Systemin is the most well-known secreted plant peptide involved in activation of defense signaling and it was the first peptide to be acknowledged as a plant hormone (Ryan and Pearce, 2003). Systemin is an 18-AA peptide which is derived from a 200-AA precursor protein called prosystemin (Pearce et al., 1991; McGurl and Ryan, 1992; Farrokhi et al., 2008; Table 1). The isolation of systemin was achieved by feeding cut Solanum lycopersicum (tomato) stems with a few microliters of tomato leaf juice, causing the plants to produce proteinase inhibitor proteins, including systemin (Pearce et al., 1991; Ryan and Pearce, 1998). Prosystemin localizes in the cytoplasm of phloem parenchyma cells, however upon wounding or death of these cells, prosystemin is processed to systemin, probably by proteinases. This allows systemin to diffuse into the apoplast and be detected by receptors on the plasma membrane of mesophyll cells (Boller, 2005). Systemin induces protease inhibitor production within local tissue in order to suppress wounding by insect proteases. However, it also leads to a systemic wound response by activating the jasmonic acid (JA) signaling pathway (Stratmann, 2003; Farrokhi et al., 2008). The production of this peptide appears to be under diversifying selection which suggests that some herbivores may have evolved defenses against systemin signaling (Boller, 2005). Prosystemin has only been found in species of the Solaneae a subtribe of the Solanaceae family e.g., tomato, potato, pepper and nightshade but not in tobacco, another Solanaceae (Scheer and Ryan, 2002; Ryan and Pearce, 2003). Since systemin has not been identified in tobacco this plant also shows no alkalization response to systemin treatment (0.0025–25 nM) (Ryan and Pearce, 2003). Therefore, it seems systemin is restricted to a few species and is not found globally in plants. This suggests that for the Solaneae plants, pre-treatment with systemin could reduce the wounding effects of insects in the field. Systemin pre-treatment also has the potential to reduce other infections, for example virus and bacterial infections transmitted by insects.

Over a decade after the first report of systemin its tomato cell surface receptor was identified (Scheer and Ryan, 2002). SlSR160 is a 160 kDa receptor belonging to the family of leucine rich repeat receptor-like kinases (LRR-RLKs), and like classical LRR-RLKs, it contains an extracellular leucine rich repeat domain, a transmembrane domain and an intracellular kinase domain. In cell culture experiments, systemin detection by SlSR160 activates a complex signaling pathway which results in activation of a mitogen-activated protein kinase (MAPK), rapid alkalinization of the extracellular medium via blockage of a proton pump in the cell membrane, the activation of phospholipases along with activation of phytodienoic acid and jasmonic acid (JA) for defense signaling (Felix and Boller, 1995; Schaller and Oecking, 1999; Ryan and Pearce, 2003). It was later discovered that the systemin receptor, SlSR160, and the brassinosteroid receptor from A. thaliana, AtBRI1, are homologs which contain all the same domains, the most conserved of which have 83–90% similarity (Montoya, 2002; Scheer and Ryan, 2002). BRI1 is the cell surface receptor for the brassinosteroid pathway which detects the hormone brassinolide (BL), a regulator of plant growth and development. This shows dual function for this receptor in both plant defense and maintenance of growth and development. In recent years however, the dual function of BRI1/SR160 has come into question. Initially dual function was demonstrated by the expression of tomato SR160/BRI1 in tobacco which allowed the transgenic plants to respond to systemin treatment (Scheer et al., 2003). A mutant tomato Slbri1 gene called cu-3, was used to further examine the dual function of SR160/BRI1. The tomato cu-3 plants showed a Brassinosteroid (BR) deficient growth phenotype i.e., stunted growth and curled leaves, however these plants also could not induce systemin signaling (Scheer et al., 2003). Confirming this receptor has a dual role in plant defense as well as in growth and development (Montoya, 2002; Scheer et al., 2003). However, more recently it was shown that the reduced responsiveness of Slcu3 plants to systemin was due to the stunted nature of these plants and Slcu3 cell cultures responded the same as the wild type (WT) to treatment with systemin (Holton et al., 2007). These results have led to some controversies in this field and a more recent study has sought to try and understand these conflicting results. Malinowski et al. (2009) showed by expression of tomato BRI1 in tobacco cells that BRI1 binds to systemin however the systemin signaling did not show any increase by the overexpression of tomato BRI1. This report confirms results previously described by Holton et al. (2007), that silencing of SlBRI1 in tomato results in a bri1 phenotype but did not affect systemin response (Malinowski et al., 2009). These results imply that BRI1 can bind systemin but it is not the ligand that initiates intracellular signaling (Malinowski et al., 2009). Therefore, there could be another ligand for systemin yet to be identified. The newest data also explains why A. thaliana which contains BRI1 cannot sense systemin treatment or activate downstream defense signals. In A. thaliana there is some redundancy within the BRI1 gene family as three other cell surface receptors have been identified with high sequence similarity, BRI1-like (BRL1, BRL2, and BRL3). It was shown that BRL1 and BRL3 can complement BRI1 but BRL2 cannot, BRL2 is required for provascular differentiation in leaves (Clay and Nelson, 2002; Caño-Delgado et al., 2004). Given there is a gene family in A. thaliana, it is highly likely that this is true in Solanacea species. Therefore, it could be that one of these other receptors is required for systemin signal transduction.

HypSys's are another group of well-known wound-induced peptides which show functional relatedness to systemin and contain multiple hydroxyproline residues. These polypeptide hormones were first identified in tobacco using cell suspension assays treated with a crude peptide fraction from tobacco leaves which resulted in the alkalinization of medium. Two 18-AA polypeptide hormones were identified in tobacco and found to be derived from each end of a 165-AA polyprotein hormone precursor, pro-TobSys-A (Pearce et al., 2001; Table 1). These two peptides were named tobacco hydroxyproline-rich systemin (TobHypSys) I and II (Ryan and Pearce, 2003). Although HypSys's may have similar downstream functions to systemin, it appears the physical properties and the processing of these two peptides differ greatly. The synthesis of HypSys potentially involves the secretory pathway given the presence of carbohydrate residues and hydroxyproline residues, while in contrast systemin contains no hydroxyprolines and no carbohydrate residues (Pearce et al., 2001). HypSys's have also been identified in a wider variety of plants than systemin including tomato (TomHypSys I, II and II), petunia (PhHypSys I, II, and III) sweet potato, black nightshade and potato (Pearce and Ryan, 2003; Pearce et al., 2007, 2009; Chen et al., 2008; Bhattacharya et al., 2013). The TomHypSys are 20, 18, and 15-AAs in length, respectively, while the PhHypSys I, II, and III are 19, 20, and 18-AAs in length, respectively (Pearce and Ryan, 2003; Pearce et al., 2007). In these identified precursor proteins for the HypSys peptides it was noted that a 30 base pair (bp) sequence around the peptidase splice site was conserved in all, this conservation and the high sequence homology over the whole protein is allowing faster and easier identification of these peptides therefore avoiding long troublesome purification from leaf extracts (Bhattacharya et al., 2013).

The 146-AA precursor protein of the three TomHypSys peptides localizes to the cell wall matrix in the phloem parenchyma cells of tomato leaves (Narváez-Vásquez et al., 2005). The tomato, tobacco and black nightshade HypSys's all induce the production of protease inhibitors when supplied through cut petioles. Furthermore, the tomato preproHypSys gene is induced upon wounding of leaves and upon systemin or JA treatment of the leaves (Pearce et al., 2001, 2009; Pearce and Ryan, 2003). In contrast, the petunia HypSys's showed no induction of protease inhibitors against herbivore attacks, however they did induce the expression of defensin 1 (Pearce et al., 2007). Defensin 1 is a gene associated with innate immunity in plants, suggesting that the petunia HypSys's still appear to function in plant defense signaling pathways. The petunia HypSys's are functionally more similar to the AtPep family of peptides than to the other HypSys's implying that there may be some functional diversity within peptide families specific to some plants (Bhattacharya et al., 2013). The StHypSys's are functionally comparable to the tobacco and tomato HypSys peptide, however they appear to have additional functions as they also activate enzymes to protect against oxidative stress and free-radical generation (Bhattacharya et al., 2013). These are both outcomes of herbivore attack and pathogen infection therefore the StHypSys peptides appear to act as defense elicitors against both insects and pathogens (Bhattacharya et al., 2013). Although the experiments were not done it could be expected that pre-treatment of plants with HypSys prior to insect attack or pathogen infection would result in reduced symptoms as the defense pathways are primed.

Pep1 from A. thaliana is another peptide which is associated with the activation of the plant defense system (Huffaker et al., 2006). This peptide consists of 23 AA residues and is derived from the C-terminus of a 92-AA precursor protein called PROPEP1 (Huffaker et al., 2006; Farrokhi et al., 2008; Table 1). AtPROPEP1 shows low level expression in all tissue types but increased expression is evident when plants are wounded or treated with methyl jasmonate (MeJA) or ethlylene (ET) (Huffaker et al., 2006). This suggests a role for this protein in plant defense. Plants treated through cut petioles with 20 nM AtPep1, showed increased expression of Plant Defensin 1.2 (PDF1.2), Pathogenesis-related Protein 1 (PR-1) and PROPEP1 along with the production of H₂O₂ (Huffaker et al., 2006; Huffaker and Ryan, 2007). The production of H₂O₂ and PDF1.2 are both associated with innate immunity in plants while PR-1 is known to be induced during Pathogen-Associated Molecular Patterns (PAMP) Triggered Immunity (PTI) however its function remains unknown. Huffaker et al. (2006) and Huffaker and Ryan (2007) used mutants for the JA and ET pathways, to show that AtPep1 functions upstream of the JA/ET pathways, as expression of PDF1.2 and PR-1, triggered by the peptides, was blocked. Transgenic plants constitutively overexpressing AtPROPEP1 showed an increase in root mass compared to WT plants. When these overexpressing plants were infected with the root pathogen Pythium irregular the increase in root growth compared to infected WT plants was still apparent however, the aerial parts of the plants show no difference in growth or disease symptoms between the transgenic and WT. This implies that overexpression of PROPEP1 provides a growth advantage to the roots even in the presence of a pathogen (Huffaker et al., 2006). This growth advantage looks to be a by-product of PROPEP1 overexpression as the aerial components of the plants do not show increased growth implying that nutrient acquisition has not increased. Six annotated homologs of the precursor protein have been identified in A. thaliana, two of which (PROPEP2 and 3) show increased expression when plants are infected with bacterial, fungal and oomycete pathogens (Huffaker et al., 2006). The expression of PDF1.2 and PR-1 was increased in plants treated with AtPep1 (Huffaker et al., 2006). All of these results indicate that PROPEP1-3 has a role in a feedback loop for defense signaling that is activated in the presence of pathogens and also increases root development (Huffaker and Ryan, 2007).

Orthologs of these A. thaliana genes have since been found in other plant species suggesting an important and ancient role for these peptides. Important crop species are among those to be investigated for the presence of Pep peptides. Recently an ortholog of AtPROPEP1 was identified in Solanum lycopersicum, SlPROPEP, which has high AA identity (96%) with the C-terminal region of the S. tuberosum gene (Trivilin et al., 2014). Silencing of the tomato ortholog increased the susceptibility of plants to Pythium dissotocum but also decreased the expression of key defense response genes e.g., PR-1, PR-5, ERF1, LOX-D, and DEF2 (Trivilin et al., 2014). These results showed that SlPROPEP plays a role in resistance against P. dissotocum. Huffaker et al. (2011) identified and characterized an ortholog of AtPROPEP1 in maize. Similar to AtPROPEP1, ZmPROPEP1 is induced in response to JA or peptide treatment, but also fungal infection. In maize plants treated with ZmPep1 it was noted that the expression of several defense genes were up-regulated, endochitinase A, PR-4, PRms, SerPIN, and Benzoxazineless1 (Huffaker et al., 2011). Interestingly, this work also showed that pre-treatment of maize leaves with 25 pmol of ZmPep1 increased resistance of leaves to Cochliobolis heterostrophus and pre-treatment of stalks with 5 nmol of ZmPep1 increased resistance to Colletotrichum graminicola (Huffaker et al., 2011). The work described above clearly shows a benefit to the plants when pre-treated with these peptides in reducing pathogenicity. Therefore, these peptides may be useful for crop plants in areas that are known to have high pathogenicity instead of the use of fungicides or pesticides by either foliar application or with the generation of transgenic crop plants.

Two cell surface receptors for AtPep1, PEPR1 and PEPR2, have been identified in A. thaliana and demonstrated to be membrane associated leucine rich repeat (LRR) receptor kinases (Yamaguchi et al., 2006, 2010). These two receptors are 76% similar at the protein level and both showed induced expression upon wounding, also following exogenous application of Pep1 peptide and MeJA (Yamaguchi et al., 2010). PEPR1 was confirmed as recognizing Pep1 resulting in the activation of plant defense responses (Yamaguchi et al., 2006). Gain of function was also observed when AtPEPR1 was overexpressed in transgenic tobacco cell lines resulting in the rapid alkalinization of the media upon treatment with AtPep1 (Yamaguchi et al., 2006). Like previously described experiments, pre-treatment of A. thaliana with exogenous Pep1 resulted in increased resistance to Pseudomonas syringae pv. tomato DC3000 (Yamaguchi et al., 2010). PEPR1 and PEPR2 show limited redundancy as single mutants of each showed reduced resistance to P. syringae pv. tomato while a double mutant is susceptible to P. syringae pv. tomato (Yamaguchi et al., 2010). This work has elegantly identified the cell surface receptor of a secreted plant peptide which induces defense responses against pathogen invasion.

Whilst tomato and tobacco have traditionally been used for peptide discovery, soybean is rapidly becoming the new dicot model as a source of new peptides. Recently many new peptide families have been identified in soybean, GmPep914 and GmPep890 are two recently identified peptides shown to cause the rapid alkalinization of cell culture medium (Yamaguchi et al., 2011). These peptides consist of 8 AA residues and are derived from the C-terminal end of 52-AA precursor proteins GmPROPEP914 and GmPROPEP890 (Yamaguchi et al., 2011; Table 1). Treatment of soybean leaves with GmPep914 and GmPep890 induced the expression of their precursor proteins as well as defense genes including CYP93A1, a cytochrome P450 gene, chitinaseb1-1 and Gmachs1 (Glycine max chalcone synthase1) which is involved in phytoalexin production (Yamaguchi et al., 2011). GMPep914 is unrelated to any of the previously identified peptides with a known function in plant defense and it is also the smallest peptide found to have such a role (Yamaguchi et al., 2011).

GmSubPep was also recently identified in soybean. The sequence of this 12-AA peptide was found within an extracellular protein family called subtilisin-like protease (subtilase), specifically in the gene Glyma18g48580 (Pearce et al., 2010; Table 1). This peptide appears unique for soybean as systemin is for tomato due to the GmSubPep sequence located in the protease-associated (PA) domain of Glyma18g48580 which is a unique region of the subtilases within legumes (Pearce et al., 2010). Also when this peptide was tested on tomato, tobacco, Arabidopsis and corn cell cultures no alkalinization was observed (Pearce et al., 2010). The receptor of Glyma18g48580 was not induced upon wounding or treatment with MeJA, however treatment of cell cultures with the peptide resulted in an increase in some defense genes (Cyp93A1, Chib-1b, and PDR12; Pearce et al., 2010). Glyma18g48580 is predicted to be an apoplastic protein of unknown function and it has been speculated that this protein comes into contact with non-self molecules, due to its apoplastic localization. This in turn could induce cleavage of the peptide resulting in activation of defense signaling pathways. No work has been done so far with these soybean peptides to determine if they are able to increase plant resistance to pathogens or insects but given that defense genes are up-regulated it is conceivable that this is the case.

The CAPE1 peptide, was recently isolated from tomato and has been shown to be a Damage-Associated Molecular Pattern (DAMP) elicitor (Chen et al., 2014). CAPE1 was identified using endogenous peptide mixtures extracted from stressed and unstressed plants, these extracts were analyzed by nanoflow ultrahigh performance liquid chromatography mass spectrometry (nanoUHPLC-MS). CAPE1 is an 11-AA peptide derived from the C-terminus of a 159-AA precursor protein called PR-1b (also known as P14a; Chen et al., 2014; Table 1). Expression analysis showed that CAPE1 was induced by wounding and that treatment of tomato plants with the CAPE1 peptide resulted in the induction of defense hormones such as JA and salicylic acid (SA) as well as increased expression of several pathogen-related marker genes, PR-2, PR-7 and PR1b (Chen et al., 2014). However, CAPE1 treated plants did not induce PTI-responsive genes, suggesting that CAPE1 may induce systemic resistance rather than PTI (Chen et al., 2014). The authors also showed that pre-treatment of tomato leaves with CAPE1 prior to infection with P. syringae pv. tomato resulted in reduced disease symptoms in the absence of Hypersensitive Response (HR) (Chen et al., 2014). CAPE1 also showed an anti-herbivore activity; Spodoptera litura larvae feeding experiments using CAPE1 pre-treated tomato leaves resulted in a 20% reduction in size and weight (Chen et al., 2014). This data demonstrates that CAPE1 has antibacterial and anti-herbivore activity. Curiously anti-fungal activity was not reported on. Comparing these pre-treatment results to those discussed above for the ZmPep1 peptide (Huffaker et al., 2011) shows that these peptides appear to prime plants for pathogen resistance.

Sequence analysis has shown that the CAPE1 peptide is conserved in many flowering plants. The authors also suggest that three AAs immediately in front of the peptide may be the cleavage site (CNYx) and this could act as a conserved motif that could indicate a bio-active peptide in other species (Chen et al., 2014). Pathogenesis-related protein 1 from A. thaliana (AtPR1) contains the CAPE1 and potential bio-active motifs (CNYx. PxGNxxxxPY) but has the least protein similarity to tomato PR-1b. However, this CAPE1 peptide from AtPR1 was shown to increase immunity of A. thaliana to P. syringae pv. tomato indicating that this peptide functions similarly to the CAPE1 from tomato PR-1b (Chen et al., 2014).

The authors suggest a motif that could be used to search other RNAseq or genome database to investigate whether the motif is used in other peptide cleavage sites, showing a downstream application of this technique. As for the plant resistance increase that is observed when the plants are pre-treated with CAPE1 it gives further information about the PR1 protein which has always been associated with plant defense but a function of this protein has been elusive. Since PR1 is present in most plant species this could be a good biocontrol agent against pathogens and shows an ancient and probably conserved role for this protein.

Inceptins fall into a small category of plant peptides as they are processed by enzymes from the invading organism, not by the plant itself. Inceptins are 11–13 AA acidic peptides which contain disulphide bridges and are derived from the chloroplastic ATP synthase γ-subunit (Schmelz et al., 2006; Table 1). These peptides were identified from the saliva of larval fall armyworms (Spodoptera frugiperda) that were feeding on cowpea (Vigna unguiculata) plants. Inceptin peptides are cleaved from the chloroplastic ATP synthase γ-subunit during feeding of the larvae on the leaves; the peptide in the saliva of S. frugiperda is then detected by the plant, resulting in the activation of defense related genes (Schmelz et al., 2006). The perception of inceptin by cowpea plants results in activation of the volatiles homoterpene (E)-4,8-dimethyl-1,3,7-nonatriene (DMNT) and cinnamic acid which are known to attract natural insect enemies but also results in induction of JA and SA hormones (Schmelz et al., 2006, 2007).

Syringolin A is a peptide secreted by the Gram-negative bacterium P. syringae pv. syringae. Syringolin A has a ring structure composed of 5-methyl-4-amino-2-hexenoic acid and 3,4-dehydrolysine (Wäspi et al., 1998; Table 2). The α-amino group is joined by a peptide bond to a valine linked to another valine via a urea moiety (Wäspi et al., 1998). Syringolin A was recovered from liquid cultures by centrifugation, filtering, ultra-filtration, followed by gel filtration chromatography. The application of HPLC-purified syringolin onto Oryza sativa (rice) detached leaves induced resistance toward the fungal plant pathogen Pyricularia oryza (rice blast disease). Plate assays showed that syringolin A did not directly affect the growth of P. oryzae. Furthermore, no visible phytotoxic effects were observed when applied on detached rice leaves at the highest concentration (0.05 mM). Gene expression analysis of detached leaves sprayed with syringolin A solution and subsequently inoculated with P. oryzae indicated the increased transcript abundance of defense genes (including Pir7b, Pir2, Pir2, and Rir1 mRNAs). Consequently, it was proposed that syringolin A elicits rice defense responses through acquired resistance rather than being directly anti-fungal (Wäspi et al., 1998). Subsequently, syringolin A was trialed in other pathosystems. Triticum aestivum (wheat) detached leaves were sprayed with a syringolin A solution 2 days prior to inoculation with Blumeria graminis f. sp tritici (powdery mildew; Wäspi et al., 2001). As previously observed with rice blast disease, wheat leaves exposed to syringolin A prior to fungal infection appeared more resistant. Curative effects of syringolin A were also reported in this study with whole wheat plants first infected with powdery mildew and then sprayed with a syringolin A solution. Fungal colonization of wheat tissues was arrested if syringolin A was exogenously applied 2 days post-inoculation. This study suggests that the mode-of-action of syringolin A either targets the host cells in a way that maintains host's hypersensitivity or reverses the suppression of host defense imposed by the pathogen (Wäspi et al., 2001).

Cyclic lipodepsipeptides (LDPs) include two other AMPs secreted by P. syringae pv. syringae, syringomycin E (SRE) and syringopeptin 25A (SP25A) (Table 2). SRE is formed by nona-peptide lactones acylated with a long-chain 3-hydroxy FA. SP25A is composed of long and highly hydrophobic peptide chain, with a polar lactonized penta- or octa-peptide moiety at the C terminus. It has been speculated that pathogen cell walls would present a natural barrier which impairs peptide anti-microbial activity, therefore making pest cell wall porous would enhance AMPs action (Fogliano et al., 2002). LDPs were extracted from centrifuged liquid cultures followed by acetone precipitation and further fractionated using reverse-phase (RP) HPLC. SRE and SP25A were identified by MS. SRE and SP25A, together with cell wall degrading enzymes (CWDEs) from Trichoderma atroviride, were incorporated into in vitro assays performed on liquid cultures of F. oxysporum, V. dahliae, Botrytis cinerea, Penicillium expansum, Phytophthora infestans, and Rhodotorula pilimanae (Fogliano et al., 2002; Table 2). These analyses revealed that while SRE or SP25A alone did not inhibit fungal growth, in the presence of endochitinase and/or glucanase SRE and SP25A prevented spore germination, thus unraveling synergism between LDPs and CWDEs. Furthermore, postharvest assays performed on wounded apple fruits first inoculated with B. cinerea, then with P. syringae, and/or with T. atroviride indicated that the least number of B. cinerea related-wounds were observed when both P. syringae and T. atroviride were combined (Fogliano et al., 2002). This study thus provided support for a synergistic interaction between P. syringae LDPs and T. atroviride CWDEs against a variety of fungi, thereby supporting the hypothesis that, in a biocontrol strategy, the efficacy of AMPs in plant pathogen resistance would be improved if pathogen cell walls were also made vulnerable (Fogliano et al., 2002). This experiment however has not validated which LDPs and CWDEs are responsible for the synergism observed in vivo.

Potent antifungal lipopeptides such as surfactins, iturins and fengycins are produced by some strains of the Gram-positive bacterium Bacillus subtilis such as strain M4 which are beneficial rhizobacteria (Ongena et al., 2005; Table 2). B. subtilis strain M4 was also found to secrete various fengycin homologs, with the great majority of them harboring a C16 or C17 FA chain (Ongena et al., 2005). An enriched lipopeptide extract was obtained by Solid Phase Extraction (SPE) of the crude cell-free culture broth of M4 strain eluted using methanol, followed by HPLC separations and ElectroSpray Ionization (ESI)-MS analyses. In vitro assays showed that the lipopeptide-enriched supernatant had a strong inhibitory effect on the growth of F. oxysporum, Pythium ultimum, Rhizoctonia solani, Rhizopus sp., and B. cinerea. The antimicrobial potential of these peptides has also been described in planta. Post-harvest assays on immature apple fruits preconditioned with M4 supernatant or the lipopeptide-enriched supernatant showed increased pathogen resistance, mostly attributable to fengycins (Ongena et al., 2005). Similarly, bean seedlings with roots pre-treated with M4 supernatant prior to B. cinerea leaf inoculation displayed reduced disease symptoms. B. subtilis M4 thus shows great potential as biocontrol agent to better manage soilborne, foliar and post-harvest diseases. The lipopeptide modes-of-action, in particular that of fengycins, rely not only on the direct inhibition of the plant pathogen on infected plant organs, but also on an indirect interaction mediated through the host plant via systemic resistance induction. Direct interaction was demonstrated through the disease control provided by treatment of fruits with lipopeptide-enriched supernatant and by in situ detection of fengycins in inhibitory amounts. Indirect interaction was mediated by the production of phenolic compounds involved in or derived from the defense-related phenylpropanoid metabolism upon pathogen attack (Ongena et al., 2005).

More recently the same set of secreted lipopeptides, fengycins, iturins, and surfactins, along with bacillomycin, was detected in the Gram-positive bacterium Bacillus amyloliquefaciens strain PPCB004. Bacteria were cultured in liquid medium; lipopeptides were extracted using n-butanol, followed by complete evaporation and resuspension in methanol. Growth assays showed that these peptides had inhibitory properties on the mycelial growth and/or spore germination of various post-harvest fungal pathogens, including Alternaria citri, Botryosphaeria sp., Colletotrichum gloeosporioides, Fusicoccum aromaticum, Lasiodiplodia theobromae, Penicillium crustosum, and Phomopsis perse (Arrebola et al., 2010; Table 2). In this study, iturin A demonstrated the strongest inhibitory effect. Post-harvest assays performed on orange fruits inoculated with A. citri and C. gloeosporioides displayed less disease incidence when treated with B. amyloliquefaciens strain PPCB004 either before or after inoculation. The mode-of-action for iturin A has been proposed to disrupt the fungal cytoplasmic membrane, creating transmembrane channels, resulting in the release of vital ions such as K⁺, thus preventing spore germination and impairing mycelium development (Arrebola et al., 2010).

The lipopeptides bacillomycin D, fengycins A, and B are secreted by B. amyloliquefaciens strain Q-426 (Zhao et al., 2014; Table 2). They were recovered from cell-free supernatant of liquid cultures, further filtered and acid precipitated, prior to drying and methanol resuspension. Using two-dimensional (2-D) HPLC separation and tandem MS (MS/MS) analyses their molecular structure was elucidated. The peptide moiety comprised of 7–10 AA residues arranged in a cyclic structure, while the lipid moiety was composed of a chain of 14–17 FAs. Fengycin A purified from strain Q-426 disrupted the germination of spores from the fungal pathogen F. oxysporum f. sp. spinaciae in a dose dependent manner, with complete inhibition above 50 ug/mL. Fengycin A also modified hyphal growth, albeit without affecting cell membrane permeability (Zhao et al., 2014).

Along with iturin A, isoforms of bacillomycin D and of the macrolactin family, macrolactin A, 7-O-malonyl macrolactin A, and 7-O-succinyl macrolactin A were identified by HPLC-MS/MS analyses from the supernatant of centrifuged liquid cultures of B. amyloliquefaciens strain NJN-6 (Yuan et al., 2011, 2012; Table 2). This strain was isolated from the root system of a healthy banana plant. Plate assays showed that both purified bacillomycin D isoforms inhibited the hyphal growth of F. oxysporum f. sp. cubense, a banana fungal pathogen. Similarly, all three macrolactins inhibited the growth of Gram-negative pathogenic bacteria Ralstonia solanacearum. The authors demonstrate that the macrolactin activities are maintained after 1 month at room temperature (Yuan et al., 2012); such shelf-life information is valuable for an in-field application strategy. Performing tests on plant tissues infected with these pathogens would have further validated the antifungal activity of these AMPs.

Fengycin A was also secreted by the Gram-positive bacterium Bacillus atrophaeus strain CAB-1 and showed antagonistic effect against airborne plant fungal pathogens B. cinerea and Sphaerotheca fuliginea. Two groups of lipopeptides were identified in strain CAB-1 by MS analyses: fengycins and unknown lipopeptides. CAB-1's lipopeptide extracts obtained by centrifugation of the culture broth followed by acid precipitation yielded no detectable iturin or surfactin compounds (Zhang et al., 2013). The fengycins produced by strain CAB-1 were a mixture of isoforms with various acyl side chain lengths from C15 to C17 (Table 2). The C16 isoform of fengycin A was secreted in greater abundance than C15 and C17 isoforms and therefore would contribute more to growth inhibition of cucumber powdery mildew, S. fuliginea, and tomato gray mold, B. cinerea (Zhang et al., 2013).

Mycosubtilin is a nonribosomal cyclic lipopeptide. Surfactin and mycosubtilin were purified from B. subtilis strains BBG131 and BBG125, respectively. Surfactin was produced through an integrated process in a bubbleless membrane bioreactor while mycosubtilin was produced using an overflowing fed-batch process. Subsequent steps of ultrafiltration, diafiltration, evaporation and freeze-dried were required for purification (Deravel et al., 2014). Surfactin and mycosubtilin combined were reported to have dose-dependent synergetic antibiotic effects on lettuce leaves sprayed with the peptide solutions prior to inoculation with its fungal obligate pathogen Bremia lactucae (Deravel et al., 2014; Table 2). This data shows that combining several AMPs opens even greater possibilities in terms of biocontrol strategies, yet it also poses challenges as to designing experiments assessing various AMP combinations and engineering plants resistant to microbial pathogens.

Fusaricidins and closely related LI-F lipopeptides are synthesized by Paenibacillus polymyxa and display inhibitory activities against plant pathogens (Table 2). These nonribosomal peptides consist of a guanidinylated β-hydroxy FA linked to a cyclic hexapeptide including four D configured-AA residues (Debois et al., 2013). Growth inhibition effects were also observed against a variety of microbes: Rhodotorula aurantica (yeast), F. oxysporum, B. cinerea, Cladosporium cucumerinum (fungi), Phythium aphanidermatum (oomycetes), and P. syringae (Gram negative bacterium). These AMPs were identified using a matrix-assisted laser desorption ionization (MALDI) imaging approach. By using a clever experimental design involving the insertion of a sterilized MALDI glass slide coated with indium tin oxide at the bottom of the Petri dish covered with a gelified sterile nutrient medium (Debois et al., 2013). P. polymyxa was streaked over the glass slide and F. oxysporum was inoculated nearby the slide and incubated for 11 days. The glass slide was removed from the Petri dish and completely dried under vacuum prior to MALDI imaging analysis. The authors were able to obtain spectra and fragmentation patterns of the compounds released by strain Pp56 and responsible for the inhibition of F. oxysporum mycelial development (Debois et al., 2013). The antagonistic interaction between fusaricidins and LI-F lipopeptides and F. oxysporum could thus be visualized by acquiring MS spectra of the different antibiotic compounds exhibiting distinct localizations along the slide. A time-course analysis revealed the early secretion of fusaricidin B, a mixture of LI-F05b/06b/08a, and LI-F08b, away from F. oxysporum hyphae, thus suggesting that their production was not triggered by the presence of the fungus. These antibiotics would be readily secreted and would accumulate in toxic amount outside P. polymyxa cells to deter any potential pathogen attack. Indeed their distribution patterns visualized by MALDI imaging coincided with the F. oxysporum hyphal inhibition zone observed on the culture plates (Debois et al., 2013).

APV was HPLC purified from a polar extract from liquid broth supernatant, analyzed using ESI-MS/MS and NMR, and identified as the main antibiotic compound of Pantoea agglomerans strain 48b/90 (Pa48b; Sammer et al., 2009). Pa48b is a Gram-negative bacterium that was isolated from soybean leaves which showed limited fire blight disease symptoms caused by the bacterium Erwinia amylovora. Using plate assays, APV inhibitory effect was tested against various microbial phytopathogen species, including Agrobacterium tumefaciens, E. amylovora, several P. syringae pathovars, Serratia marcescens, and B. subtilis (Table 2). APV successfully inhibited the growth of pathogens on the minimum synthetic medium in a dose dependent manner, but not on the complex medium, likely due to the presence in the latter of N-acetylglucosamine which would compensate for APV inhibitory effects (Sammer et al., 2009). This result suggests that the antagonistic effect of APV and N-acetylglucosamine on phytopathogen colonization depends on nutrient availability. In a follow-up study, the APV biosynthesis gene cluster was analyzed, and located onto a megaplasmid in Pa48b (Sammer et al., 2012). In this cluster, two genes are likely to be involved in APV biosynthesis regulation, whose transcription is tightly coordinated with translation to avoid precursor cytotoxicity. In silico sequence analysis of the APV gene cluster revealed a 99% identity to the diaminopropionate-peptide biosynthesis cluster of P. agglomerans strain CU0119. One of the genes of this cluster, DdaI, is predicted to be a transmembrane efflux pump mediating self-resistance to antibiotics (Sammer et al., 2012). In planta assays were then undertaken in which soybean leaves were first inoculated with P. syringae pv. glycinea prior to applying an APV solution onto the infected wounds. Consistent with the previous study, APV inhibitory effect was dose-dependent, yet only led to minor decrease of the disease symptoms. Therefore, APV was not confirmed to be the key antibiotic factor in the antagonism (Sammer et al., 2012).

The trichokonin (TK) family is composed of three major peptaibols, TKs VI, VII and VIII, produced by the ascomycota fungus Trichoderma pseudokoningii strain SMF2. Peptaibols are characterized by the presence of an unusual AA, a-aminoisobutyric acid, a C-terminal-hydroxylated and a N-terminal-acetylated AA. Because of their linear and amphipathic nature, peptaibols can form voltage-dependent ion channels in lipid bilayer membranes. TKs were obtained from solid state fermentation of T. pseudokoningii strain SMF2, followed by gel filtration of the crude extract and preparative HPLC separation. In vitro plate assays showed that TK VI exhibited antimicrobial activities against various pathogenic fungi and oomycetes, Ascochyta citrullina, B. cinerea, F. oxysporum, Phytophthora parasitica, and V. dahlia (Shi et al., 2012, Table 2). Toxicity assays using F. oxysporum protoplasts treated with TK VI showed the appearance of ROS, fragmentation of nuclear DNA, along with a change of fungal membrane permeability and disintegration of subcellular structures. The antimicrobial efficacy of TK was also tested on Chinese cabbage (Brassica rapa) leaves inoculated with the bacterial pathogen, Pectobacterium carotovorum subsp. carotovorum (Li et al., 2014). As observed by Shi et al. (2012), TK treatment led to an increase in the production of ROS, along with increased activities of pathogenesis-related proteins, and the activation of SA signaling pathway in the cabbage host. These data show that the TK family can induce cell death of pathogenic fungi and oomycetes as well as induce the activation of plant defense pathways.

The antimicrobial peptide Penaeidin 4-1 (Pen4-1) was isolated from Atlantic white shrimp (Litopenaeus setiferus) under pathogen challenge. Pen4-1 is composed of 47 AAs including six cysteine residues forming three disulfide bridges (Cuthbertson et al., 2004; Table 1). Pen4-1 was purified using affinity chromatography as follows: pooled L. setiferus haemocyte extracts were concentrated using a SPE and applied to an affinity resin containing a Pen4-1-specific antibody. Pen4-1 can inhibit multiple plant pathogenic fungal species, such as B. cinerea, Penicillium crustosum, and F. oxysporum. Penaeidins harbor a unique two-domain structure, a proline rich N-terminal domain (PRD) and a cysteine-rich domain (CRD) with a stable alpha-helical structure, which might have contributed to its broad range of microbial targets, primarily Gram-positive bacteria and fungi. Transgenic lines of a commercial creeping bentgrass (Agrostis stolonifera) displayed enhanced resistance to the fungal pathogens Sclerotinia homoecarpa and R. solani as a result of Pen4-1 ectopic expression (Zhou et al., 2011; Table 3). Testing Pen4-1 into other plant hosts of S. homoecarpa and R. solani would confirm its potential as a biocontrol agent.

Metchnikowin (mtk) is an immune-inducible peptide synthesized in the fat body of Drosophila melanogaster as a 52-AA pre-pro-peptide upon microbial challenges (Levashina et al., 1995; Table 3). In vitro assays demonstrated that mtk inhibits the growth of drosophila pathogens, the Gram-positive bacterium Micrococcus luteus and the ascomycete fungus Neurospora crassa. The mtk AA sequence was determined by use of the Edman sequencing method as a proline-rich 26-AA peptide. MS analyses revealed two mtk isoforms of 3025 and 3045 Da, respectively, due to AA substitution (Rahnamaeian et al., 2009; Table 3). Low concentrations of mtk inhibited the in vitro growth of the pathogenic fungi Fusarium graminearum and Fusarium culmorum. Transgenic barley plants expressing the D. melanogaster mtk gene in its 52-AA pre-propeptide form under the control of the inducible mannopine synthase (mas) gene promoter were produced. Mas promoter's induction was triggered by wounding, plant growth hormones, as well as fungal infection. Mtk was successfully processed into its mature form and targeted to the apoplastic compartment in transgenic plants. When inoculated with F. graminearum, transgenic plants displayed higher frequencies of typical defense responses such as HR of attacked cells and the development of callose deposition at the cell wall underneath attempted penetration sites. This enhanced plant innate immune response was substantiated by the up-regulation of Pathogenesis-Related genes PR-1 and PR-5 (Rahnamaeian et al., 2009). The activation of systemic acquired resistance (SAR) was confirmed at the transcript level in a more recent follow-up study in which the mtk barley transgenic plants were infected with B. graminis f. sp. hordei and subjected to RT-PCR analyses (Rahnamaeian and Vilcinskas, 2012; Table 3). When the transgenic barley plants were infected with F. graminearum, mtk treatment impeded the development of functional haustorium, whose formation is crucial for commencement of biotrophic interaction.

Thanatin is produced by the stinkbug Podisus maculiventri; it consists of 21 AA residues (2.4 kDa) which form an internal disulphide bond important for its antimicrobial activity (Imamura et al., 2010; Table 3). A recombinant thanatin gene under the constitutive control of cauliflower mosaic virus 35S (CaMV35S) gene promoter was introduced into rice plants which were then exposed to blast disease (Magnaporthe oryzae). Thanatin was extracted from crude extracts of the transformants, purified using 2-D HPLC (cation exchange chromatography followed by RP chromatography) and its identity confirmed by MS. Although blast disease symptoms were observed on both the transgenic lines and the WT plants, diseased areas on the transformants were significantly smaller than those on the WT plants. While transgenic rice plants were not fully protected against M. oryzae, they had acquired partial resistance (Imamura et al., 2010). Very recently, thanatin gene was introduced into maize and controlled by the ubiquitin-1 promoter which targets the expressed AMP to the plant apoplastic space (Schubert et al., 2015; Table 3). Transgenic maize plants were grown until ears were fully developed and the mature ears were then exposed to the pathogenic fungus Aspergillus flavus. The expression of thanatin in maize transformants led to a significant reduction in fungal biomass of A. flavus relative to that of WT plants (Schubert et al., 2015). The mode-of-action of thanatin in plants has yet to be fully understood.

Cecropin AMPs were first isolated from the haemolymph of the moth Hyalophora cecropia; cecropin A (37 AAs, 4 kDa; Table 3) exhibits a rapid, potent and long-lasting lytic activity against prominent bacterial and fungal phytopathogens (Bundó et al., 2014). Rice transgenic plants were obtained in which cecropin A expression was targeted to the seed endosperms by using the tissue-specific promoters Glutelin B1 or Glutelin B4. Seeds from rice transgenic plants exhibited resistance to infection by fungal (Fusarium verticillioides) and bacterial (Dickeya dadantii) pathogens (Bundó et al., 2014). The mode-of-action of cecropin A in rice seeds remains to be investigated.

Magainins are ribosomal AMPs produced in the skin of Xenopus frogs. Growth assays in the presence of the 23-AA magainin-2 showed that the peptide inhibited the growth of various Gram-positive and Gram-negative bacteria, a few fungi and protozoa (Zasloff, 1987). The mag gene encoding for the MAGAININ-2 protein was ectopically over-expressed in pearl millet (Pennisetum glaucum) under the control of the constitutive CaMV 35S gene promoter and challenged with three strains of downy mildew (Sclerospora graminicola viz. Sg 384, Sg 445, and Sg 492; Ramadevi et al., 2014; Table 3). While some pearl millet transgenic lines exhibited a slight decrease in disease incidence 7 days post inoculation relative to the controls, none demonstrated full resistance (Ramadevi et al., 2014). Performing statistical analyses however would have helped test the significance of these results. The authors attribute the minor response of their genetically modified plants to the complexity of cell wall and cell membrane components of the oomycete pathogen, which are similar in composition and structure to those of host plant cells. Perhaps combining MAGAININ-2 with cell wall degrading enzymes, as tested by Fogliano et al. (2002) whom combined SRE or SP25A peptides with endochitinase and/or glucanase, would prove an efficient strategy.

Esculentins are highly potent AMPs of 46 AAs exclusively found in the skin secretion of the frog Rana esculenta. Esc28L is a variant of esculentin-1b artificially created by substituting the Methionine residue at position 28 with a Leucine and an additional Methionine in position 47 (Ponti et al., 1999). The idea of using variant peptides to increase plant pathogen resistance while targeting them to the extracellular space was first explored by Ponti et al. (1999, 2003; Table 3). Extracellular targeting not only eliminates potential toxicity of the variant AMP toward host cells, while permitting direct contact with pathogens growing and multiplying in the extracellular space. In a follow-up study, Esc28L was fused to the Signal Peptide (SP) sequence of Phaseolus vulgaris EndopolyGalacturonase-Inhibiting Protein (PGIP) to target Esc28L to the secretory pathway. Transgenic tobacco (Nicotiana tabacum) lines were produced in which the constitutive expression of Esc28L conferred enhanced resistance to the bacterial pathogens of tobacco, P. syringae pv. tabaci and Pseudomonas aeruginosa, as well as against the fungal pathogen Phytophthora nicotianae, and moreover demonstrated insecticidal effect against drosophila (Ponti et al., 2003).

The human cathelicidin antimicrobial protein hCAP18 is synthesized in neutrophils as a preproprotein which comprises a conserved cathelin prodomain, and a non-conserved C-terminal peptide. The latter is enzymatically cleaved after secretion forming LL-37, a 37-AA functional antimicrobial peptide. Inspired by the work of Ponti et al. (2003), a mutated variant of LL-37 was created by substituting the Methionine residue at position 37 with a Leucine (LL-37 Met37Leu) and was fused to the SP sequence of PGIP to elicit secretion (Jung et al., 2012; Table 3). LL-37 Met37Leu was then overexpressed in Chinese cabbage plants. Transgenic lines were subsequently inoculated with various bacterial (P. carotovorum subsp. carotovorum) and fungal (F. oxysporum f. sp. lycopersici, Colletotrichum higginsianum, R. solani) pathogens. The transgenic plants displayed enhanced pathogen resistance with decreased disease symptoms relative to the controls (Jung et al., 2012). No mode-of-action was proposed in this study but since LL-37 Met37Leu is targeted to the extracellular space of the transgenic plants, it could be assumed that a direct interaction occurs between the peptide and the pathogen.

Mammal cathelicidins as a whole are worth investigating for plant engineering programs aiming at improving pathogen resistance. Fourteen, twelve, and eight divergent cathelicidin genes were identified in the wallaby, possum and platypus genomes, respectively. Of these, the proteins WAM 1 and WAM2, and Platypus AntiMicrobial (PAM) 1 and PAM2, were tested against various bacterial and yeast pathogens, and shown to be much more potent than LL-37 (Wang et al., 2011). A phylogenetic approach was then used to design an archaic WAM predicted to have originated 59 million years ago and ancestral to the major clade of marsupial AMPs including the modern WAM1 and WAM2 (Wang et al., 2011). In theory, it should be more difficult for the modern pathogens to overcome plant resistance mediated by an archaic AMPs.