
53. Temperature-induced production of recombinant human insulin in high-cell density cultures of recombinant Escherichia coli - ScienceDirectsciencedirect.com
54. Metabolic engineering of Escherichia coli: a sustainable industrial platform for bio-based chemical production - PubMedpubmed.ncbi.nlm.nih.gov
55. Engineering metabolic pathways in Escherichia coli for constructing a “microbial chassis” for biochemical production - ScienceDirectsciencedirect.com
56. Escherichia coli engineered for isobutyraldehyde productionimage-ppubs.uspto.gov
57. Genetic manipulations restored the growth fitness of reduced-genome Escherichia coli - ScienceDirectsciencedirect.com
58. Early settlers: which E. coli strains do you not want at birth?journals.physiology.org
59. Prenatal versus Postnatal Initial Colonization of Healthy Neonates’ Colon Ecosystem by the Enterobacterium Escherichia coli - PMCpmc.ncbi.nlm.nih.gov
60. Characteristics of Escherichia coli Isolated from Intestinal Microbiota Children of 0–5 Years Old in the Commune of Abomey-Calavincbi.nlm.nih.gov
61. From Gut Flora to Global Threat: The Evolutionary Story of <i>Escherichia coli</i>intechopen.com
62. Menaquinones, Bacteria, and Foods: Vitamin K2 in the Dietintechopen.com
63. Frontiersfrontiersin.org
64. Metabolic engineering of Escherichia coli BW25113 for the production of Vitamin K2 based on CRISPR/Cas9 mediated gene knockout and metabolic pathway modificationncbi.nlm.nih.gov
65. Vitamin K2en.wikipedia.org
66. Commensal and Pathogenic Escherichia coli Metabolism in the Gutjournals.asm.org
67. Identification of Commensal Escherichia coli Genes Involved in Biofilm Resistance to Pathogen Colonizationjournals.plos.org
68. Nutritional Basis for Colonization Resistance by Human Commensal Escherichia coli Strains HS and Nissle 1917 against E. coli O157:H7 in the Mouse Intestinejournals.plos.org
69. Deciphering the Role of Colicins during Colonization of the Mammalian Gut by Commensal E. colidoi.org
70. The commensal Escherichia coli CEC15 reinforces intestinal defences in gnotobiotic mice and is protective in a chronic colitis mouse modelnature.com
71. Commensal Escherichia Coli - an overviewsciencedirect.com
72. Escherichia coli, Diarrheagenicwwwnc.cdc.gov
73. Diarrheagenic Escherichia Coli - an overviewsciencedirect.com
74. Diarrheagenic Escherichia coli pathotypes frequency in Khuzestan province of Iranncbi.nlm.nih.gov
75. Understanding Escherichia coli Pathotypes: How the ATCC Collection Advances Infectious Disease Researchatcc.org
76. Kinds of E. colicdc.gov
77. Frontiersfrontiersin.org
78. New Strategies for the Prevention of Urinary Tract Infections by Uropathogenic Escherichia coliintechopen.com
79. Research Progress on perinatal E. coli infection - PMCpmc.ncbi.nlm.nih.gov
80. High-risk Escherichia coli clones that cause neonatal meningitis and association with recrudescent infectionelifesciences.org
81. High-risk Escherichia coli clones that cause neonatal meningitis and association with recrudescent infection - PMCpmc.ncbi.nlm.nih.gov
82. Outbreak Investigation of E. coli O157:H7: Raw Cheddar Cheese (March 2026)fda.gov
83. Notes from the Field: Multistate Outbreak of Escherichia coli O157:H7 Infections Linked to a National Fast-Food Chain — United States, 2022ncbi.nlm.nih.gov
84. Investigation Update: E. coli Outbreak, July 2026cdc.gov
85. Investigation Update: E. coli Outbreak, September 2026cdc.gov
86. Diarrheagenic E. coli: ETEC, EPEC, EIEC, EHEC, and EAEC Explained — Microbe Onlinemicrobeonline.com
87. Enteroinvasive Escherichia Coli - an overviewsciencedirect.com
88. E. coli long-term evolution experimenten.wikipedia.org
89. Rapid Evolution of Citrate Utilization by Escherichia coli by Direct Selection Requires citT and dctAjournals.asm.org
90. Innovation in an E. coli evolution experiment is contingent on maintaining adaptive potential until competition subsides - PMCpmc.ncbi.nlm.nih.gov
91. Genomic and phenotypic evolution of Escherichia coli in a novel citrate-only resource environment - PMCpmc.ncbi.nlm.nih.gov
92. Insights into the low-temperature adaptation and nutritional flexibility of a soil-persistent Escherichia coliacademic.oup.com
93. Survival of Salmonella and Escherichia coli in Two Different Soil Types at Various Moisture Levels and Temperatures - ScienceDirectsciencedirect.com
94. Environmental Drivers for Persistence of Escherichia coli and Salmonella in Manure-Amended Soils: A Meta-Analysis - ScienceDirectsciencedirect.com
95. Long-Term Persistence and Leaching of Escherichia coli in Temperate Maritime Soils - PMCpmc.ncbi.nlm.nih.gov
96. Survival of Escherichia coli in the environment: fundamental and public health aspects - PMCpmc.ncbi.nlm.nih.gov
97. Persistence of Escherichia coli O157:H7 and Its Mutants in Soilsncbi.nlm.nih.gov
98. Physical Factors Impacting the Survival and Occurrence of Escherichia coli in Secondary Habitatsmdpi.com
99. Molecular testing devices for on-site detection of E. coli in water samplesncbi.nlm.nih.gov
100. Methods for detection and enumeration of coliforms in drinking water: a reviewiwaponline.com
101. Rapid microbiology instrumentation for high-accuracy quantitative water safety assessments in remote and low-resource locationsbiorxiv.org
102. Fecal coliform and E. coli Analysis in wastewater and water byohiowea.org
103. Microbiological Water Quality Testing Methods for E. coli ...aquagenx.com
104. Risk Factors and Molecular Features of Sequence Type (ST) 131 Extended-spectrum β-Lactamase-producing Escherichia coli in Community-onset Bacteremiancbi.nlm.nih.gov
105. The Epidemic of Extended-Spectrum-β-Lactamase-Producing Escherichia coli ST131 Is Driven by a Single Highly Pathogenic Subclone, H30-Rxjournals.asm.org
106. The higher prevalence of extended spectrum beta-lactamases among Escherichia coli ST131 in Southeast Asia is driven by expansion of a single, locally prevalent subclonenature.com
107. The role of plasmids in carbapenem resistant E. coli in Alameda County, Californiancbi.nlm.nih.gov
108. Carbapenem-resistant Enterobacteralesecdc.europa.eu
109. Carbapenem triggers dissemination of chromosomally integrated carbapenemase genes via conjugative plasmids in Escherichia colijournals.asm.org
110. An allosteric transport mechanism for the AcrAB-TolC multidrug efflux pumpncbi.nlm.nih.gov
111. Expression of multidrug efflux pump gene acrAB in Escherichia coli: a systematic review and meta analysis - PMCpmc.ncbi.nlm.nih.gov
112. Deciphering the roles of AcrAB-TolC efflux pump in promoting the transmission of antibiotic resistance - ScienceDirectsciencedirect.com
113. AcrAB-TolC efflux pump system plays a role in carbapenem non-susceptibility in Escherichia colincbi.nlm.nih.gov
114. Emergence of carbapenemase-producing Escherichia coli in acute care hospitals in 32 European countries (the CCRE survey): a prospective, multicentre, cross-sectional, epidemiological, microbiological, and genomic surveillance study - PMCpmc.ncbi.nlm.nih.gov
115. National genomic surveillance integrating standardized quantitative susceptibility testing clarifies antimicrobial resistance in Enterobacteralesncbi.nlm.nih.gov