Identification of Novel Cryptic Multifunctional Antimicrobial Peptides from the Human Stomach Enabled by a Computational-Experimental Platform.

Novel approaches are needed to combat antibiotic resistance. Here, we describe a computational-experimental framework for the discovery of novel cryptic antimicrobial peptides (AMPs). The computational platform, based on previously validated antimicrobial scoring functions, indicated the activation peptide of pepsin A, the main human stomach protease, and its N- and C-terminal halves as antimicrobial peptides. The three peptides from pepsinogen A3 isoform were prepared in a recombinant form using a fusion carrier specifically developed to express toxic peptides in Escherichia coli. Recombinant pepsinogen A3-derived peptides proved to be wide-spectrum antimicrobial agents with MIC values in the range 1.56-50 µM (1.56-12.5 µM for the whole activation peptide). Moreover, the activation peptide was bactericidal at pH 3.5 for relevant foodborne pathogens, suggesting that this new class of previously unexplored AMPs may contribute to microbial surveillance within the human stomach. The peptides showed no toxicity toward human cells and exhibited anti-infective activity in vivo, reducing by up to 4 orders of magnitude the bacterial load in a mouse skin infection model. These peptides thus represent a promising new class of antibiotics. We envision that computationally guided data mining approaches such as the one described here will lead to the discovery of antibiotics from previously unexplored sources.

Understanding the mechanism of prosegment-catalyzed folding by solution NMR spectroscopy.

Multidomain protein folding is often more complex than a two-state process, which leads to the spontaneous folding of the native state. Pepsin, a zymogen-derived enzyme, without its prosegment (PS), is irreversibly denatured and folds to a thermodynamically stable, non-native conformation, termed refolded pepsin, which is separated from native pepsin by a large activation barrier. While it is known that PS binds refolded pepsin and catalyzes its conversion to the native form, little structural details are known regarding this conversion. In this study, solution NMR was used to elucidate the PS-catalyzed folding mechanism by examining the key equilibrium states, e.g. native and refolded pepsin, both in the free and PS-bound states, and pepsinogen, the zymogen form of pepsin. Refolded pepsin was found to be partially structured and lacked the correct domain-domain structure and active-site cleft formed in the native state. Analysis of chemical shift data revealed that upon PS binding refolded pepsin folds into a state more similar to that of pepsinogen than to native pepsin. Comparison of pepsin folding by wild-type and mutant PSs, including a double mutant PS, indicated that hydrophobic interactions between residues of prosegment and refolded pepsin lower the folding activation barrier. A mechanism is proposed for the binding of PS to refolded pepsin and how the formation of the native structure is mediated.

Sitafloxacin resistance in Helicobacter pylori isolates and sitafloxacin-based triple therapy as a third-line regimen in Japan.

The third-line treatment regimen for Helicobacter pylori after failure of clarithromycin- and metronidazole-based therapies is not yet established. Sitafloxacin (STX) is a quinolone that possesses potent in vitro activity against H. pylori. In this study, the susceptibility of H. pylori isolates to STX was examined and the efficacy of STX-based triple therapy as a third-line regimen was evaluated. STX showed minimum inhibitory concentrations (MICs) of ≤1 µg/mL against all 100 H. pylori isolates, and the MIC(90) (MIC for 90% of the organisms) of STX was 5 log(2) dilutions lower than that of levofloxacin (LVX). The MIC(50) (MIC for 50% of the organisms) of STX against gyrA mutants was 0.12 µg/mL and was significantly lower than that of LVX (8 µg/mL). The activity of STX at pH 5.5 was significantly less than that at pH 7.0. In the clinical trial, 28 patients with two eradication failures were treated with STX-based triple therapy [rabeprazole 10 mg twice daily (b.i.d.), amoxicillin 750 mg b.i.d. and STX 100mg b.i.d. for 7 days]. The eradication rate was 75% using intention-to-treat analysis and 80% using per-protocol analysis. Two gyrA mutant strains were eradicated. Amongst participants, a low pepsinogen I/II ratio was associated with successful eradication. These results suggest that STX could be active against most clinical H. pylori isolates and that STX-based triple therapy is a promising and safe third-line therapy.

Rational redesign of porcine pepsinogen containing an antimicrobial peptide.

A novel strategy for the controlled release and localization of bioactive peptides within digestive and immunity-related enzymes was developed. The N-terminus of porcine pepsinogen A was fused to the basic amino acid-rich region of bovine lactoferricin B termed 'tLfcB', a cationic antimicrobial/anticancer peptide. Recombinant tLfcB-porcine pepsinogen A was expressed in soluble form in Escherichia coli as a thioredoxin (Trx) fusion protein. Thioredoxin-tLfcB-porcine pepsinogen A was found to activate autocatalytically under acidic conditions. Recombinant pepsin A derived from the activation of the fusion protein had a catalytic rate and substrate affinity similar to that derived from the recombinant thioredoxin-porcine pepsinogen A control. Pepsin-treated thioredoxin-tLfcB-porcine pepsinogen A yielded increased antimicrobial activity against the Gram-negative bacteria E.coli relative to control suggesting that a second function (antimicrobial activity) was successfully engineered into a functional peptidase. The novel design strategy described herein presents a potential strategy for targeted delivery of antimicrobial or therapeutic peptides in transgenic organisms via re-engineering native proteins critical to plant and animal defense mechanisms.

Lineage-specific duplication and loss of pepsinogen genes in hominoid evolution.

Fourteen different pepsinogen-A cDNAs and one pepsinogen-C cDNA have been cloned from gastric mucosa of the orangutan, Pongo pygmaeus. Encoded pepsinogens A were classified into two groups, i.e., types A1 and A2, which are different in acidic character. The occurrence of 9 and 5 alleles of A1 and A2 genes (at least 5 and 3 loci), respectively was anticipated. Respective orthologous genes are present in the chimpanzee genome although their copy numbers are much smaller than those of the orangutan genes. Only A1 genes are present in the human probably due to the loss of the A2 gene. Molecular phylogenetic analyses showed that A1 and A2 genes diverged before the speciation of great hominoids. Further reduplications of respective genes occurred several times in the orangutan lineage, with much higher frequencies than those occurred in the chimpanzee and human lineages. The rates of non-synonymous substitutions were higher than those of synonymous ones in the lineage of A2 genes, implying the contribution of the positive selection on the encoded enzymes. Several sites of pepsin moieties were indeed found to be under positive selection, and most of them locate on the surface of the molecule, being involved in the conformational flexibility. Deduced from the known genomic structures of pepsinogen-A genes of primates and other mammals, the duplication/loss were frequent during their evolution. The extreme multiplication in the orangutan might be advantageous for digestion of herbaceous foods due to the increase in the level of enzymes in stomach and the diversification of enzyme specificity.

The prosegment catalyzes pepsin folding to a kinetically trapped native state.

Investigations of irreversible protein unfolding often assume that alterations to the unfolded state, rather than the nature of the native state itself, are the cause of the irreversibility. However, the present study describes a less common explanation for the irreversible denaturation of pepsin, a zymogen-derived aspartic peptidase. The presence of a large folding barrier combined with the thermodynamically metastable nature of the native state, the formation of which depends on a separate prosegment (PS) domain, is the source of the irreversibility. Pepsin is unable to refold to the native state upon return from denaturing conditions due to a large folding barrier (24.6 kcal/mol) and instead forms a thermodynamically stable, yet inactive, refolded state. The native state is kinetically stabilized by an unfolding activation energy of 24.5 kcal/mol, comparable to the folding barrier, indicating that native pepsin exists as a thermodynamically metastable state. However, in the presence of the PS, the native state becomes thermodynamically stable, and the PS catalyzes pepsin folding by stabilizing the folding transition state by 14.7 kcal/mol. Once folded, the PS is removed, and the native conformation exists as a kinetically trapped state. Thus, while PS-guided folding is thermodynamically driven, without the PS the pepsin energy landscape is dominated by kinetic barriers rather than by free energy differences between native and denatured states. As pepsin is the archetype of a broad class of aspartic peptidases of similar structure and function, and many require their PS for correct folding, these results suggest that the occurrence of native states optimized for kinetic rather than thermodynamic stability may be a common feature of protein design.

Contribution of the intra- and intermolecular routes in autocatalytic zymogen activation: application to pepsinogen activation.

Taking as the starting point a recently suggested reaction scheme for zymogen activation involving intra- and intermolecular routes and the enzyme-zymogen complex, we carry out a complete analysis of the relative contribution of both routes in the process. This analysis suggests the definition of new dimensionless parameters allowing the elaboration, from the values of the rate constants and initial conditions, of the time course of the contribution of the two routes. The procedure mentioned above related to a concrete reaction scheme is extrapolated to any other model of autocatalytic zymogen activation involving intra- and intermolecular routes. Finally, we discuss the contribution of both of the activating routes in pepsinogen activation into pepsin using the values of the kinetic parameters given in the literature.

A novel fusion protein system for the production of native human pepsinogen in the bacterial periplasm.

Human pepsinogen is the secreted inactive precursor of pepsin. Under the acidic conditions present in the stomach it is autocatalytically cleaved into the active protease. Pepsinogen contains three consecutive disulfides, and was used here as a model protein to investigate the production of aspartic proteases in the Escherichia coli periplasm. Various N-terminal translocation signals were applied and several different expression vectors were tested. After fusion to pelB, dsbA or ompT signal peptides no recombinant product could be obtained in the periplasm using the T7 promoter. As a new approach, human pepsinogen was fused to E. coli ecotin (E. coli trypsin inhibitor), which is a periplasmic homodimeric protein of 142 amino acids per monomer containing one disulfide bridge. The fusion protein was expressed in pTrc99a. After induction, the ecotin-pepsinogen fusion protein was translocated into the periplasm and the ecotin signal peptide was cleaved. Upon acid treatment, the fusion protein was converted into pepsin, indicating that pepsinogen was produced in its native form. In shake flasks experiments, the amount of active fusion protein present in the periplasm was 100 microg per litre OD 1, corresponding to 70 microg pepsinogen. After large scale cultivation, the fusion protein was isolated from the periplasmic extract. It was purified to homogeneity with a yield of 20%. The purified protein was native. Acid-induced activation of the fusion protein proceeded very fast. As soon as pepsin was present, the ecotin part of the fusion protein was rapidly digested, followed by a further activation of pepsinogen.

Purification and partial characterization of feline pepsinogen.

OBJECTIVE: To purify and partially characterize feline pepsinogen (fPG) from the gastric mucosa and compare fPG with PGs of other species. SAMPLE POPULATION: Stomachs of 6 cats. PROCEDURE: A crude protein extract was prepared from the gastric mucosa of feline stomachs. Feline PG A was purified by ammonium sulfate precipitation, weak-anion-exchange chromatography, size-exclusion chromatography, and strong-anion exchange chromatography. Partial characterization consisted of estimation of molecular weights (MWs) and isoelectric points, N-terminal amino acid sequencing, and investigation of susceptibility to pepstatin inhibition. RESULTS: Several fPG A-group isoforms were identified. The MWs of the isoforms ranged from 37,000 to 44,820. Isoelectric points were all < pH 3.0. The proteolytic activity of the activated PGs was inhibited completely by pepstatin in a range of equimolar to 10-fold molar excess. The specific absorbance of fPG A was 1.29. The N-terminal amino acid sequence of the first 25 residues of the predominant fPG A7 had 75%, 72%, 64%, and 56% homology with PG A of dogs, rabbits, cattle, and humans, respectively. Sequences of 4 other fPG A-group isoforms were similar to fPG A7. All isoforms were immunologically cross-reactive with sheep anti-fPG A7 antiserum. CONCLUSIONS AND CLINICAL RELEVANCE: PG A is the only identified type of PG in cats and, similar to pg in other species, comprises multiple isoforms. The availability of fPG A may be used to facilitate the development of an immunoassay to quantify serum fPG A as a potential marker for gastric disorders in cats.

Extracellular calcium acts as a "third messenger" to regulate enzyme and alkaline secretion.

It is generally assumed that the functional consequences of stimulation with Ca2+ -mobilizing agonists are derived exclusively from the second messenger action of intracellular Ca2+, acting on targets inside the cells. However, during Ca2+ signaling events, Ca2+ moves in and out of the cell, causing changes not only in intracellular Ca2+, but also in local extracellular Ca2+. The fact that numerous cell types possess an extracellular Ca2+ "sensor" raises the question of whether these dynamic changes in external [Ca2+] may serve some sort of messenger function. We found that in intact gastric mucosa, the changes in extracellular [Ca2+] secondary to carbachol-induced increases in intracellular [Ca2+] were sufficient and necessary to elicit alkaline secretion and pepsinogen secretion, independent of intracellular [Ca2+] changes. These findings suggest that extracellular Ca2+ can act as a "third messenger" via Ca2+ sensor(s) to regulate specific subsets of tissue function previously assumed to be under the direct control of intracellular Ca2+.

Increased breath nitrous oxide after ingesting nitrate in patients with atrophic gastritis and partial gastrectomy.

BACKGROUND: Toxic nitrite and N-nitroso compounds due to gastric bacterial growth are often detected in the stomach of patients with atrophic gastritis and partial gastrectomy. The aim of this study is to investigate whether breath N2O, a major metabolite of denitrification, detected after ingestion of nitrate is associated with atrophic gastritis and partial gastrectomy. METHODS: Nine young, 16 normal older, nine atrophic gastritis and six partial gastrectomy subjects ingested 100 g lettuce, equal to 130 mg nitrate, and breath N2O was measured at 15-min intervals for 5 h. N2O was analyzed using an infrared-photoacoustic analyzer, and atrophic gastritis was diagnosed by pepsinogen test. RESULTS: The mean breath N2O concentrations were higher in the following order at all times: partial gastrectomy>atrophic gastritis>normal>young. The maximum N2O concentrations in the patients with partial gastrectomy and atrophic gastritis were 1655 +/- 296 and 1350 +/- 200 (mean +/- S.E.) ppb, respectively, which were higher than that of the normal subjects, 827 +/- 91 ppb (P < 0.05). The maximum N2O concentration in young people was 527 +/- 86 ppb, which was lower than that of the normal older people (P < 0.051). CONCLUSION: These higher N2O concentrations in gastric patients reflect bacterial growth in the stomach due to the reduction of gastric acid.

N-terminal modifications increase the neutral-pH stability of pepsin.

A structure-function study was undertaken to determine the effects of N-terminal mutations in pepsin designed to introduce the Lys-X-Tyr motif and increase N-terminal flexibility. At pH 7.0, E7K/T12A/E13Q pepsin was inactivated more slowly compared to WT, whereas the mutants E7K and T12A/E13Q were not stabilized. Far-UV circular dichroism revealed that changes in secondary structure accompanied the inactivation process, and that the structural changes occurred at approximately the same rate as inactivation. All of the inactivated pepsin forms showed retention of substantial secondary structure, more than previously determined for pepsin denatured at pH 7.2 and 8.0, suggesting the presence of a structural intermediate at pH 7.0. The coupled mutations at positions 12 and 13 impacted the pH dependence of activity at pH 0.9, lowered affinity for a synthetic substrate, and lowered the turnover number. The introduction of Lys at position 7 apparently destabilized the interaction between prosegment-enzyme body as evidenced by activation at higher pH (>or= 4.0) compared to WT, but showed no change for pH dependence of activity, nor a statistically significant change in affinity for the synthetic substrate.

Purification and partial characterization of canine pepsinogen A and B.

OBJECTIVE: To purify and partially characterize various isoforms of canine pepsinogen (PG) from gastric mucosa. SAMPLE POPULATION: Stomachs obtained from 6 euthanatized dogs. PROCEDURE: Mucosa was scraped from canine stomachs, and a crude mucosal extract was prepared and further purified by use of weak anion-exchange chromatography, hydroxyapatite chromatography, size-exclusion chromatography, and strong anion-exchange chromatography. Pepsinogens were characterized by estimation of molecular weights, estimation of their isoelectric points (IEPs), and N-terminal amino acid sequencing. RESULTS: Two different groups of canine PG were identified after the final strong anion-exchange chromatography: PG A and PG B. Pepsinogens differed in their molecular weights and IER Pepsinogen B appeared to be a dimer with a molecular weight of approximately 34,100 and an IEP of 4.9. Pepsinogen A separated into several isoforms. Molecular weights for the various isoforms of PG A ranged from 34,200 to 42,100, and their IEPs ranged from 4.0 to < 3.0. The N-terminal amino acid sequence for the first 25 amino acid residues for PG A and B had good homology with the amino acid sequences for these proteins in other species. CONCLUSIONS AND CLINICAL RELEVANCE: Canine PG B and several isoforms of canine PG A have been purified. Availability of these PGs will facilitate development of immunoassays to measure PG in canine serum as a potential diagnostic marker for gastric disorders in dogs.

Soluble expression and purification of porcine pepsinogen from Pichia pastoris.

This paper presents a new system for the soluble expression and characterization of porcine pepsinogen from the methylotrophic yeast Pichia pastoris. The cDNA that encodes the zymogenic form of porcine pepsin (EC 3.4.23.1) was cloned into the EcoRI site of the vector pHIL-S1 downstream from the AOX1 alcohol oxidase promoter. After P. pastoris transformation, colonies were screened for expression of pepsinogen based on enzyme activity of the active form, pepsin. The recombinant enzyme was purified 138-fold by anion exchange and affinity column chromatography. Homogeneity was confirmed through SDS-PAGE, Western blot, and N-terminal sequencing. When compared to commercial pepsin, the recombinant pepsin had similar kinetic profiles, pH/temperature stability, and secondary/tertiary conformation. A glycosylated form was also isolated and found to exhibit kinetic and structural characteristics similar to those of the commercial and wild-type pepsin, but was slightly more thermal stable. The above results indicate that the P. pastoris expression system offers a convenient and efficient means to produce and purify a soluble form of pepsin(ogen).

A basic residue at position 36p of the propeptide is not essential for the correct folding and subsequent autocatalytic activation of prochymosin.

Position 36p in the propeptides of gastric aspartic proteinases is generally occupied by lysine or arginine. This has led to the conclusion that a basic residue at this position, which interacts with the active-site aspartates, is essential for folding and activation of the zymogen. Lamb prochymosin has been shown by cDNA cloning to possess glutamic acid at 36p. To investigate the effect of this natural mutation which appears to contradict the proposed role of this residue, calf and lamb prochymosins and their two reciprocal mutants, K36pE and E36pK, respectively, were expressed in Escherichia coli, refolded in vitro, and autoactivated at pH 2 and 4.7. All four zymogens could be activated to active chymosin and, at both pH values, the two proteins with Glu36p showed higher activation rates than the two Lys36p forms. Glu36p was also demonstrated in natural prochymosin isolated from the fourth stomach of lamb, as well as being encoded in the genomes of sheep, goat and mouflon, which belong to the subfamily Caprinae. A conserved basic residue at position 36p of prochymosin is thus not obligatory for its folding or autocatalytic activation. The apparently contradictory results for porcine pepsinogen A [Richter, C., Tanaka, T., Koseki, T. & Yada, R.Y. (1999) Eur. J. Biochem. 261, 746-752] can be reconciled with those for prochymosin. Lys/Arg36p is involved in stabilizing the propeptide-enzyme interaction, along with residues nearer the N-terminus of the propeptide, the sequence of which varies between species. The relative contribution of residue 36p to stability differs between pepsinogen and prochymosin, being larger in the former.

The impact of internal parasites on the productivity of young cattle organically reared on semi-natural pastures in Sweden.

A grazing experiment with young cattle was conducted over two consecutive (1997, 1998) grazing seasons on semi-natural pasturelands in central-eastern Sweden. Comparisons were made between groups of animals that were either untreated and set-stocked, ivermectin bolus treated and set-stocked or untreated but moved in mid-summer (mid-July) to ungrazed pasture. The whole experimental area had remained virtually free of cattle during the previous two seasons and the cattle had been raised indoors since birth. To introduce low-levels of parasite infection into the experimental system, each animal received a 'priming dose' of approximately 10, 000 infective trichostrongylid larvae at the time of turnout for both years. Results of the first year study showed that the level of parasitism was so low that it failed to induce any productivity losses in both groups of untreated cattle, which grew as well as those given boluses at turnout. In contrast, in 1998 both groups of untreated cattle suffered varying degrees of sub-clinical and clinical parasitism to result in an average of 30kg liveweight depression, compared with the bolus treated cattle, at the end of the season. The only major departure between the two years was that in the latter, the cattle in the untreated groups were exposed to infective larval pickup, which had overwintered on pasture. Cattle in the move treatment grazed in the same sequence on pastures used by similar classes of animals during the previous year. That is, their pastures at turnout had not been grazed since mid-summer of the previous year. Clearly this early season (1997) grazing by young cattle resulted in sufficient overwintered larvae at the start of the following year (1998) to cause productivity losses of the same magnitude as those recorded for young cattle grazing on pastures contaminated for the entire grazing season of the previous year. This was confirmed by tracer tests that were carried out on all treatments, at the time of turnout and the mid-summer move in 1999. These results have major significance to organic cattle producers in Sweden who have a much higher tendency to practice a variety of grazing management techniques aimed at controlling nematode parasite infections in young cattle, than their conventional farming colleagues. It has been identified that one of these strategies is to simply use summer/autumn saved pastures for young stock at turnout, which if grazed by young stock prior to this, could prove to be counter-productive.

The pepsin residue glycine-76 contributes to active-site loop flexibility and participates in catalysis.

Glycine residues are known to contribute to conformational flexibility of polypeptide chains, and have been found to contribute to flexibility of some loops associated with enzymic catalysis. A comparison of porcine pepsin in zymogen, mature and inhibited forms revealed that a loop (a flap), consisting of residues 71--80, located near the active site changed its position upon substrate binding. The loop residue, glycine-76, has been implicated in the catalytic process and thought to participate in a hydrogen-bond network aligning the substrate. This study investigated the role of glycine-76 using site-directed mutagenesis. Three mutants, G76A, G76V and G76S, were constructed to increase conformational restriction of a polypeptide chain. In addition, the serine mutant introduced a hydrogen-bonding potential at position 76 similar to that observed in human renin. All the mutants, regardless of amino acid size and polarity, had lower catalytic efficiency and activated more slowly than the wild-type enzyme. The slower activation process was associated directly with altered proteolytic activity. Consequently, it was proposed that a proteolytic cleavage represents a limiting step of the activation process. Lower catalytic efficiency of the mutants was explained as a decrease in the flap flexibility and, therefore, a different pattern of hydrogen bonds responsible for substrate alignment and flap conformation. The results demonstrated that flap flexibility is essential for efficient catalytic and activation processes.

Isolation and characterization of pepsinogen from Trimeresurus flavoviridis (Habu snake).

Pepsinogen was isolated from the gastric mucosa of Trimeresurus flavoviridis (Habu snake) by DEAE-cellulose and DEAE-Sepharose ion-exchange chromatographies, and Sephacryl S-200 gel-chromatography. The yield calculated from the crude extract was 29% with 6.2-fold purification. The purified pepsinogen gave a single band on both native- and SDS-PAGE. As no other active enzyme was detected on the chromatographies, it was concluded that the Habu snake has one major pepsinogen. The molecular mass of the pepsinogen was estimated to be 38 kDa by SDS-PAGE. The sequence of the N-terminal 26 amino acid residues was determined and compared with those of other pepsinogens. The N-terminal structure of Habu snake pepsinogen was more homologous with those of mammalian pepsinogens C than those of mammalian pepsinogens A. The pepsinogen was rapidly converted to pepsin by way of an intermediate form induced by acidification. The optimum pH of Habu snake pepsin for bovine hemoglobin was 1.5-2.0, and it retained full activity at pH 6.2 and 30 degrees C on incubation for 30 min. The optimum temperature for the snake pepsin was 50 degrees C and it was stable at 40 degrees C on incubation for 10 min. The proteolytic activity of the pepsin toward bovine hemoglobin was about two times higher than that of porcine pepsin A, however, the activity toward oxidized bovine insulin B-chain was lower than that of porcine pepsin A, and it did not hydrolyze oligopeptides. The specificity for oxidized bovine insulin B-chain of the pepsin was different from that of porcine pepsin A. Habu snake pepsin was inhibited by pepstatin A but not by serine, cysteine, or metallo protease inhibitors.