Search

Mutants of Escherichia coli unable to use fructose by means of the phosphoenolpyruvate/glycose phosphotransferase system mutate further to permit growth on that ketose by derepression of a manno(fructo)kinase (Mak(+) phenotype) present in only trace amounts in the parent organisms (Mak-o phenotype). The mak gene was located at min 8.8 on the E. coli linkage map as an ORF designated yajF, of hitherto unknown function; it specifies a deduced polypeptide of 344 aa. The derepression of Mak activity was associated with a single base change at position 71 (codon 24) of the gene, where GCC (alanine) in Mak-o has been changed to GAC (aspartate) in Mak(+). By cloning selected portions of the total 1,032-bp mak gene into a plasmid that also carried a temperature-sensitive promoter, we showed that the mutation resided in a 117-bp region that does not specify sequences necessary for Mak activity but was located 46 bp upstream of a 915-bp portion that does. Mak(+) and Mak-o strains differ greatly in the heat stability of the enzyme: at 61 degrees C, mak-o cloned into a mak-o recipient loses 50% of its activity in approximately 6 min, whereas it takes over 30 min to achieve a similar reduction in the activity of mak(+) cloned into a mak-o strain. However, the Mak activity of the cloned fragment specifying the enzyme without the regulatory region lost activity with a half-life of 29 min irrespective of whether it was derived from a mak(+) or a mak-o donor, which indicates that the A24D mutation contributes to the high enzyme activity of Mak(+) mutants by serving to protect Mak from denaturation.

There are three main routes for the utilization of fructose by Escherichia coli. One (Route A) predominates in the growth of wild-type strains. It involves the functioning of the phosphoenolpyruvate:glycose phosphotransferase system (PTS) and a fructose operon, mapping at min. 48.7, containing genes for a membrane-spanning protein (fruA), a 1-phosphofructose kinase (fruK) and a diphosphoryl transfer protein (fruB), under negative regulation by a fruR gene mapping at min. 1.9. A second route (Route B) also involves the PTS and membrane-spanning proteins that recognize a variety of sugars possessing the 3,4,5-D-arabino-hexoseconfiguration but with primary specificity for mannose(manXYZ), mannitol (mtlA) and glucitol (gutA) and which, if over-produced, can transport also fructose. A third route (Route C), functioning in mutants devoid of Routes A and B, does not involve the PTS: fructose diffuses into the cell via an isoform (PtsG-F) of the major glucose permease of the PTS and is then phosphorylated by ATP and a manno(fructo)kinase (Mak+) specified by a normally cryptic 1032 bp ORF (yajF) of hitherto unknown function (Mak-o), mapping at min. 8.8 and corresponding to a peptide of 344 amino acids. Conversion of the Mak-o to the Mak+ phenotypeinvolves an A24D mutation in a putative regulatory region.

From mutants of Escherichia coli unable to utilize fructose via the phosphoenolpyruvate/glycose phosphotransferase system (PTS), further mutants were selected that grow on fructose as the sole carbon source, albeit with relatively low affinity for that hexose (K(m) for growth approximately 8 mM but with V(max) for generation time approximately 1 h 10 min); the fructose thus taken into the cells is phosphorylated to fructose 6-phosphate by ATP and a cytosolic fructo(manno)kinase (Mak). The gene effecting the translocation of fructose was identified by Hfr-mediated conjugations and by phage-mediated transduction as specifying an isoform of the membrane-spanning enzyme II(Glc) of the PTS, which we designate ptsG-F. Exconjugants that had acquired ptsG(+) from Hfr strains used for mapping (designated ptsG-I) grew very poorly on fructose (V(max) approximately 7 h 20 min), even though they were rich in Mak activity. A mutant of E. coli also rich in Mak but unable to grow on glucose by virtue of transposon-mediated inactivations both of ptsG and of the genes specifying enzyme II(Man) (manXYZ) was restored to growth on glucose by plasmids containing either ptsG-F or ptsG-I, but only the former restored growth on fructose. Sequence analysis showed that the difference between these two forms of ptsG, which was reflected also by differences in the rates at which they translocated mannose and glucose analogs such as methyl alpha-glucoside and 2-deoxyglucose, resided in a substitution of G in ptsG-I by T in ptsG-F in the first position of codon 12, with consequent replacement of valine by phenylalanine in the deduced amino acid sequence.

The Escherichia coli genome sequencing project in the 47 to 48 centisome region has resulted in the sequencing of the complete fructose operon (fruBKA). Due to a single base insertion, the presence of the fruB gene went unnoticed. The revised nucleotide sequence of the fruB gene, the deduced amino acid sequence of its protein product, the diphosphoryl transfer protein of the phosphoenolpyruvate: sugar phosphotransferase system, and putative transcriptional regulatory signals of the fru operon of E. coli are here presented and compared with that from Salmonella typhimurium.

Mutants of Escherichia coli devoid of the membrane-spanning proteins PtsG and PtsMP, which are components of the phosphoenolpyruvate-dependent phosphotransferase system (PTS) and which normally effect the transport into the cells of glucose and mannose, do not grow upon or take up either sugar. Pseudorevertants are described that take up, and grow upon, mannose at rates strongly dependent on the mannose concentration in the medium (apparent Km > 5 mM); such mutants do not grow upon glucose but are derepressed for the components of the fructose operon. Evidence is presented that mannose is now taken up via the fructose-PTS to form mannose 6-phosphate, which is further utilized for growth via fructose 6-phosphate and fructose 1,6-bisphosphate.

The sequence was determined of the 936 nucleotides that compose the fruK gene of Escherichia coli K12, together with the final 310 bases of fruF and the initial 224 bases of fruA, which flank fruK. These genes specify proteins that effect the uptake of fructose and its PEP-dependent conversion to fructose 1-phosphate (fruA and fruF), and the ATP-dependent phosphorylation of that product to fructose 1,6-bisphosphate (fruK); together, these genes form the fruFKA operon. The deduced amino acid sequence of the fruK product exhibits little similarity to the major 6-phosphofructokinase of E. coli (pfkA) and the 6-phosphofructokinases present in a number of pro- and eukaryotic organisms, but there is 27%, 25% and 22% identify of sequence respectively with the minor 6-phosphofructokinase (pfkB) of E. coli, the lacC gene product of Staphylococcus aureus and the ribokinase of E. coli.

The utilization of fructose by Escherichia coli involves, as first step, the uptake of the sugar, normally via the phosphoenolpyruvate-dependent phosphotransferase system (PTS). This fructose-specific PTS differs in several ways from that effecting the uptake of other sugars that also possess the 3,4,5-D-arabino-hexose configuration: these differences are discussed. Mutants that lack the genes ptsI and ptsH, which specify components of the PTS common to most PT-sugars, can mutate further to regain the ability to utilize fructose when this is present in relatively high concentration (i.e. greater than 2 mM) in the medium. Some of the properties of this unusual uptake system is discussed.

A lysogen of Escherichia coli K12 with lambda cI857 S7 xis6 nin5 b515 b519 integrated into ptsI was induced and the lysates plated on a Pel- host [on which lambda strains with less than the wild-type amount of DNA form plaques at low frequency (Cameron et al., 1977)]. All of the 40 plaques examined contained phage able to transduce at least two of the genes known from bacteriophage P1 transduction experiments to be closely linked to ptsI. Assuming that each specialized transducing phage arose by a single illegitimate recombination event, the distribution of phage types showed that the gene order is cysA gsr ptsI (ptsH, iex) cysZ lig; both gsr+ and iex+ were dominant. Analysis of restriction endonuclease digests of the transducing phage confirmed that no unexpected DNA rearrangements had taken place and allowed the construction of a map of the sites of action of the restriction endonucleases EcoRI, HindIII, BamI and Kpn for over 20 kilobases of E. coli DNA. In an Appendix, we show cysA and cysZ mutants to be deficient in sulphate assimilation.

Galactitol-positive strains of Escherichia coli K12 are inhibited by the galactitol analogues L-fucitol and 2-deoxy-D-galactitol, but not by D-fucitol; Salmonella typhimurium LT2 is not inhibited by these compounds. Most mutants selected as resistant to either toxic compound are unable to utilize galactitol as carbon source, but a relatively rare class is inducible for the Enzyme II of the galactitol:phosphoenolpyruvate phosphotransferase system, the product of which is D-galactitol 6-phosphate. The lesion in one such mutant maps near metG at about min 45 on the E. coli genome.

L-Sorbose is phosphorylated by Escherichia coli by two distinct Enzymes II of the phosphoenolpyruvate-dependent phosphotransferase system. The glucose Enzyme II (specified by the gene ptsG) phosphorylates L-sorbose with an apparent Km of 0.08 +/- 0.03 mM and V of 31.8 +/- 3.5 nmol . mg-1 . min-1 whilst the fructose Enzyme II (specified by the gene ptsF) phosphorylates it with an apparent Km of 28.9 +/- 2.7 mM and V of 20.2 +/- 0.8 nmol . mg-1 . min-1. L-Sorbose induces neither of these Enzymes II, but sorbose inhibits the growth of strains expressing either of these functions constitutively. Mutants that have lost their sensitivity to L-sorbose are found to have lost either the glucose or the fructose phosphotransferase Enzyme II.U

Cells of the wild-type yeast (Saccharomyces cerevisiae) strain Y185, grown under conditions that de-repress the formation of a general amino acid permease ('Gap') system, bind delta-N-chloroacetyl[1-(14)C]ornithine; L- and D-amino acid substrates of the general amino acid permease system protect against this binding. The protein responsible is released from the cells by homogenization or by preparation of protoplasts; it is not released by osmotic shock. This protein is virtually absent from the wild-type strain when it is grown under conditions that repress the general amino acid permease system, and is also absent from a Gap- mutant Y185-His3, selected by its resistance to D-amino acids. This mutant and repressed wild-type cells also fail to form a number of membrane proteins elaborated by de-repressed wild-type cells. It is possible that all these proteins are components of the general amino acid permease system.

Glucosamine, mannose and 2-deoxyglucose enter Escherichia coli by the phosphotransferase system coded for by the gene ptsM. The glucosamine- and mannose-negative, deoxyglucose-resistant phenotype of ptsM mutants can be suppressed by a mutation mapping near ptsG that allows constitutive expression of the glucose phosphotransferase coded for by the gene ptsG. N-Acetylglucosamine enters E. coli by two distinct phosphotransferase systems (White, 1970). One of these is the PtsM system, the other is coded for by a gene which maps near the nagA,B genes at about min 15 on the E. coli chromosome. We propose that this gene be designated ptsN. Strains with either of these components of the phosphotransferase system will utilize N-acetylglucosamine as sole carbon source.

Strains of Escherichia coli devoid of systems for the active transport of galactose (galP mgl) still grow on galactose but at rates that are a function of the galactose concentration of the medium: half-maximal growth rates require more than 2 mM-galactose to be present. Evidence is presented that galactose is taken up by such strains by facilitated diffusion on a carrier specified by the umg gene (or by a gene highly co-transducible with it) which is thus a part of, or closely associated with, an enzyme II for glucose of the phosphoenolpyruvate-phosphotransferase system. However, the entry of galactose does not require phosphotransferase activity, and the sugar taken up appears in the cells as free galactose.

Mutants of Escherichia coli K12 defective in the gene iex (crr) no longer utilize glucose or N-acetylglucosamine in preference to lactose, but competition between either of these sugars and another that also enters by a phosphotransferase (PT) mechanism is not affected. In this they differ from gsr (tgs) mutants. In gsr mutants, glucose does not exclude any other sugar, though N-acetylglucosamine still does so. In gsr mutants that are also ptsM the phosphoenolpyruvate-dependent phosphorylation of glucose or methyl alpha-glucoside is reduced by 90%: N-acetylglucosamine phosphorylation is not affected. The iex mutation does not affect the phosphorylation of either of these compounds. The wild-type alleles iex+ and gsr+ are dominant in lambda heterozygotes. Glucose inhibits the lactose permease of wild-type cells, but only when the permease is present in low amounts. The inhibition is also relieved (1) by induction of another transport system that is subject to regulation by the iex system or (2) by an iex mutation. We suggest that the iex gene specifies a protein that, in cells transporting certain sugars by a PT mechanism, acts to inhibit active transport systems. The protein is present in limiting concentration in the cell, sufficient only to inhibit the basal, uninduced, level of the active transport systems. In consequence the inducer (or its precursor) may be excluded from the cell and induction thus prevented.

The general amino acid permease ('Gap') system of the wild-type yeast (Saccharomyces cerevisiae) strain Y185 is inhibited by the uptake and accumulation of its substrate amino acids. Surprisingly, this inhibition persists even after 'pools' of amino acids, accumulated initially, have returned to normal sizes. Recovery from this inhibition depends on a supply of energy and involves the synthesis of a membrane protein component of the Gap system.

A mutant, DAI, has been isolated from the Escherichia coli K12, strain K2. 1t, as a colony resistant to 2-deoxyglucose (DG) when growing on fructose but still sensitive to DG when growing on other sugars. The mutation in DAI specifically affects the catabolite inhibition of fructose utilization by glucose and glucose-6-phosphate; the affected gene (designated cif) is located at min 41 on the E. coli linkage map and is highly co-transducible with the genes that specify the uptake of fructose (ptsF) and enzymic conversion of fructose-1-phosphate to fructose-1,6-bisphosphate (fpk).

In contrast to active transport, the uptake of carbohydrates via the phosphoenolpyruvate-dependent phosphotransferase system (PTS) leads to the appearance in the cell of the sugar initially as a 1- or 6-phosphate ester. The components of the PTS that transfer phosphate to the sugar are not absolutely specific for any one sugar. Both their synthesis and their activity are controlled; in the latter, "fine" control, glucose-6-phosphate appears to play an important role. Studies of growth on, and uptake of, galactose by E.coli mutants devoid of components of the PTS and also devoid of active transport systems for galactose, suggest that proteins effecting facilitated diffusion of hexoses may be part of, or be closely associated with, the sugar-specific components of the PTS.

The Enzyme IIfru of the phosphoenolpyruvate- (PEP-) dependent phosphotransferase system (PTS), which catalyses the uptake of fructose and its concomitant phosphorylation to fructose 1-phosphate by Escherichia coli, is specified by a gene designated fruA. The nucleotide sequence of a 2.5 kb PvuII restriction fragment spanning fruA+, cloned on a plasmid, was determined. This fragment contained three open reading frames (ORFs) but only one complete ORF, 1689 base pairs long, which was preceded by a well-defined Shine-Dalgarno sequence and ended with a rho-independent transcription terminator. The amino acid sequence deduced from this DNA corresponds to that of a protein of 563 amino acids (57.5 kDa), which has the hydropathic profile expected of an integral membrane protein (average hydropathy = 0.40) and which is characterized by a number of well-marked hydrophobic loops that may correspond to membrane-spanning regions. There is relatively little overall homology between this protein and those of other Enzymes II of the PTS but there is considerable correspondence between the region surrounding one of the six histidine residues (His381) of Enzyme IIfru and those surrounding the particular histidines of other Enzymes II, and of HPr, known to be involved in phosphorylation. A plasmid carrying the complete fruA+ nucleotide sequence, but not that of any other functional protein, fully restored the ability of fruA mutants to grow on fructose and of extracts of fruA mutants to phosphorylate fructose, which confirms that the nucleotide sequence determined species Enzyme IIfru.

1. From Escherichia coli strain K2.1.5(c).8.9, which is devoid of 6-phosphogluconate dehydrogenase (gnd) and 6-phosphogluconate dehydratase (edd) activities, a mutant R6 was isolated that was tolerant to gluconate though still edd(-), gnd(-). 2. Measurements of the fate of labelled gluconate, of the conversion of gluconate into 6-phosphogluconate, and of the induction of gluconate kinase by the two organisms show that, although both inducibly form a gluconate-transport system, strain R6 is impaired in its ability to convert the gluconate thus taken up into 6-phosphogluconate; it was therefore used for study of the kinetics and energetics of gluconate uptake. 3. Suspensions of strain R6 induced for gluconate uptake took up this substrate via a ;high affinity' transport process, with K(m) about 10mum and V(max.) about 25nmol/min per mg dry mass; a ;low affinity' system demonstrated to occur in certain E. coli mutants was not induced under the conditions used in this work. 4. The uptake of gluconate was inhibited by lack of oxygen and by inhibitors of electron transport; such inhibitors also promoted the efflux of gluconate taken up. 5. Membrane vesicles prepared from strain R6 also manifested these properties when incubated with suitable electron donors, at rates similar to those observed with whole cells. 6. The results indicate that the active transport of gluconate into the cells is the rate-limiting step in gluconate utilization by E. coli, and that the mechanism of this process can be validly studied with membrane vesicles.

Recombinant plasmids were constructed that carried various fragments of the DNA specifying the Escherichia coli genes ptsH and (part of) ptsI, the genes for the common components of the phosphoenolpyruvate: sugar phosphotransferase. Expression of plasmid-specified functions in minicells showed that ptsH and ptsI were transcribed clockwise. Most of the transcription of ptsI was from the ptsH promoter, but some was from a second site within or after ptsH.

Strains of Escherichia coli that are unable to grow on fructose because they lack the phosphoenolpyruvate: fructose phosphotransferases specified by ptsF and ptsX mutate to grow on media containing fructose as sole carbon source, but do not regain the function of either of the missing phosphotransferases. Instead, fructose is taken up and phosphorylated to fructose 6-phosphate by a phosphoenolpyruvate: sorbitol phosphotransferase which, in wild-type cells, is induced by sorbitol but not by fructose, but which is constitutively expressed in these mutants. The regulatory gene srlC controlling enzymes of sorbitol uptake and catabolism has been located on the E. coli genome as part of the linkage group cysI srlC attI86 pheA.

Dilute cultures of wild-type Escherichia coli K12 and of derivatives impaired in one or other Enzyme-II component of the glucose phosphotransferase system were grown in continuous culture under glucose limitation. Cells harvested from the chemostat took up [U-14C]glucose from 0.1 mM solutions at rates directly related to the rates at which those cells had grown; the activity of the phosphotransferase system in those cells, rendered permeable with optimal accounts of toluene, parallels the ability of the cells to take up glucose. The capacity of these systems was rate-limiting for growth under the negligibly low glucose concentration in the chemostat, but was adequate to account for the stimulation of respiration observed when the cells were presented suddenly with excess glucose.

The DNA sequences for some of the genes involved in the phosphoenolpyruvate-dependent phosphotransferase system (PTS) of Escherichia coli and Salmonella typhimurium have been reported. Comparison of the deduced amino acid sequences of enzyme IIBgi, enzyme IIMtl, and enzyme IIGlc/enzyme IIIGlc, which catalyze the uptake and concomitant phosphorylation of beta-glucosides, mannitol, and glucose, respectively, reveals considerable sequence homology. In particular, the carboxyl-terminal region of enzyme IIBgl is so homologous to the whole of enzyme IIIGlc as to suggest a common function. We postulate that His-547 of enzyme IIBgl receives a phosphate group directly from the cytoplasmic protein HPr and transfers this phosphate to His-306 located in the amino-terminal half of enzyme IIBgl. This latter histidine is conserved in enzyme IIBgl and enzyme IIGlc and, in both proteins, occurs in a region that shows homology with the His-15 region of HPr, which is known to act as the phosphate carrier. An equivalent histidine residue, His-195, is also present in enzyme IIMtl, although here the flanking sequence is different. None of these specified histidine residues is likely to be buried within the membrane.

The active transport of carbohydrates by Escherichia coli is discussed with particular reference to (1) identification of an uptake process as 'active transport', (2) nature and control of transport proteins, and (3) mechanisms of energy transduction. (1) The use of substrate analogues, of mutants blocked in metabolism and of subcellular vesicles in the isolation of the transport process from interference by subsequent metabolic reactions is described. Criteria are outlined for establishing that the solute is taken up against a concentration gradient and that this is energy-dependent. Three types of poisons for energy systems that act primarily on respiration, on ATP formation and as uncoupling ('proton conducting') agents are considered. (2) Methods are described for the selection of mutants impaired in the active uptake of specific carbohydrates. (3) Results show that the uptake of galactose, D-fucose and arabinose by appropriate strains of E. coli is inducible, specific and accompanied by proton uptake. Such and other data support a model based on a chemiosmotic theory of active transport.