11. Diversity among Escherichia coli.

(a) The Clonal Nature of Escherichia coli.

As has been discussed previously E. coli cells their natural environment have to survive a variety of environmental challenges. Natural populations of E. coli have have been shown to possess extensive genetic diversity. By means of various studies, it has been suggested, that E. coli may be organised into a limited number of genetically distinct clones. While most microbiologists know (or believe they know) what is meant by the term clone, it is a very difficult concept to explain. In its simplest terms, it can be concieved as all the descendants from one particular bacterium. While they may undergo various forms of change their interrelationships can still be seen. However, when looking at a group of bacteria like a population of E. coli it is often not easy to determine which may be derived from a single cell and have just undergone a few 'minor' changes, or which are derived from different cells.

However, such clones may be sufficiently stable to have achieved a broad geographic distribution and this especially be true for some of the pathogenic forms. It may also apply to the non-pathogenic forms. The clonal concept can particularly be applied to the discussions earlier on the interactions within the restricted environment of the intestinal tract of a host animal. It can be viewed as a diversity of clones, which will be maintained as new clones enter the system. Some of these new clones may establish and some even become dominant, while others may become extinct.

While the ability to carry virulence factor genes may be of selective advantage within an animal host, it would probably be disadvantageous in the inanimate environment. A recent study has demonstrated that 13% of E. coli isolates from treated and raw waters carried DNA sequences for one or more virulence factors including toxins and the adherence factors. Obviously the pathogenic E. coli clones can survive in the inanimate environment. Such virulence factors are often coded for on plasmids, which means they can survive and be maintained in small numbers within their host populations of E. coli, and can be readily transmitted throughout the E. coli population if the need arises.

(b) How Escherichia coli changes its Environment.

As has been discussed earlier, E. coli is a typical facultative anaerobe, being able to grow both in the presence and absence of oxygen. this property means that it is partly responsible for maintaining the strictly anaerobic conditions prevailing in the intestines. These permit the strict anaerobes to grow. In the intestinal environment E. coli multiplies close to epithelial cells, where there may be oxygen available from the blood after passage through the cells to the attached microbial mat. By mopping up the oxygen, the E. coli cells provide the anaerobic environment.

The ability to multiply aerobically provides E. coli with a distinct selective advantage over the predominant intestinal anaerobes with which it is shed into the natural environment. E. coli and only a few other bacterial species, synthesizes menaquinone (vitamin K). This substance is a component of the respiratory chain as well as being a precurser of prothrombin in the host liver. By being able to survive on a relatively limited number of low molecular weight compounds, E. coli is particularly well adapted to its environment. As these substances may only be present in low amounts and often transiently, E. coli can be considered as having a predominantly scavenging role This must be maintained by potent active transport systems and the energy of these must be provided by oxydoreduction reactions, which must be able utilize a variety of different electron donors and acceptors.

Many strains of E. coli are able to decarboxylate the amino acids ornithine and/or arginine to form the respective amines. While the amino acids are neutral having both a basic and acidic part in the molecule, the amines are bases. These abilities have been used to biotype strains of E. coli. It was noted that under semianaerobic conditions at low pH, i.e. when the acidity was high, then "inducible" ornithine and arginine decarboxylases can appear, especially if there is an excess of those amino acids present. It has been found that these inducible enzymes are quite distinct from the biosynthetic ones and are involved in pH control of the organisms. Neither enzyme is inhibited by physiological concentrations of polyamines. Thus, when E. coli ferments a carbohydrate such as glucose under semianaerobic conditions then acids are formed as has been mentioned earlier. These resulting acids increase the acidity of the medium or reduce the pH. If an amino acid such as arginine, lysine or ornithine is present then these inducible decarboxylases will convert the amino acids to the basic amines agmatine, cadaverine, and putrescine respectively with the loss of CO2. In this way the pH effect of the fermentation of glucose is reversed.

(c) Escherichia coli Serotypes.

A very important way of differentiating E. coli is by serotyping. This technique can currently differentiate up to around 10,000 different types, which are known as the serotypes of E. coli. The principles of serotyping are based on the fact that when a certain specific carbohydrate or protein molecule is injected into an animal, the animal will produce antibodies, which will specifically bind to the substance injected, which is known as the antigen. Serotyping is based on the specific interaction of carbohydrate or protein antigens with their respective antibodies. The most important parts of the E. coli, which are used as antigens for the purpose of serotyping are the carbohydrate part of the lipopolysaccharide cell wall and the protein flagellae. Also used for certain situations are the capsules, fimbrae, toxins and theoretically any substance that ellicits an antibody response can be used as an antigen.

The development of the principles of serology and serotyping of E. coli closely followed those of the Salmonella group, whose serotyping scheme was established in the 1930's as the Kauffmann-White scheme. In the early 1940's Kauffmann established a viable serotyping scheme for the E. coli group, which became the basis of the currently internationally accepted serotyping scheme for E. coli. Currently this is based on two classes of antigens; the 'O' and 'H' antigens.

The 'O' antigens are somatic antigens not inactivated by heat at 100°C or 121°C. They consist of the lipopolysaccharide component of the E. coli cell wall. This consists of a lipid and core polysaccharide, which is common to all E. coli. Pointing outwards from this structure like the spikes of a porcupine are a series of carbohydrate repeating units. Each of these is unique for each 'O' antigen. While the core lipopolysaccharide provides the basic structure of the cell wall, these 'O' antigen specific side chains can be seen as the decorations on the wall. Originally only 20 different 'O' antigens were described and given numbers O1 to O20. As more were described further numbers were added such that the highest described 'O' antigen now is O173. Most likely further types will be described in future. As a new type is described it is given the next available number so the next 'O' antigen will be O174. A few numbers have been deleted for a variety of reasons and some have been sub-typed such as O18a,b and O18a,c, thus, there are still around 170 different 'O' types known.

There are some strains of E. coli isolated which lack any of these specific 'O' antigen structures. As these tend to agglutinate spontaneously in salt solution and clump together they are known as 'rough' and such strains are designated OR.

The 'H' antigens are the flagellar antigens and are inactivated by heat at 100°C. Being the flagellar antigens, E. coli have to be motile to carry them. There are non-motile variants of E. coli and they would be considered as lacking an 'H' antigen. They are desgnated H-.

The 'H' antigens comprise the protein flagellin, which makes up the flagellae of E. coli. The 'H' antigens are given consecutive numbers in order of them being discovered. Currently there are 55 different 'H' antigens, which are desgnated H1, H2 etc. to H56. One 'H' antigen, H50, has been deleted from the scheme because it was too similar to H10.

From this description it can be seen that with around 170 different 'O' types and 55 different 'H' types at least 9000 different O:H serotypes of E. coli can be identified. It should be noted that if both the 'O' and the 'H' type of a strain have been characterised, one can describe this strain's serotype as being e.g. O4:H5. Other important serotypes which have been described over the years are O55:H7, which was frequently associated with infantile diarrhoea in the past. Serotype O148:H28 is an enterotoxigenic serotype associated with travellers' diarrhoea and serotype O157:H7 has been associated with many outbreaks of haemolytic ureamic syndrome recently.

Originally a third group of antigens were described. These are the 'K' antigens. These are somatic or surface antigens, which occur as envelopes, sheaths or capsules. They act as masking antigens for the 'O' antigens, inhibiting the binding of antibody and subsequent agglutination of living suspensions in 'O' antisera. Depending on their lability to heat, they were subdivided as follows.

'K' antigens of type L were defined by the fact that their antigenicity, agglutinability and the antibody binding power is destroyed at 100°C for 1 hour. They can occur as envelopes, sheaths or capsules.

'K' antigens of type A were defined by the fact that their agglutinability and antigenicity is destroyed at only at 121°C for 2.5 hours The antibody binding power is not inactivated at 100°C for 2.5 hours or at 121°C for 2 hours. A antigens can occur as capsules.

'K antigens type B were defined by the fact that their agglutinability of B antigens is destroyed at 121°C for 2.5 hours. The antibody binding power of B antigens is not inactivated at 100°C for 2.5 hours or at 121°C for 2 hours and the anitigenicity is inactivated at 100°C for 2 hours. B antigens can occur as envelopes or sheaths.

For full serotyping of a strain of E. coli to be achieved its 'O', 'K' and 'H' antigens must be determined. 'K' antigen determination is now generally only done for specific purposes, nevertheless just 'O' and 'H' determination is a very time consuming procedure. A number of cross-reactions have been noted among the 'O' antigens, hence serotyping E. coli tends to be restricted to special reference laboratories.

There is a chemical structural basis to the serological characteristics of the E. coli antigens, particularly the 'O' antigens. Specific chemical structures of carbohydrate molecules have been shown to determine the 'O' antigenic specificity. These 'O' antigen specific chemical arrangements were called the chemotypes. When these chemotypes were examined in a more detailed way it was found that in some cases the lipopolysaccharide of other bacterial species had the same chemotype as some of the E. coli 'O' types and they were also closely related serologically. This may be due to pure coincidence, or the different organisms acquired the ability to make that chemotype from one another or from another common source. As an example E. coli O111 and Salmonella 'O' group O (O35) have been known to cross-react strongly serologically and it has since been shown that the chemical structures of the two antigens are identical.

(c) Escherichia coli Biotypes.

E. coli are able to ferment a variety of carbohydrate substrates, generally by converting them to glucose or to a substrate on the fermentative chain of the breakdown of glucose. The various fermentable carbohydrates include substances such as compounds of glucose with other sugars. Thus maltose, trehalose and cellobiose are compounds of two glucose molecules linked in different ways. Lactose is a compound of glucose and the related sugar galactose, sucrose is a compound of glucose and the related sugar fructose. Raffinose is a compound of glucose, fructose and galactose. Other carbohydrate like substances which can be fermented include sugars like mannose, sorbose, arabinose and rhamnose and the related sugar alcohols ducitol, mannitol and sorbitol. Also compounds of sugars with other substances can be fermented particularly if the sugar molecule can be separated from the other part. An example is the substance salicin.

The ability to ferment a given sugar of the types described above by a strain of E. coli is dependent on the strain having the requisite enzymes to convert it to glucose or to a substance on the degradative chain from glucose. It has been found that different strains of E. coli differ in their ability to perform these conversions. Thus while all strains will ferment glucose and over 90% will ferment mannitol and lactose, with many of the other sugars mentioned above they will vary. This is the basis of biotyping E. coli. These tests are also easy to perform, by determining, whether a strain of E. coli will produce acid following growth in the presence of the carbohydrate.

In addition, as was discussed earlier in respect to the maintenance of neutrality, E. coli differ in their abilities to degrade the amino acids arginine, lysine and ornithine.

Extensive studies on the developments of biotyping schemes for E. coli have shown that tests for the ability of strains to ferment dulcitol, raffinose and sorbose and to decarboxylate ornithine gave optimal discrimination among strains. With these four tests it was possible to establish 16 primary biotypes, which could be further subdivided by means of a secondary set of tests. These included the fermentation of rhamnose, decarboxylation of lysine, hydrolysis of aesculin, motility development within 24 hours, production of type-1 fimbriae, requirement for nicotinamide, thiamine and/or another growth factor for growth on minimal medium.

When strains from a variety of sources were tested then good discrimination was achieved, however, when the sources were limited discrimination was not so good. This suggests that strains may well be related within clones or subclones.

A number of rapid and automated tests have been developed over the years by commercial companies. Generally these are designed to be used in a clinical laboratory for the identification of isolates from patients. In a number cases these groups of tests will include some of the tests, which have been discussed above which can differentiate between different types of E. coli. These can also be used for biotyping. While serotyping is an internationally accepted mode of subtyping E. coli and identifying clones, biotyping, which can achieve the same ends has not been internationally systematised.

Like many organisms E. coli produce a variety of esterases, which are enzymes that hydrolyse multiple substrates. If an extract from E. coli is prepared containing the proteins found within the cytoplasm, and these are subjected to an electric field in a gel, through which they can diffuse, then the different proteins will align themselves according to their size, shape and particularly electric properties. This electrophoretic diversity of the esterases has been shown give reliable information for both the specific as well as the subspecific differentiation of bacteria and been successfully used to differentiate strains of E. coli. Good discrimination between strains was achieved, when a large number of strains of both human and animal origin were studied. Certain pathogenic forms were shown to produce distinctive patterns.

Extensive studies have also demonstrated that determination of the variety found among the outer membrane proteins (OMP) of E. coli can be used to differentiate strains and characterise specific clones. These investigations have demonstrated that OMP patterns discriminate among the genetically different clonal groups within the specific 'O' serogroups.

By permitting motile strains of E. coli to swarm towards each other it was found that while the swarms of identical or related strains will merge when they meet, those of different strains will form a distinct zone of demarcation. This phenomenon has been termed "Colony Incompatibility". Further studies have found that by permitting strains to swarm towards each other and determining whether they are compatible or not, they could be subdivided and these subdivision used for ecological investigations.

(e) Escherichia coli Phagetypes.

Bacteriophages, or phages in short, are viruses that attack and infect bacteria. They can cause the host bacterium to be destroyed, or they can be carried by the bacterium without affecting it for some time, even over many generations, and then when there is a change in the outside environment the phage destroys its host. Generally the phage attaches itself to the outside of the bacterium, its nucleic acid is injected into the host bacterium, where it takes over the metabolic activities, such that the host makes multiple copies of the phages, which are released on lysis of the bacterium. These new phages can then continue to infect neighbouring bacteria.

There are predominantly two mechanisms, which can prevent a phage from infecting and destroying a bacterium. Either the phage cannot attach itself effectively to the bacterium and thus all subsequent actions fail, or there is a block for the phage replication, in which case it is destroyed by the bacterium. Both these mechanisms are very specific and thus the whether a given phage will be able to lyse a given bacterium is very specific. It is precisely due to this specificity that the subtyping (or subdivision) of bacterial groups on the basis of their sensitivity or resistance to phages is based. This type of subtyping is known as phagetyping and has been used for the subtyping of some organisms such as Salmonella typhi for the purpose of epidemiological investigations for many decades.

The same system can also be applied to many subgroups of E. coli and in fact has been successfully used. Some phages are specific for certain 'O' serogroups, while others will attach to other parts of the outer surface of the E. coli cell. While phage typing has been used as a means of discriminating between strains of E. coli in general, it is much more useful to discriminate among a specific 'O' serogroup, and most useful studies have been done with these groups, especially the enteropathogenic serogroup O111 and the enterohaemorrhagic serogroup O157. Thus phagetyping really is only used for pathogenic E. coli.

Generally, in order to obtain the phages required for a phagetyping scheme cultures are taken from the type of environment, in which the test organisms are likely to be found. These are placed on lawns of the E. coli to be tested and when they cause lysis, the phages are purfied and used. When a sufficient number of phages which have varied characteristics for lysis are collected then a phagetyping system can be set up. As a simple example if one takes three phages with different specificities labelled A, B and C, one strain of E. coli might be lysed by all three, it would be considered to belong to phagetype 1 in this system. Another strain which is only lysed by phages A and B would be labelled phagetype 2, a strain lysed by phages A and C would be type 3, one lysed by phages B and C would be phagetype 4. Phagetypes 5, 6, and 7 would only be lysed by one of the phages A, B, and C respectively and finally phagetype 8 would not be lysed by any of the three phages. Thus, using three phages, strains of E. coli could theoretically be subtyped into eight phagetypes. As required further phages can always be added to the scheme. With standard sets of phages being used in reference laboratories around the world these phagetyping schemes have international comparability.

(f) Escherichia coli Colicintypes.

Colicins are protein compounds produced by bacteria, which act as toxins or antibiotics to closely related species of bacteria or within a species of bacteria. While the term bacteriocin is used to cover most types of such agents affecting many different bacterial groups, colicins are those affecting E. coli and related species. The mode of action of the different colicins on the target bacteria is quite different. Some colicins such as the B-colicins act by forming pores in the in cell membranes of affected cells, causing leakage of various cellular components, leading to death of the cells. Other colicins like E2 and E5 kill the cells by entering through the mebrane into the cell and degrading the nucleic acids.

At one time it was considered that due to their high specificity for certain strains colicins might be useful as very specific antibiotics, however due the rapid rate at which many organisms in the wild develop resistance to them, this has not been found to practicable. Bacteria which produce a certain type of colicin are generally resistant to it, although there may also be strains which are resistant without being producers. These strains might simply lack the specific receptor for the colicin. Other strains are sensitive. Strains may produce one type of colicin but be sensitive to another. This is the basis on which colicin typing used. Different strains may produce colicins with different activity patterns and thus other strains can be subdivided according to their sensitivity to these colicins. The subtyping of E. coli with colicins is currently only used in specific investigations and not of widespread use.

(g) Escherichia coli Pulsed-field gel

Electrophoresis types.

With the advent of molecular biology and the many new sophisticated techniques, the technique of Pulsed-field gel Electrophoresis (PFGE) has been brought into extensive use to subdivide many different groups of bacteria including E. coli. Conventional gel electrophoresis was based on the principle that application of an eletrical field to a biological preparation, such as proteins or nucleic acids in a supporting gel caused the different molecules to migrate in this field according to their electric charge and also size and shape. The determination of esterase patterns described above, was based on this technique.

PFGE is different of these conventional techniques, by being able to separate much larger DNA fragments, than these conventional methods can achieve. This is due to the procedure of switching the current between two separate sets of electrodes which are at an obtuse angle to each other. The result of this is that whole chromosomes can be separated. However, with bacteria such as E. coli, which have only one chromosome, this chromosome is cut with a specific DNA-cutting enzyme, which yields large fragments, which are then separated by PFGE. Obviously different patterns will be obtained, depeding on the DNA-cutting enzyme, used.

This technique has been applied particularly to distinguish strains belonging to serogroup O157 as a means of determining possible epidemiological linkages between strains, especially to find the potential source of the outbreak. Unfortunately, the PFGE types do not necessarily aggree with the phagetypes and thus there is a certain amount of unresolved controversity as to which technique gives the more appropriate answer in an investigation. In a number of investigations it has been found that strains belonging to one phagetype will segregate into a number of PFGE types and vice versa.

(h) Conclusions

These discussions on subtyping of E. coli should have clearly demonstrated the variety of types and subtypes of this organism, which can be defined. In their natural environment, whether it is the intestinal content of an animal or human or whether it is in a natural inanimate environment, these types will be in constant competion, but also assisting each other. While one strain may produce a colicin or phage which kills other E. coli, a strain may also produce a haemolysin, which makes iron available for the other other non-haemolytic strains.

E. coli should therefore not be considered as one unit, they are a family of bacteria, which are in constant flux, not only amongst each other but also with the other bacterial and non-bacterial groups within their environment. It is probably that out of this competitive environment that the various pathogenic E. coli which will be discussed later have emerged. However, it is only by knowing what E. coli really are, rather than just looking at the pathogens, that one gets the understanding for the pathogens. But first we must examine the genetic basis of these organisms, which ultimately holds the key to a full understanding of them.

12. Genetics of Escherichia coli.

(a) Introduction

When E.coli is in its natural habitat, the human intestine, it grows very quickly, when food is passing through the intestinal tract, but otherwise may not grow at all. Also as has been mentioned earlier, within the intestinal tract it has to respire anaerobically. It must also be able to survive and multiply in an environment in which it is outnumbered by a variety of competitors.

Most research on the genetic aspects of E.coli has been done on laboratory derivatives of a strain of E.coli K12, or on strains associated with intestinal disease and on representatives of some of the 'O' and 'K' serotypes. While much of this information has been shown to be generally applicable to all E. coli there always has to be excersised a degree of caution when such extrapolations are done. The strain K12 had been originally isolated from the faeces of a healthy human and been in laboratory culture for many decades so it may have altered significantly from wild strains, which may currently be isolated. It lacks the specific components of the bacterial cell wall, which provide the O-antigenic specificity and is considered rough. It carries the H antigen H48. The designation K12 is unrelated to the K antigen typing system and is merely a name given to the strain. It does not carry the capsular antigen K12.

(b) The Escherichia coli Genome.

There have been more investigations into the genome of E.coli over the years, than on ony other living organism on this planet, including the human. These studies have shown that the genome of E.coli is a single circular DNA molecule consisting of about 4 x 106 base pairs with a molecular weight of 4 x 109 and a total length of 1.4mm. Being packed into a cell of about 2µm in length makes the effective DNA concentration in a typical growing cell about 17mg/ml. Within the cell it is compacted more than 1000-fold into a distinct structure, which has been termed the nucleoid. This is equivalent to the nucleus of the eucaryotic cell. These nucleoids have been visualised by appropriate staining and also by electron microscopy. Those regions of the chromosome, which are being copied appear as single stranded elements at the edges of the nucleoid, where the nascent RNA and ribosomes are to be found.

A linkage map of E.coli genes was first constructed as long ago as 1964. At that time approximately 100 genes had been identified. Two decades later some 1000 genes have been mapped and the rate of identifying and locating the position of genes increased exponentially so that now virtually all the genes of E. coli are mapped. For convenience, the circular E.coli chromosome has been subdivided into 100 minutes with each minute comprising approximately 47 kilobase pairs.

In the past 25 years, there has been a quantum leap in knowledge of the E.coli chromosome, which has been made possible by the great advances in molecular genetics, which have occurred. It is noteworthy that strains of E.coli, particulary E. coli K12, played a predominant role in these advances. Important roles in these developments were played by such agents as specialized transducing bacteriophages, which carry gene sequences from one strain to another. The utilization of transposons, which confer resistance to various antibiotics as easily selectable markers, in both matings and transductions removed the need to select for the actual mutation, permitted many more genes to be mapped. These studies have resulted in the development of various catalogues such as the linkage map, a physical map, and an ordered phage clone bank, the gene-protein index and the various catalogues of genomic insertions.

Genetic studies currently in progress suggest that only a very limited number of clones of E.coli may be associated with certain pathogenic mechanisms. Thus the various serotypes, biotypes, toxin types and other variants associated with a certain specific condition such as e.g. colibacillosis in birds are minor variants within only a small number of clones. These studies support the hypothesis that strains of E. coli associated with a certain condition represent special groups of pathogenic clones, and confirm earlier suggestions on the clonal nature of enteropathogenic E. coli. Similar studies on pathogenic E. coli of serogroup O157 have revealed wide genetic diversity and indicated that the relatively recently recognized pathogenic serotype O157.H7 was derived from an old lineage rather than having only recently emerged. An analysis of strains causing urinary tract infections in humans and animals has similarly indicated that these organisms belong widely disseminated uropathogenic clones. Such genetic analyses may lead to greater understanding of the taxonomy of E. coli and by being able to define specific pathogenic clones it may be possible to determine whether strains are pathogenic or not.

Maps showing the positions of all the available genetic loci are regularly published. By 1990 over 1400 such loci had been published and the number is rapidly increasing. It has to be constantly borne in mind that the E. coli genome is a circular strand of DNA about 1.4 mm long coiled within a cell which is maximally only 2µm long. This DNA can replicate itself with an extraordinaryly high degree of accuracy within 20 minutes, at the same time specifically uncoiling those parts of itself, which are required to be copied onto the messenger RNA so that all the appropriate enzymes and other proteins required for the continued growth of the E. coli can be made.

(c) Replication of the Escherichia coli Genome.

When it is arranged that E. coli cells are growing synchronously, i.e. they are all replicating at the same time, it has been found that the cell mass increases exponentially during the cell cycle. Under those conditions DNA replication is initiated at the same age in all cells. Following the completion of the DNA replication, the two genomes separate, a septum is formed between them and at that moment both the cells mass as well as the cell numbers have doubled and the cycle is ready to start again.

The replication of this 1.4 mm long double stranded DNA molecule with a molecular weight of about 2.5 x 109, comprising about 4 x 106 base pairs has been worked out in principle some time ago. It starts at a unique site on the genome, known as the oriC and proceeds in both directions. The synthesis is discontinuous using replication intermediate fragments about 1000 nucleotides long.

The oriC of the DNA is positioned within a three-dimensional structure and the first stage of DNA replication is that probably around 20 to 30 molecules of the protein DnaA bind to the DNA at that point. This is considered the first stage in the initiation of replication. This complex binds strongly to ATP or ADP but it is the oriC-DnaA-ATP, which is the essential prerequiste for the next stage.

Then two proteins DnaB and DnaC form a protein complex, which on being delivered to the oriC-DnaA-ATP becomes the preprepriming complex. The DnaB protein has been shown to be the replicative helicase of the E. coli. In this role it travels progressively along the lagging-strand template and by joining with the primase to synthesize the nascent fragments. An additional protein appears to be involved at this stage. This is the HU protein, which is similar to the histones of the eucaryotic cells. It seems to be the helicase DnaB to be the preprepriming complex at oriC. There are at the oriC site three repeats of the sequence 5'-GATCTNTTNTTTT-3', where the DnaB proteins attach to the DNA.

The next stage in the process is that the DnaB and the DnaC initiator proteins unwind the duplex DNA near the oriC-site and the two sister strands are separated by the action of the DnaB helicase activity. The Single Strand Binding (SSB) protein now binds to the single strands, which prevents their reassociating. At the same time the action of the DnaB helicase activity, which is obligatorily coupled to the hydrolysis of the ribosyl nuleotide triphosphates (rNTPs) proceeds as stage II in the process. This complex on the now partially unwound oriC has been termed the prepriming complex.

This process would necessarily soon come to a halt if the supercoiled DNA is not unwound. The enzyme DNA gyrase provides the required swivelling action. This structure of the nucleoprotein complex situated on this partially unwound oriC has been named the prepriming complex. The E. coli enzyme primase (Stage III) can now synthesize the primers required for the replication of the DNA.

For the synthesis of the oriC-specific RNA primers (Stage IV), the primase as well as the four proteins DnaA, DnaB, DnaC and DNA gyrase are essential prerequisites. In addition, while not essential the SSB and HU proteins have a strong stimulatory action on the RNA synthesis. It is predominantly the level of primase activity, which determines the comparative rates of RNA and DNA synthesis.

The actual rate of DNA chain elongation (Stage V) is limited by the rate of unwinding of the the DNA chain and is mediated by the enzyme DNA polymerase III. This process now can proceed quite fast. The termination of replication a oriC (Stage VI) is again a very slow process. The movement of the replication fork slows down perceptibly, as it approaches the terminus. If DNA topoisomerse is absent then there develop multiple intertwinings among the DNA circles. Also enzymes such as DNA ligase and DNA polymerase I must be present to correct the gaps or interruptions formed during the process.

(d) RNA Synthesis in Escherichia coli.

The ultimate objective of RNA synthesis is the synthesis of proteins, both structual and enzymic proteins, which respectively maintain the integrity of the E. coli cell and permit it to function and carry out the various metabolic acitivities associated with a living organism. There are basically three types of RNA involved in these processes. These are transfer RNA (tRNA), ribosomal RNA (rRNA) and messenger RNA (mRNA). The tRNA molecules are small RNA structures which each act as an adaptor between the amino acids and the mRNA. The mRNA is the transcript of the message as taken from the DNA and the rRNA provides the nucleic acid backbone of the ribosome in which the protein is synthesised.

The basis of the formation of both these RNA moieties as well as ultimately the proteins is the genetic code. According to this code an amino acid is specified by three nucleotides in the mRNA. This triplet of nucleotides is called a codon. These codons are arranged contiguously but not overlapping along the mRNA. With four different bases available on the mRNA to form each triplet there are theoretically 64 different triplets possible, which can form a codon. However, there are only 20 amino acids, therefore most amino acids can be specified by more than one codon. The reverse does not occur, meaning that no codon will specify more than one amino acid. There are three codons: UAA, UAG and UGA which do not specify for an amino acid but signal termination of protein synthesis.

***************needs to be completed

The main enzymes involved in the synthesis of RNA from the DNA templates of the genome are the RNA polymerases. These enzymes consist of five subunits. There are two identical subunits, known as the a-subunits with a molecular weight of 39,000, two related but different subunits known as the b- and b'-subunits with molecular weights of 155,000 and 165,000 respectively, which all together form the so-called 'core' enzyme. When the fifth subunit, known as the s-factor is added then the so-called 'holoenzyme' has been formed. The degree of binding of s-factor is dependent on a number of physiological factors. In addition there are a number of different s-factors, which confer different specificities on the RNA polymerase.

For a given stretch of DNA to be copied onto RNA, the RNA polymerase must find it and has to begin at the right point. It has to be remembered that that this particular stretch of apparently random DNA actually consists of a unique sequence. The area on this stretch, which the RNA polymerase seeks is known as the promoter sequence for the given gene or set of genes to be copied.

***************needs to be completed

Each tRNA molecule has at one specific point the complementary triplet to the codon on the mRNA. By combining with its specific amino acid, the tRNA effectively converts the amino acid into an RNA derivative. This can can be placed in the correct order according to the mRNA code. The tRNA thus must have two specificities so that the correct amino acid-tRNA complex is formed and secondly that the tRNA locates the correct position along the mRNA chain, within the ribosome.

(e) Synthesis of Ribosomal RNA.

The production of ribosomal RNA (rRNA) has to be seen as part of the whole productivity of the three main RNA types: rRNA, transferRNA (tRNA) and messengerRNA (mRNA). The rRNA coding operons in E. coli (rrn) have been shown to transcribed to a remarkable extent. During rapid growth, more than half the cells RNA being produced is rRNA.

The DNA sequences on the chromosome for rRNA in E. coli are seven nontiguous operons. These are asymmetrically situated around the origin of replication. The seven operons known as rrnA-E and rrnG-H are at approximately the following positions: rrnC, rrnA, rrnB and rrnE in that order between 84-90; rrnD at 72.1; rrnG at 56.1 and rrnH at 5.1. It has been noted that the rRNA genes are contranscribed from large precurser RNA molecules. These are then cleaved by specific RNases and thus the specific rRNA molecules are synthesized.

These seven rrn sequences code for the three rRNA types in the order promoter - 16S RNA - 23S RNA - 5S RNA. In the spacer regions between the 16S and the 23S RNAs and also at the distal ends of some of the rrn operons there are found genes for some of the tRNAs. There are two tandem s70 promoter regions P1 and P2 separated by about 100 base pairs in each operon. Between the 16S and 23S coding regions of rrnB and rrnG there is a ribosomal spacer loop and the coding region for tRNAGlu-2, at the same position in rrnC and rrnE is only the coding region for tRNAGlu-2, and for rrnA, rrnD and rrnH are the coding regions for tRNAIle-1 and tRNAAla-1B. After the coding regions for the 5S RNA there is in rrnC coding regions for tRNAAsp and tRNATrp. After the coding regions for the 5S RNA there is in rrnD coding regions for tRNAThr-1 and another 5S RNA and after the coding regions for the 5S RNA there is in rrnH a coding region for tRNAAsp. Each operon unit also has a complex terminator region (ter). There is r-independent terminator or terminators followed by a r-dependent terminator.

Apart from the promoters P1 and P2 there are further upstream promoters in some operons. Especially it has been demonstrated that two promoters P3 and P4 are present about a kilobase upstream on the rrnB operon. Curiously these promoters could be transcribing a mRNA for a 289-amino acid protein involved in vitamin B12 metabolism.

(f) Regulation of Ribosomal RNA synthesis.

The predominant part of the regulatory mechanism for rRNA synthesis occurs at the level of transcription initiation. However, it is also important that a high level of elongatioon rate be maintained. It appears that the promoter P1 rather than P2 is the one which is more highly regulated. The control is at two levels. They are the so-called stringent response, determined by amino acid availability and control in relation to the general growth rate of the E. coli. In addition there are other control elements upstream of P1, which include binding sites for the transcription activator Fis and also the so-called UP element. This latter is an AT-rich extension of the promoter region and has intrinsic activation capabilities through contacts with the a-subunit of RNA polymerase. There is also an antitermination sequence beteen P1 and P2..

When the E. coli is starved of amino acids, a rapid shutdown of both rRNA and tRNA ensues, combined with an increase in the production of two unusual guanine derived nucleotides. These are guanosine tetraphosphate (ppGpp) and guanosine pentaphosphate (pppGpp). In addition the elongation rate of RNA polymerase transcribing mRNA molecules decreases. The first result of the amino acid starvation is an increase in the numbers of uncharged (or nonaminoacylated) tRNA molecules. It is the build up of these uncharged tRNA molecules, which triggers the stringent response, by causing an increase in the two guanosine phosphates ((p)ppGpp). Their synthesis is catalyzed by a protein, the RelA protein, which is also known as the strigent factor. Following an initial upsurge in the concentration of the (p)ppGpp, they attain a level of 10- to 20-fold higher than before within about 15 minutes.

The RelA protein, coded for by the relA gene at about 60' on the E. coli genome is the so-called (p)ppGpp synthetase. It is 77 kDa protein. It only synthesises (p)ppGpp in the presence of ribosomes subsequent to addition of mRNA, codon specific uncharged tRNA, ATP and GTP. The reaction involves the transfer of pyrophosphate from the ATP to GTP or GDP. There is some disaggreement of how (p)ppGpp inhibit rRNA synthesis. One suggestion is that ppGpp somehow induces inhibition of binding RNA polymerase at the initiation sites for rRNA at P1 and P2. However, others believe it does not effect this binding. It has been suggested that there are two forms of RNA polymerase, one bound to (p)ppGpp and incapable of initiating transcription and a free or unbound enzyme, which is capable of initiating transcription. RNA polymerase may undergo a conformational change following binding to promoters subject to stringent control, as a result of which it can bind (p)ppGpp. The most commonly assumed site for the binding of the (p)ppGpp molecules is b-subunit of the the RNA polymerase, although the b'- and s-subunits have also been implicated.

While the stringent control model discussed above considers that the reduction of the rRNA synthesis is a direct result of the action of the increased concentration of (p)ppGpp, there is also considered a model based on ribosome feedback to explain the reduction of rRNA synthesis during starvation. This latter model would purely reflect the fact that reduced translational activity causes reduced rRNA synthesis. Both models may actually be in operation with the (p)ppGpp being only an extreme case of a general phenomenon. It is known that many factors of E. coli metabolism are under more than one control mechanism so there really is no reason why this important system should not be as well.

Of the other regulators mentioned earlier, less is known at present. The Fis seems predominantly to act in a fine-tuning capacity. Possibly it acts by priming the rRNA synthesis so that it can adapt more rapidly to the changed growth conditions. The UP elements, which are the AT-rich sites upstream of a numner of promoters have also been observed upstream of the rrnB P2 promoter. The role of these UP elements have still to be clarified.

There is a close association between mRNA transcription and its translation by the ribosomes into protein. The picture of an mRNA slowly being peeled off the DNA, with ribosomes attached, from each of which a poplypeptide chain emerges, is the view which is the accepted model for protein synthesis. This model not only demonstrates the great efficiency with which E. coli biosynthesizes and reproduces itself, but is also an example of control. In this model the ribosomes are also cast in the role of restricting access of the termination factor r to the mRNA. The question is how transcription of rrn avoid premature transcription termination, as they are not subject to this ribosomal effect. It has been suggested that there is an antitermination mechanism in operation within the E. coli rrn operons. There are very strong r-dependent terminators at the end of the rrn operons, which may counteract the early termination. Many aspects of these systems have yet to be clarified.

+*+*RNA replication

© Copyright Microbionet.

All literary matter in Microbionet is covered by copyright, and must not be reproduced, stored in a retrieval system, or transmitted in any form by electronic or mechanical means, photocopying, or recording, without written permission. This page is to be read in conjunction with the Disclaimer.

Use of this site signifes your agreement to the Legal Notices. Legal Notices © Copyright Microbionet. All rights reserved.