Bacteriology at UW-Madison
Unstained cells of E. coli viewed by phase microscopy. about 1000X magnification.
Theodor Escherich first described E. coli in 1885, as Bacterium coli commune, which he isolated from the feces of newborns. It was later renamed Escherichia coli, and for many years the bacterium was simply considered to be a commensal organism of the large intestine. It was not until 1935 that a strain of E. coli was shown to be the cause of an outbreak of diarrhea among infants.
E. coli and its relatives are known to microbiologists as "enteric bacteria", because they live in the intestinal tract of humans and other animals. The best known other enteric bacteria are Salmonella, which includes the agent of typhoid fever, and Shigella, which is the bacterial cause of dysentery.
E. coli is in the bacterial family Enterobacteriaceae, which is made up of Gram-negative, nonsporeforming, rod-shaped bacteria that are often motile by means of flagella. The majority of strains grow well on the usual laboratory media in both the presence and absence of oxygen, and metabolism can be either by respiration or fermentation.
Gram stain of E. coli viewed by conventional light microscopy. about 1000X magnification. E coli is usually seen as a unicellular Gram-negative organism about 1 micrometer in width and 2-4 micrometers in length.
Detection and Isolation of Escherichia coli
E. coli as an Indicator of Fecal Pollution
For most of the 20th century, E. coli has been used as the principal indicator of fecal pollution in both tropical and temperate countries. E. coli comprises about 1% of the total fecal bacterial flora of humans and most warm-blooded animals. Sewage is always likely to contain E. coli in relatively large numbers. In addition, E. coli, being a typical member enteric bacterium is presumed to have survival characteristics very similar to those of the well-known pathogens such as Salmonella and Shigella. Thus, E. coli has been used world-wide as an indicator of fecal microbiological contamination. As such an indicator organism, its value is significantly enhanced by the ease with which it can be detected. and cultured.
Tests to identify isolates as E. coli have, of necessity, been simple tests designed predominantly to differentiate them from organisms normally associated with uncontaminated water. Since full biochemical analyses are not generally performed, the term "coliform" has been coined to describe E. coli-like organisms that satisfy these limited tests. As a result, regulations are promulgated throughout the world defining standards for water based on the so-called "coliform count." For example, in the U.S., according to a regulation published in the Federal Register (1986), there is a requirement that there be 0 coliforms/100 ml drinking water, as determined by any method for any sampling frequency.
Since not all organisms which meet the criteria of a coliform are associated with the intestinal tract (some may be free-living), a further distinction must be made between "fecal coliforms" (E. coli) and "nonfecal coliforms" (e.g. Klebsiella and Enterobacter). The nonfecal coliforms are regularly found in soil and water and in associations with plants, so that their occurrence does not necessarily indicate fecal pollution.
In order to distinguish E. coli from related species likely to be found naturally in the environment, a battery of tests called the IMViC reactions was developed in order to differentiate fecal coliforms from nonfecal coliforms. IMViC is an acronym in which the capital letters stand for Indole, Methyl red, Voges-Proskauer, and Citrate.) The IMViC set of tests examines: the ability of an organism to (1) produce Indole; (2) produce sufficient acid to change the color of Methyl red indicator; (3) produce acetoin, an intermediate in the butanediol fermentation pathway (a positive result of the Voges-Proskauer test); and (4) the ability to grow on Citrate as the sole source of carbon. E. coli is positive in the first two tests and negative in the second two; nonfecal coliforms are the opposite - negative in the first two tests and positive for the second two.
If E. coli is detected in water, it is an indication of fecal pollution. Most-likely the strain of E. coli is a harmless non pathogen, but the indication is that other pathogenic intestinal microbes could also be present. The pathogenic fecal coliforms (e.g. Salmonella and Shigella ) can be readily distinguished from strains of E. coli on the basis of a lactose fermentation test. All strains of E. coli ferment the sugar lactose while those of Salmonella and Shigella do not.
Detection of E. coli in Food
The International Commission on Microbiological Specifications for Foods (ICMSF, 1978) has adopted a set of standard techniques for the enumeration of E. coli in food products, accepted by the International Standards Organization (ISO, 1984). This method employs the use of lauryl sulfate tryptose broth at 35 or 37¢XC as a mildly selective-enrichment medium. This is followed by growth in EC broth containing 0.15% bile salts at 45¢XC as a second selective step. The ability to produce indole from tryptophan (in tryptone broth) at 45¢XC defines the strains as E. coli. These tests miss some types of E. coli, such as those most closely related to the Shigella group, but it is the detection of possible fecal contamination that is important in these tests rather than the presence of specific types.
Detection of E. coli in Water
There is no method for the detection of E. coli in water that is accepted throughout the world. In the US, a standard method using membrane filter enumeration for both total and thermotolerant coliforms has been established (American Public Health Association (1986). Further IMViC tests on selected isolates can then be performed as described above.
In the UK, the definition of E. coli in water microbiology is also based on the ability to produce gas from lactose and produce indole from tryptophan at 44¢XC. A method for enumeration employs a standard multiple tube test with a modified glutamate synthetic medium at 37¢XC as a first selective step, followed by further cultivation in standard media at 44¢XC.
Detection of E. coli in Clinical Specimens
While large numbers of E. coli will be found in fecal specimens or specimens contaminated with feces or intestinal contents, most other clinical specimens are usually not contaminated with E. coli. The major exception is urine, which requires special attention in the clinical situation. From those specimens in which E. coli is likely to be present in large numbers, direct plating on media such as MacConkey agar or Eosin Methylene Blue (EMB) agar is sufficient. If the number of E. coli is likely to be very low or the amount of specimen is limited, enrichment in a rich nutrient medium such as brain heart infusion broth may be used. A number of different commercially available kits are generally used to identify the isolates as E. coli.
From specimens likely to contain only a few viable E. coli cells, such as blood from patients suspected of having E. coli bacteremia, various enrichment procedures are used. Identification follows standard bacteriological techniques.
Left: Escherichia coli microcolony on agar surface. Right: E. coli colonies on EMB Agar. On EMB agar E. coli colonies develop a purple color and have a characteristic green sheen. Colonies of Salmonella and Shigella would appear colorless, so the medium can be used to differentiate E. coli from the pathogens.
Rapid Methods for Detecting E. coli
A fluorogenic detection method has been developed based on the cleavage of methylumbelliferyl-D-glucuronide (MUG) to the free methylumbelliferyl moiety, which fluoresces a blue color after irradiation with long-wave ultraviolet radiation. Although strains of E. coli are generally positive in this test, some strains of Salmonella, Shigella, and Yersinia are also capable of splitting MUG; the latter two genera are usually not present in food. A disadvantage is that enterohemorrhagic E. coli (EHEC) strains are generally negative in this test. MUG can be added to various selective media, so there is a great potential in its use for detecting E. coli.
Automated or semi-automated systems are also being used for the detection of E. coli as part of the detection methods for Enterobacteriaceae. Techniques involving impedance measurements have shown promise. Other techniques such as immunoassays and nucleic acid hybridization studies can also be used to enumerate E. coli, and DNA probes directed at a number of genes have also been developed.
Physiology of E. coli
Physiologically, E. coli is versatile and well-adapted to its characteristic habitats. In the laboratory it can grow in media with glucose as the sole organic constituent. Wild-type E. coli has no growth factor requirements, and metabolically it can transform glucose into all of the molecular components that make up the cell. The bacterium can grow in the presence or absence of O2. Under anaerobic conditions it will grow by means of fermentation, producing characteristic "mixed acids and gas" as end products. However, it can also grow by means of anaerobic respiration, since it is able to utilize NO3 or fumarate as final electron acceptors for respiratory electron transport processes. In part, this adapts E. coli to its intestinal (anaerobic) and its extraintestinal (aerobic or anaerobic) habitats.
In the ecological niches that E. coli occupies and its abilities to grow both aerobically and anaerobically are important. E. coli is well adapted to its intestinal environment as it is able to survive on a relatively limited number of low-molecular weight substances, which may only be available transiently and at relatively low concentrations. The generation time for E. coli in the intestine is thought to be about 12 hours. The type of nutrients available there to E. coli consist of mucus, desquamated cells, intestinal enzyme secretions, and incompletely digested food. Given the absorption capacity and efficiency of the intestine, there are probably only small amounts free carbohydrates or other easily absorbable forms of nutrients, and there is competition from hundreds of other types of bacteria. A similar situation probably also applies to sources of nitrogen.
In its natural environment, as well as the laboratory, E. coli can respond to environmental signals such as chemicals, pH, temperature, osmolarity, etc., in a number of very remarkable ways considering it is a single cell organism. For example, it can sense the presence or absence of chemicals and gases in its environment and swim towards or away from them. Or it can stop swimming and grow fimbriae that will specifically attach it to a cell or surface receptor. In response to changes in temperature and osmolarity, it can vary the pore diameter of its outer membrane porins to accommodate larger molecules (nutrients) or to exclude inhibitory substances (e.g. bile salts). With its complex mechanisms for regulation of metabolism the bacterium can survey the chemical content its environment in advance of synthesizing any enzymes necessary to use these compounds. It does not wastefully produce enzymes for degradation of carbon sources unless they are available, and it does not produce enzymes for synthesis of metabolites if they are available as nutrients or growth factors in the environment.
Escherichia coli in the Gastrointestinal Tract
The commensal E. coli strains that inhabit the large intestine of all humans and warm-blooded animals comprise about 1% of the total bacterial biomass. This E. coli flora is in constant flux. One study on the distribution of different E. coli strains colonizing the large intestine of women during a one year period (in a hospital setting) showed that 52.1% yielded one serogroup, 34.9% yielded two, 4.4% yielded three, and 0.6% yielded four. The most likely source of new serotypes of E. coli is acquisition by the oral route. To study oral acquisition, the carriage rate of E. coli carrying antibiotic resistance (R) plasmids was examined among vegetarians, babies, and non vegetarians. It was assumed that non vegetarians might carry more E. coli with R factors due to their presumed high incidence in animals treated with growth-promoting antimicrobial agents. However, omnivores had no higher an incidence of R-factor-containing E. coli than vegetarians, and babies had more resistant E. coli in their feces than non vegetarians. No suitable explanation could be offered for these findings. Besides, investigation of the microbial flora of the uninhabited Krakatoa archipelago has shown the presence of antibiotic-resistant E. coli associated with plants.
Infections Caused by Pathogenic E. coli
E. coli is responsible primarily for three types of
infections in humans: urinary tract infections, neonatal
meningitis, and intestinal diseases. These conditions
depend on a specific array of pathogenic (virulence) determinants
possessed by the organism. Pathogenic E. coli are
discussed elsewhere in the text in more detail at Pathogenic E.
coli: Gastroenteritis, Urinary tract Infections and Neonatal
Urinary Tract Infections
Uropathogenic E. coli cause 90% of the urinary tract infections (UTI) in anatomically-normal, unobstructed urinary tracts. The bacteria colonize from the feces or perineal region and ascend the urinary tract to the bladder. Bladder infections are 14 times more common in females than males by virtue of the shortened urethra. The typical patient with uncomplicated cystitis is a sexually-active female who was first colonized in the intestine with a uropathogenic E. coli strain. The organisms are propelled into the bladder from the periurethral region during sexual intercourse. With the aid of specific fimbriae they are able to colonize the bladder.
The frequency of the distribution of the host cell receptor for the bacterial fimbriae plays a role in susceptibility and explains why certain individuals have repeated UTI caused by E. coli. Uncomplicated E. coli UTI virtually never occurs in individuals lacking the receptors.
Neonatal meningitis affects 1/2,000-4,000 infants. Eighty percent of E. coli strains involved synthesize K-1 capsular antigens (K-1 is only present 20-40% of the time in intestinal isolates).
E. coli strains invade the blood stream of infants from the nasopharynx or GI tract and are carried to the meninges.
Epidemiological studies have shown that pregnancy is associated with increased rates of colonization by K-1 strains and that these strains become involved in the subsequent cases of meningitis in the newborn. Probably, the infant GI tract is the portal of entry into the bloodstream. Fortunately, although colonization is fairly common, invasion and the catastrophic sequelae are rare.
Neonatal meningitis requires antibiotic therapy that usually includes ampicillin and a third-generation cephalosporin.
Lysis of a dividing pair of E. coli cells in the presence of a beta-lactam antibiotic. Some beta lactam antibiotics, such as ampicillin and cephalosporin, are effective in the treatment of meningitis caused by strains of E. coli (above). The beta lactam antibiotics prevent cell wall synthesis and assembly in the bacterium. When the bacterium grows in the presence of the antibiotic, the cell wall becomes progressively weaker and weaker, so the the organism eventually ruptures or "lyses", pouring out its cytoplasmic contents as shown here.
As a pathogen, E. coli, of course, is best known for its ability to cause intestinal diseases. Five classes (virotypes) of E. coli that cause diarrheal diseases are now recognized: enterotoxigenic E. coli (ETEC), enteroinvasive E. coli (EIEC), enterohemorrhagic E. coli (EHEC), enteropathogenic E. coli (EPEC), and enteroaggregative E. coli (EAggEC). Each class falls within a serological subgroup and manifests distinct features in pathogenesis.
Enterotoxigenic E. coli (ETEC)
ETEC are an important cause of diarrhea in infants and travelers in underdeveloped countries or regions of poor sanitation. The diseases vary from minor discomfort to a severe cholera-like syndrome. ETEC are acquired by ingestion of contaminated food and water, and adults in endemic areas evidently develop immunity. The disease requires colonization and elaboration of one or more enterotoxins. Both traits are plasmid-encoded.
Enterotoxins produced by ETEC include the LT (heat-labile) toxin and/or the ST (heat-stable) toxin, the genes for which may occur on the same or separate plasmids. The LT enterotoxin is very similar to cholera toxin in both structure and mode of action. It binds to the same intestinal receptors that are recognized by the cholera toxin, and its enzymatic activity is identical to that of the cholera toxin.
The ST enterotoxin is actually a family of toxins which are peptides of molecular weight about 2,000 daltons. Their small size explains why they are not inactivated by heat. ST causes an increase in cyclic GMP in host cell cytoplasm. This leads to secretion of fluid and electrolytes resulting in diarrhea.
Symptoms ETEC infections include diarrhea without fever. The bacteria colonize the GI tract by means of a fimbrial adhesin, e.g. CFA I and CFA II, and are noninvasive, but produce either the LT or ST toxin.
Enteroivasive E. coli (EIEC)
EIEC closely resemble Shigella in their pathogenic mechanisms and the kind of clinical illness they produce. EIEC penetrate and multiply within epithelial cells of the colon causing widespread cell destruction. The clinical syndrome is identical to Shigella dysentery and includes a dysentery-like diarrhea with fever. Like Shigella, EIEC are invasive organisms. but hey do not produce LT or ST toxin and, unlike Shigella, they do not produce the shiga toxin.
Enteropathogenic E. coli (EPEC)
EPEC induce a watery diarrhea similar to ETEC, but they do not possess the same colonization factors and do not produce ST or LT toxins. They produce a non fimbrial adhesin designated intimin, an outer membrane protein, that mediates the final stages of adherence. Although they do not produce LT or ST toxins, there are reports that they produce an enterotoxin similar to that of Shigella. Other virulence factors may be related to those in Shigella.
Adherence of EPEC strains to the intestinal mucosa is a very complicated process and produces dramatic effects in the ultrastructure of the cells resulting in rearrangements of actin in the vicinity of adherent bacteria. The phenomenon is sometimes called "attaching and effacing" of cells. EPEC strains are said to be "moderately-invasive" meaning they are not as invasive as Shigella, and unlike ETEC or EAggEC, they cause an inflammatory response. The diarrhea and other symptoms of EPEC infections probably are caused by bacterial invasion of host cells and interference with normal cellular signal transduction, rather than by production of toxins.
Some types of EPEC are referred to as Enteroadherent E. coli (EAEC), based on specific patterns of adherence. They are an important cause of traveler's diarrhea in Mexico and in North Africa.
Enteroaggregative E. coli (EAggEC)
The distinguishing feature of EAggEC strains is their ability to attach to tissue culture cells in an aggregative manner. These strains are associated with persistent diarrhea in young children. They resemble ETEC strains in that the bacteria adhere to the intestinal mucosa and cause non-bloody diarrhea without invading or causing inflammation. This suggests that the organisms produce a toxin of some sort. Recently, a distinctive heat-labile plasmid-encoded toxin has been isolated from these strains, called the EAST (EnteroAggregative ST) toxin. They also produce a hemolysin related to the hemolysin produced by E. coli strains involved in urinary tract infections. The role of the toxin and the hemolysin in virulence has not been proven. The significance of EAggEC strains in human disease is controversial.
Enterohemorrhagic E. coli (EHEC)
EHEC are represented by a single strain (serotype O157:H7), which causes a diarrheal syndrome distinct from EIEC (and Shigella) in that there is copious bloody discharge and no fever. A frequent life-threatening situation is its toxic effects on the kidneys (hemolytic uremia).
EHEC has recently been recognized as a cause of serious disease often associated with ingestion of inadequately cooked hamburger meat. Pediatric diarrhea caused by this strain can be fatal due to acute kidney failure (hemolytic uremic syndrome [HUS]). EHEC are also considered to be "moderately invasive". Nothing is known about the colonization antigens of EHEC but fimbriae are presumed to be involved. The bacteria do not invade mucosal cells as readily as Shigella, but EHEC strains produce a toxin that is virtually identical to the Shiga toxin. The toxin plays a role in the intense inflammatory response produced by EHEC strains and may explain the ability of EHEC strains to cause HUS. The toxin is phage encoded and its production is enhanced by iron deficiency.
E. coli O157:H7 Transmission EM. American Society for Microbiology
Biotechnological Applications of E. coli
The advances in molecular biology, genetics and biochemistry during the past four decades have led to an enormous development in the field of biotechnology. Studies with E. coli have played a major role in these developments, and the bacterium has been in the forefront of many technological advances.
In the early days of biotechnology (1960s), emphasis was placed on
improvements of established procedures of bioprocessing, such as the
production of yeast, vaccines, and antibiotics. These
investigations stimulated genetic research of microbes to increase
their potential to produce a wide variety of products in the service
of humanity. Although much was being learned about E. coli
and its genetics, the direct use of the bacterium in the industry was
limited. The industrial production of the amino acid threonine by E.
coli mutants, begun in 1961, is an exception.
At this time, organisms were generally subjected to mutagenic agents, which produced a series of random mutations, from which the specifically required mutants were selected.
In the last two decades, procedures have evolved which permit the preparation of strains that have very specific productive capabilities. As the genetic structure of E. coli was well known, and it is an organism which can grow on simple media (mineral salts and glucose) under aerobic and anaerobic conditions, the bacterium became the basis for most developments in genetic manipulations leading to genetic engineering.
The basic principle of these genetic manipulations is gene cloning, which enables the isolation and replication of individual DNA fragments. This consists of a series of linked steps, involving the isolation of the desired gene as double-stranded DNA (dsDNA), insertion of the gene into a suitable vector, and using the vector to introduce the DNA into a cell which will express the desired genetic information. In the case of cloning a gene in E. coli, first the DNA of suitable character is isolated, then it is joined to the DNA of a suitable vector producing a series of recombinant molecules. Then the recombinant molecules are introduced into the bacterium in which the target gene becomes established. Recombinants are selected in various ways with the purpose of expressing the desired genetic information.
The source for DNA cloning can be genomic DNA fragments, cDNA fragments produced by the action of reverse transcriptase on mRNA molecules, chemically synthesized oligonucleotides, or amplified DNA from the products of the polymerase chain reaction (PCR). Plasmids, phages, and cosmids have all been successfully used as vectors, and transformation, transfection, and transduction have all been used to introduce the foreign DNA into the E. coli cell. Plasmids are among the most widely used vectors for the insertion of foreign DNA into an E. coli. Plasmids lend themselves very well as vectors since they are independent replicons which are stabily inherited in an extrachromosomal state and can be made to carry easily identifiable phenotypic markers such as antibiotic resistance or sugar fermentation.
An example of the use of plasmids to introduce a foreign gene into E. coli in order to produce a useful product is illustrated by the use of the E. coli plasmid pBR322 to clone the gene for production of the human growth hormone, somatostatin. In this case, the gene for the small polypeptide hormone was produced by synthetic means. The double-stranded DNA coding for the 15 amino acids of somatostatin was synthesized with the addition of a translation a stop signal at the end. The synthetic gene was then recombined with the plasmid within the beta-galactosidase structural gene and introduced into E. coli. In this way, the production of the somatostatin peptide could be controlled by the lac operon. In a similar manner, the genes for human insulin production were inserted into E. coli which was then able to synthesize the human hormone.
Such general techniques of molecular biology and bacterial genetics are now being applied within research laboratories and industry to produce a wide variety of strains of genetically engineered E. coli from which a number of useful products can be produced. Likewise, the problems associated with the expression of eukaryotic DNA by a procaryotic promoter in E. coli were solved by construction of a fusion gene. In this system, the control region and the N-terminal coding sequence of an E. coli gene are ligated to a eukaryotic sequence so that translation of the chimeric protein can occur. The only condition is that the eukaryotic sequence must be in the correct reading frame. The desired protein is then enzymatically or chemically cleaved from the E. coli product.
E. coli strains are been genetically engineered to produce a variety of mammalian proteins, especially products of medical or veterinary interest including enzymes and vaccine components. E. coli has also been used to manufacture other substances including enzymes that are useful in the degradation of cellulose and aromatic compounds and enzymes for ethanol production. There may be no limit to what E. coli can produce through recombinant DNA technology as long as the substance is a natural product for which a genetic sequence can be found.