Conformational limitations of glycylsarcosine as a prototypic substrate for peptide transporters.

Peptide transporters are present in all species to absorb the small peptides that occur ubiquitously as products of proteolysis. The broad substrate specificities of these systems allow them to be exploited therapeutically for delivery of peptidomimetic drugs in microbes and man. To this end, glycylsarcosine is currently used as a standard substrate for assaying peptidomimetic transport by peptide transporters. However, in this study we find it is unsuitable as a general substrate, based on assays of its transport by model bacterial peptide transporters and computer-based conformational analysis of its structure. Of the two generic transporters for di- and tripeptides, exemplified by Dpp and Tpp in Escherichia coli, only Dpp can transport glycylsarcosine. The explanation for this finding came from molecular modelling, which indicated that glycylsarcosine can adopt only a restricted range of conformers compared with typical dipeptides, and that of the conformers with a trans peptide bond, the majority have the specific psi and phi backbone torsion angles needed for molecular recognition and transport by Dpp but none possessed psi and phi torsions required for recognition by Tpp; moreover, 38% of its conformers have cis peptide bonds that are not substrates for any peptide transporter. Thus, using glycylsarcosine as substrate in competition assays with compounds that typically form conformers recognised by both types of peptide transporter will underestimate their transport. These findings have implications for assays of oral availability of peptidomimetic drugs such as beta-lactams, ACE inhibitors and anti-viral compounds, for which glycylsarcosine is routinely used.

Structural basis for recognition of dipeptides by peptide transporters.

Our objective in this work was to identify the structural basis for the molecular recognition of peptides by peptide transporters. Various assays for dipeptide transport by the dipeptide and tripeptide permeases of Escherichia coli were performed, together with measurements of thermodynamic parameters of substrate binding to the dipeptide binding protein using isothermal titration calorimetry. Computer-based conformational analysis of the test dipeptides was performed to define the repertoire of conformers that each dipeptide adopts in solution. Strict correlations were identified between the complement of particular conformers adopted by a peptide and its bioactivity as a substrate for each transporter. Details of the structural and electronic parameters that define the molecular recognition templates (MRTs) of the dipeptide substrates of these transporters are presented; similar MRTs are likely to apply with dipeptidases. These MRTs provide the essential information for the rational design of peptide-based drugs tailored for exploitation of peptide transporters in microorganisms and man.

Substrate specificity of the periplasmic dipeptide-binding protein from Escherichia coli: experimental basis for the design of peptide prodrugs.

Pure dipeptide-binding protein (DppA) from Escherichia coli was studied in a filter binding assay to determine its binding specificity. A substrate:DppA stoichiometry of 1:1 was found with both [14C]AlaAla and Ala[14C]Phe. Surprisingly, substrate binding did not vary over the pH range pH 3-9.5. Different dipeptides yielded liganded protein with various pI values, implying that DppA can undergo subtly different conformational changes to accommodate different substrates. Using [125I]Tyr-peptides as substrates in competition assays, the relative binding affinities for a range of dipeptides were found to parallel their overall transport rates into E. coli through the dipeptide permease (Dpp), showing that DppA alone controls the specificity of Dpp. With a series of substituted glycyl peptides, binding affinity was progressively enhanced by alkylation (with methyl to butyl) of the N-terminal alpha-amino group. Thus, results from this approach provide an essential experimental basis, which complements the information from the crystal structure of DppA, for the design of peptidomimetic antibacterials targeted for transport through Dpp.

A mutational hot-spot in the hisM gene of the histidine transport operon in Salmonella typhimurium is due to deletion of repeated sequences and results in an altered specificity of transport.

Duplicated sequences within hisM, a gene coding for a membrane-bound component of histidine transport, result in frequent deletions which, being in frame, allow production of an altered protein with apparent changed specificity of transport. While the wild-type transport system does not transport L-histidinol but does transport L-histidine and several of its analogs, the hisM deletion mutants do not transport the latter compounds but do transport L-histidinol. These results are interpreted as supporting the hypothesis (Ames and Higgins 1983) that transport through periplasmic systems involves binding of the substrate by the cytoplasmic membrane-bound components.

Transport and hydrolysis of antibacterial peptide analogues in Escherichia coli: backbone-modified aminoxy peptides.

Aminoxy analogues of di- and tripeptides in which the peptide linkage is replaced by -CO-NHO-, either as an L- or D-2-aminoxypropionic acid (L or D-OAla) residue, have been examined for antibacterial activity in vitro and for uptake into Escherichia coli. Isolation of analogue-resistant mutants and cross-resistance tests with peptide-transport mutants indicate that all three peptide permeases can transport these backbone-modified analogues. A number of mutants with defects in particular intracellular peptidases show decreased sensitivity to a range of these analogues, allowing identification of the enzymes responsible for their cleavage and confirming that hydrolysis is essential for their toxicity. Ala-OAla is a bacteriostatic agent that inhibits nucleic acid and protein synthesis within 1 min of being added to an exponentially growing culture. In crude extracts Ala-OAla inhibits transaminase activity but only after liberation of OAla by endogenous peptidases. These antibacterial agents illustrate an approach to drug targeting in which peptide carriers are used to promote uptake of essentially impermeant toxic moieties.

gamma-Glutamyltransferase is not involved in the bulk uptake of amino acids, peptides or gamma-glutamyl-amino acids in yeast (Saccharomyces cerevisiae).

gamma-Glutamyltransferase activity has been measured in yeast (Saccharomyces cerevisiae) and shown to be associated mainly with the membrane fraction. A similar level of activity is found in a wild-type strain and in gap and gpp strains, the latter mutants being defective in the general amino acid and peptide permeases respectively. The activity is inhibited in whole cells by 6-diazo-5-oxo-L-norleucine (N2O-Nle), azaserine and serine-borate complex; this inactivation seemingly acts from without, for it is similar in (i) control and dicyclohexylcarbodi-imide-treated cells and in (ii) the wild-type and a gap mutant, a treatment and a mutation that it has been shown prevents uptake of the inhibitors. Thus a major portion of the gamma-glutamyltransferase activity appears to exist in a membrane-bound form that is orientated with its gamma-glutamyl-binding site facing the outside. Yeast cells in which gamma-glutamyltransferase has been inactivated by N2O-Nle show no significant change in their rates of uptake of a variety of amino acids, dipeptides and gamma-glutamyl-amino acids. The results preclude a major, direct role for gamma-glutamyltransferase in the transport of these substrates.

Campylobacter enteritis in Nottingham.

The incidence of campylobacter gastroenteritis in the population of Nottingham over a period of 3 years was studied. There was a seasonal variation with the highest number of cases occurring in the summer months. Campylobacter sp. were isolated from the stools of a total of 780 patients over this period. Of these 160 patients with gastroenteritis required admission to hospital. These patients' illness had an acute onset, and the predominant features were diarrhoea, severe abdominal pain, nausea and bright red blood with the stool. However not all the patients had diarrhoea. More than a third of the patients studied were less than 10 years old. The mean duration of symptoms was 4 days and the average stay in hospital was 5 days; some patients required prolonged admission (14 days). In a few cases campylobacter enteritis mimicked other clinical conditions including acute appendicitis. This study emphasises the importance of campylobacter enteritis as a cause of gastroenteritis in the community and the degree of morbidity associated with this illness.

Monitoring enzyme synthesis as a means of studying peptide transport and utilization in Escherichia coli.

A new method has been developed for measuring peptide transport in aminoacid auxotrophs of Escherichia coli by following induction of beta-galactosidase. Appearance of the enzyme was determined after addition of inducer and peptides to amino-acid starved bacteria. For a given number of lysine equivalents, the rate and the extent of enzyme synthesis were the same for lysine and lysyl peptides; similar results were found for glycine and glycl peptides. Saturation constants for peptide transport were determined from the exogenous peptide concentration that gave half maximal rates of enzyme synthesis. The saturation constants, studies with mutants defective in peptide transport, and detection of competition between peptides for uptake, all endorsed earlier conclusions from growth tests about the structural specificities for peptide transport. The new method is quicker, more sensitive and more informative than growth tests.