Synthetic Antimicrobial Peptides as Agricultural Pesticides for Plant-Disease Control

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1225

REVIEW Synthetic Antimicrobial Peptides as Agricultural Pesticides for Plant-Disease Control by Emilio Montesinos a ) and Eduard Bardaj" b ) a

) Institute of Food and Agricultural Technology-CIDSAV-XaRTA, University of Girona, Campus Montilivi, E-18071 Girona (phone: þ 34 972418476; fax: þ 34 972418399; e-mail: [email protected]) b ) Department of Chemistry-LIPPSO, University of Girona, Campus Montilivi, E-18071 Girona (e-mail: [email protected])

There is a need of antimicrobial compounds in agriculture for plant-disease control, with low toxicity and reduced negative environmental impact. Antimicrobial peptides are produced by living organisms and offer strong possibilities in agriculture because new compounds can be developed based on natural structures with improved properties of activity, specificity, biodegradability, and toxicity. Design of new molecules has been achieved using combinatorial-chemistry procedures coupled to high-throughput screening systems and data processing with design-of-experiments (DOE) methodology to obtain QSAR equation models and optimized compounds. Upon selection of best candidates with low cytotoxicity and moderate stability to protease digestion, anti-infective activity has been evaluated in plant – pathogen model systems. Suitable compounds have been submitted to acute toxicity testing in higher organisms and exhibited a low toxicity profile in a mouse model. Large-scale production can be achieved by solution organic or chemoenzymatic procedures in the case of very small peptides, but, in many cases, production can be performed by biotechnological methods using genetically modified microorganisms (fermentation) or transgenic crops (plant biofactories).

1. Introduction. – Plant diseases caused by viruses, bacteria, and fungi are responsible for losses of 16% production in agriculture, or affect the quality and safety of derived foods [1]. Crop protection against plant diseases has been traditionally achieved with chemical pesticides, but is now oriented towards a rational use of pesticides with lower intrinsic toxicity and reduced negative environmental impact [2]. The European Union (Directive 91/414/CEE) and the USA (Insecticide, Fungicide, and Rodenticide Act) have undertaken regulatory changes in pesticide registration requirements based on toxicology, traceability, and environmental impact. Several compounds have been banned because of problems of environmental toxicity or development of resistance in the target pathogen like in the case of antibiotics [3]. The use of antibiotics in agriculture is questioned in several countries because of the possibility of horizontal transfer of cross-resistance from plant-associated bacteria to closely related human and animal pathogens [4]. Unfortunately, the reduction in the number of pesticide types available is currently generating additional problems in some plant diseases of economic importance that lack effective methods of control. As an example, the control of bacterial diseases currently relies in Europe almost on copper biocides, because antibiotics (e.g., streptomycin) are not authorized as a standard H 2008 Verlag Helvetica Chimica Acta AG, ZJrich

1226

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

practice. Therefore, antimicrobial compounds are necessary for plant-disease protection that satisfies the new requirements. Antimicrobial peptides (AMPs) offer possibilities for use as new pesticides because they are short-sequence peptides, produced by bacteria, fungi, plants, and animals, that cover a broad spectrum of activities against viruses, bacteria, fungi, and parasites. Since the discovery of the small bacteriocins in prokaryotes, and defensins in plants and animals, ca. 900 AMPs have been reported [5]. Several reviews are available on AMPs produced by bacteria [6 – 10], fungi [11], plants [12 – 14], and animals [15 – 17]. Exploitation of AMPs obtained from living organisms is difficult in many cases because of the low amounts naturally present or a relatively high intrinsic toxicity. Therefore, new designed molecules have been developed, and produced by synthetic or biotechnological methods. The potential use of antimicrobial peptides in plant-disease control have been recently reviewed [18]. 2. AMP Design. – Natural AMPs show an extraordinary molecular diversity, and have been naturally designed by evolutive combinatorial bio-chemistry, resulting in specific targeted molecules directed mainly towards one of the most ancient and wellestablished cell systems: the membrane. Non-ribosomally synthesized peptides contain proteinogenic and non-proteinogenic amino acids (N-methylation, side-chain crosslinking, cyclization, glycosylation, etc.) and differ in terms of molecular-structure organization. Peptides originated from ribosomal synthesis are essentially made of lamino acids, and some of them include building blocks based on side-chain modifications (dehydration, cross-linking). Structural differences are based on amino acid composition, all are, in general, amphipathic, and have particular distribution of hydrophobic and hydrophilic residues. Classifications of antimicrobial peptides have been made on the basis of secondarystructure organization [19] or according to their primary structure [20]. Regardless of their structural organization, AMPs are made of neutral amino acids combined with cationic (magainin and cecropin), anionic (dermicidin), aromatic/tryptophan (indolicidin), or hydroxylated amino acids that define specific electrostatic properties. The reason why certain residues should be located in the peptide sequence for efficient antimicrobial activity is one of the main concerns about the rationale of these structures, in terms of molecular design of new antimicrobial peptides. Several AMPs have been produced according to different strategies; some of the most relevant structures are summarized in Table 1. Shortening natural sequences from linear peptides with an a-helical structure (such as magainin and cecropin A) is one of those strategies. There are several examples including fragments from magainin (MSI99, pexiganan, ESF1, and ESF12), lactoferricin B (17 – 31 and 20 – 25), protegrins (iseganan), tachyplesins (TPY), and cecropins (D4E1). The construction of peptide hybrids made of fragments of natural peptides has also provided an important source of new structures. Hybrid mellitin and cecropin fragments provided a large series of improved peptides [47] [48] that finally allowed to obtain shorter peptides like PEP11 and PEP3. Also these hybrids resulted in more active and less hemolytic peptides like P18 or CAMEL. Apart from the modification of natural sequences, one promising strategy arises for de novo synthesized peptides, especially those extracted from chemical libraries oriented for this purpose.

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1227

Table 1. Synthetic Antimicrobial Peptides Active Against Plant Pathogens Sequence

Name

Number Net Hydro- Activity a ) Referof residues charge philicity ence

C16-KLLK FRLHF FRLKFH Ac-RKKWFW-NH2 Ac-RRWQWR- NH2 FRLKFHF cyclo( KKLKKFKKLQ ) WKLFKKILKVL-NH2 KKLFKKILKYL-NH2 KWKLFKKIGAVLKVL-NH2 RGGLCYCRGRFCVCVGR-NH2 FKLRAKIKVRLRAKIKL KWVFRVNYRGIKYRRQR MASRAAGLAARLARLALR KWKLFKKIPKFLHLAKKF MASRHMFLPLIGRVLSGIL ARHGSCNYVFPAHKCICYF MASRAAGLAARLARLALRAL GIGKFLKKAKKFGKAFVKILKK-NH2 GIGKFLKSAKKFGKAFVKILNS FAKKFAKKFKKFAKKFAKFAFAF MGIGKFLREAGKFGKAFVGEIMKP MKWKLFKKIGIGAVLKVLTTGLPALKLTK MAMWKDVLKKIGTVALHAGKAALGAVADTISO MALEHMKWKLFKKIGIGAVLKVLTTGLPALKLTK MPKWKVFKKIEKVGRNIRNGIVKAGPAIAVLGEAKALG MPRWRLFRRIDRVGKQIKQGILRAGPAIALVGDARAVG HQPKWKVFKKIEVVGRNIRNGIVKAGPAIAVLGEAKALG HSSGYTRPLRKPSRPIFIRPIGCDVCYGIPSSTARLCCFRYGDCHL-NH2 KSCCRNTWARNCYNVCRLPGTISREICAKKCRCKIISGTTCPSDYPK

– PPD1 66 – 10 PAF 26 LfcinB20–25 77 – 3 BPC194 PEP3 BP100 CAMEL Iseganan D4E1 TPY ESF12 P18 MsrA3 MBG01 ESF1 Pexiganan MSI-99 D2A21 Myp30 CEMA MsrA2

4 5 6 6 6 7 10 11 11 15 17 17 17 18 18 19 19 20 22 23 23 24 29 31

2.0 1.1 2.1 3.0 3.0 2.1 6.0 5.0 6.0 6.0 4.8 8.0 7.0 4.0 7.1 2.1 2.0 4.0 10.0 6.0 9.0 3.0 7.0 2.1

0.50 0.20 0.33 0.50 0.67 0.29 0.70 0.36 0.45 0.33 0.24 0.47 0.53 0.28 0.39 0.21 0.21 0.25 0.41 0.41 0.39 0.29 0.24 0.23

FþB F F F F F B F B B B FþB F F FþB FþB F F B FþB FþB FþB FþB FþB

[21] [22] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [30] [38] [39] [39] [40] [41]

MsrA1

34

6.1

0.24

FþB

[42]

SB-37

38

7.0

0.34

B

[43]

Shiva-1

38

7.0

0.34

B

[43]

MB-39

39

6.1

0.33

FþB

[44]

Pen4 – 1

46

6.0

0.30

F

[45]

D32R

47

7.6

0.40

FþB

[46]

a

) F, Antifungal; B, antibacterial; F þ B, antifungal and antibacterial.

Some general rules have been accepted with respect to structure – activity requirements for AMPs. Their cationic charge and the ability to adopt amphipathic conformation upon membrane interaction are the key structural features to promote membrane binding and disruption. This is mainly a consequence of the hydrophobic/ hydrophilic balance of amino acids and their distribution along the peptide sequence. Apart from their antimicrobial activity, their resistance to proteolytic enzymes is another important issue that accounts for potential application of a given sequence.

1228

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

Resistance to proteolysis can be increased by means of introducing d-amino acids [49], with fatty acid modifications [50], or by cyclization (BPC194) or simply fine-tuning discrete positions on the peptide sequence (BP100, see below). Also, a specificity for microbial and fungal cells, and a reduced toxicity to mammalian or plant cells is of crucial importance to obtain AMPs with potential applications in human, animal, or plant protection. An excess of lipidic residues or a high content in tryptophan residues increases both antimicrobial and cytotoxic activities. However, among all the reported structures, it is quite difficult to extract clear rules for cell specificity of peptides. Interestingly, the finding of the cyclic decapeptide BPC194 is a good example on how little changes in the sequence and type of amino acid residues in certain positions affect specificity against three plant pathogenic bacteria [26] [51]. Combinatorial chemistry allows the production of large collections of compounds that can be screened for biological activity. The combinatorial approach can be carried out using a positional scanning strategy (where multiple sequences are made fixing one position at a time and randomizing the other positions) or using a parallel method (where specific compounds are made in single reactors). The first approach allows the simultaneous synthesis of a large number of variations for a given peptide, but requires the application of deconvolution methods for the determination of the most important amino acids at each position [52 – 54]. The use of parallel methods provides final pure samples of single compounds, facilitating the interpretation of biological results but requiring more efforts for their production. At this point, it is worth considering the available tools for directing the synthesis of parallel libraries with the aim to reduce the large lists of compounds to be made. For AMPs oriented to agriculture, the final goal is to find the shortest, most active, and less toxic peptide. The application of a combinatorial approach led to short linear peptides effective against phytopathogenic fungi like PAF26 [55] and PEP6 [37], and, in our case, linear BP100 and cyclic BPC194 [26] [51] to produce improved antibacterial peptides. The development of BP100 started with the modification of WKLFKKILKVL-NH2 (PEP3) [27], an undecapeptide previously reported to display antifungal and antibacterial activities with low cytotoxicity [27] [37]. With the aim to adapt its sequence for increased activity and/or selectivity for the target pathogens, our first studies resulted in a series of undecapeptides. These peptides were based on changes in positions 1 and 10, amino acids located at the edge frontier positions of the theoretical a-helical structure predicted from Edmundson wheel projections, in order to modify the total amphipatic balance of the molecule while preserving both cationic and lipophilic sides in two blocks. The best hit compounds were BP15, BP20, and BP76. Further modifications conducted by combinatorial-chemistry methods led to BP100 (H-KKLFKKILKYL-NH2 ) as the best lead compound (Fig. 1). Fig. 2 shows the modification of the cationic side of the a-helical structure from PEP3 to BP100, with the placement of a cationic lysine in position 1 and an aromatic-polar tyrosine in position 10. This example of fine tuning of a given peptide sequence using a design based on the modification of the a-helical amphipatic balance can be of great interest for the development of new improved peptides. The development of BPC194 started from the idea to make de novo-designed peptides using natural amino acids (Leu-Lys balance) and exploring by combinatorial chemistry the development of short and non-toxic antimicrobial peptides. Synthetic

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1229

Fig. 1. Development of the antibacterial undecapeptide BP100 using a combinatorial chemistry procedure consisting of changes in amino acids at positions 1 and 10. Edmondson wheel representation of initial (upper left) and final (upper right) structures, and sphere molecular representation of BP100 using the ChemBioDraw 11.0 software.

Fig. 2. Evolution from the linear undecapeptide PEP3 to BP100 consisting of adjusting the polarity of residue at position 10 and placing a lysine residue at position 1, thus increasing the cationic surface of the molecule

1230

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

peptides with interesting antimicrobial activities, and containing Lys and Leu residues have been described [52] [56 – 59]. The construction of short Leu-Lys-containing peptides in different sizes and amino acid proportions gave us a preliminary background where the most relevant facts were that the linear sequences had nonsignificant activities on bacterial growth inhibition, but the cyclized structures showed relevant inhibitory profiles [51]. From a list of the best compounds, a cyclic decapeptide was used as best hit to improve four positions by means of combinatorial chemistry [26] (Fig. 3).

Fig. 3. Development of the antibacterial cyclic decapeptide BPC194 using a combinatorial-chemistry procedure consisting of changes in amino acids at four positions. Representation of initial (upper left) and final (upper right) structures, and sphere molecular representation of BPC194 using the ChemBioDraw 11.0 software.

Design of experiments (DOE) constitutes a well-known statistical methodology which permits to determine the inner rules of a process from relevant experimental data. This methodology has been proven to be useful in quantitative structure – activity relationships (QSAR) studies. The DOE approach uses a factorial design to grasp simultaneous synergic and nonlinear effects among experimental factors. Starting from partitioning an initial set of compounds with given activities, two sets of structures are

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1231

established. A training set is selected in order to extract all the possible information from a minimal number of molecules. Then, a two-level factorial DOE is applied over this set of peptides in order to obtain a QSAR equation model, which finally can be used to predict the properties of a larger test set of compounds [60]. With the orientation given by the DOE analysis of the results obtained from the first library of 56 cyclic decapetides, we developed a short libray of 16 compounds that provided the twelve best lead compounds [26], the most relevant being BPC194 (Fig. 3). 3. Screening. – Finding suitable AMPs requires in vitro testing of antimicrobial activity before a reduced number of lead compounds is assayed in plant – pathogen interaction tests. The most popular methods of screening are based on the inhibition of cell multiplication using a biostatic approach. The procedure consists of adding the AMP at different concentrations to the pathogen cell suspension and measure cell multiplication in a suitable culture medium. This can be achieved by monitoring absorbance of culturable cells within time. Methods based on absorbance have the advantage that they can be optimized using well microplates and automatic readers. However, several problems may arise with filamentous growth (e.g., fungi) and due to sedimentation that can be solved with the appropriate equipment (shaking, thermostatizing, etc.), or using gelled liquid medium. The antimicrobial activity also may depend on media composition (e.g., ionic strengh, binding of the AMP) and the developmental stage of target cells (vegetative cells, spores). Combinatorial-chemistry approaches to optimize lead molecules often require massive screening of a great number of compounds (from hundreds to thousands), and High Throughput Screening Systems (HTS) are needed. Several rapid and highly sensible methods have been developed to assess antimicrobial activity based on vital staining or cell performance reporters. The degree of permeation of bacterial and fungal cell membranes can be monitored with increase of fluorescence due to the entrance of the cationic dye Sytox Green that binds to nucleic acids. Quantitative assays of antimicrobial activity based on this method have been developed for plant defense studies [61] and antibiotic susceptibility testing [62]. The activity and growth of cells can also be measured using engineered target pathogens expressing constitutively the green fluorescent protein (gfp) or luciferase (lux) genes, thus permitting the assessment of activity on the basis of fluorescence or bioluminescence changes. gfp and lux reporters have been used for antibiotic susceptibility studies in clinical pathogens [63 – 65]. Due to the higher sensitivity of fluorescence and bioluminescence measurements, the analysis can be performed faster than with absorbance, and relatively sophisticated microplate readers have been developed that permit massive screening of hundreds of AMPs at different concentrations against several target pathogens in short time periods. As in many functional peptides, biochemical and biophysical cellular processes affected by AMPs can be used as target. Although most AMPs act through a membrane-damage mechanism, several have been reported to interfere with other cellular processes. Binding assays have been developed that can be used for screening and are based on the fact that some AMPs interact with bacterial lipopolysaccharides or theicoic acids [66], ergosterol [31], DNA or RNA [67], or chitine synthesis [68]. Calcein release from small unilamelar vesicles composed of different phospholipids

1232

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

that mimic the membrane composition of fungi and bacteria has been also used [21]. Selection of the best AMPs cannot be based exclusively on antimicrobial activity, because the microbiocidal activity, cytotoxicity to non-target cells, and protease susceptibility are additional properties that have to be taken into account. Microbiostatic compounds exert the inhibitory effect on the target cells only when the compound is present, and growth is restored upon its dilution or inactivation. Contrary to that, the microbiocidal compounds kill the cells and, therefore, the inhibitory effect persists after releasing the treatment. A liquid-contact test based on the interaction for given periods of time of the target pathogen and AMPs, and subsequent measurement of surviving cells upon sufficient dilution below MIC values is used to evaluate biocidal activity. Obviously, a candidate molecule would be better if the activity is microbiocidal rather than microbiostatic. Linear and cyclic AMPs developed in our laboratory showed a potent bactericidal effect against plant pathogenic bacteria like Erwinia amylovora PMV6076, Xanthomonas vesicatoria 2133-2, and Pseudomonas syringae pv. syringae EPS94, with decimal reduction times of less than an hour at concentrations around the minimum inhibitory concentration (MIC) [69]. Screening of cytotoxicity to animal or plant cells is always a problem in AMPs targeted to bacteria or fungi when the main mechanism of action is based on interaction with cell membranes. Cytotoxicity can be evaluated using animal or plant cell model systems. Due to the strong development of AMPs in the area of pharmaceutical products, hemolysis in red blood cells (RBCs) has been the method commonly used for cytotoxicity assessment. Generally, minimal hemolytic concentrations in the range of 50 – 100 mm are considered as adequate. However, methods specifically adapted to plant-protection-product requirements have been developed like plant protoplast phytotoxicity [70] or pollen germination – growth inhibition [71]. A slightly higher toxicity is observed in plant cell tests than in RBCs. Tobacco protoplast toxicity in the range of 30 – 35 mg/ml for a cecropin – melittin hybrid peptide have been reported [70]. Toxic values of 10 – 50 mm have been reported for cecropin – magainin hybrid peptides in pollen inhibition germination assays [72]. The values of cytotoxicity to non-target cells considered as suitable are generally around 3 – 30-times higher than the MIC values observed against the target pathogens. The best linear and cyclic peptides developed in our laboratory exhibit very low hemolytic activity, with minimal hemolytic concentrations in the range of 100 – 150 mm. Several studies have shown that an increase in the peptide hydrophobicity and amphipathicity are related to an increase of cytotoxicity to mammalian cells, but data are not available for plant cells. Interestingly, certain tetrapetides without significant antimicrobial activity became active upon linking to fatty acid chains, and the cytotoxic activity increased with the length of the hydrophobic chain [50]. Obviously, during the optimization of AMPs the objective is to minimize hemolytic activity. We have found that, in the linear undecapeptides, the less hemolytic compounds incorporated a Lys residue, either in position 1 or 10, and N-terminal derivatization and the presence of Trp increased the hemolytic activity [28]. In the case of cyclic decapeptides, the simultaneous presence of a Lys in positions 2 and 3 gave the less hemolytic compounds [26]. Stability to protease digestion is also a desired property in antimicrobial peptides to assure a reasonable half-life of the molecules in the plant environment, because

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1233

proteases from epiphytic microorganisms or from the plant tissues may degrade antimicrobial peptides [27] [37]. Although this stability is a desired property, a reasonable level of protease susceptibility is necessary to guarantee a certain degree of biodegradability in the natural environment. We optimized the structure of our peptides to improve protease stability, and we found that cyclic peptides are more stable to proteinase K digestion than the linear counterparts. Typical degradation half times for the best linear (BP100) and cyclic (BPC194) antibacterial peptides are in the range of 45 – 60 min [26] [69] (Table 2). Table 2. Relevant Characteristics of Two Synthetic Peptides Active against Plant Pathogenic Bacteria Characteristics Minimum inhibitory concentration ( MIC [mm]) against Pseudomonas syringae pv. syringae EPS94 Xanthomonas vesicatoria 2133-2 Erwinia amylovora PMV6076 Bactericidal activity at MIC value [decimal reduction time in min] Hemolytic activity in human red blood cells [%] Proteinase K stability [% degradation after 45 min] Acute oral toxicity in mice ( LD50 [mg/kg of body weight]) a

BP100

BPC194

2.5 – 5.0 5.0 – 7.5 2.5 – 5.0 36

3.1 – 6.2 3.1 – 6.2 6.2 – 12.5 nd a )

22 b ) 44 > 2000

17 c ) stable –

) Not determined. b ) At 150 mm. c ) At 375 mm.

4. Evaluation of Anti-Infective Activity in Plant – Pathogen Model Systems. – Finding antimicrobial activity of a compound using in vitro tests is not predictive of its capacity for inhibition of infection by the pathogen on host tissues, because several host components can interfere with the activity of AMPs. Therefore ex vivo methods are used as plant – pathogen infection inhibition tests. These methods are closer to the real situation in the whole plant and have the advantage of facilitating massive screening procedures. Several ex vivo systems have been reported based on detached plant organs susceptible to pathogen (leaves, flowers, fruits, roots) [23] [37] [69] [73 – 75]. Ex vivo methods are specially necessary when working with plant quarantine pathogens that require biosafety containment levels in the laboratory greenhouse (e.g., Erwinia amylovora in the EU). Plant model systems can also be used. Most promising for massive AMPs screening is Arabidopsis thaliana, a plant model system which is susceptible to several plant pathogens including bacteria of the genera Pseudomonas, Erwinia, Ralstonia, and Xanthomonas, and fungi like Erysiphe cichoracearum, Phytophthora parasitica, Alternaria brassicola, Pythium spp., and B. cinerea [76] [77]. Data have been collected in several pathosystems as for BP100 against E. amylovora in apple and pear [69], PAF26 against Penicillium digitatum in citrus fruits [24], and ultrashort lipopeptides against B. cinerea on cucumber [50]. Generally, the effective doses are in the range of 50 – 200 mm. 5. Toxicity to Higher Organisms. – Many AMPs either from natural sources (e.g., microbial, animal, or plant origin) or synthetic can be toxic to animals or plants. In spite

1234

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

of a certain indication of toxicity given by the in vitro cytotoxicity tests (RBCs, plant protoplasts), plant or animal assays have to be performed. A good indication of the lack of phytotoxicity is provided by the fact that plants expressing AMPs do not show signs of toxicity in many cases, and they complete the biological cycle. However, based on experience in the pharma field and for human safety reasons, animal toxicity testing is commonly used. Due to the fact that the main way of interaction of pesticides with animals and humans is by oral ingestion through food, it seems more appropiate to perform oral toxicology assays. The most frequent toxicological tests in animal models are acute oral toxicity in mice or rats, with the objective of determining the median lethal dose (LD50) and the lower limit lethal dose (LLD; OECD Test Guidelines). Complementary tests to evaluate other ways of interaction are dermal, eye, and inhalation irritation tests in guinea pig or rabbit. Unfortunately, there is a lack of information on toxicity issues in animal tests for most AMPs [78]. One of the best peptides developed in our laboratory, BP100, has been submitted to oral acute toxicity testing in mice, and the LD50 is higher than 2,000 mg/kg of body weight, thus it can be considered to exhibit very low toxicity. 6. Large-Scale Production. – The MIC values for most AMPs developed against plant pathogens as measured with in vitro assays are in the range of 1 – 10 mm. These values compare well with standard antibiotics, but the practical rates needed for field use are at 50 – 200 g a.i./ha. Development of AMPs for agricultural use has several constrains, mainly as a consequence of the need for cheap products in plant protection. Based on existing commercial pesticides, an economic threshold for a commercial exploitation can be situated around 5 – 10 E/g of active ingredient. This threshold is far from the 50 – 400 US$ estimated for a pharmacological product [78] [79] and does not seem economically feasible at these costs that can be assumed only by the pharmaceutical industry. A low-cost production is difficult to be achieved by standard organic chemistry procedures based on solid-phase synthesis that are commonly employed for small-scale laboratory peptide design. Therefore, strategic areas of interest to bring AMPs into the plant pesticide market have to deal with decreasing production costs. Short AMPs in the range of 4 – 6 amino acid residues can be synthetized at a reasonable cost by solution organic synthesis and/or chemoenzymatic approaches, as in the case of several lipopeptides or hexapeptides [80]. However, this is not possible for larger peptide sizes. Another approach is the production by biotechnological procedures based on microbial systems (large-scale fermentation) or transgenic crops (plant biofactories), but is restricted to AMPs containing proteinogenic amino acids that are suitable for ribosomal synthesis. Biotechnological expression of a vast amount of peptides of relatively large size has been achieved in microbial [81] and plant models [82]. However, AMPs have intrinsically several technological problems due to self-toxicity for the producer cell and difficulties of expression because of the small size in many cases. Strategies of cloning are available to counteract these problems, based on tandem polypeptides, fusion proteins, or directed secretion as inclusion bodies (intracellular) or tissuespecific location [83]. Also, depending on the approach used, post-processing of the product is required that limits an economically feasible production (downstream or

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1235

cellular self-processing) [84]. Obviously, microbial and plant-based production costs can be reduced, because, in theory, a plant-protection product based on AMPs does not necessarily need to be obtained in a highly purified state like for the pharmaceutical products. Biotechnological production based on genetically engineered microorganisms is available for several proteins and for the production of pharmaceuticals [81] at costs that have been estimated between 50 – 100 US$/g [79]. Yields of AMPs are reported in the range of one to several hundred mg/l of fermented broth [83]. For example, cecropin, lactoferricin, and indolicidin have been expressed in E. coli [85 – 87], and penaedin, CAMA, and ABP-CM4 have been expressed in Pichia pastoris [88 – 90]. Therefore, the feasibility for an agricultural exploitation at the rates required for field use is difficult to be achieved if it is taken into account that, to treat one hectar of crop, an equivalent amount of up to 1,000 – 10,000 l fermentor would be needed. AMPs produced using plants as biofactories offer good expectations for agricultural purposes, because a great part of the technology is already developed to obtain selfprotected plants against pathogens [18], and because a great reduction in production costs to 10 – 20 US$/g has been estimated [79]. Interestingly, high yields have been obtained for peptides and proteins of different sizes and of pharmaceutical application [82]. An example of this potential was the expression of a small cedar pollen allergen protein of 13 kDa that accumulated in the grain endosperm of a transgenic rice at rates of 3 g/kg of seed [91]. A yield of around 10 – 20 kg/ha is expected, assuming a standard production for a grain crop of 6,000 kg/ha. However, the main limitation of the use of plants as biofactories is the risk of out-crossing, thus requiring large costs on risk minimization as the use of greenhouse containment systems. REFERENCES [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]

E. C. Oerke, J. Agric. Sci. 2006, 144, 31. G. N. Agrios, OPlant PathologyP, Academic Press, London, 2005, p. 635. A. K. Vidaver, Clin. Infect. Dis. 2002, 34, S107. P. S. McManus, V. O. Stockwell, V. O. Sundin, A. L. Jones, Annu. Rev. Phytopathol. 2002, 40, 443. R. E. W. Hancock, H. G. Sahl, Nat. Biotechnol. 2006, 24, 1551. R. W. Jack, G. Jung, Curr. Opin. Chem. Biol. 2000, 4, 310. T. Degenkolb, A. Berg, W. Gams, B. Schlegel, U. GrRfe, J. Pept. Sci. 2003, 9, 666. P. D. Cooter, C. Hill, P. Ross, Curr. Protein Pept. Sci. 2005, 6, 61. T. Stein, Mol. Microbiol. 2005, 56, 845. J. M. Raaijmakers, I. de Bruijn, M. J. D. de Kock, Mol. Plant-Microbe Interact. 2006, 19, 699. T. B. Ng, Peptides 2004, 25, 1055. B. Bechinger, Crit. Rev. Plant Sci. 2004, 23, 271. F. T. Lay, M. A. Anderson, Curr. Protein Pept. Sci. 2005, 6, 85. N. Farrokhi, J. P. Whitelegge, J. A. Brusslan, Plant Biotechnol. J. 2008, 6, 105. M. Zasloff, Nature 2002, 415, 389. H. G. Boman, J. Intern. Med. 2003, 254, 197. P. Bulet, R. Stçcklin, L. Menin, Immunol. Rev. 2004, 198, 169. E. Montesinos, FEMS Microbiol. Lett. 2007, 270, 1. W. VanPt Hof, E. C. Veerman, E. J. Helmerhorst, N. Amerongen, Biol. Chem. 2001, 382, 597. K. A. Brogden, Nat. Rev. Microbiol. 2005, 3, 238. A. Makovitzki, D. Avrahami, Y. Shai, Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15997. J. D. Reed, D. L. Edwards, C. F. Gonzalez, Mol. Plant-Microbe Interact. 1997, 10, 537.

1236

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

[23] B. Lo´pez-GarcUa, L. Gonza´lez-Candelas, E. Pe´rez-Paya´, J. F. Marcos, Mol. Plant-Microbe Interact. 2000, 13, 837. [24] A. Mun˜oz, J. F. Marcos, J. Appl. Microbiol. 2006, 101, 1199. [25] C. F. Gonzalez, E. M. Provin, L. Zhu, D. J. Ebbole, Phytopathol. 2002, 92, 917. [26] S. Monroc, E. Badosa, E. Besalu, M. Planas, E. Bardaji, E. Montesinos, L. Feliu, Peptides 2006, 27, 2575. [27] L. Cavallarin, D. Andreu, B. San Segundo, Mol. Plant-Microbe Interact. 1998, 11, 218. [28] R. Ferre´, E. Badosa, L. Feliu, M. Planas, E. Montesinos, E. Bardaji, Appl. Environ. Microbiol. 2006, 72, 3302. [29] W. Kamysz, A. Krolicka, K. Bogucka, T. Ossowski, J. Lukasiak, E. Lojkowska, J. Phytopathol. 2005, 153, 313. [30] J. Chen, T. J. Falla, H. Liu, M. A. Hurst, C. A. Fujii, D. A. Mosca, J. R. Embree, D. J. Loury, P. A. Radel, C. C. Chang, L. Gu, J. C. Fiddes, Biopolymers 2000, 55, 88. [31] A. J. DeLucca, T. J. Walsh, Antimicrob. Agents Chemother. 1999, 43, 1. [32] A. G. Rao, Arch. Biochem. Biophys. 1999, 361, 127. [33] W. A. Powell, C. M. Catranis, C. A. Maynard, Mol. Plant-Microbe Interact. 1995, 8, 792. [34] D. G. Lee, K. S. Hahm, S. Y. Shin, Biotechnol. Lett. 2004, 26, 337. [35] M. Osusky, L. Osuska, R. E. Hancock, W. W. Kay, S. Misra, Transgenic Res. 2004, 13, 181. [36] W. M. M. Schaaper, G. A. Posthuma, R. H. Meloen, H. H. Plasman, L. Sijtsma, A. van Amerongen, F. Fant, F. A. M. Borremans, K. Thevissen, W. F. Broekaert, J. Pept. Res. 2001, 57, 409. [37] G. S. Ali, A. S. N. Reddy, Mol. Plant-Microbe Interact. 2000, 13, 847. [38] A. R. Alan, E. D. Earle, Mol. Plant-Microbe Interact. 2002, 15, 701. [39] D. Rioux, V. Jacobi, M. Simard, R. C. Hamelin, Can. J. Bot. 2000, 78, 462. [40] R. E. W. Hancock, M. H. Brown, K. Piers, US Pat. appl. No. 07/913,492, filed August 21, 1992. [41] M. Osusky, L. Osuska, W. Kay, S. Misra, Theor. Appl. Genet. 2005, 111, 711. [42] M. Osusky, G. Zhou, L. Osuska, R. E. W. Hancock, W. W. Kay, S. Misra, Nat. Biotechnol. 2000, 18, 1162. [43] J. M. Jaynes, P. Nagpala, L. Destefano-Beltran, J. H. Huang, J. H. Kim, T. Denny, S. Cetiner, Plant Sci. 1993, 89, 43. [44] L. D. Owens, T. M. Heutte, Mol. Plant-Microbe Interact. 1997, 10, 525. [45] B. J. Cuthbertson, E. E. Bullesbach, J. Fievet, E. Bachere, P. S. Gross, Biochem. J. 2004, 381, 79. [46] M. Vila-Perello´, A. Sa´nchez-Vallet, F. GarcUa-Olmedo, A. Molina, D. Andreu, FEBS Lett. 2003, 536, 215. [47] D. Andreu, J. Ubach, A. Boman, B. Wa˚hlin, D. Wade, R. B. Merrifield, H. G. Boman, FEBS Lett. 1992, 296, 190. [48] D. Wade, D. Andreu, S. A. Mitchell, A. M. V. Silvera, A. Boman, H. G. Boman, R. B. Merrifield, Int. J. Pept. Protein Res. 1992, 40, 429. [49] R. B. Merrifield, E. L. Merrifield, P. Juvvadi, D. Andreu, H. G. Boman, Ciba Found. Symp. 1994, 186, 5. [50] A. Makovitzki, A.Viterbo, Y. Brotman, I. Chet, Y. Shai, Appl. Environ. Microbiol. 2007, 73, 6629. [51] S. Monroc, E. Badosa, L. Feliu, M. Planas, E. Montesinos, E. Bardaji, Peptides 2006, 27, 2567. [52] S. E. Blondelle, R. A. Houghten, Trends Biotechnol. 1996, 14, 60. [53] C. Pinilla, J. Appel, S. Blondelle, C. Dooley, B. Doerner, J. Eichler, J. Ostresh, R. A. Houghten, Biopolymers 1995, 37, 221. [54] S. Y. Hong, J. E. Oh, M. Kwon, M. J. Choi, J. H. Lee, B. L. Lee, H. M. Moon, K. H. Lee, Antimicrob. Agents Chemother. 1998, 42, 2534. [55] B. Lo´pez-GarcUa, E. Pe´rez-Paya´, J. F. Marcos, Appl. Environ. Microbiol. 2002, 68, 2453. [56] Z. Oren, Y. Shai, Biochemistry 2000, 39, 6103. [57] N. Papo, Z. Oren, U. Pag, H. G. Sahl, Y. Shai, J. Biol. Chem. 2002, 277, 33913. [58] R. M. Epand, B. G. Sayer, R. H. Epand, Biochemistry 2003, 42, 14677. [59] N. Papo, Y. Shai, J. Biol. Chem. 2005, 280, 10378. [60] J. M. Barroso, E. Besalu´, J. Mol. Struct. 2005, 727, 89. [61] K. Thevissen, F. R. G. Terras, W. F. Broekaert, Appl. Environ. Microbiol. 1999, 65, 5451.

CHEMISTRY & BIODIVERSITY – Vol. 5 (2008)

1237

[62] B. L. Roth, M. Poot, S. T. Yue, P. J. Millard, Appl. Environ. Microbiol. 1997, 63, 2421. [63] L. A. Collins, M. N. Torrero, S. G. Franzblau, Antimicrob. Agents Chemother. 1998, 42, 344. [64] J. S. Webb, S. R. Barrat, H. Sabev, M. Nixon, I. M. Eastwood, M. Greenhalch, P. S. Handley, G. D. Robson, Appl. Environ. Microbiol. 2001, 67, 5614. [65] M. Tenhami, K. Hakkila, M. Karp, Antimicrob. Agents Chemother. 2001, 45, 3456. [66] M. G. Scott, H. Yan, R. E. W. Hancock. Infect. Immun. 1999, 67, 2005. [67] C. Marchand, K. Krajewski, H. F. Lee, S. Antony, A. A. Johnson, R. Amin, P. Roller, M. Kvaratskhelia, Y. Pommier, Nucleic Acids Res. 2006, 34, 51557. [68] S. Oita, M. Ohnishi-Kameyama, T. Nagata, Biosci. Biotechnol. Biochem. 2000, 64, 958. [69] E. Badosa, R. Ferre´, M. Planas, L. Feliu, E. Besalu´, J. Cabrefiga, E. BardajU, E. Montesinos, Peptides 2007, 28, 2276. [70] D. P. Yevtushenko, R. Romero, B. S. Forward, R. E. Hancock, W. W. Kay, S. Misra, J. Exp. Bot. 2005, 56, 1685. [71] U. Kristen, Toxicol. in Vitro 1997, 11, 181. [72] V. Jacobi, A. Plourde, P. J. Charest, R. C. Hamelin, Can. J. Bot. 2000, 78, 455. [73] J. Cabrefiga, E. Montesinos, Phytopathology 2005, 95, 1430. [74] J. France´s, A. Bonaterra, M. C. Moreno, J. Cabrefiga, E. Badosa, E. Montesinos, Postharvest Biol. Technol. 2006, 39, 299. [75] E. Montesinos, A. Bonaterra, Phytopathology 1996, 86, 464. [76] B. N. Kunkel, D. M. Brooks, Curr. Opin. Plant Biol. 2002, 5, 325. [77] W. E. Durrant, X. Dong, Annu. Rev. Phytopathol. 2004, 42, 185. [78] A. K. Marr, W. J. Gooderham, R. E. Hancock, Current Opin. Pharmacol. 2006, 6, 468. [79] A. Elbehri, AgBioforum 2005, 8, 18. [80] Y. Shai, Biopolymers 2002, 66, 236. [81] A. B. Ingham, R. J. Moore, Biotechnol. Appl. Biochem. 2007, 47, 1. [82] M. E. Horn, S. L. Woodard, J. A. Howard, Plant Cell Rep. 2004, 22, 711. [83] L. Metlitskaia, J. E. Cabralda, D. Suleman, C. Kerry, J. Brinkman, D. Bartfield, M. M. Guarna, Biotechnol. Appl. Biochem. 2004, 39, 339. [84] K. Graumann, A. Premstaller, Biotechnol. J. 2006, 1, 164. [85] X. Xu, F. Jin, X. Yu, S. Ji, J. Wang, H. Cheng, C. Wang, W. Zhang, Protein Expr. Purif. 2007, 53, 293. [86] X. Feng, J. Wang, A. Shang, D. Teng, Y. Yang, Y. Yao, G. Yang, Y. Shao, S. Liu, F. Zhang, Protein Expr. Purif. 2006, 47, 110. [87] K. M. Morin, S. Arcidiacono, R. Beckwitt, C. M. Mello, Appl. Microbiol. Biotechnol. 2006, 70, 698. [88] L. Li, J. Wang, X. Zhao, C. Kang, N. Liu, J. Xiang, F. Li, S. Sueda, K. Kondo, Protein Expr. Purif. 2005, 39, 144. [89] F. Jin, X. Xu, L. Wang, W. Zhang, D. Gu, Protein Expr. Purif. 2006, 50, 147. [90] J. Zhang, S. Q. Zhang, X. Wu, Y. Q. Chen, Z. Y. Diao, Process Biochem. 2006, 41, 251. [91] H. Takagi, S. Saito, L. Yang, S. Nagasaka, H. Nishizawa, F. Takaiwa, Plant Biotechnol. J. 2005, 3, 521. Received January 3, 2008