• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. May 2002; 68(5): 2453–2460.
PMCID: PMC127571

Identification of Novel Hexapeptides Bioactive against Phytopathogenic Fungi through Screening of a Synthetic Peptide Combinatorial Library

Abstract

The purpose of the present study was to improve the antifungal activity against selected phytopathogenic fungi of the previously identified hexapeptide PAF19. We describe some properties of a set of novel synthetic hexapeptides whose d-amino acid sequences were obtained through screening of a synthetic peptide combinatorial library in a positional scanning format. As a result of the screening, 12 putative bioactive peptides were identified, synthesized, and assayed. The peptides PAF26 (Ac-rkkwfw-NH2), PAF32 (Ac-rkwhfw-NH2), and PAF34 (Ac-rkwlfw-NH2) showed stronger activity than PAF19 against isolates of Penicillium digitatum, Penicillium italicum, and Botrytis cinerea. PAF26 and PAF32, but not PAF34, were also active against Fusarium oxysporum. Penicillium expansum was less susceptible to all four PAF peptides, and only PAF34 showed weak activity against it. Assays were also conducted on nontarget organisms, and PAF26 and PAF32 showed much-reduced toxicity to Escherichia coli and Saccharomyces cerevisiae, demonstrating selectivity towards certain filamentous fungi. Thus, the data showed distinct activity profiles for peptides differentiated by just one or two residue substitutions. Our conclusion from this observation is that a specificity factor is involved in the activity of these short peptides. Furthermore, PAF26 and PAF32 displayed activities against P. digitatum, P. italicum, and B. cinerea similar to that of the hemolytic 26-amino acid melittin, but they did not show the high toxicity of melittin towards bacteria and yeasts. The four peptides acted additively, with no synergistic interactions among them, and PAF26 was shown to have improved activity over PAF19 in in vivo orange fruit decay experiments.

Effective control of fungal plant pathogens is achieved by application of fungicides. Currently available antifungal compounds raise human health and environmental safety concerns that have resulted in the withdrawal of some of them. In addition, resistance problems have been encountered and could result in loss of effectiveness of fungicides (17, 22). New agents, which could substitute for or complement available fungicides, should therefore have properties that differentiate them from existing compounds.

Antimicrobial peptides (antifungals and antibacterials) have been isolated from very diverse organisms (2, 14, 16), and a number of them have received attention because of their low toxicity against mammalian cells (35). In plants, there are examples that relate the production of antimicrobial peptides to host defense and pathogen virulence (24, 36). In this scenario, the use of antimicrobial peptides in agriculture has been proposed (13, 26, 31, 36, 39). We have considered extending such use to protection of fruits and vegetables against fungal infection during postharvest handling and storage (23), an application that does not necessarily rely on the production of transgenic plants.

Natural defense peptides are limited in their practical use because of their levels of activity and poor bioavailabilities. Moreover, a lack of correlation between peptide sequences and specific activities makes it difficult to predict their properties against desired pathogens. A challenge ahead is the design of potent microbicidal peptides that have in vivo specific activities against selected pathogens. Engineering of natural peptides has resulted in improved antimicrobial activities against plant pathogens (8, 10, 25, 30, 42). Importantly, some designed peptides recently have been demonstrated to be active in vivo in transgenic plants (7, 21, 27). However, the process involved in the development of such engineered peptides, i.e., their isolation and characterization and the synthesis of analogs required for activity optimization, is time-consuming and limited by the number of individual peptides which can be generated and screened. The diversity available through combinatorial chemistry provides a powerful tool to identify peptides with potentially improved or new properties (4, 5). Synthetic combinatorial libraries (SCLs) allow the rapid and systematic examination of millions of novel peptides. Indeed, an SCL in an iterative format has been screened for the identification of peptides with in vitro activity against phytopathogenic fungi (32).

We are interested in the identification of short peptides with specific activity against (postharvest) fungal plant pathogens. We have previously demonstrated that a hexapeptide called 19 (PAF19 in this report) shows activity against fungi that cause postharvest decay in fruits (23). The objective of the present work was the improvement of the properties of PAF19 through combinatorial chemistry. We present data on the use of a soluble SCL for the identification of individual peptides with specific antifungal activities. An all-d-amino acid hexapeptide SCL in a positional scanning (PS) format was screened for in vitro growth inhibition of the citrus green mold fungus Penicillium digitatum. Several peptides were identified and were shown to have different activity profiles against selected fungi.

MATERIALS AND METHODS

Microbial strains and isolates.

The fungi used in this study were P. digitatum isolate PHI-26, Penicillium italicum isolate PHI-1, and Penicillium expansum isolate CMP-1 (23); Botryotinia fuckeliana CECT2100 (anamorph Botrytis cinerea) (Colección Española de Cultivos Tipo [Spanish National Collection of Type Cultures]), and Fusarium oxysporum f. sp. lycopersici CECT2866. All fungi were cultured on potato dextrose agar (Difco, Detroit, Mich.) plates for 7 to 10 days at 24°C. Conidia were collected from the agar and transferred to distilled water, filtered, titrated with a hemocytometer, adjusted to the appropriate concentration, and used in the antifungal assays.

Control experiments used laboratory strains of Escherichia coli (DH5α) and Saccharomyces cerevisiae (W303-1A). S. cerevisiae was grown in YPD medium (1% yeast extract, 1.5% peptone, 2% dextrose) at 24°C and E. coli was grown in Luria-Bertani medium at 37°C; cultures were grown to stationary phase, diluted to the appropriate concentration, and used in the antimicrobial assays.

Synthesis of peptide libraries and individual peptides.

The peptide library and individual peptides were synthesized by solid-phase methods and simultaneous multiple peptide synthesis using N-(9-fluorenyl) methoxycarbonyl chemistry (12). The all-d-amino acid hexapeptide SCL was synthesized in the so-called PS format as previously described (29) (see Fig. Fig.11 and Results). The mixture (X) positions were incorporated by coupling a mixture of 19 d-amino acids (cysteine was omitted) with the relative ratios suitably adjusted to yield close to equimolar incorporation. Peptide mixtures were solubilized in 5% dimethyl sulfoxide, aliquoted at a final concentration of 3 mg/ml, and stored at −20°C.

FIG. 1.
Antifungal activity of the all-d-amino acid hexapeptide PS-SCL against P. digitatum PHI-26. Each graph represents the activity of each sublibrary, wherein each position of the peptide is defined by the d-amino acid labeled on the x axis. Individual bars ...

Individual peptides were purified by preparative reversed-phase high-pressure liquid chromatography with a Waters Delta Pak C18 column. Matrix-assisted laser desorption ionization-time-of-flight mass spectrometry was used to confirm peptide identity. Stock solutions of each peptide were prepared at a 1 mM concentration in 5 mM 3-(N-morpholino)-propanesulfonic acid (pH 7.0) buffer. Peptide concentrations were determined by measuring the absorbance at 280 nm. After their selection in the course of this study, peptides PAF26, PAF32, and PAF34 were purchased from an external company (DiverDrugs, Barcelona, Spain) to confirm their properties from independent chemical syntheses.

In vitro antimicrobial activity assays.

The in vitro antimicrobial activities of the peptides were determined using a microtiter plate assay as previously described (8, 23). Growth was determined by measuring absorbance at 492 nm (A492) in a Titertek Multiskan Plus microplate reader (Labsystems, Helsinki, Finland). In all of the experiments, three replicates were prepared for each treatment; the blank mean A492 value from one row of mock inoculations was subtracted from each well's A492 measurement, and the mean and standard deviation (SD) were then calculated for each treatment.

In the assay of the peptide library, 100 μl of potato dextrose broth (Difco) containing 0.003% (wt/vol) chloramphenicol was combined with 50 μl of potato dextrose broth containing 105 P. digitatum PHI-26 conidia/ml and 50 μl of peptide mixtures from the PS-SCL, added from 4× stock solutions. Two concentrations of peptide mixtures were assayed: 200 and 400 μg/ml. Control experiments were performed to confirm that the presence of chloramphenicol did not significantly affect fungal growth under our assay conditions.

The assay of sequence-defined peptides was essentially the same, and peptides were added from stocks to reach final concentrations of 5, 10, 20, 40, 60, 80, and 160 μM. For in vitro yeast and bacteria growth, the inoculum was 50 μl of a freshly diluted overnight culture (A600 of 0.01 for yeast and A600 of 0.001 for bacteria) brought to 180 μl of the appropriate medium and subsequent addition of the peptides from stocks. For bacterial assays, the plates were incubated at 37°C.

The MIC was defined as the lowest peptide concentration that showed no growth at the end of the experiment (usually after 3 days of incubation). The IC50 was defined as the concentration required to obtain 50% inhibition and was calculated from the experimental values of inhibition at different peptide concentrations and after an incubation period at which 50% of final growth was reached in the controls (ca. 40 to 48 h for fungi and 20 to 24 h for bacteria and yeasts).

Fruit decay test.

Freshly harvested oranges (Citrus sinensis L.) were surface sterilized in a 5% commercial bleach solution for 5 min, followed by extensive washing in tap water. The fruits were air dried and wounded by making punctures at four sites around the equator. Suspensions containing 105 conidia/ml were mixed with peptides (at desired final concentrations) and 10 μl of the solution was applied onto each wound. For each treatment three replicas (five fruits per replica, four wounds per fruit) were inoculated. Fruits were maintained at 20°C and 85% relative humidity and scored daily. Mock inoculations without fungus did not produce a significant percentage of rot symptoms. Statistical analyses were carried out with the software package StatGraphics Plus 4.0 (Manugistics Inc., Rockville, Md.).

RESULTS

Screening of a PS-SCL for in vitro growth inhibition of P. digitatum.

The PS-SCL consisted of six sublibraries; in each sublibrary, a single position of the hexapeptide was defined with each one of the 19 d-amino acids (i.e., 19 peptide mixtures/sublibrary; cysteine was omitted), while the remaining positions were composed of mixtures of all d-amino acids except cysteine. Thus, the library consisted of 114 peptide mixtures and is represented as Ac-O1XXXXX-NH2, Ac-XO2XXXX-NH2, Ac-XXO3XXX-NH2, Ac-XXXO4XX-NH2, Ac-XXXXO5X-NH2, and Ac-XXXXXO6-NH2 (O, defined by one of 19 d-amino acids; X, close-to-equimolar mixture of 19 d-amino acids) (5, 29) (Fig. (Fig.1).1). Each mixture contained approximately 2.5 × 106 (195) different peptides. Each sublibrary provides information about the most important amino acid residue at each position. Such segregation permits information to be obtained in a single assay, which is then used to synthesize individual peptides directly from the screening results.

The PS-SCL was tested for in vitro growth inhibition of P. digitatum PHI-26. The percentages of inhibition were then calculated (Fig. (Fig.1)1) (0% inhibition indicates growth equal to the control sample). A number of mixtures exhibited significant inhibition, and clear differences were observed between the most and least inhibitory, suggesting that activity is due to an individual peptide(s) present in the mixture. The so-called deconvolution process of the library consisted, first, in selecting the putative relevant amino acid residues for each position. In our case, the greatest differences in activity between mixtures were observed within positions 1 and 2, while very similar (and significant) activities were observed among the more active mixtures of position 4. The latter indicates that no clear-cut assignment could be made at position 4 of the peptide. The initially selected residues were arginine, tryptophan, and lysine at positions 1 and 2; tryptophan, arginine, lysine, leucine, phenylalanine, and histidine at position 3; tryptophan, arginine, phenylalanine, histidine, isoleucine, lysine, and leucine at position 4; glycine, arginine, tryptophan, lysine, phenylalanine, and leucine at position 5; and arginine, tryptophan, lysine, phenylalanine, and histidine at position 6. In general, positively charged residues (arginine or lysine) and residues with aromatic rings (tryptophan or phenylalanine) rendered antifungal activity, as already observed in assays of SCLs against different microorganisms (4, 5, 18, 32).

The activity against P. digitatum of the more effective peptide mixtures was reproduced in a second experiment (data not shown). These active mixtures were also assayed with the gram-negative bacterium E. coli DH5α in order to discard those mixtures toxic to a nontarget organism and to introduce specificity to the selection. Most of the antifungal mixtures were also antibacterial (data not shown), with a predominance of basic and aromatic residues as previously described (5). Of particular note was the high antibacterial activity of the mixture corresponding to glycine at position 5. Given these data, an alternative approach was taken as follows.

Selection of amino acid residues.

We had earlier identified a synthetic hexapeptide (Ac-rktwfw-NH2; PAF19) with activity against selected fungi that cause postharvest decay (23). PAF19 was initially synthesized in a search of an SCL for antivirals (28). Paradoxically, the mixture corresponding to threonine at position 3 (as in PAF19) was not antifungal, implying that the amino acid sequence of PAF19 could not be identified from the assay described here (Fig. (Fig.1).1). Our approach was thus redirected to consider PAF19 as a lead compound and to use the data from the assay of the PS-SCL to improve its activity in terms of both potency and specificity.

Positions 1, 2, 5, and 6 of PAF19 were conserved, whereas positions 3 and 4 were combinatorialized. Residues selected for substitutions have chemical properties representative of the active mixtures. A restraint was imposed in that the number of tryptophan residues of the peptide was fixed to two, given the structure of the antimicrobial center of lactoferricin B (33), a hexapeptide with analogy to PAF19 (23). The residues used in both positions (3 and 4) were arginine and lysine (as positively charged residues), phenylalanine, leucine (as an example of the hydrophobic valine and isoleucine), and methionine (due to its intermediate activity in both positions). In addition, glutamine was used only in position 3, given the intermediate activity of the polar glutamine and asparagine in that sublibrary, and histidine was only in position 4.

As a result of our focused deconvolution process, we selected 12 d-amino acid hexapeptide sequences with putative antifungal activity (Table (Table1,1, hexapeptides 25 to 36); a first series (from peptides 25 to 30) where position 3 of hexapeptide PAF19 was replaced, and a second one (from peptides 31 to 36) where position 3 was changed to tryptophan and position 4 was replaced with the amino acids described above.

TABLE 1.
Amino acid sequences of peptidesa

Activities of the defined hexapeptides.

The set of 12 individual hexapeptides (Table (Table1)1) was synthesized, purified, and assayed against P. digitatum and E. coli (Fig. (Fig.2).2). In these experiments, we used PAF19 as a reference control and the hexapeptide P20 (in which a substitution in PAF19 of Trp for Pro at position 4 abolished its activity) (23) as a negative control. As predicted, given the lack of antibacterial activity of the lead compound PAF19, much higher overall activity was obtained against the fungus.

FIG. 2.
Antimicrobial activity against either P. digitatum PHI-26 (grey bars) or E. coli DH5α (white bars) of the defined peptides, labeled on the x axis. Each graph represents the percentage of in vitro growth at 10 (A), 20 (B), or 80 (C) μM ...

We chose those peptides that exhibited strong activity against P. digitatum combined with low toxicity to E. coli. By doing so, our selection criterion was directed to specificity once the activity requirement was fulfilled. The hexapeptides selected (and renamed) for further analyses were PAF26 (Ac-rkkwfw-NH2), PAF32 (Ac-rkwhrw-NH2), and PAF34 (Ac-rkwlfw-NH2) (Table (Table1).1). Their MIC and IC50 values for P. digitatum were determined (Table (Table2),2), based on inhibition responses to increasing peptide concentrations and growth over time (see representative experiments in Fig. Fig.33 and and4),4), and showed an improved activity over PAF19, thus fulfilling one of the objectives of the present study. Peptide concentrations below the MIC did not prevent sporulation after very long incubation times (data not shown).

FIG. 3.
Activity of hexapeptides P20 (black circles), PAF19 (white circles), PAF26 (white diamonds), PAF32 (white triangles), or PAF34 (white squares) on the in vitro growth of P. digitatum PHI-26 (A), P. expansum CMP-1 (B), or F. oxysporum CECT2866 (C). Data ...
FIG. 4.
Activity of hexapeptides on the in vitro growth of F. oxysporum CECT2866 (A) or S. cerevisiae W303-1A (B). Data are shown as the mean (± SD) A492 over time (in hours), in the absence (black circles) or presence of a 160 μM concentration ...
TABLE 2.
Effect of selected synthetic peptides on the in vitro growth of fungi, bacteria, and yeast

Antimicrobial compounds may interact synergistically, additively, or antagonistically. An assay with P. digitatum was carried out following the Abbot approach, as described in reference 15, and demonstrated additive interactions between PAF19, PAF26, PAF32, and PAF34 (data not shown). This additive mode of action has been shown for other natural (34) and synthetic (1) peptides and suggests that the peptides act on the same (or very similar) cellular process(es). Not surprisingly, given the lack of synergism among peptides, it was found that PAF26 inhibited conidial germination (data not shown), as shown previously for PAF19 (23).

Activity profiles of hexapeptides PAF26, PAF32, and PAF34.

The antifungal assays were extended to test the peptides against other phytopathogenic fungi, such as two other Penicillium pathogens of fruits, P. italicum and P. expansum, and against B. cinerea and F. oxysporum (Table (Table22).

The MIC and IC50 values of PAF26, PAF32, and PAF34 against P. italicum were also lower than those of PAF19 (Table (Table2),2), confirming their higher potencies. Also, PAF26, PAF32, and PAF34 were more active than PAF19 against B. cinerea (Table (Table2),2), although the improvement was less marked in this case, especially for PAF26 and PAF34, which suggests a certain degree of peptide specificity. This suggestion was strengthened by the distinct behaviors against some fungi (Table (Table2;2; Fig. Fig.33 and and4).4). For instance, PAF26, PAF32, and PAF34 showed similar activities against P. digitatum (high activity; Fig. Fig.3A)3A) and P. expansum (very low activity; Fig. Fig.3B),3B), while the effect of PAF34 on F. oxysporum was markedly different from that of PAF26 and PAF32 (Fig. (Fig.3C3C).

An isolate of P. italicum (CECT2294) that was previously found to be resistant to PAF19 (23) was also insensitive to the hexapeptides PAF26 and PAF32 (data not shown). Indirectly, this result would also suggest that all four peptides disrupt similar cellular function(s), as shown above by their lack of synergism.

In addition to E. coli, we selected a laboratory strain of the yeast S. cerevisiae for testing nonspecific toxicity. At 24 h of growth at concentrations above 80 μM, some inhibition was observed for PAF26 (Table (Table22 and Fig. Fig.4B),4B), and a significant activity was observed for PAF32 and PAF34 (Table (Table2).2). After incubation for longer times, however, it was observed that PAF26 (Fig. (Fig.4B)4B) had delayed, but not inhibited, the growth of S. cerevisiae and that its effect at later time points was not distinguishable from that of P20. On the other hand, PAF34 at 160 μM remained as inhibitory (Fig. (Fig.4B).4B). The reversal sensitivity of S. cerevisiae and F. oxysporum to PAF26 and PAF34 (Fig. (Fig.4)4) reinforced the different activity profiles of PAFs, depending of the target microorganism, and the putative involvement of specificity factor(s) in their action.

In vivo activity of PAF26.

Previously, controlled inoculations showed that the in vitro properties of PAF19 correlated with a retard in the decay caused by fungal infection (23). We used the same bioassay to compare the activities of PAF19 and PAF26 (Fig. (Fig.5).5). At 3 days postinoculation (dpi), 100% of the wounds were infected in the control, 73% were infected in the presence of PAF19, and only 38% were infected in the presence of PAF26. However, all of the wounds became infected by later time points--at 5 and 6 dpi in the presence of PAF19 and PAF26, respectively (Fig. (Fig.5).5). The limited effectiveness of PAFs in preventing decay was hypothesized to be due to a few mycelium tips moving ahead of the inoculation site to infect surrounding tissue (where peptides are absent) (see also reference 23). It is worth mentioning that thiabendazole (a commercial postharvest fungicide) did not delay progression of disease when used in parallel inoculations at a concentration equal to five times its in vitro MIC (B. López-García and J. F. Marcos, unpublished data), and thus it performed worse than PAFs. The lower incidence of infection in the presence of PAF26 correlated with a smaller size of the infected area (data not shown). The data confirm that PAF26 is more potent than PAF19 on fruit tissue, as occurs in in vitro growth assays.

FIG. 5.
Effect of PAF19 and PAF26 on orange fruit decay. Fruits were infected with P. digitatum PHI-26 either alone (black circles) or in the presence of a 100 μM concentration of PAF19 (white circles) or PAF26 (white triangles). Data are shown as the ...

DISCUSSION

In this study, we have identified a set of novel antifungal synthetic peptides derived from the combinatorial optimization of the previously described lead compound, PAF19 (23). The all-d-amino acid hexapeptides PAF26 (Ac-rkkwfw-NH2), PAF32 (Ac-rlwjfw-NH2), and PAF34 (Ac-rkwlfw-NH2) were shown to be more active than PAF19 against a group of fungal phytopathogens (Table (Table22 and Fig. Fig.3).3). In fact, some of the newly described peptides were as potent in vitro as the natural toxic peptide melittin but, very remarkably, they did not share with melittin its high nonspecific toxicity to E. coli and S. cerevisiae (Table (Table2;2; compare the IC50 and MIC values of PAF32 and melittin for all of the microorganisms tested).

The last observation points to the need in assays to discard nondesired toxic activities and select for specificity rather than mere antimicrobial potency. For instance, peptides 25 and 31 showed the strongest activities against P. digitatum, but they were initially discarded after the assay with E. coli because they were also toxic to the bacterium at 80 μM (Fig. (Fig.2).2). In fact, peptide 31 (Ac-rkwffw-NH2) showed a noteworthy >95% inhibition of P. digitatum at 10 μM (Fig. (Fig.2A)2A) and thus could be considered a potent antifungal. PAF26, PAF32, and PAF34 were selected and further studied. The additional studies advise selection of PAF26 over PAF32 and PAF34 due to the higher residual activity of the latter against the yeast (Fig. (Fig.4B4B and data not shown). Remarkably, PAF26 had improved activity (compared to PAF19) in protection against decay caused on orange fruits (Fig. (Fig.5),5), thus paralleling the in vitro data.

We demonstrated that PAF19 synthesized with the l-enantiomers of amino acids is also active against P. digitatum (23). Plant defense peptides produced in transgenic plants through biotechnology confer protection against disease (6, 13, 26, 36). Ongoing research is exploring the use of genetic transformation of microorganisms for the production and delivery of PAFs. Due to their short length, strategies such as fusion to (or cleavage from) longer polypeptides will be evaluated. Once produced, their use in combination with other control approaches could be tested. Indeed, the peptide nisin acts synergistically with biocontrol yeasts to reduce postharvest decay of apples (11). A necessary property of antimicrobial peptides if they are to be used in conjunction with, or be produced by, biocontrol microorganisms is their lack of nonspecific toxicity. The absence of toxicity of PAF19 and PAF26 to bacteria and yeasts (this work and also reference 23) and their sequence similarities with the antimicrobial center of bovine lactoferrin (a natural protein from bovine milk) (23, 33) look promising for their development as lead compounds for plant protection.

Soluble nonsupport-bound SCLs have the advantage that they can be used in complex cell-growth-based assays for microbicidal compounds. An additional advantage of an SCL in a PS format (29), over the iterative format (19), is that it can be used for the very rapid (i.e., a single-day experiment) identification of peptides. It should be noted that the use of the alternative iterative format led to a hexapeptide (i.e., peptide 66-10, Ac-frlkfh-NH2, in reference 32) different from those described here. This could be explained partly by the different fungus used in the primary assay of the SCL, considering that we showed some specificity in PAF activity. For instance, phenylalanine at position 1 (as occurs in peptide 66-10) was not identified in our assay, and none of our peptides were as potent as 66-10 against F. oxysporum (one of the fungi used in the iterative screening). However, the sequences of the inhibitors identified here are similar to those obtained using the iterative process (32) in that they contain positively charged and hydrophobic residues. Similar results were also reported through the screening of iterative or positional SCLs for melittin inhibitors: the sequences obtained were similar in nature, but not identical (3).

Our work confirms and extends previous data on how minor amino acid changes affect antimicrobial properties (4, 18), and it also allows some conclusions on sequence-activity relationships. Antimicrobial activity does not merely result from the addition of cationic charge and hydrophobicity properties (23), and arginine residues seemed to confer broader and higher toxicity than did lysine residues (Fig. (Fig.2,2, compare equal activities against P. digitatum in the couples 26 and 29 and 33 and 36, and also higher activity against E. coli of 29 and 36). It also seems that there is some permissibility to residue permutation since, for instance, peptides 25 (Ac-rkfwfw-NH2) and 31 (Ac-rkwffw-NH2) share very similar antimicrobial properties.

Our data show distinct activity profiles for peptides differentiated by just one or two residues. A paradigm example is the reversal activities of PAF26 and PAF34 against some microorganisms (Fig. (Fig.4).4). Another example is PAF32 and PAF34, with a single histidine-to-leucine substitution involved: PAF32 shows activity against F. oxysporum and very low activity against P. expansum and S. cerevisiae (Table (Table2),2), whereas PAF34 is inactive against F. oxysporum (Fig. (Fig.3)3) and is the most active of all the peptides tested against P. expansum and S. cerevisiae (Fig. (Fig.4).4). These data clearly go beyond the previously described inactivation of PAF19 by a Trp-to-Pro substitution at position 4 (23). Our conclusion from these observations is that a factor conferring specificity would be involved in the mode of action of these peptides. It is anticipated that combinations of antimicrobial peptides with different specificities will provide protection towards distinct pathogens (39).

Cationic antimicrobial peptides with an α-helical (magainins, cecropins, or melittin) or β-sheet (defensins) structure are known to interact with cell membranes, and their activity profiles correlate with variations in the structure and complexity of the membranes (16, 20, 38). Upon interaction, some of these peptides would form pores, which disrupt membrane function and cell viability (41). However, it is difficult to fit such a membrane-spanning mechanism with the size of very short antimicrobial peptides. Also, an interaction with biological membranes is unlikely to be mediated by chiral receptors, as it is well documented that d- and l-enantiomers of antimicrobial peptides are equally active (40), a property also demonstrated for PAF19 (23). The phospholipid composition of artificial liposomes has been related to synthetic peptide folding properties and antimicrobial activity (4, 20) and leakage of dyes (18). It is noteworthy in this regard that the so-called plant lipid transfer proteins, which bind (phospho)lipids and locate in the cell wall, are recognized as one of the families of natural antimicrobial peptides in plants (14). Recently, it has been shown that a gene encoding an enzyme involved in the synthesis of a particular sphingolipid confers sensitivity in S. cerevisiae to a plant antimicrobial peptide (a defensin) (37). Moreover, the synthetic antifungal peptide D4E1 was shown complexed with ergosterol, a sterol present in conidial walls of phytopathogenic fungi (9). It is tempting to speculate that subtle lipid composition differences of fungal membranes (or conidial walls) are responsible for the different activity and specificity profiles of the synthetic peptides reported in this work. The occurrence of P. italicum isolates (23) which are not sensitive to PAFs will be a useful tool to test and further study such a hypothesis as a way to characterize the antimicrobial action of short synthetic peptides and improve their efficacy. Obviously, another alternative to improvement could be the use of the most promising peptides described here (for instance, PAF26) as lead compounds in a further iterative process.

Acknowledgments

Belén López-García is the recipient of a predoctoral fellowship from Generalitat Valenciana (Spain). This work was supported by grant BIO4-CT97-2086 from the European Union.

The synthesis of the peptide library was carried out with the collaboration of T. Carbonell (Departament de Bioquímica i Biologia Molecular, Universitat de València). We acknowledge A. Izquierdo for her excellent technical assistance, A. Veyrat and L. González-Candelas for their helpful comments during the course of this investigation, and L. González-Candelas for critical reading of the manuscript.

REFERENCES

1. Ali, G. S., and A. S. Reddy. 2000. Inhibition of fungal and bacterial plant pathogens by synthetic peptides: in vitro growth inhibition, interaction between peptides and inhibition of disease progression. Mol. Plant-Microbe Interact. 13:847-859. [PubMed]
2. Andreu, D., and L. Rivas. 1998. Animal antimicrobial peptides: an overview. Biopolymers 47:415-433. [PubMed]
3. Blondelle, S. E., R. A. Houghten, and E. Pérez-Payá. 1996. Identification of inhibitors of melittin using nonsupport-bound combinatorial libraries. J. Biol. Chem. 271:4093-4099. [PubMed]
4. Blondelle, S. E., and K. Lohner. 2000. Combinatorial libraries: a tool to design antimicrobial and antifungal peptide analogues having lytic specificities for structure-activity relationship studies. Biopolymers 55:74-87. [PubMed]
5. Blondelle, S. E., E. Pérez-Payá, and R. A. Houghten. 1996. Synthetic combinatorial libraries: novel discovery strategy for identification of antimicrobial agents. Antimicrob. Agents Chemother. 40:1067-1071. [PMC free article] [PubMed]
6. Carmona, M. J., A. Molina, J. A. Fernández, J. J. López-Fando, and F. García-Olmedo. 1993. Expression of the alpha-thionin gene from barley in tobacco confers enhanced resistance to bacterial pathogens. Plant J. 3:457-462. [PubMed]
7. Cary, J. W., K. Rajasekaran, J. M. Jaynes, and T. E. Cleveland. 2000. Transgenic expression of a gene encoding a synthetic antimicrobial peptide results in inhibition of fungal growth in vitro and in planta. Plant Sci. 154:171-181. [PubMed]
8. Cavallarín, L., D. Andreu, and B. San Segundo. 1998. Cecropin A-derived peptides are potent inhibitors of fungal plant pathogens. Mol. Plant-Microbe Interact. 11:218-227. [PubMed]
9. De Lucca, A. J., J. M. Bland, C. Grimm, T. J. Jacks, J. W. Cary, J. M. Jaynes, T. E. Cleveland, and T. J. Walsh. 1998. Fungicidal properties, sterol binding, and proteolytic resistance of the synthetic peptide D4E1. Can. J. Microbiol. 44:514-520. [PubMed]
10. De Samblanx, G. W., I. J. Goderis, K. Thevissen, R. Raemaekers, F. Fant, F. Borremans, D. P. Acland, R. W. Osborn, S. Patel, and W. F. Broekaert. 1997. Mutational analysis of a plant defensin from radish (Raphanus sativus L.) reveals two adjacent sites important for antifungal activity. J. Biol. Chem. 272:1171-1179. [PubMed]
11. El-Neshawy, S. M., and C. L. Wilson. 1997. Nisin enhancement of biocontrol of postharvest diseases of apple with Candida oleophila. Postharvest Biol. Technol. 10:9-14.
12. Fields, G. B., and R. L. Noble. 1990. Solid phase peptide synthesis utilizing 9-fluorenylmethoxycarbonyl amino acids. Int. J. Pept. Protein Res. 34:161-214. [PubMed]
13. Gao, A. G., S. M. Hakimi, C. A. Mittanck, Y. Wu, B. M. Woerner, D. M. Stark, D. M. Shah, J. Liang, and C. M. Rommens. 2000. Fungal pathogen protection in potato by expression of a plant defensin peptide. Nat. Biotechnol. 18:1307-1310. [PubMed]
14. García-Olmedo, F., A. Molina, J. M. Alamillo, and P. Rodríguez-Palenzuela. 1998. Plant defense peptides. Biopolymers 47:479-491. [PubMed]
15. Gisi, U. 1996. Synergistic interaction of fungicides in mixtures. Phytopathology 86:1273-1279.
16. Hancock, R. E. W., and D. S. Chapple. 1999. Peptide antibiotics. Antimicrob. Agents Chemother. 43:1317-1323. [PMC free article] [PubMed]
17. Holmes, G. J., and J. W. Eckert. 1999. Sensitivity of Penicillium digitatum and P. italicum to postharvest citrus fungicides in California. Phytopathology 89:716-721. [PubMed]
18. Hong, S. Y., J. E. Oh, M. Y. Kwon, M. J. Choi, J. H. Lee, B. L. Lee, H. M. Moon, and K. H. Lee. 1998. Identification and characterization of novel antimicrobial decapeptides generated by combinatorial chemistry. Antimicrob. Agents Chemother. 42:2534-2541. [PMC free article] [PubMed]
19. Houghten, R. A., J. R. Appel, S. E. Blondelle, J. H. Cuervo, C. T. Dooley, and C. Pinilla. 1992. The use of synthetic peptide combinatorial libraries for the identification of bioactive peptides. BioTechniques 13:412-421. [PubMed]
20. Huang, H. W. 2000. Action of antimicrobial peptides: two-state model. Biochemistry 39:8347-8352. [PubMed]
21. Huang, Y., R. O. Nordeen, M. Di, L. D. Owens, and J. H. McBeath. 1997. Expression of an engineered cecropin gene cassette in transgenic tobacco plants confers resistance to Pseudomonas syringae pv. tabaci. Phytopathology 87:494-499. [PubMed]
22. Knight, S. C., V. M. Anthony, A. M. Brady, A. J. Greenland, S. P. Heaney, D. C. Murray, K. A. Powell, M. A. Schulz, C. A. Spinks, P. A. Worthington, and D. Youle. 1997. Rationale and perspectives in the development of fungicides. Annu. Rev. Phytopathol. 35:349-372. [PubMed]
23. López-García, B., L. González-Candelas, E. Pérez-Payá, and J. F. Marcos. 2000. Identification and characterization of a hexapeptide with activity against phytopathogenic fungi that cause postharvest decay in fruits. Mol. Plant-Microbe Interact. 13:837-846. [PubMed]
24. López-Solanilla, E., F. García-Olmedo, and P. Rodríguez-Palenzuela. 1998. Inactivation of the sapA to sapF locus of Erwinia chrysanthemi reveals common features in plant and animal bacterial pathogenesis. Plant Cell 10:917-924. [PMC free article] [PubMed]
25. Marcos, J. F., R. N. Beachy, R. A. Houghten, S. E. Blondelle, and E. Pérez-Payá. 1995. Inhibition of a plant virus infection by analogs of melittin. Proc. Natl. Acad. Sci. USA 92:12466-12469. [PMC free article] [PubMed]
26. Molina, A., and F. García-Olmedo. 1997. Enhanced tolerance to bacterial pathogens caused by the transgenic expression of barley lipid transfer protein LTP2. Plant J. 12:669-675. [PubMed]
27. Osusky, M., G. Zhou, L. Osuska, R. E. Hancock, W. W. Kay, and S. Misra. 2000. Transgenic plants expressing cationic peptide chimeras exhibit broad-spectrum resistance to phytopathogens. Nat. Biotechnol. 18:1162-1166. [PubMed]
28. Pérez-Payá, E., B. López-García, M. Pastor, M. Vilar, and J. F. Marcos. 1999. Inhibition of plant virus infection by synthetic peptides produced by multiple and combinatorial synthesis, p. 846-847. In S. Bajusz and F. Hudecz (ed.), Peptides 1998. Proceedings of the 25th European Peptide Symposium. Akadémiai Kiadó, Budapest, Hungary.
29. Pinilla, C., J. R. Appel, P. Blanc, and R. A. Houghten. 1992. Rapid identification of high affinity peptide ligands using positional scanning synthetic peptide combinatorial libraries. BioTechniques 13:901-905. [PubMed]
30. Powell, W. A., C. M. Catranis, and C. A. Maynard. 1995. Synthetic antimicrobial peptide design. Mol. Plant-Microbe Interact. 8:792-794. [PubMed]
31. Rao, A. G. 1995. Antimicrobial peptides. Mol. Plant-Microbe Interact. 8:6-13. [PubMed]
32. Reed, J. D., D. L. Edwards, and C. F. Gonzalez. 1997. Synthetic peptide combinatorial libraries: a method for the identification of bioactive peptides against phytopathogenic fungi. Mol. Plant-Microbe Interact. 10:537-549. [PubMed]
33. Schibli, D. J., P. M. Hwang, and H. J. Vogel. 1999. The structure of the antimicrobial active center of lactoferricin B bound to sodium dodecyl sulfate micelles. FEBS Lett. 446:213-217. [PubMed]
34. Segura, A., M. Moreno, F. Madueño, A. Molina, and F. García-Olmedo. 1999. Snakin-1, a peptide from potato that is active against plant pathogens. Mol. Plant-Microbe Interact. 12:16-23. [PubMed]
35. Steiner, H., D. Andreu, and R. B. Merrifield. 1988. Binding and action of cecropin and cecropin analogues: antibacterial peptides from insects. Biochim. Biophys. Acta 939:260-266. [PubMed]
36. Terras, F. R., K. Eggermont, V. Kovaleva, N. V. Raikhel, R. W. Osborn, A. Kester, S. B. Rees, S. Torrekens, F. Van Leuven, J. Vanderleyden, B. P. Cammue, and W. F. Broekaert. 1995. Small cysteine-rich antifungal proteins from radish: their role in host defense. Plant Cell 7:573-588. [PMC free article] [PubMed]
37. Thevissen, K., B. P. Cammue, K. Lemaire, J. Winderickx, R. C. Dickson, R. L. Lester, K. K. Ferket, F. Van Even, A. H. Parret, and W. F. Broekaert. 2000. A gene encoding a sphingolipid biosynthesis enzyme determines the sensitivity of Saccharomyces cerevisiae to an antifungal plant defensin from dahlia (Dahlia merckii). Proc. Natl. Acad. Sci. USA 97:9531-9536. [PMC free article] [PubMed]
38. Tossi, A., L. Sandri, and A. Giangaspero. 2000. Amphipathic, alpha-helical antimicrobial peptides. Biopolymers 55:4-30. [PubMed]
39. van der Biezen, E. A. 2001. Quest for antimicrobial genes to engineer disease-resistant crops. Trends Plant Sci. 6:89-91. [PubMed]
40. Wade, D., A. Boman, B. Wahlin, C. M. Drain, D. Andreu, H. G. Boman, and R. B. Merrifield. 1990. All-D amino acid-containing channel-forming antibiotic peptides. Proc. Natl. Acad. Sci. USA 87:4761-4765. [PMC free article] [PubMed]
41. Yang, L., T. M. Weiss, R. I. Lehrer, and H. W. Huang. 2000. Crystallization of antimicrobial pores in membranes: magainin and protegrin. Biophys. J. 79:2002-2009. [PMC free article] [PubMed]
42. Zhong, L., R. J. Putnam, W. C. J. Johnson, and A. G. Rao. 1995. Design and synthesis of amphipathic antimicrobial peptides. Int. J. Pept. Protein Res. 45:337-347. [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...