Identification and Rational Design of Novel Antimicrobial Peptides for Plant Protection

ANRV351-PY46-12 ARI 22 June 2008 11:13

Identication and RationalDesign of Novel AntimicrobialPeptides for Plant ProtectionJose F. Marcos,1 Alberto Munoz,1

Enrique Perez-Paya,2 Santosh Misra,3

and Belen Lopez-Garca11Departamento de Ciencia de los Alimentos, Instituto de Agroqumica y Tecnologa deAlimentos (IATA)-CSIC, 46100 Burjassot, Spain; email: [email protected];[email protected]; [email protected] de Investigacion Prncipe Felipe (CIPF), 46013 Valencia, Spain; and Instituto deBiomedicina de Valencia (IBV)CSIC; email: [email protected] of Biochemistry and Microbiology, University of Victoria,Victoria BC V8W 3P6, Canada; email: [email protected]

Annu. Rev. Phytopathol. 2008. 46:273301

First published online as a Review in Advance onApril 25, 2008

The Annual Review of Phytopathology is online atphyto.annualreviews.org

This articles doi:10.1146/annurev.phyto.121307.094843

Copyright c 2008 by Annual Reviews.All rights reserved

0066-4286/08/0908/0273$20.00

Key Words

bioactive peptides, antibacterial, antifungal, peptide libraries, plantresistance, cell penetrating peptides, peptide production

AbstractPeptides and small proteins exhibiting antimicrobial activity have beenisolated from many organisms ranging from insects to humans, includ-ing plants. Their role in defense is established, and their use in agricul-ture was already being proposed shortly after their discovery. However,some natural peptides have undesirable properties that complicate theirapplication. Advances in peptide synthesis and high-throughput activityscreening have made possible the de novo and rational design of novelpeptides with improved properties. This review summarizes ndingsin the identication and design of short antimicrobial peptides withactivity against plant pathogens, and will discuss alternatives for theirheterologous production suited to plant disease control. Recent studiessuggest that peptide antimicrobial action is not due solely to microbepermeation as previously described, but that more subtle factors mightaccount for the specicity and absence of toxicity of some peptides. Theelucidation of the mode of action and interaction with microbes will as-sist the improvement of peptide design with a view to targeting specicproblems in agriculture and providing new tools for plant protection.

273

Click here for quick links to Annual Reviews content online, including:

Other articles in this volume Top cited articles Top downloaded articles Our comprehensive search

FurtherANNUALREVIEWS

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

AMP: antimicrobialpeptides

INTRODUCTION

Pathogenic microorganisms, the leading causeof plant diseases and crop losses, require thecontinued use of chemicals for control as wellas to meet world food needs. However, theemergence of resistant isolates and pathogens,the limited spectrum of action, and negativelong-term repercussions on human health andthe environment have propelled the search fornew alternatives as substitutes for the chemicalscurrently in use.

Peptides and small proteins exhibiting directantimicrobial activity have been characterizedfrom a vast number of organisms ranging frominsects to humans (21, 191). The general bi-ological role of antimicrobial peptides (AMP)in defense against challenging microbes isrecognized. Antimicrobial peptides look likepromising alternatives as novel therapeutics incombating the increasing incidence of antibi-otic resistances in pathogenic microbes; andseveral examples are undergoing clinical trials(55, 57, 191).

Plants have specic defense mechanismsthat include small antimicrobial proteins orpeptides with activity against phytopathogens(19, 50, 176). In addition to their in vitro an-timicrobial activity, some of these endogenousproteins are involved in active defense againstdisease. A recent study has suggested that theseproteins might account for up to 2%3% of thegenes predicted in plant genomes (162).

In parallel to what has occurred in medicine,antimicrobial peptides and proteins have beenproposed for potential use in agriculture (24,67, 114, 152, 175). However, initial studies soonrevealed that some natural peptides have unde-sirable properties such as nonspecic toxicity,low stability, and poor bioavailability that wouldcompromise their application on crops.

Advances over the past decade have madepossible the rational design of novel nonnat-ural AMP with the goal of improving theirproperties. The short sequence length of AMPfavors structure/activity studies in a holistic ap-proach to enhance their stability, potency, andspecicity toward certain microbes. Their se-

quence length also makes feasible the chemicalsynthesis to high purity and facilitates the de-sign of synthetic genes for their heterologousproduction through biotechnology. All thesefeatures boost scientic interest in this class ofmolecules.

In this review, we summarize and describethe processes of identication and characteri-zation of AMP active toward phytopathogens,improvement of peptide properties by the ra-tional modication of the amino acid sequence,and the different strategies for use of the result-ing peptide in plant protection.

OVERVIEW ON NATURALANTIMICROBIAL PEPTIDESAND PROTEINS

AMP are a broad class of peptides and smallproteins of short size (below 3040 aminoacid residues), which have antimicrobial activityagainst microorganisms, at least under in vitroassay conditions. Most of the conclusions fromprevious reviews on the identication, charac-terization, and activity of AMP of interest inmedicine (21, 57, 68, 140, 191) can be extrap-olated to plant pathology. We deal only withthe application of AMP to plant protection, anddiscuss only examples/topics outside of our eldthat are relevant.

Although posttranslational modication,the inclusion of nonnatural amino acids, andpeptide-like (mimetic) compounds can mod-ulate (and increase) substantially the activ-ity and stability of AMP (57), we focus ourdiscussion mostly on ribosomally synthesizedpeptides. The properties and applications ofnonribosomal or posttranslationally modiedpeptides in agriculture and food science havebeen reviewed elsewhere (32, 114).

Properties of Antimicrobial Peptides

AMP are diverse (see examples in Table 1) andcan be subdivided into several groups based ontheir origin, composition, and structure (21,41). However, they also share certain com-mon structural characteristics such as (i ) amino

274 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

Table 1 Representative natural short antimicrobial peptides

Peptide Amino acid sequencea SourceApidaecin GNNRPVYIPQPRPPHPRI InsectCecropin A KWKFKKIEKMGRNIRDGIVKAGPAIEVIGSAKAI InsectCecropin B KWKVFKKIEKMGRNIRNGIVKAGPAIAVLGEAKAL InsectIb-AMP1 QWGRRCCGWGPGRRYCVRWC PlantIndolicidin ILPWKWPWWPWRR BovineMagainin 2 GIGKFLHSAKKFGKAFVALKAL FrogMelittin GIGAVLKVLTTGLPALISWIKRKRQQ InsectPR-39 RRRPRPPYLPRPRPP PorcineTritrpticin VRRFPWWWPFLRR Porcine

aOne letter amino acid code is shown to indicate peptide sequences. Positive charged residues (R, K) are indicated in red;aliphatic hydrophobic (I, L, V) in blue; and aromatic hydrophobic (W, F) in green.

acid composition: the residues most abun-dant in AMP are cationic (arginine and lysine)and hydrophobic (tryptophan, phenylalanine,leucine, and isoleucine) (see colored residues inTable 1); (ii ) net charge: most AMP are pos-itively charged at physiological pH, althougha minor subgroup includes anionic peptides;(iii ) amphipathicity: conferred by their aminoacid composition and arrangement; and (iv) aremarkable diversity of structures and confor-mations, including -helices, -sheets, non-conventional structures, or even extendedconformations (the latter two are speciallyabundant in short AMP). Most of these struc-tures are amphipathic and are induced underspecic experimental conditions that also facil-itate their interaction with lipid bilayers. In fact,biological membrane-induced amphipathicityof AMP is a hallmark property related to an-timicrobial activity (see below).

Some cationic AMP are enriched in certainamino acids. A highly abundant class is that ofpeptides rich in cysteines, which can form disul-de bonds that make peptide structure compactand remarkably stable against adverse biochem-ical conditions and protease degradation. Mostof the antimicrobial peptides found in plants be-long to this class (Ib-AMP1, Table 1). There ismuch sequence and structural diversity withinthis group, which includes animal, insect, andfungal defensins; plant proteins such as thion-ins, defensins, and lipid transfer proteins (LTP)(19, 50, 176); and antifungal proteins (AFP)

Amphipathic: anamphipathic peptidecontains both polarand nonpolar domains

AFP: antifungalprotein

Analog: a sequencederivative of a givenpeptide

isolated from certain fungal genera includingAspergillus or Penicillium (102, 123).

Other relevant groups rich in specicresidues are short cationic peptides with a highproportion in tryptophan (29), among whichrepresentative examples are Indolicidin, withactivity against phytopathogens, or Tritrpticin(Table 1). Likewise, there are peptides enrichedin histidines such as histatins, or in prolines suchas PR-39.

DESIGN OF ANTIMICROBIALPEPTIDES AGAINSTPLANT PATHOGENS

The development of efcient methods for thesynthesis of peptides and their analogs, pep-tide collections and libraries, has resulted insubstantial advances in the identication andcharacterization of bioactive peptides. Ratio-nally designed peptide modications, additions,or deletions can be assayed for their effectson peptide activity with a view to improvingtheir properties in terms of specicity againstpathogens, reduced toxicity against plant or an-imal cells, greater stability, or modulation of thespectrum of action. In the rst steps of the pro-cedure, assays are performed by using suitablein vitro methods.

In most cases, the primary screen for pep-tide activity is a cell-based assay of in vitrogrowth inhibition that enables miniaturizationin the high-throughput formats required for the

www.annualreviews.org Rational Design of Novel Antimicrobial Peptides 275

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

analysis of complex collections of compounds(i.e., libraries) (20, 153). Most of the AMP as-sayed in vitro have 50% inhibitory and min-imum completely inhibitory concentrations(IC50 and MIC, respectively) in the range of1 to 20 M. Additional data such as those rel-ative to inhibition of fungal spore germinationor to killing activity have been also reported.These in vitro procedures, if conducted on non-target microorganisms or a panel of differentpathogens, can also be used as a means to eval-uate peptide specicity or the activity prole(96). Although still limited, there are also ex-amples of peptide screens that were targetedto specic pathogen functions (see below) (12,111, 157, 158).

Peptide toxicity has been evaluated mostlyby assaying the cytolysis toward human redblood cells. Although membrane-active pep-tides such as melittin are highly cytolytic inthese assays, caution should be taken in inter-preting the results, as the salt concentration ofthe buffers used to evaluate hemolysis is ex-pected to block the electrostatic interactionsthat most peptides need to interact with mi-crobes (see below). Toxicity evaluations shouldtherefore be conducted under conditions mim-icking the environment that the peptides willencounter in the host plants. In selected exam-ples, assays have been used that are based onin vitro pollen or seed germination and proto-plast viability in the presence of peptides (34,65, 187).

Regarding peptide stability, extracts ob-tained from different plants or plant tissues (26,133, 143), supernatants of cultures of pathogens(34, 127), or commercially available proteases(46) are well suited to determine peptide sus-ceptibility to protease degradation and inu-ence on peptide activity.

Modification of Natural Peptides

A primary strategy in the design of novel AMPis the modication of peptides found in na-ture. Numerous studies have focused on at-tempts to diminish nonspecic toxicity whileretaining the desirable bioactivity through

the design of sequence analogs or hybridpeptides.

Cecropins, magainins, and melittin(Table 1) are well-characterized membrane-active AMP (9) whose toxic properties havemade their practical use difcult. Cecropins,a broad class of antibacterial peptides, wereamong the rst in which attempts were made tomodulate the ratio of activity against microbesvs toxicity to plant cells. Two sequence analogsof cecropin B with reduced toxicity to plantprotoplasts, SB-37 and Shiva-1 (Table 2),were engineered that differed in their primarysequences. Each maintained the amphipathicproperties of the parental molecule andshowed differential activity against distinctplant pathogenic bacteria (66, 126).

Melittin is a cytolytic peptide from honeybee venom (Table 1). It has been widely stud-ied as a model for the interaction of lytic pep-tides with biological membranes, and also toidentify determinants of toxicity to human cellsand selectivity to bacteria (147). It has alsogiven rise to a signicant number of AMPused in plant protection. The screening of non-hemolytic analogs of melittin resulted in theidentication of peptides capable of inhibitingthe infection of tobacco leaves by tobacco mo-saic tobamovirus (TMV) (100). These peptidesshowed limited sequence homology with theTMV capsid protein and interacted abnormallywith TMV RNA.

Fusions of fragments of natural AMP havebeen designed in an attempt to endow the re-sulting chimeras with desirable properties fromthe parentals. A well-documented instance isthat of cecropin::melittin hybrids (Table 2). Ex-amples are the peptides CEMA and its deriva-tive MsrA1 (132); CAMEL with activity againstdifferent species of Pectobacterium (74); or Pep1which is active against distinct phytopathogenicfungi (26). Likewise, cecropin A::magainin hy-brids have been designed that show antibacte-rial activity but do not produce hemolysis (81).

Cecropins are not stable in the presenceof plant extracts (107), and indolicidin orLL-37 is degraded by bacterial proteases (129,159). The N terminus of cecropin A is cleaved

276 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

Table 2 Antimicrobial peptide analogs with activity against phytopathogens

Peptidea Amino acid sequenceb Pathogen ReferencesSB-37 (Cec B) MPKWKVFKKIEKVGRNIRNGIVKAGPAIAVLGEAKALG Bacteria (126)Shiva-1 (Cec B) MPRWRLFRRIDRVGKQIKQGILRAGPAIALVGDARAVG Bacteria (66)MrsA1 (Cec A::Mel) MALEHMKWKLFKKI::GIGAVLKVLTTGLPALKLTK E. carotovora, F. solani, P. cactorum (132)CAMEL (Cec A::Mel) KWKLFKKI::GAVLKVL Bacteria (74)Pep1 (Cec A::Mel) KWKLLKKI::GAVLKVL Fungi, P. infestans (26)P18 (Cec A::Mag) KWKLFKKI::PKFLHLAKKF Bacteria, F. oxysporum (81)Pep3 (Cec A) WKLFKKILKVL Fungi, P. infestans (26)BP76 (Cec A) KKLFKKILKFL Bacteria (46)MB39 (Cec B) HQPKWKVFKKIEKVGRNIRNGIVKAGPAIAVLGEAKALG Bacteria, fungi (133)Myp30 (Mag) MGIGKFLREAGKFGKAFVGEIMKP E. carotovora, P. tabacina (84)MSI-99 (Mag) GIGKFLKSAKKFGKAFVKILNS Bacteria, fungi (2)Rev4 (Ind) RRWPWWPWKWPLI P. tabacina (83)10R (Ind) RRPWKWPWWPWRR E. carotovora, fungi (11, 42)11R (Ind) RWRRWPWWPWRRK E. carotovora, fungi (11, 42)MsrA2 (Dermaseptin B1) MAMWKDVLKKIGTVALHAGKAALGAVADTISQ E. carotovora, fungi (131)MsrA3 (Temporin A) MASRHMFLPLIGRVLSGIL Bacteria, fungi (130)PV5 (Polyphemusin I) MRRYCYRKCYKGYCYRKCR E. carotovora, fungi (10)MBG01 (Rs-AFP1) ARHGSCNYVFPAHKCICYF F. culmorum (158)LfcinB17-31 (LF) FKCRRWQWRMKKLGA Fungi (122)

aPeptide name and in parenthesis peptide parental. Cec: cecropin; Mag: magainin; Mel: melittin; Ind: indolicidin; LF: lactoferrin.bResidues changed relative to the parental sequence are underlined. Other amino acid sequence details as in Table 1.

by extracellular proteases from the conidia ofA. avus, which may explain the lack of activityagainst the spores of this fungus but not againstothers (13). Distinct analogs of natural AMPwith greater resistance to in vitro degradationhave been described. Examples include MB39from cecropin B, Pep3 from cecropin A, andMyp30 from magainin 2 (Table 2). A furtherderivative of Pep3 is BP76, which exhibits im-proved activity against phytopathogenic bacte-ria, reduced hemolysis, and lower susceptibilityto proteinase K (46). Notably, the indolicidinreverse sequence Rev4 showed increased sta-bility in the presence of plant extracts and com-mercial protease cocktails, as well as activity invitro toward Peronospora tabacina (83).

Sequence modication of natural peptideshas also resulted in increased potency against(selected) pathogens. The magainin 2 deriva-tive MSI-99 (Table 2) has more positive chargeand antibacterial activity than the parental pep-tide, while activity against lamentous fungi

and oomycetes was maintained (2). The indoli-cidin analog CP-11 (42) has increased positivecharge and amphipathic properties that resultin greater antifungal and antibacterial activities.Further derivatives of this are peptides 10R and11R (Table 2).

Peptide cyclization is a remarkable modi-cation of synthetic peptides that in some caseshas a dual effect. It has resulted in increasedstability and also selectivity against microbesdue to conformational constraints that facili-tate the amphipathic separation of residues inthe molecule. Initially, this approach was ap-plied mostly to cationic tryptophan-rich pep-tides (33, 156), but also has been extended toother peptide classes (112, 113). Although thistype of modication precludes peptide produc-tion through biotechnological approaches, theincreases in stability and antibacterial activityare remarkable and should be considered if pep-tides of this class are to be used as phytosanitaryproducts for eld or postharvest treatments.


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

Peptide library: anordered collection ofpeptides or peptidemixtures that containsand represents thesequence diversity ofpeptides of a given sizeor property. Theintrinsic order andorganization of apeptide library permitindividual bioactivesequences to beidentied underappropriate assayconditions

Rational Designof Antimicrobial Peptides

The rational design of novel peptides is basedon knowledge of the biophysical and structuralproperties of natural AMP. There are numerousexamples in the biomedical eld, but not manyhave been tested for their utility in plant protec-tion. ESF1 and ESF12 (Table 3) are -helicalpeptides that mimic the charge distribution andstructure of magainins. They were developed instudies that showed that a reduction in positivecharge in their structural scaffold decreases an-timicrobial activity, whereas an increase in hy-drophobic residues is correlated with unspeciccytotoxicity and therefore should be avoided(143, 144). Peptides GR7 and SA3 (Table 3)were derived from ESF1 and have higher am-phipathicity and net charge, and reduced theinhibitory concentration toward F. oxysporumdown to 0.1 M (40).

A similar rational design approach was takento develop D4E1 and D2A21 (Table 3). Theactivities of these peptides were compared withthose of cecropin B and magainin II, showingthat they are more active and have no adverseeffect on pollen and seed germination (65).In contrast to most of the examples describedabove that are -helical, D4E1 presents -sheet

folding. Interestingly, it is also one of the pep-tides that have been successfully expressed in alarger number of plant species (see below).

Synthetic Peptide Libraries

The tools of synthetic combinatorial chemistryapproaches have been harnessed to synthesizeand assay vast sources of molecular diversity forthe identication of lead bioactive compounds(14). From hundreds to millions of compoundscan be screened in much shorter times thanwith traditional methods. Despite the disadvan-tage of their high cost, reports of their use inagriculture and food applications are increas-ing (76, 86).

The so-called nondened syntheticpeptide libraries do not have sequence re-strictions and represent all possible sequencecombinations of a peptide of a given size. Theyare collections arranged as mixtures of peptidesand a deconvolution procedure is needed toidentify the individual active peptide(s) (14).There are two main deconvolution strategies:iterative and positional (see SupplementalFigure 1. Follow the Supplemental Materiallink from the Annual Reviews home page athttp://www.annualreviews.org). A cell-based

Table 3 Rationally designed AMP against phytopathogens

Peptide Amino acid sequencea Pathogen ReferencesESF1 MASRAAGLAARLARLALRAL Bacteria, fungi (143)ESF12 MASRAAGLAARLARLALR Bacteria, fungi (143)GR7 MASRAARLAARLARLALRAL Bacteria, fungi (40)SA3 MAARAARLAARLARLALRAL Bacteria, fungi (40)D4E1 FKLRAKIKVRLRAKIKL Bacteria, fungi (34)D2A21 FAKKFAKKFKKFAKKFAKFAFAF Bacteria, fungi (155)PAF26 RKKWFW Fungi (96, 120)BM0 RFWWFRRR Fungi (111, 120)PPD1 frlhf Fungi (154)66-10 frlkfh Fungi (154)77-3 frlkfhf F. sambucinum (53)BP100 KKLFKKILKYL Bacteria (8)ACHE-I-7.1 SINWRHH Nematodes (182)Pc87 ADRPSMSPT P. capsici (12)

aAmino acid residues in lower case indicate D-estereoisomers. Other amino acid sequence details as in Table 1.

278 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

assay of in vitro growth inhibition towardfour different phytopathogenic fungi was usedon an iterative library to identify one pen-tapeptide (PPD1) and one hexapeptide (66-10)(Table 3). Likewise, a positional hexapeptidelibrary was assayed against the postharvestfungal pathogen Penicillium digitatum, iden-tifying PAF26 (Table 3). In this latter casepeptides were also assayed against nontargetmicroorganisms (the bacteria Escherichia coliand the unicellular fungus Saccharomyces cere-visiae) to identify peptides with distinct spectraof antimicrobial activity.

In these approaches it is common to usepeptides synthesized with the d-enantiomers ofamino acids because of their resistance to degra-dation (14). The d- or l-versions of this shortAMP do not present signicant differences interms of antimicrobial potency (111, 120, 181),which could be related to their mode of ac-tion and is also a prerequisite for productionthrough biotechnology.

To reduce costs and restrict the biological,sequence, or structural properties of peptides,one option is the use of dened libraries (15).For instance, a dened octapeptide library thatincluded an amino acid motif for attachment tothe yeast surface was screened for yeast growthinhibition and, secondarily, toward a S. cerevisiaeplasma membrane ATPase as a target, to iden-tify the octapeptide BM0 (Table 3) (111),which later showed activity against fungalphytopathogens (120).

In other cases, dened libraries are builtupon lead peptides of known activity previouslyidentied by nondened approaches; this canbe considered as a further step in the pipelineof combinatorial AMP identication (seeSupplemental Figure 1). Heptapeptides 77-3(Table 3) and 77-12 are derivatives of 66-10,and were identied against the potato pathogenFusarium sambucinum (53). Likewise, a seriesof heptapeptide derivatives of PAF26 was as-sayed against a panel of phytopathogenic fungito show distinct activity proles of specic pep-tides (121). That Magnaporthe grisea was thefungus most dissimilar in terms of sensitivity topeptides indicates the existence of distinct fun-

gal components that modulate the interactionwith peptides. A remarkable recent example isthe identication of BP100 (Table 3) from a125-member library derived from BP76, whichwas assayed for antibacterial activity, low cyto-toxicity and protease degradation, and in vivoprotection in a detached ower assay (8).

In a distinct approach of a dened libraryuse, scrambling of positions 14 of a cyclicdecapeptide allowed the identication of im-proved AMP against Erwinia amylovora thatshowed low hemolysis (112).

Peptide Libraries Producedthrough Biotechnology

Peptide libraries can also be generated andproduced in vivo through biotechnologicalmethodologies. Peptide library production andactivity may then be induced and identied ei-ther in cis, against the producer microorganism(suicide strategy) (30), or in trans, against a tar-get microbe different from the producer (153).The strain producer of the bioactive peptidecan be subsequently cloned and used to iso-late the peptide sequence. Theoretically, thesestrategies enable the bioassay of more pep-tide complexity than the synthetic procedures(i.e., longer peptides), but to our knowledge noexamples of use in plant protection have yetbeen reported. These strategies can also incor-porate codon-shufing methodologies for thedirected evolution and improvement of antimi-crobial proteins (151), a procedure for whichtheir small size seems to be particularly wellsuited.

Biotechnological approaches for peptide li-brary production also allow bioassays otherthan cell-based ones to be used for peptide iden-tication. Phage-display techniques are broadlyknown for the selection of peptides and pro-teins with binding afnity toward biomolecules.The peptide ACHE-I-7.1 (Table 3) was iden-tied from a phage library as an inhibitorof acetylcolinesterase, a target of pesticidesthat disrupt chemosensing in nematodes (182),and was successfully expressed in potato tosuppress parasitism by cyst nematodes (87).


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

Peptides with afnity for zoospores of Phytoph-thora capsici were also isolated by phage displayunder the assumption that binding to spore en-velopes would alter the normal interaction withplants and thus the life cycle of the pathogen(12) (Table 3), as was conrmed later (44).

The yeast two-hybrid system to identifyinteracting protein domains in vivo was usedto isolate aptamers with afnity for a replica-tion protein of tomato golden mosaic gemi-nivirus (98). A rich diversity of peptides (around20 residues in length) that successfully inhib-ited viral DNA accumulation in tobacco proto-plasts was obtained. Their sequence compari-son allows the identication of peptide motifsthat would confer broad protection againstgeminivirus. In a similar but more dened strat-egy, a random peptide library was generatedfrom fragments of the capsid protein of tomatospotted wilt tospovirus (TSWV), and the yeastsystem enabled the identication of a 29-residue aptamer that conferred resistance totospoviruses (157).

Antimicrobial Peptides Derivedfrom Proteins

Short AMP whose sequence belongs to plantantimicrobial proteins were identied in exper-iments to map the minimal active domains ofthe full-length proteins. Peptide MBG01 froma Raphanus defensin (Table 2) is as potent as thefull protein and allowed the design of sequencemodications to improve its properties. A sim-ilar strategy was followed with a thionin andafterwards generated a series of rationally de-signed derivates with strong activity (178, 179).Most of these designed sequence modicationsare nonnatural and thus probably of interest tothe biomedical eld, but their application tospecic plant protection problems should notbe discarded.

In recent years, food-related proteins andtheir hydrolysates have been intensively stud-ied as source of AMP for their obvious prac-tical implications (138). Lactoferricin (Lfcin)is the antimicrobial core of the milk pro-tein lactoferrin (LF) (45). Distinct peptides de-

rived from Lfcin have antimicrobial propertiesagainst bacteria, fungi, or viruses (45, 59), andhave demonstrated antifungal activity againstplant pathogens (Table 2). There are exam-ples of recombinant LF-related proteins pro-duced in plants, with the objective of generatingtransgenic plants either with resistance pheno-types or as production factories for molecularfarming (82, 124, 192).

Novel Strategies

Although not yet applied to plant protection,alternative in silico strategies could be a novelsource of AMP not found in nature. Virtual as-say approaches, for example, are capable of pre-dicting the antimicrobial activity of peptide se-quences, thus reducing the number of bioassaysto be conducted in the real world (153, 163).The rational design of new peptides has beenalso achieved by dening grammar rules thatdescribe the presence of amino acid residueswithin an AMP (94). Both examples are bioin-formatic tools that take advantage of the wealthof information in AMP public databases, re-garding peptide sequences, structure activityrelationships, and mode of action (see RelatedResources).

The accompanying SupplementalFigure 1 (Follow the Supplemental Materiallink from the Annual Reviews home page athttp://www.annualreviews.org) summarizesthe strategies described to identify novelantimicrobial peptides (AMP) by rationaldesign.

MODE OF ACTION AGAINSTMICROBIAL CELLS

A detailed knowledge of the mode of action ofAMP is essential if the goal of applying them toplant protection is ever to be reached. In someof the examples cited above, the activity screensconducted implicitly indicate the mechanism bywhich the peptides affect the pathogen. How-ever, for most of the cases in which activity wasdetermined as cell-based growth inhibition as-says of bacteria or fungi, the peptide mode of

280 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

action needs further investigation. Of note isthe elucidation of the mechanisms that conferpeptide specicity toward phytopathogenic mi-croorganisms and that avoid toxicity to animalsor plants. Such information should aid furtheridentication and design of novel peptides. Sev-eral excellent reviews address these questions inthe biomedical eld (21, 191); here we summa-rize only those specic examples that are rele-vant to phytopathology.

Interaction with Microorganisms

The rst step in the AMP mode of action isthe physical interaction with outer structuresthat surround the microbial cell. In general, itis assumed that such interaction is not stereo-specic since it has been shown in selected ex-amples that peptides synthesized with eitherthe l- or the d-enantiomers of amino acids donot differ substantially in antimicrobial activity(181).

The cationic properties of many AMP andproteins result in an electrostatic attraction tothe negative-charged microbial envelopes, suchas lipopolysaccharide (LPS) of gram-negativebacteria, which is a plausible explanation forpeptide specicity to microbes and lower tox-icity to animal and plant cells (41, 191). Dis-tinct experimental data support such an electro-static attraction. The peptide CEMA (Table 2)binds LPS, and this binding is correlated withthe differential antimicrobial activity of an ana-log with lower afnity for LPS (141). Mutantsof Ralstonia solanacearum with increased sensi-tivity to thionins and LTP present alterationsin a LPS biosynthetic gene (172). Regardingamino acid sequence, an increase in net posi-tive charge is linked with increased antimicro-bial activity (121, 143). Likewise, a set of ala-nine substitution analogues of PAF26 showedthat the higher contribution to its antifungalactivity results from its three cationic residues(121), thus reinforcing the importance of ionicattraction.

In general, the electrostatic attraction isperturbed by the increase of ionic strengthand, as occurs with plant antimicrobial proteins

(19, 50), peptide antimicrobial activity dimin-ishes with the addition of salt ions to the invitro assay medium. However, the activity ofselected AMP also depends on an ionic milieucomparable to that of mammalian body uids(38, 140). This issue must be taken into accountin relation to the composition of the microen-vironment, cell, or tissue location in which thepeptides are expected to act.

However, there are examples of antimicro-bial peptides and proteins whose interactionwith microbes either does not rely primarily onelectrostatic attraction or is aided by bindingto specic components. The use of the modelfungus S. cerevisiae has allowed the identica-tion of specic cell wall or membrane compo-nents that mediate the action of antimicrobialproteins. Thus, the plant antimicrobial proteinPR-5 (osmotin) effect on yeast is inuenced bycell wall glycoproteins (189). Also, the sensi-tivity of S. cerevisiae to distinct plant defensinsis determined by glycolipids specic to fungalmembranes (168, 170), and similar conclusionswere drawn from a study with the lamen-tous fungus Neurospora crassa (135). Mutantsof F. oxysporum and Aspergillus oryzae in chitinbiosynthesis had lowered sensitivity to the AFPfrom Aspergillus giganteus and its membranepermeabilizing effect (54).

Regarding AMP, the case of nisin is signif-icant inasmuch as it shares the membrane dis-rupting capabilities of many AMP (see below)that are promoted by a specic interaction ofnisin with a Lipid II precursor (18). Also, pep-tide D4E1 (Table 3) preferentially binds er-gosterol, an esterol that is a hallmark of fungalmembranes (34). Taken together, all these nd-ings indicate that although surface net charge isundoubtedly relevant to cationic AMP action,other factors can also modulate peptide inter-action with microbes.

Interaction with Synthetic Membranesand Membrane Mimetics

Most AMP have amphipathic properties thatenable their insertion into and disruption ofmodel lipid membranes, as determined by


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

structural and biophysical techniques (93, 145).Many studies correlate peptide antimicrobialactivity with effective interaction with articiallipid bilayers that would mimic those of targetmicroorganisms. Topological models have beenproposed to explain the interaction of AMPwith membranes (21). In summary, the propen-sity of a given peptide to interact and disrupt alipid membrane and the concrete topology ofthe membrane disruption depend on the sizeand structure of the peptide, membrane lipidcomposition, a ratio of peptide to lipid concen-tration, and folding capabilities of the peptide,among others.

The conclusions derived from such studiesand models have allowed the rational design ofsome AMP (34, 143, 155). However, the im-plicit rationale that peptide activity is a directconsequence of membrane-disturbing capabil-ities and thus microorganism cell permeation isa generalization that cannot always be used toexplain antimicrobial action (see below). Thisview stems from the fact that among the rstAMP characterized are examples of peptideswith a high propensity to disrupt lipid bilayers.Peptides that have such lytic properties clearlyalso have a higher probability of being toxic tonontarget cells.

In any case, it should be considered that therst line of interaction between a given AMPand the target microorganism is not the lipidbilayer but rather the outside structures and cellenvelopes, which can vary from one type of or-ganism to another.

Morphological Alterations

A number of studies have described distinctphysical alterations in the microbial cells afterexposure to antimicrobial peptides. Generally,the two most relevant and broadly documentedare alterations of shape and growth, and cellpermeation.

Distinct microscopy techniques have al-lowed the visualization of morphological andgrowth alterations in bacterial cell shape af-ter exposure to distinct AMP (21), and in themycelium of phytopathogenic fungi exposed

to short AMP (3, 4, 26, 89, 119, 149, 155).In selected examples, alterations described atthe ultrastructural level are indicative of celldegradation (79, 127, 155).

At concentrations above those that are com-pletely inhibitory, collapsed fungal hyphae werealso indicative of cell death. At lower peptideconcentrations, partial fungal growth inhibi-tion was reected in shorter interseptum dis-tances, thick hyphae, cell enlargement, mycelialaggregates, and/or abnormal branching pat-terns. The latter could be either hyperbranch-ing (3, 4, 89), or abnormal tip dichotomousbranching and aborted lateral branching (89,119). In some cases, fungal mycelium reactswith alterations in the deposition of chitin (116,119). Of note is that some of these alterationsare phenocopies of mutations in genes that areinvolved in controlling polar growth or biosyn-thesis of cell wall components (58, 101).

It is difcult to draw comparative conclu-sions among studies from different laboratoriesand thus it is open to debate whether distinc-tive alterations are indicative of distinct modesof action. Microscopic analyses have shownthat different AMP have different effects onPseudomonas aeruginosa cells, which indicatesthat they would have different targets or mech-anisms of activity (21, 73). In an analogous studyconducted in fungi, two sequence-related plantdefensins induced distinctive alterations andmolecular responses in F. graminearum (150).

Cell Permeabilization

Permeation of synthetic membranous vesiclescan be easily measured through the release of(uorescent) probes. For a substantial numberof AMP, their capability to insert into and desta-bilize lipid bilayers has been shown to correlatewith permeation of articial vesicles as well aswith antimicrobial activity (75, 169).

The use of specic uorescent probes allowsthe visualization of cell permeation of microor-ganisms after exposure to peptides. Assays arebased on coincubation of the microbe with thepeptide and the probe, which can penetrate thecell only if the plasma membrane is disturbed,

282 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

and once inside binds to specic cell compo-nents and emits uorescence. This emission canbe quantied and visualized. Data of this typehave been obtained with plant antimicrobialproteins (169), AFP proteins from Penicillium(115, 167), and with synthetic AMP (53, 119,122, 127, 154, 155), acting on phytopathogenicfungi. In most of these examples, the observa-tion of cell permeation is interpreted as the di-rect and primary effect of peptides that leads tocell killing.

However, other studies have argued thatthere is not always a complete correlation be-tween cell permeation and antimicrobial ac-tivity (41, 183). Signicant conclusions can bedrawn when permeation is evaluated at differentpeptide concentrations, since it becomes clearthat at the minimal lethal concentration notall peptides kill through membrane disruption(105). For instance, the cationic tryptophan-rich peptide BM2 did not cause yeast perme-ation at low concentrations that partially inhibitgrowth (111), and areas of fungal mycelium thatshow morphological alterations and growth in-hibition due to peptide exposure do not havedetectable permeation (119, 122). Specic pep-tides kill bacteria, but they either do not per-meate bacterial membranes or have additionalmodes of action (16, 25, 164).

These and other results (see below) sug-gest that specic AMP could have more subtlemechanisms of action not necessarily or primar-

CPP: cell-penetratingpeptides

ily linked to microbe permeation (21, 56, 93,185). A hallmark property in this regard is theability of some AMP to cross membranes andhave an intracellular mode of action (21, 55).

Intracellular Modes of Action

Specic cationic peptides rich in arginine oraromatic residues have the propensity to crossbiological membranes in a nondestructive man-ner, and translocate to the cell inside; they havebeen named penetratins or cell-penetratingpeptides (CPP) (61, 69). Initially, they were de-scribed as part of viral or homeotic intercellularsignaling proteins. Their biotechnological useas shuttles to deliver bioactive proteins to cellshas been proposed (160).

This translocation capability to either fun-gal or bacterial cells is being demonstrated foran increasing number of antimicrobial pep-tides, some of which with activity against plantpathogens (Table 4). In selected examples,internalization to fungal mycelium has beenshown at low concentrations at which no effecton growth inhibition or cell permeation is de-tected (119). Cell-penetrating properties havebeen also shown for AFP in phytopathogenicfungi (116) and also for plant defense proteins(27, 92).

The reverse direction has also been taken,and recent studies have derived antimicrobialproperties from peptides previously known onlyas CPP (72, 134, 193). These relationships

Table 4 Cell-penetrating antimicrobial peptides

Peptide Amino acid sequence AMPa CPPb

Penetratin RQIKIWFQNRRMKWKK (134) (37)Pep-1 KETWWETWWTEWSQPKKKRKV (193) (118)Tat (47-58) GRKKRRQRRRPPQ (72) (180)Apidaecin GNNRPVYIPQPRPPHPRI (24, 24a) (25)Buforin II TRSSRAGLQFPVGRVHRLLRK (136) (78)Indolicidin ILPWKWPWWPWRR (43, 80, 120) (165)LfcinB FKCRRWQWRMKKLGAPSITCVRRAF (122, 173) (60)Magainin 2 GIGKFLHSAKKFGKAFVALKAL (2, 80, 190) (103)PAF26 RKKWFW (96) (119)Polyphemusin I RRWCFRVCYRGFCYRKCR (109) (146)

aAMP: antimicrobial peptides. Literature references in parentheses report antimicrobial activity.bCPP: cell-penetrating peptides. Literature references in parentheses report cell translocation activity.


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

CP-AMP: cell-penetratingantimicrobial peptides

raise questions about the real differences be-tween specic groups of antimicrobial and cell-penetrating peptides (61), which in fact wouldbelong to a unique class for which the proposedname of cell-penetrating antimicrobial peptides(CP-AMP) seems appropriate. This questiondeserves to be explored in the near future sinceit perhaps holds the key to a novel source ofAMP sequences and information on the de-sign of peptides with improved activity and newproperties against (plant) pathogens.

Once inside the cell, some AMP may actas antimicrobials through specic mechanismsthat include binding to nucleic acids or the in-hibition of enzymatic activity or synthesis ofmacromolecules (proteins, nucleic acids, or cellwall components) (21, 52). Relevant examplesof peptides active against phytopathogens arethose of indolicidin and lactoferricin (165, 174).Due to the cationic nature of many antimi-crobial peptides and proteins, their afnity foranionic nucleic acids is expected and in vitrobinding has been shown in a number of cases(62, 116, 119). Although such activity in vivocould alter cell homeostasis severely, it remainsto be determined to what extent this propertymediates peptide antimicrobial action.

The steps reviewed above in the interac-tion of AMP with microorganisms and mech-anisms of action are modeled in Figure 1.Although interaction of AMP with lipid bi-layers and cell (cytoplasmic) membranes is anessential property and step in the killing pro-cess, other questions might be more descrip-tive of peptide specicity toward microbes, suchas the interaction with outer cell structures,ability to translocate across membranes, andwhether putative intracellular targets exist, andshould receive more attention in the futuredevelopment of novel peptides. For simplicitybut also because detailed knowledge is lack-ing, the model does not differentiate amongbacterial or fungal microorganisms. Selectedmicrographs with examples of some of thesteps described in the model are shown inSupplemental Figure 2 (Follow the Supple-mental Material link from the Annual Reviewshome page at http://www.annualreviews.org)

for the interaction between the synthetic pep-tide PAF26 (Table 3) and the phytopathogenicfungus P. digitatum.

RESPONSE TO ANTIMICROBIALPEPTIDES

It was initially assumed that the lytic mode ofAMP action, signicantly nonspecic, wouldprevent the development of resistance amongmicrobes, as occurs with other more specictherapeutics. However, this is not the case (137,140). In fact, it has been proposed that the greatdiversity of AMP in nature results from the co-evolution of hosts and microbes in a continuousprocess in which microorganisms mount resis-tance responses as a countermeasure to peptideaction and hosts diversify peptide structures tocircumvent resistance mechanisms (140, 185).Microbial resistance mechanisms to AMP canbe summarized as the secretion of extracellu-lar proteases or proteins capable of binding toand inactivating AMP, alteration of microbialsurface composition and net charge to preventattachment, and induction of active membranetransporters capable of extruding peptides fromcells. Conceptually, each of these mechanismscan be envisioned as blocking the different stepsleading to peptide interaction and action as de-scribed in Figure 1. Pathogenic bacteria couldactively induce these responses because theyhave specic sensor systems that are able to rec-ognize the presence of antimicrobial peptidesand activate signaling cascades with the poten-tial to change gene expression (7, 91, 104).

There are numerous reports related tohuman pathogen resistance and response toAMP, although examples on microbial phy-topathogens are still scarce. However, evidenceindicates that both plant and animal pathogensdeal in a similar way with host AMP (51). Ithas been suggested that the microorganism re-sponse could be specic to an antimicrobialpeptide type, which might be relevant in thestudy of plant-pathogen interactions (51). Ce-cropin A is degraded by conidial extracellularproteases from certain plant pathogenic fungibut not from others (13, 35). The ABC class of

284 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

e

b

a

c

d

i

g

Cell permeationand lysis

Intracellular action

Sen

sing

of A

MP

CW/CE

Out

In

f

Cell wall synthesis

Res

pons

e to

AM

P

h

Lipidbilayer

AMP

DNA RNA Protein

Figure 1Model of peptide interaction with microorganisms and mode of antimicrobial action. The proposal is basedon previous reports and considers recent ndings summarized in the text. As presented, it can accommodateprokaryotic and eukaryotic cells. Steps involved are outlined as follows. (a) AMP (in red, may beunstructured) interact with the outer cell envelope, being the electrostatic attraction the initial driving forceof the interaction. (b) The AMP can diffuse toward and bind to deeper lipid bilayers, where the biophysicalproperties of most AMP enable insertion into the bilayer (c), a process that is dependent of lipidcomposition. During this progression, the AMP folds owing to changes in the microenvironment. FoldedAMP might enhance their amphipathicity and thus their antimicrobial properties. At this point, AMP caneither translocate the bilayer and penetrate to the cytosol (d ), disrupt the membrane architecture and lead tocell permeation (e), or both; the balance among these alternatives is dependent on time and peptideconcentration, as well as on properties of peptide and cell components. ( f ) Internalized AMP can bind toDNA, RNA, and/or proteins, and disrupt DNA replication, RNA synthesis, or enzyme activity, dependingon the peptide. Intracellular action can lead to cell killing and subsequent cell permeation. ( g) Other modesof AMP action are disruption of cell wall synthesis, architecture, or cell morphology due to abnormalinteraction with outer cell components. Microorganisms have sensing mechanisms to detect and signalpeptide presence (h), a process that is required to change gene expression and protein activities to counteractpeptide action at the cytosol, membrane, cell envelope, or cell environment (i ).

membrane proteins is involved in the detoxi-cation of compounds by active extrusion fromcells. Some of the corresponding bacterial genesare related to susceptibility to AMP and in ad-dition involved in bacterial virulence to planttissues (99). Also, as mentioned above, modi-cation of the net surface charge can affect sen-sitivity to peptides (139, 172). A more general

approach has been taken to screen a collectionof mutants from the alfalfa symbiont Sinorhizo-bium meliloti for their increased sensitivity to anantimicrobial peptide (125). Of the seven genesidentied, three are involved in the biosynthesisof an exopolysaccharide accumulated on the cellsurface, and one is, again, an ABC membranetransporter.


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

The ndings that microorganisms can de-velop resistance to AMP, as they do to otherantimicrobials, could diminish the interest intheir use. However, knowledge of the mech-anisms affecting sensitivity/resistance to AMPwill surely help in designing more specic andeffective peptides, which could exploit multi-ple modes of action within a single molecule;for instance, membrane-permeating propertiescombined with binding to specic targets (140)(Figure 1). It must also be borne in mind thatAMP and proteins are an ancient and effectivedefense mechanism that has persisted duringevolution, most likely because hosts can depletean arsenal of different peptides with distinctmodes of action acting on a given microorgan-ism (140). Similarly, multiple peptides or mul-tidomain AMP could be used or engineered tomimic natures solution to this question.

The genomic characterization of the tran-scriptional changes of selected microorganismsshould increase our knowledge on the specicmicrobial response to the action of peptides (38,48). Likewise, the isolation of mutants with al-tered sensitivity to peptides will continue toprovide information on peptide targets and in-teracting counterparts, and thus will help toevaluate the feasibility of selection of resistantphenotypes.

USE OF ANTIMICROBIALPEPTIDES IN PLANTPROTECTION

Most of the initial work on AMP is based onin vitro data showing the activity of syntheticpeptides in growth media. These approachesoffer obvious advantages since they allow high-throughput formats to screen complex collec-tions of peptides quantitatively, with a mini-mum amount of peptide use, and thus exploitingone of the advantages of the work with shortpeptides, i.e., the capacity to analyze high di-versities and dened sequence modications.As such, these strategies are needed to identifycandidates and lead compounds, which subse-quently must be tested in plants to complete theidentication of potentially useful peptides.

Control of Plant Diseases throughAddition of Synthetic AMP

Interaction between peptides and pathogens inthe plant microenvironment is thought to bemodulated by complex factors that must affectpeptide activity. A rst step in introducing thesefactors in peptide selection and evaluating thecontrol of plant diseases through AMP is toconduct experimental coinoculations in whichpeptides and pathogen are applied to suscep-tible tissues. Such experiments can be carriedout on different candidate peptides of poten-tially useful in vitro properties.

Assays based on detached leaves or leak disks(2, 4), owers (8), fruits (90, 95, 96), or potatotubers (4, 74) have been conducted with pos-itive results. The protective effect reported islimited in most of these examples, likely dueto the fact that peptides are applied locallyonto the plant surface or the inoculation pointand, therefore, the pathogen escapes the AMPaction if it grows outside this area. In general,in all these reports the reduction of disease in-cidence or progression correlates with the invitro activity of a given peptide. However, a re-cent comparative study showed that in vitro in-hibitory activity of a set of eight distinct AMP isnot correlated with their capacity to retard fruitdecay caused by fungi (120).

There are examples in which these exper-imental bioassays are close to the real plantpathology application. A case study is posthar-vest fruit diseases, for which AMP could beused as an additional postharvest additive (89,90, 95), and synthetic peptides that incorporatenonnatural peptide modications to increasetheir properties (as noted above) could be con-sidered provided that the production costs areminimized.

Combined Action of AMPwith Other Antimicrobials

Some AMP are active against natural iso-lates of fungi that have developed resistanceagainst common commercial fungicides (53,97), a property that reinforces their poten-tial use as an alternative to the phytochemicals

286 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

currently used. Moreover, the synergistic ac-tivity reported for synthetic peptides andthiabendazole against Fusarium suggests thatAMP could be tools to lower the dosageand thus the residue level of fungicides (53).Although still unexplored, the combined actionof distinct antimicrobial strategies to arrive atmore environmentally friendly control prac-tices could incorporate the use of AMP at sub-lethal concentrations.

A signicant example is the combination ofAMP with biocontrol microorganisms, a pre-requisite being the absence of toxicity of thepeptide against them (95). In one case, a ge-netically modied biocontrol agent could beeven used to produce and release the AMP(see below). The model yeast S. cerevisiae trans-formed with a gene fusion that secreted pep-tide Pep3 (Table 2) behaved as an antagonisticorganism that protected tomato fruits againstColletotrichum decay (71).

Transgenic Expression of AMPin Plants and Disease Resistance

An obvious advantage to the use of peptidesas antimicrobials is their in situ productionthrough biotechnology. The overexpression oftransgenes encoding plant antimicrobial pro-teins has been demonstrated as a successful ap-proach to protect plants against diseases causedby microorganisms (22, 49, 110). Likewise,genes from nonplant origins have been trans-ferred, including fungal AFP (128) and LF fromhuman or bovine origin (166, 192).

To date, the structure and properties ofmany AMP have been studied, and a num-ber of them have been analyzed for the effectsof heterologous expression on plant resistanceto bacterial and fungal phytopathogens. Trans-genic plants expressing AMP found in naturehave been shown to exhibit broad-spectrumdisease resistance (5, 47, 67, 142). Likewise,rationally designed AMP also showed stableexpression and antimicrobial activity in planttissues (Table 5).

Different synthetic peptides provided resis-tance against E. carotovora in potato (130, 131,

132). In the case of the chimera MrsA1, ab-sence of toxicity was inferred by feeding micewith transgenic potato tubers (132). Two sig-nicant examples of peptides that have beenexpressed in different plants and conferred pro-tection against distinct microbial pathogensare D4E1 and MSI-99 (see Table 5). Alsoremarkable are indolicidin and polyphemusinvariants that showed broad-spectrum enhancedresistance to different bacterial, fungal, andviral pathogens when expressed in tobacco(10, 11, 126).

One potential problem with transgenic ex-pression of peptides is low stability due to smallsize and susceptibility to protease degradation.Some contradictory published data on the pro-tective effect of peptide transgenes in plantsmay be attributable to different rates of peptidedegradation by endogenous peptidases (133).Differences of stability are likely in relationto peptide structure and plant species. There-fore, strategies are needed to optimize produc-tion through stability of expressed peptides intransgenic plants. Rational design and molecu-lar modeling of peptides are effective methodsto increase peptide stability without compro-mising activity (see above) (130132). Anotheralternative is directing the peptide to speciclocations where protease activity is presumedto be low (31, 84, 161). An additional interest-ing approach is to mimic natural plant systemsfor the delivery of small-sized peptides, as engi-neered with the peptide sequence Pep11, whichreplaced systemin in the prosystemin polypep-tide and was successfully expressed in tomato(70).

In vitro data does not have to be completelydenitive in order to evaluate the potential util-ity of a given AMP in planta. The peptide MsrA3had only a modest potency in vitro but gave pos-itive results in transgenic potatoes that allowedprolonged tuber storage (130). By contrast, inthe case of highly toxic (unspecic) peptides,the use of plant gene promoters that induce ex-pression only under pathogen attack has pro-duced positive results in different pathosystems(88, 108, 186188). In fact, nonconstitutive pro-duction at specic times or plant tissues could


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

Table 5 Rational-designed antimicrobial peptides expressed in plants

Peptide Host Pathogen(s) ReferencesShiva-1 Tobacco (N. tabacum) P. solanacearum (66)

Potato (S. tuberosum) E. carotovora (188)

Paulownia tomentosa Phytoplasma (39)MB39 Tobacco (N. tabacum) P. syringae (63)

Apple (M. domestica) E. amylovora (88)SB-37 Potato (S. tuberosum) E. carotovora (6)MsrA1 Potato (S. tuberosum) E. carotovora, P. cactorum, F. solani (132)D4E1 Tobacco (N. tabacum) C. destructivum (23)

Poplar (Populus sp.) A. tumefaciens, X. populi (106)

Cotton (G. hirsutum) T. basicota (148)Myp30 Tobacco (N. tabacum) E. carotovora, P. tabacina (84)ESF12 Poplar (Populus sp.) S. musiva (85)MSI-99 Tobacco (N. tabacum) Bacteria, fungi (28, 36)

Grapevine (V. vinifera) Bacteria, fungi (177)

Banana (Musa sp.) F. oxysporum, M. musicota (28)

Tomato (L. esculentum) P. syringae (1)MsrA3 Potato (S. tuberosum) E. carotovora, P. infestans, P. erythroseptica (130)Pep11 Tomato (L. esculentum) P. infestans (70)CEMA Tobacco (N. tabacum) F. solani (187)MsrA2 Potato (S. tuberosum) E. carotovora, fungi (131)

Tobacco (N. tabacum) Bacteria, fungi, oomycetes (186)ACHE-I-7.1 Potato (S. tuberosum) G. pallida (87)Rev4 Tobacco (N. tabacum) Bacteria, oomycetes (184)

Arabidopsis thaliana10R, 11R Tobacco (N. tabacum) E. carotovora, fungi, TMV (11)PV5 Tobacco (N. tabacum) E. carotovora, fungi, TMV (10)

minimize the risk for the emergence of newstrains of microorganisms that are resistant tothese candidate peptides.

Future Alternativesto the Production of AMP

To exploit the activity of AMP for the treatmentof human and animal diseases, large amountsof peptides must be produced efciently andat reasonable cost (57). To reduce the highproduction costs of synthetic AMP researchershave turned to recombinant expression in het-erologous systems such as microorganism ascell factories. Genetically modied microor-ganisms can be engineered to produce and ob-tain suitable amounts of AMP (64), including

peptides that have demonstrated activity againstplant pathogens (77, 117). In the plant pro-tection scenario, peptides obtained in this waycould be part of formulates of phytosanitaryproducts used in the eld during plant cultiva-tion or postharvest additives used on harvestedcommodities.

AMP-producing transgenic plants not onlyare protected against pathogen infection butalso could be used as production platforms ofantimicrobial proteins in molecular farming ofcrops. Data obtained from AMP transgenic ex-pression for plant disease resistance can also beexploited for this purpose. Several factors re-sponsible for functional peptide expression intransgenic plants have been tested includingpromoter sequences (186, 187), codon usage

288 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

(49, 132), fusion to signal peptide sequences (10,11), and toxicity to the expressing plants (11).As a result, signicant progress is being made inovercoming the problems involved during theheterologous production of AMP in plants, andyields in the range of 1 to 10 g of peptide pergram of fresh leaf tissue have been achieved. Fu-ture options for improvement exist. Many plantantimicrobial proteins are produced and storedin seeds (171), which have obvious advantagesas plant-based production systems (17). A sig-nicant example is that of the antimicrobial LF,which was produced in rice grains from whichit can be extracted with signicant purity andyield (124).

The incorporation of AMP into plantsthrough genetic engineering offers a means toprevent disease-associated losses as well as to

protect the environment. However, public re-sistance to the production of genetically modi-ed plants is still an important factor. SelectedAMP have been expressed in the chloroplastgenome in order to conne the synthetic genewithin the transgenic plant (36). Although thepotential undesirable toxic effects of AMP, ifany, need to be further investigated, it is em-phasized that the small size of this class ofcompounds and the knowledge generated inthe studies reviewed herein and in future con-tributions should help to modulate this andother peptide properties through peptide se-quence modication. The successful applica-tion of AMP to plant protection will likely helperadicate certain plant diseases, reduce the envi-ronmental degradation of intensive agriculture,and improve the quality and safety of our food.

SUMMARY POINTS

1. Antimicrobial peptides are promising alternative strategies for use in plant diseasecontrol.

2. Peptide libraries and the rational design of peptide analogs are tools that can exploitpeptide sequence diversity to identify novel and improved peptides specic against plantpathogens.

3. Transmembrane pore formation and cell permeation are not the sole mechanisms ofmicrobial killing exerted by antimicrobial peptides. Interaction with cell envelopes andmore subtle intracellular modes of action could be equally important in explaining peptidespecicity and potency.

4. Antimicrobial peptides have been demonstrated to confer disease protection when pro-duced by transgenic plants.

FUTURE ISSUES

1. Isolation, identication, and improved design by rational methods of novel AMP targetedto plant protection, with higher specicity toward plant pathogens, greater stability inplant microenvironments, and reduced nonspecic toxicity.

2. The increased use of suitable high-throughput technologies to boost the number anddiversity of peptides that can be analyzed.

3. A shift from cell-based growth inhibition toward more specic screenings using molecularapproaches to enhance the properties of new AMP.


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

4. Elucidation of the mode of action of selected AMP to facilitate the knowledge-baseddesign of peptides.

5. Application of functional genomics tools such as array technology to monitor gene ex-pression changes, the isolation of mutants, or screens of genome scale collections of genedeletions, to aid in the identication of the modes of action, cell targets, and microberesponses to AMP.

6. Identication in phytopathogenic microorganisms of biochemical/molecular targets thatare related with their virulence or pathogenicity to plants and also amenable to screensof peptide collections.

7. Increased number and diversity of experiments to achieve the safe transgenic expressionof AMP in plants.

8. Heterologous production of AMP using microorganisms as cell factories or in plants bymolecular farming.

DISCLOSURE STATEMENT

The authors are not aware of any biases that might be perceived as affecting the objectivity of thisreview.

ACKNOWLEDGMENTS

We apologize to all the investigators whose research could not be appropriately cited because ofspace limitations. We extend special thanks to our collaborators and laboratory members for helpand thoughtful discussions. Work on the use of antimicrobial peptides in plant protection has beensupported by grants BIO2003-00927 and BIO2006-09523 (Spain) to J.F.M.; BIO4-CT97-2086(European Union) to E.P.-P.; the National Centres of Excellence and AFMnet (Canada) to S.M.;and GV06-283 (Comunidad Valenciana) to B.L.-G.

LITERATURE CITED

1. Alan AR, Blowers A, Earle ED. 2004. Expression of a magainin-type antimicrobial peptidegene (MSI-99) in tomato enhances resistance to bacterial speck disease. Plant Cell Rep.22:38896

2. Alan AR, Earle ED. 2002. Sensitivity of bacterial and fungal plant pathogens to the lyticpeptides, MSI-99, magainin II, and cecropin B. Mol. Plant-Microbe Interact. 15:7018

3. Ali GS, Harman GE, Reisch BI. 2003. The interaction of endochitinase, a synthetic peptideand resveratrol in controlling fungi in vitro. Eur. J. Plant Pathol. 109:63944

4. Ali GS, Reddy AS. 2000. Inhibition of fungal and bacterial plant pathogens by syntheticpeptides: in vitro growth inhibition, interaction between peptides and inhibition of diseaseprogression. Mol. Plant-Microbe Interact. 13:84759

5. Allefs SJHM, Dejong ER, Florack DEA, Hoogendoorn C, Stiekema WJ. 1996. Erwiniasoft rot resistance of potato cultivars expressing antimicrobial peptide tachyplesin I. Mol.Breed. 2:97105

290 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

6. Arce P, Moreno M, Gutierrez M, Gebauer M, DellOrto P, et al. 1999. Enhanced resistanceto bacterial infection by Erwinia carotovora subsp atroseptica in transgenic potato plantsexpressing the attacin or the cecropin SB-37 genes. Am. J. Potato Res. 76:16977

7. Bader MW, Sanowar S, Daley ME, Schneider AR, Cho US, et al. 2005. Recognition ofantimicrobial peptides by a bacterial sensor kinase. Cell 122:46172

8. Badosa E, Ferre R, Planas M, Feliu L, Besalu E, et al. 2007. A library of linear undecapeptideswith bactericidal activity against phytopathogenic bacteria. Peptides 28:227685

9. Bechinger B. 2004. Structure and function of membrane-lytic peptides. Crit. Rev. Plant Sci.23:27192

10. Bhargava A, Osusky M, Forward BS, Hancock REW, Kay WW, Misra S. 2007. Expressionof a polyphemusin variant in transgenic tobacco confers resistance against plant pathogenicbacteria, fungi and a virus. Plant Cell Tiss. Organ Cult. 88:30112

11. Bhargava A, Osusky M, Hancock REW, Forward BS, Kay WW, Misra S. 2007. Antiviralindolicidin variant peptides: evaluation for broad-spectrum disease resistance in transgenicNicotiana tabacum. Plant Sci. 172:51523

12. Bishop-Hurley SL, Mounter SA, Laskey J, Morris RO, Elder J, et al. 2002. Phage-displayedpeptides as developmental agonists for Phytophthora capsici zoospores. Appl. Environ. Micro-biol. 68:331520

13. Bland JM, De Lucca AJ. 1998. Identication of Cecropin A proteolytic cleavage sitesresulting from Aspergillus avus extracellular protease(s). J. Agric. Food Chem. 46:532427

14. Blondelle SE, Pinilla C, Boggiano C. 2003. Synthetic combinatorial libraries as an alterna-tive strategy for the development of novel treatments for infectious diseases. In Methods inEnzymology: Combinatorial Chemistry, Part B, ed. GA Morales, BA Bunin, 369:32244. SanDiego: Academic

15. Blondelle SE, Takahashi E, Houghten RA, Perez-Paya E. 1996. Rapid identication ofcompounds with enhanced antimicrobial activity by using conformationally dened com-binatorial libraries. Biochem. J. 313:14147

16. Boman HG, Agerberth B, Boman A. 1993. Mechanisms of action on Escherichia coli of ce-cropin P1 and PR-39, two antibacterial peptides from pig intestine. Infect. Immun. 61:297884

17. Boothe JG, Saponja JA, Parmenter DL. 1997. Molecular farming in plants: oilseeds asvehicles for the production of pharmaceutical proteins. Drug Dev. Res. 42:17281

18. Breukink E, Wiedemann I, van Kraaij C, Kuipers OP, Sahl HG, de Kruijff B. 1999. Use ofthe cell wall precursor lipid II by a pore-forming peptide antibiotic. Science 286:236164

19. Broekaert WF, Cammue BPA, De Bolle MF, Thevissen K, De Samblanx GW, OsbornRW. 1997. Antimicrobial peptides from plants. Crit. Rev. Plant Sci. 16:297323

20. Broekaert WF, Terras FRG, Cammue BPA, Vanderleyden J. 1990. An automated quanti-tative assay for fungal growth-inhibition. FEMS Microbiol. Lett. 69:5559

21. Brogden KA. 2005. Antimicrobial peptides: pore formers or metabolic inhibitors in bacte-ria? Nat. Rev. Microbiol. 3:23850

22. Carmona MJ, Molina A, Fernandez JA, Lopez-Fando JJ, Garca-Olmedo F. 1993. Ex-pression of the alpha-thionin gene from barley in tobacco confers enhanced resistance tobacterial pathogens. Plant J. 3:45762

23. Cary JW, Rajasekaran K, Jaynes JM, Cleveland TE. 2000. Transgenic expression of a geneencoding a synthetic antimicrobial peptide results in inhibition of fungal growth in vitroand in planta. Plant Sci. 154:17181

24. Casteels P, Ampe C, Jacobs F, Vaeck M, Tempst P. 1989. Apidaecinsantibacterial peptidesfrom honeybees. EMBO J. 8:238791


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

24a. Casteels P, Romagnolo J, Castle M, Casteels-Josson K, Erdjument-Bromage H, Tempst P.1994. Biodiversity of apidaecin-type peptide antibiotics. J. Biol. Chem. 269:2610715

25. Casteels P, Tempst P. 1994. Apidaecin-type peptide antibiotics function through a nonpore-forming mechanism involving stereospecicity. Biochem. Biophys. Res. Commun. 199:33945

26. Cavallarin L, Andreu D, San Segundo B. 1998. Cecropin A-derived peptides are potentinhibitors of fungal plant pathogens. Mol. Plant-Microbe Interact. 11:21827

27. Chadha P, Das RH. 2006. A pathogenesis related protein, AhPR10 from peanut: an insightof its mode of antifungal activity. Planta 225:21322

28. Chakrabarti A, Ganapathi TR, Mukherjee PK, Bapat VA. 2003. MSI-99, a magaininanalogue, imparts enhanced disease resistance in transgenic tobacco and banana. Planta216:58796

29. Chan DI, Prenner EJ, Vogel HJ. 2006. Tryptophan- and arginine-rich antimicrobial pep-tides: Structures and mechanisms of action. Biochim. Biophys. Acta 1758:1184202

30. Choi KC, Kim HR, Park YS, Park SM, Kim JH. 2002. Design and screening of in vivoexpressed antimicrobial peptide library. Biotechnol. Lett. 24:25156

31. Coca M, Penas G, Gomez J, Campo S, Bortolotti C, et al. 2006. Enhanced resistance tothe rice blast fungus Magnaporthe grisea conferred by expression of a cecropin A gene intransgenic rice. Planta 223:392406

32. Cotter PD, Hill C, Ross RP. 2005. Bacteriocins: developing innate immunity for food.Nat. Rev. Microbiol. 3:77788

33. Dathe M, Nikolenko H, Klose J, Bienert M. 2004. Cyclization increases the antimicrobialactivity and selectivity of arginine- and tryptophan-containing hexapeptides. Biochemistry43:914050

34. De Lucca AJ, Bland JM, Grimm C, Jacks TJ, Cary JW, et al. 1998. Fungicidal properties,sterol binding, and proteolytic resistance of the synthetic peptide D4E1. Can. J. Bot. 44:51420

35. De Lucca AJ, Bland JM, Jacks TJ, Grimm C, Cleveland TE, Walsh TJ. 1997. Fungicidalactivity of Cecropin A. Antimicrob. Agents Chemother. 41:48183

36. DeGray G, Rajasekaran K, Smith F, Sanford J, Daniell H. 2001. Expression of an antimi-crobial peptide via the chloroplast genome to control phytopathogenic bacteria and fungi.Plant Physiol. 127:85262

37. Derossi D, Joliot AH, Chassaing G, Prochiantz A. 1994. The third helix of the antennapediahomeodomain translocates through biological membranes. J. Biol. Chem. 269:1044450

38. Dorschner RA, Lopez-Garca B, Peschel A, Kraus D, Morikawa K, et al. 2006. The mam-malian ionic environment dictates microbial susceptibility to antimicrobial defense peptides.FASEB J. 20:3542

39. Du T, Wang Y, Hu QX, Chen J, Liu S, et al. 2005. Transgenic Paulownia expressing shiva-1gene has increased resistance to Paulownia witches broom disease. J. Integr. Plant Biol.47:15006

40. Dykes GA, Aimoto S, Hastings JW. 1998. Modication of a synthetic antimicrobial peptide(ESF1) for improved inhibitory activity. Biochem. Biophys. Res. Commun. 248:26872

41. Epand RM, Vogel HJ. 1999. Diversity of antimicrobial peptides and their mechanisms ofaction. Biochim. Biophys. Acta 1462:1128

42. Falla TJ, Hancock REW. 1997. Improved activity of a synthetic indolicidin analog.Antimicrob. Agents Chemother. 41:77175

43. Falla TJ, Karunaratne DN, Hancock REW. 1996. Mode of action of the antimicrobialpeptide indolicidin. J. Biol. Chem. 271:19298303

292 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

44. Fang ZD, Laskey JG, Huang S, Bilyeu KD, Morris RO, et al. 2006. Combinatoriallyselected defense peptides protect plant roots from pathogen infection. Proc. Natl. Acad. Sci.USA 103:1844449

45. Farnaud S, Evans RW. 2003. Lactoferrina multifunctional protein with antimicrobialproperties. Mol. Immunol. 40:395405

46. Ferre R, Badosa E, Feliu L, Planas M, Montesinos E, Bardaj E. 2006. Inhibition of plant-pathogenic bacteria by short synthetic cecropin A-melittin hybrid peptides. Appl. Environ.Microbiol. 72:33028

47. Florack D, Allefs S, Bollen R, Bosch D, Visser B, Stiekema W. 1995. Expression of giantsilkmoth cecropin-B genes in tobacco. Transgenic Res. 4:13241

48. Gamberi T, Cavalieri D, Magherini F, Mangoni ML, De Filippo C, et al. 2007. An inte-grated analysis of the effects of Esculentin 1-21 on Saccharomyces cerevisiae. Biochim. Biophys.Acta 1774:688700

49. Gao AG, Hakimi SM, Mittanck CA, Wu Y, Woerner BM, et al. 2000. Fungal pathogenprotection in potato by expression of a plant defensin peptide. Nat. Biotechnol. 18:130710

50. Garca-Olmedo F, Molina A, Alamillo JM, Rodrguez-Palenzuela P. 1998. Plant defensepeptides. Biopolymers 47:47991

51. Garca-Olmedo F, Rodrguez-Palenzuela P, Molina A, Alamillo JM, Lopez-Solanilla E,et al. 2001. Antibiotic activities of peptides, hydrogen peroxide and peroxynitrite in plantdefence. FEBS Lett. 498:21922

52. Gifford JL, Hunter HN, Vogel HJ. 2005. Lactoferricin: a lactoferrin-derived peptidewith antimicrobial, antiviral, antitumor and immunological properties. Cell. Mol. Life Sci.62:258898

53. Gonzalez CF, Provin EM, Zhu L, Ebbole DJ. 2002. Independent and synergistic activityof synthetic peptides against thiabendazole-resistant Fusarium sambucinum. Phytopathology92:91724

54. Hagen S, Marx F, Ram AF, Meyer V. 2007. The antifungal protein AFP from Aspergillusgiganteus inhibits chitin synthesis in sensitive fungi. Appl. Environ. Microbiol. 73:212834

55. Hancock REW. 2001. Cationic peptides: effectors in innate immunity and novel antimi-crobials. Lancet Infect. Dis. 1:15664

56. Hancock REW, Lehrer R. 1998. Cationic peptides: a new source of antibiotics. TrendsBiotechnol. 16:8288

57. Hancock REW, Sahl HG. 2006. Antimicrobial and host-defense peptides as new anti-infective therapeutic strategies. Nat. Biotechnol. 24:155157

58. Harris SD, Momany M. 2004. Polarity in lamentous fungi: moving beyond the yeastparadigm. Fungal Genet. Biol. 41:391400

59. Haug BE, Strom MB, Svendsen JSM. 2007. The medicinal chemistry of short lactoferricin-based antibacterial peptides. Curr. Med. Chem. 14:118

60. Haukland HH, Ulvatne H, Sandvik K, Vorland LH. 2001. The antimicrobial peptideslactoferricin B and magainin 2 cross over the bacterial cytoplasmic membrane and residein the cytoplasm. FEBS Lett. 508:38993

61. Henriques ST, Melo MN, Castanho MARB. 2006. Cell-penetrating peptides and antimi-crobial peptides: How different are they? Biochem. J. 399:17

62. Hsu CH, Chen CP, Jou ML, Lee AYL, Lin YC, et al. 2005. Structural and DNA-bindingstudies on the bovine antimicrobial peptide, indolicidin: evidence for multiple conforma-tions involved in binding to membranes and DNA. Nucleic Acids Res. 33:405364

63. Huang Y, Nordeen RO, Di M, Owens LD, McBeath JH. 1997. Expression of an engineeredcecropin gene cassette in transgenic tobacco plants confers resistance to Pseudomonas syringaepv. tabaci. Phytopathology 87:49499


Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

by U

nive

rsity

of O

klah

oma,

Hea

lth S

cien

ces C

ente

r on

11/1

0/08

. For

per

sona

l use

onl

y.

ANRV351-PY46-12 ARI 22 June 2008 11:13

64. Ingham AB, Moore RJ. 2007. Recombinant production of antimicrobial peptides in het-erologous microbial systems. Biotechnol. Appl. Biochem. 047:19

65. Jacobi V, Plourde A, Charest PJ, Hamelin RC. 2000. In vitro toxicity of natural and designedpeptides to tree pathogens and pollen. Can. J. Bot. 78:45561

66. Jaynes JM, Nagpala P, Destefanobeltran L, Huang JH, Kim JH, et al. 1993. Expressionof a cecropin-B lytic peptide analog in transgenic tobacco confers enhanced resistance tobacterial wilt caused by Pseudomonas solanacearum. Plant Sci. 89:4353

67. Jaynes JM, Xanthopoulos KG, Destefanobeltran L, Dodds JH. 1987. Increasing bacterialdisease resistance in plants utilizing antibacterial genes from insects. BioEssays 6:26370

68. Jenssen H, Hamill P, Hancock REW. 2006. Peptide antimicrobial agents. Clin. Microbiol.Rev. 19:491511

69. Joliot A, Prochiantz A. 2004. Transduction peptides: from technology to physiology. Nat.Cell Biol. 6:18996

70. Jones RW, Ospina-Giraldo MD, Clemente T. 2004. Prosystemin-antimicrobial-peptidefusion reduces tomato late blight lesion expansion. Mol. Breed. 14:8389

71. Jones RW, Prusky D. 2002. Expression of an antifungal peptide in Saccharomyces: a newapproach for biological control of the postharvest disease caused by Colletotrichum coccodes.Phytopathology 92:3337

72. Jung HJ, Park Y, Hahm KS, Lee DG. 2006. Biological activity of Tat (4758) peptide onhuman pathogenic fungi. Biochem. Biophys. Res. Commun. 345:22228

73. Kalfa VC, Jia HP, Kunkle RA, McCray PB, Tack BF, Brogden KA. 2001. Congenersof SMAP29 kill ovine pathogens and induce ultrastructural damage in bacterial cells.Antimicrob. Agents Chemother. 45:325661

74. Kamysz W, Krolicka A, Bogucka K, Ossowski T, Lukasiak J, Lojkowska E. 2005.Antibacterial activity of synthetic peptides against plant pathogenic Pectobacterium species.J. Phytopathol. 153:31317

75. Kang JH, Shin SY, Jang SY, Lee MK, Hahm KS. 1998. Release of aqueous contents fromphospholipid vesicles induced by cecropin A (1-8) magainin 2 (1-12) hybrid and its ana-logues. J. Pept. Res. 52:4550

76. Khan JA, Vulfson EN. 2003. Combinatorial chemistry in food research. Comb. Chem. HighThroughput Screen. 6:56974

77. Kim HK, Chun DS, Kim JS, Yun CH, Lee JH, et al. 2006. Expression of the cationicantimicrobial peptide lactoferricin fused with the anionic peptide in Escherichia coli. Appl.Microbiol. Biotechnol. 72:33038

78. Kobayashi S, Chikushi A, Tougu S, Imura Y, Nishida M, et al. 2004. Membrane transloca-tion mechanism of the antimicrobial peptide buforin 2. Biochemistry 43:1561016

79. Kristyanne ES, Kim KS, Stewart JM. 1997. Magainin 2 effects on the ultrastructure of veplant pathogens. Mycologia 89:35360

80. Kuzina LV, Miller TA, Cooksey DA. 2006. In vitro activities of antibiotics and antimicro-bial peptides against the plant pathogenic bacterium Xylella fastidiosa. Lett. Appl. Microbiol.42:51420

81. Lee DG, Hahm KS, Shin SY. 2004. Structure and fungicidal activity of a synthetic antimi-crobial peptide, P18, and its truncated peptides. Biotechnol. Lett. 26:33741

82. Lee TJ, Coyne DP, Clemente TE, Mitra A. 2002. Partial resistance to bacterial wilt intransgenic tomato plants expressing antibacterial lactoferrin gene. J. Am. Soc. Hortic. Sci.127:15864

83. Li QS, Lawrence CB, Davies HM, Everett NP. 2002. A tridecapeptide possesses bothantimicrobial and protease-inhibitory activities. Peptides 23:16

294 Marcos et al.

Ann

u. R

ev. P

hyto

path

ol. 2

008.

46:2

73-3

01. D

ownl

oade

d fro

m ar

journa

ls.an

nualr

eview

s.org

Identification and Rational Design of Novel Antimicrobial Peptides for Plant Protection

Documents

Transcript of Identification and Rational Design of Novel Antimicrobial Peptides for Plant Protection