Chemical and biological characterization ofsclerosin, an antifungal lipopeptide

Chrystal L. Berry, Ann Karen C. Brassinga, Lynda J. Donald,W.G. Dilantha Fernando, Peter C. Loewen, and Teresa R. de Kievit

Abstract: Pseudomonas sp. strain DF41 produces a lipopeptide, called sclerosin that inhibits the fungal pathogen Sclerotiniasclerotiorum. The aim of the current study was to deduce the chemical structure of this lipopeptide and further characterizeits bioactivity. Mass spectrometry analysis determined the structure of sclerosin to be CH3-(CH2)6-CH(OH)-CH2-CO-Dhb-Pro-Ala-Leu/Ile-Ala-Val-Val-Dhb-Thr-Val-Leu/Ile-Dhp-Ala-Ala-Ala-Val-Dhb-Dhb-Ala-Dab-Ser-Val-OH, similar to corpeptinsA and B of the tolaasin group, differing by only 3 amino acids in the peptide chain. Subjecting sclerosin to various ringopening procedures revealed no new ions, suggesting that this molecule is linear. As such, sclerosin represents a new mem-ber of the tolaasin lipopeptide group. Incubation of S. sclerotinia ascospores and sclerotia in the presence of sclerosin inhib-ited the germination of both cell types. Sclerosin also exhibited antimicrobial activity against Bacillus species. Conversely,this lipopeptide demonstrated no zoosporicidal activity against the oomycete pathogen Phytophthora infestans. Next, we as-sessed the effect of DF41 and a lipopeptide-deficient mutant on the growth and development of Caenorhabditis elegans lar-vae. We discovered that sclerosin did not protect DF41 from ingestion by and degradation in the C. elegans digestive tract.However, another metabolite produced by this bacterium appeared to shorten the life-span of the nematode compared toC. elegans growing on Escherichia coli OP50.

Key words: lipopeptide, sclerosin, biocontrol, Pseudomonas, Sclerotinia sclerotiorum.

Résumé : Pseudomonas sp. souche DF41 produit un lipopeptide appelé sclérosine qui inhibe le pathogène fongique Sclero-tinia sclerotiorum. Le but de cette étude était de déduire la structure chimique de ce lipopeptide et de caractériser davantageson activité biologique. Une analyse en spectrométrie de masse a permis de déterminer que la structure de la sclérosine était :CH3-(CH2)6-CH(OH)-CH2-CO-Dhb-Pro-Ala-Leu/Ile-Ala-Val-Val-Dhb-Thr-Val-Leu/Ile-Dhp-Ala-Ala-Ala-Val-Dhb-Dhb-Ala-Dab-Ser-Val-OH, similaire à celle des corpeptines A et B du groupe tolaasine, ne différant que par trois acides aminésde la chaine peptidique. En soumettant la sclérosine à des procédés variés d’ouverture d’anneau, aucun nouvel ion n’aété détecté, suggérant que cette molécule est linéaire. Ainsi, la sclérosine représente un nouveau membre du groupe delipopeptides tolaasine. L’incubation d’ascospores et de sclérotes de S. sclerotinia en présence de sclérosine inhibait lagermination des deux types de cellules. La sclérosine montrait aussi une activité antimicrobienne envers les espèces deBacillus. À l’inverse, ce lipopeptide n’exerçait aucune activité zoosporicide envers le pathogène oomycète Phytophthorainfestans Par la suite, nous avons évalué l’effet de DF41 et d’un mutant dépourvu de lipopeptide sur la croissance etle développement de larves de Caenorhabditis elegans. Nous avons découvert que la sclérosine ne protégeait pas DF41de l’ingestion par C. elegans et de la dégradation dans le tractus digestif. Cependant, un autre métabolite produit parcette bactérie semblait diminuer la longévité du nématode comparativement à C. elegans cultivé sur Escherichia coliOP50.

Mots‐clés : lipopeptide, sclérosine, contrôle biologique, Pseudomonas, Sclerotinia sclerotiorum.

[Traduit par la Rédaction]

Introduction

Sclerotinia sclerotiorum is an economically important soil-borne pathogen capable of infecting over 400 plant species(Purdy 1979). In canola, S. sclerotiorum causes stem rot,and breeding for resistance has generally been unsuccessful,resulting in heavy reliance on fungicides for disease control.

However, increasing concern over the use of agrochemicalsand the coincident development of fungicide resistance hasled to interest in alternative disease management strategies,including biological control. One such biocontrol agent,Pseudomonas species strain DF41 originally isolated fromthe canola rhizosphere, has demonstrated strong antagonismtoward S. sclerotiorum (Berry et al. 2010; Savchuk and

Received 26 April 2012. Revision received 1 June 2012. Accepted 4 June 2012. Published at www.nrcresearchpress.com/cjm on 27 July2012.

C.L. Berry, A.K.C. Brassinga, P.C. Loewen, and T.R. de Kievit. Department of Microbiology, University of Manitoba, 411 BullerBuilding, Winnipeg, MB R3T 2N2, Canada.L.J. Donald. Department of Chemistry, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.W.G.D. Fernando. Department of Plant Science, University of Manitoba, Winnipeg, MB R3T 2N2, Canada.

Corresponding author: Teresa R. de Kievit (e-mail: dekievit@cc.umanitoba.ca).

1027

Can. J. Microbiol. 58: 1027–1034 (2012) doi:10.1139/W2012-079 Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

Fernando 2004). We have previously reported the characteriza-tion of a mutant strain, DF41-1278, harbouring a transposoninsertion in a gene required for synthesis of a lipopeptide,called sclerosin (Berry et al. 2010). Strain DF41-1278 ex-hibited no antifungal activity in both in vitro and in green-house assays, indicating that lipopeptide production isessential for DF41-mediated inhibition of S. sclerotiorum(Berry et al. 2010).Pseudomonas species, and in particular plant-associated

pseudomonads, are prolific producers of lipopeptides(Raaijmakers et al. 2006), many of which demonstrate phyto-toxic (Bender et al. 1999), antibacterial (Gerard et al. 1997),antifungal (Nielsen et al. 2002), or antiprotozoan (Mazzola etal. 2009) activity. The primary mode of action of lipopep-tides is through pore formation in target membranes, whichdisrupts the flow of ions, leading to membrane collapse(Dalla Serra et al. 1999; Hutchison and Gross 1997; Mottand Takemoto 1989; van de Mortel et al. 2009). Lipopeptidesalso influence bacterial motility and colonization, thereby af-fecting biofilm formation, proliferation, and survival(Raaijmakers et al. 2010). In addition, amphisin, producedby Pseudomonas sp. strain DSS73, was found to facilitatebacterial passage through the digestive tract of Caenorhabdi-tis elegans but did not protect against nematode grazing(Bjørnlund et al. 2009).Lipopeptide molecules consist of a short peptide moiety

linked to a fatty acid tail and are synthesized via a nonriboso-mal thiotemplate mechanism on peptide synthetases that uti-lize both standard and nonstandard amino acids to produce abroad range of lipopeptide products (Berti et al. 2007; Finkingand Marahiel 2004; Marahiel 1997). Many pseudomonad-pro-duced lipopeptides also contain a lactone ring in the peptideportion, although linear lipopeptides, including syringafactinsA–F and peptin31, have been isolated from Pseudomonassyringae (Berti et al. 2007; Fiore et al. 2008).

This report describes the structure of sclerosin, determinedusing tandem mass spectrometry (MS/MS), as being a newlipopeptide resembling members of the tolaasin group. In ad-dition, it describes the influence of sclerosin on S. sclerotio-rum ascospore and sclerotial germination, on the growth ofBacillus species, on the zoospores of the oomycete pathogenPhytophthora infestans, and on the growth and developmentof the nematode C. elegans.

Materials and methods

Bacterial strains and growth conditionsThe bacterial strains and plasmids used in this study are

listed in Table 1. Escherichia coli and Bacillus strains werecultured at 37 and 28 °C, respectively, on Lennox agar(Difco Laboratories). Pseudomonas species were cultured onKing’s B (KB) medium at 28 °C or in M9CAgly (M9 mini-mal salts medium amended with 1% casamino acids (Difco)and 240 mmol/L glycerol). Antibiotics (Research ProductsInternational Corp.) were added to the media at the followingconcentrations: gentamicin (50 µg/mL), ampicillin (100 µg/mL),and kanamycin (50 µg/mL) for E. coli; tetracycline (15 µg/mL),piperacillin (80 µg/mL), rifampicin (50 µg/mL), kanamycin(5 µg/mL), gentamicin (40 µg/mL) for DF41.

Mass spectrometry analysis of sclerosinPseudomonas sp. strain DF41 and the sclerosin-deficient

mutant DF41-1278 were grown in 900 mL volumes ofM9CAgly medium for 4 days at 28 °C with shaking. Cellswere removed by centrifugation (10 000g for 10 min atroom temperature), and metabolites, including sclerosin,were extracted from culture supernatants with ethyl acetate,dried, dissolved in methanol, and fractionated by high-pressure liquid chromatography (HPLC) (Berry et al. 2010).Fractions were analysed on a matrix-assisted laser desorption

Table 1. Bacterial strains and plasmids used in this study.

Strain or plasmid Relative genotype or phenotype Source or referenceStrainsPseudomonasDF41 Rifr wild type (canola root tip isolate) Savchuk and Fernando 2004DF41-1278 Rifr, Kanr nrps::Tn5-1063 genomic fusion Berry et al. 2010DF41-rfp DF41 containing mCherry expressed from pMCh-23 This studyDF41-1278-rfp DF41-1278 containing mCherry expressed from pMCh-23 This studyP. syringae pv. syringaeB728a

Wild type Loper and Lindow 1987

Escherichia coliDH5a supE44 DlacU169 (f80 lacZDM15) hsdR17 recA1 endA1

gyrA96 thi-1 relA1Gibco

OP50 ura, Strr, common C. elegans food source Brenner 1974BacillusB. megaterium ATCC14581

Wild-type strain American Type Culture Collection

B. mycoides Wild-type strain Fernando laboratory collectionB. thuriengensis Wild-type strain Fernando laboratory collection

PlasmidspUCP23 Broad-host range vector; IncP Ori T, Apr, Gmr West et al. 1994pRSET-B mCherry mCherry expression vector, f1ori, Apr Shaner et al. 2004pMCh-23 pUCP23 carrying the mCherry red fluorescent protein gene This study

Note: Rif, rifampicin; Kan, kanamycin; Str, streptomycin; Ap, ampicillin; Gm, gentamicin.

1028 Can. J. Microbiol. Vol. 58, 2012

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

ionization quadrupole-quadrupole-time-of-flight (MALDIQqTOF) instrument maintained by the Department of Physicsat the University of Manitoba (Loboda et al. 2000). A 0.5 µLaliquot was mixed with an equal volume of saturated 2,5-di-hydroxybenzoic acid in water – acetonitrile (3:1 v/v) and 2%formic acid and were spotted onto the target. Spectra werecollected in positive mode for 40 s. Spectra from matchingfractions were compared to identify the ions unique toDF41, which were subjected to MS/MS on the same instru-ment (Shevchenko et al. 2000). Samples from DF41 HPLCfractionation were also analysed by electrospray ionization(ESI) on an ESI-TOF instrument built and maintained in theDepartment of Physics at the University of Manitoba(Kozlovski et al. 2005). ESI and MALDI spectra were ana-lysed with TOFMA, an in-house software program.

Chemical treatments for lipopeptide hydrolysisTo determine if a lactone ring was present, aliquots of

100 µL were dried, dissolved in 100 µL of 0.5 mol/L ammo-nia in methanol, and incubated overnight at 37 °C to inducering hydrolysis (Kuiper et al. 2004). Peptide fragments weregenerated by drying 100 µL aliquots of extract, dissolving in100 µL of 25% trifluoroacetic acid (Zhong et al. 2005), andheating in a microwave oven for 5 m at high power. Afterheating, the extract was dried and resuspended in 5 µL ofmethanol before spotting onto the MALDI target. Selectedions were analysed by MS/MS for sequence determination.Parallel experiments were conducted using the cyclic lipo-peptide surfactin (Sigma–Aldrich), for which MS data areavailable (Yang et al. 2006), to verify that the ring hydrolysisconditions were effective. All chemicals for MS were pur-chased from Fisher Scientific except trifluoroacetic acid,which was obtained from Pierce.

Sclerosin inhibits ascospore and sclerotial germinationThe inhibitory effect of purified sclerosin on ascospore

germination was determined by drying a 300 µL aliquot ofthe methanol extract containing sclerosin (methanol alone forthe control) in a tissue culture dish and then dissolving it in3 mL of molten potato dextrose agar (PDA, Difco). A sterilepaper filter containing S. sclerotiorum ascospores (dipped ina solution of 2 × 104 ascospores/mL) was placed in thecentre of the plate and incubated at room temperature for3 days, after which ascospore germination was measured.Sclerotia germination was assayed after washing sclerotia in10% bleach for 30 s, rinsing, dipping into purified sclerosinin 80% methanol (methanol alone was used for the control),and incubating on PDA plates. Germination was monitoredfor 72 h.

The effect of sclerosin on P. infestans zoosporesPhytophthora infestans US11 was maintained on V8-PDA

agar (per litre: 150 mL V8 juice, 10 g PDA, 3 g CaCO3, 10 gbacto agar (Difco)). Zoospores were released from sporangiausing the method described by Rohwer and colleagues(Rohwer et al. 1987). In brief, P. infestans was grown onV8-PDA for approximately 6 days until fungal hyphae cov-ered approximately ¾ of a 35 mm × 10 mm Petri plate. Theplate was then flooded with sterile water, and fungal hyphaewere flattened using the bottom of a sterile glass test tube.The plates were dried and incubated overnight at room tem-

perature to induce zoospore formation. Zoospores were har-vested the next day by flooding the plate with sterile water.The suspension was removed from the plate and filteredthrough cheesecloth to eliminate any adherent mycelia. To re-lease zoospores from sporangia, the filtrate was incubated at4 °C for 2 h. A 20 µL aliquot of the zoospore suspension wasadded to a glass slide and microscopically observed for lysisupon addition of an equal volume of the purified lipopeptide.

Generation of DF41 and DF41-1278 expressing themCherry red fluorescent proteinDerivatives of DF41 and DF41-1278 harbouring the

mCherry red fluorescent protein (RFP) were created. To con-struct a plasmid that could be stably maintained in Pseudo-monas, an 825 bp BamHI–EcoRI fragment containing therfp gene from pRSET-B mCherry was cloned into the samesites of pUCP23. The resulting plasmid, pMCh-23, was thenelectroporated into DF41 and DF41-1278, creating DF41-rfpand DF41-1278-rfp. The optimal excitation and emissionwavelength for mCherry RFP are 587 and 610 nm, respec-tively (Shaner et al. 2004).

Ingestion and digestion of DF41 and DF41-1278 byC. elegansCaenorhabditis elegans laboratory strain Bristol N2 nem-

atodes were maintained at 15 °C on nematode growth mediaspotted with benign live or UV-killed E. coli OP50 as a foodsource and were manipulated using established techniques(Hope 1999). To determine if LP41 causes any delay in C. el-egans development from the larval stage to egg-laying adultstage, 5 larval stage 1 (L1) nematodes hatched on UV-killedE. coli OP50 were placed on lawns of DF41-rfp and DF41-1278-rfp. A 100 µL aliquot of an overnight bacterial culturegrown in 1/10 KB was spotted onto a Petri plate (35 mm ×10 mm) containing 1/10 KB agar. The plates were incubatedfor 6 h at 28 °C to produce thin lawns of either DF41-rfp orDF41-1278-rfp. Assay plates were cooled to room tempera-ture prior to seeding with C. elegans, followed by incubationat 25 °C for 2 weeks. For microscopic examination, nema-todes were mounted on 2% agarose pads on glass microscopeslides and anesthetized with 10 mmol/L Levamisole (Sigma–Aldrich) in M9 buffer. Nematodes were examined by Nomar-ski differential interference microscopy with an ApoTome-equipped Zeiss Axio Imager.

Antibacterial activity of DF41 and DF41-1278The antibacterial activity of DF41 and DF41-1278 was

tested using Bacillus megaterium, Bacillus thuringiensis, andBacillus mycoides. Aliquots (5 µL) of DF41 and DF41-1278grown in KB broth were spotted onto PDA plates and incu-bated for 48 h at 28 °C. Overlays containing 3 mL of anovernight Bacillus culture in 10 mL of 0.7% agar wereadded, and the plates were incubated at room temperaturefor 48 h before the zones of inhibition were measured.

Results and discussion

Sclerosin is structurally similar to the tolaasin group oflipopeptidesThe structure of sclerosin was determined using MS. Pre-

liminary MALDI-TOF MS analysis of secondary metabolites

Berry et al. 1029

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

extracted from the growth medium of both DF41 and DF41-1278 revealed multiple components (Fig. 1), most of whichwere clustered at m/z < 1500, but 3 major ions at m/z2095.3, 2123.3, and 2145.3 were unique to the DF41 extract.On HPLC fractionation, the ion at m/z 2095 eluted at 27 min,and the ions at m/z 2123 and 2145 both eluted at 29 min, ap-parently differing only by a sodium ion. Electrospray ioniza-tion of the samples from 27 and 29 min gave multiplycharged ions of exactly the same mass as seen in the MALDIspectra (data not shown).MALDI MS/MS applied to determine the amino acid se-

quences yielded similar patterns of fragments for the ions atm/z 2095, 2123, and 2145. Fragmentation in the MALDIQqTOF instrument produces mainly y and b ions (Biemann1990) from which the sequence was manually assembled by“walking” along the spectrum (Fig. 2). Application ofTOFMA identified the standard amino acids, leaving 3 un-usual amino acids to be identified as Dab (2,4-diaminobuty-ric acid), Dhb (2-amino-butenoic acid), and Dhp (dehydro-2-aminopropanoic acid) (Table 2). A second b-type series (b*)lacking the N-terminal fatty acid and Dhb (253 Da smaller)was used to confirm the sequence, which was then tested us-ing pTOOL. The absence of aromatic amino acids in the pep-tide portion was confirmed by the lack of absorbance in the260–280 nm range. Because the m/z 2095 and m/z 2123 and2145 ions yield the same peptide sequence, the 28 Da differ-

ence lies in the fatty acid chains — 3-hydroxyoctanoic acidin 2095 and 3-hydroxydecanoic acid in 2123 and 2145 —consistent with the difference in retention times on theHPLC. The sequence determined for the 2123 and 2145 ionsis CH3-(CH2)6-CH(OH)-CH2-CO-Dhb-Pro-Ala-Leu/Ile-Ala-Val-Val-Dhb-Thr-Val-Leu/Ile-Dhp-Ala-Ala-Ala-Val-Dhb-Dhb-Ala-Dab-Ser-Val-OH.While cyclic lipopeptides are common, a linear sclerosin

was suggested by the clear identification of ions in the regionwhere cyclization would be expected. The linear nature wasconfirmed by the lack of new ions appearing after treatmentwith ammonium–methanol solution (Kuiper et al. 2004;Nutkins et al. 1991) or heating in 25% trifluoroacetic acid.As a control, the cyclic lipopeptide surfactin yielded C-terminus ions with an additional 18.01 Da when treated inthe same way (Yang et al. 2006). Thus the 3 ions m/z 2095and 2123 and 2145 are most likely linear.Strain DF41-1278 contains a Tn5 insertion in a region

with high sequence identity to an enzyme involved in thesynthesis of syringopeptins (Berry et al. 2010), a class oflipopeptides in the tolaasin group (Raaijmakers et al. 2006).Like sclerosin, this group of lipopeptides contains a high pro-portion of hydrophobic as well as unusual amino acids withinan 18–25 amino acid peptide portion. In particular, the ionsin sclerosin are similar in size and sequence to those of cor-peptin A (m/z 2094) and corpeptin B (m/z 2120), 2 lipopep-tides from the tolaasin group, produced by Pseudomonascorrugata that exhibit phytotoxic activity (Emanuele et al.1998). Indeed, the peptide sequence of sclerosin differs fromcorpeptins A and B by only 3 amino acids, which are shown

Fig. 1. MALDI (matrix-assisted laser desorption ionization) spectraof methanol-extracted metabolites from DF41 (A) and DF41-1278(B). Total counts after 40 s of accumulation are shown for the com-plete spectra, and for expansions of the region m/z 2075–2175 seeinset in both panels.

Fig. 2. MALDI MS/MS (matrix-assisted laser desorption ionizationtandem mass spectrometry) spectrum of the singly charged ion at m/z2123.3 from Fig. 1A. The spectrum is shown in 2 parts: from m/z100 to 1200 in the top panel, and from m/z 1100 to 2200 in the bot-tom panel (4× the scale of the top panel). The inferred sequence isshown above and uses the conventional labelling of y ions from thecarboxyl end, and b ions from the fatty acid (amino) end. The b*ions are from a second series lacking the N-terminal fatty acid andDhb (2-amino-butenoic acid). Only the most intense ions arelabelled. Full details of all ions and their expected values are listed inTable 2.

1030 Can. J. Microbiol. Vol. 58, 2012

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

in boldface and are underlined (3-hydroxydecanoic acid-Dhb-Pro-Ala-Leu/Ile-Ala-Val-Val-Dhb-Thr-Val-Leu/Ile-Dhp-Ala-Ala-Ala-Val-Dhb-Dhb-Ala-Dab-Ser-Val). The formation ofpaired homologues differing only in the fatty acid componentis common among lipopeptides including corpeptins A and B(a 26 Da difference) (Emanuele et al. 1998) and syringopeptins(a 28 Da difference) (Ballio et al. 1991; Grgurina et al. 2005).

Sclerosin inhibits S. sclerotiorum ascospore and sclerotialgerminationTo further characterize the inhibitory effect of sclerosin on

the growth of S. sclerotiorum, ascospores and sclerotia weretreated with the methanol extracts from DF41 (containingsclerosin) and DF41-1278 (no sclerosin) (Fig. 3). Both asco-spore and sclerotial germination were completely inhibited bythe sclerosin-containing extract but were unaffected by theextract of DF41-1278. That sclerosin is strongly antagonisticto the fungal pathogen S. sclerotiorum sets it apart from othermembers of the tolaasin group of lipopeptides, which mainlyexhibit phytotoxic activity, except for tolaasin itself, which istoxic to mushrooms. Clearly, sclerosin is not phytotoxic, atleast to canola plants (Berry et al. 2010), and the effectiveinhibition of ascospore and mycelial germination suggeststhe mechanism by which DF41 suppresses S. sclerotiorumproliferation.

DF41 displays antibacterial activity against Bacillus speciesAntibacterial activities have been ascribed to lipopeptides

and several members of the tolaasin group, particularly thesyringopeptins, that inhibit the growth of B. megaterium(Emanuele et al. 1998; Grgurina et al. 2005; Lavermicoccaet al. 1997). Consistent with inclusion of sclerosin in the tol-aasin group, DF41, but not DF41-1278, displayed strong in-hibition of B. megaterium and slightly less antagonismtowards B. thuringiensis and B. mycoides (Table 3).

Sclerosin does not affect zoospores of the oomycetepathogen P. infestansThe zoospore membranes of P. infestans, an oomycete

pathogen causing late blight in potato and tomato plants

Table 2. Observed ions from the MS/MS (tandem mass spectrometry) spectrum shown in Fig. 2 comparedwith those predicted (expected) by pTOOL.

Observed (expected)

bn bn*a yn

Ion m/z Ion m/z Ion m/z Predicted residueb

b0 NO (1.0) — — y23 2123.3 (2123.3) 3-OH Decanoic acidb1 NO (171.1) — — y22 1953.1 (1953.1) Dhbb2 254.2 (254.2) b0 NO (1.0) y21 1870.1 (1870.0) Prob3 351.2 (351.2) b1 NO (98.1) y20 1773.0 (1773.0) Alab4 422.2 (422.3) b2 169.1 (169.1) y19 1702.0 (1702.0) Leu/Ileb5 535.3 (535.4) b3 282.2 (282.2) y18 1588.9 (1588.9) Alab6 606.4 (606.4) b4 353.2 (353.2) y17 1517.9 (1517.8) Valb7 705.4 (705.5) b5 452.3 (452.3) y16 1418.8 (1418.8) Valb8 804.5 (804.5) b6 551.3 (551.4) y15 1319.7 (1319.7) Dhbb9 887.5 (887.6) b7 634.4 (634.4) y14 1236.7 (1236.6) Th/Hseb10 988.6 (988.6) b8 735.4 (735.4) y13 1135.7 (1135.6) Valb11 1087.6 (1087.7) b9 834.5 (834.5) y12 1036.6 (1036.5) Leu/Ileb12 1200.7 (1200.8) b10 947.6 (947.6) y11 923.5 (923.5) Dhpb13 NO (1269.8) b11 1016.6 (1016.6) y10 854.5 (854.4) Alab14 1340.7 (1340.8) b12 1087.6 (1087.7) y9 783.4 (783.4) Alab15 1411.8 (1411.9) b13 1158.7 (1158.7) y8 712.4 (712.4) Alab16 1482.9 (1482.9) b14 1229.7 (1229.7) y7 641.4 (641.3) Valb17 1581.9 (1582.0) b15 1328.7 (1328.8) y6 542.3 (542.3) Dhbb18 NO (1665.0) b16 1411.8 (1411.8) y5 459.3 (459.2) Dhbb19 NO (1748.0) b17 1494.8 (1494.9) y4 376.2 (376.2) Alab20 NO (1819.1) b18 NO (1565.9) y3 305.2 (305.1) Dabb21 1919.1 (1919.1) b19 1665.9 (1666.0) y2 205.1 (205.1) Serb22 2006.1 (2006.2) b20 1753.0 (1753.0) y1 118.1 (118.1) Valb23 2105.0 (2105.2) b21 1852.0 (1852.1) y0 NO (19.0)

Note: NO, not observed.ab* ions lack the 253 Da of the N-terminal fatty acid and Dhb.bUnusual amino acids include Dab (2,4-diaminobutyric acid — C4H8N2O, 100.06 Da), Dhb (2-amino-butenoic acid or

2,3-dehydro-2-aminobutyric acid — C4H5NO, 83.04 Da), and Dhp (dehydro-2-aminopropanoic acid — C3H3NO,69.02 Da).

Table 3. Antimicrobial activity of DF41 and the sclerosin-deficientDF41-1278 against Bacillus species.

Zone of inhibition (mm)

Organism DF41 DF41-1278P. syringae pv.syringae B728a

B. megaterium 9.0 (1.5) 0.0 (0.0) 9.0 (1.0)B. thuringiensis 4.0 (2.0) 0.0 (0.0) 4.5 (1.5)B. mycoides 6.0 (3.0) 0.0 (0.0) 12.5 (2.0)

Note: Values are the mean (standard deviation) obtained from 6 replicates.

Berry et al. 1031

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

(Govers and Latijnhouwers 2004), suffer lipopeptide-mediated disruption (de Bruijn et al. 2007; de Souza et al.2003). When sclerosin was tested similarly, it exhibited noeffect on Phytophthora zoospores. The lipopeptides exhib-iting zoosporicidal activity, including massetolide, viscosi-namide, and viscosin, are all smaller than sclerosin andother tolaasin-like lipopeptides. None of these latter mole-cules have been associated with lytic activity against zoo-spores, suggesting that the size of the lipopeptide is adeterminant.

Caenorhabditis elegans feeding on Pseudomonas sp.strains DF41 and DF41-1278 exhibit decreased survivalcompared with E. coli-fed larvaeIn nature, bacteria are constantly at risk of predation and

have evolved defence strategies, including lipopeptide pro-duction, to evade grazing by protozoa and nematodes(Jousset et al. 2010; Mazzola et al. 2009). However, amphisinproduced by Pseudomonas sp. strain DSS73 does not preventC. elegans ingestion but does enhance the flow of bacteriathrough the nematode intestine (Bjørnlund et al. 2009).When C. elegans was grown on DF41 and the sclerosin-deficient DF41-1278, larvae were not produced; in contrastto E. coli OP50-fed nematodes, which did generate offspring(Fig. 4). Microscopic examination revealed that rfp-expressing DF41 and DF41-1278 did not accumulate in thedigestive tract and were degraded after passage through thepharyngeal mechanical grinder (data not shown). Thus, un-like amphisin, sclerosin does not influence the passage ofbacteria through the nematode. At present, it is not knownwhether DF41 is producing metabolites that are toxic to thenematodes or whether this bacterium represents a suboptimalfood source for C. elegans and as such doesn’t support repro-duction. Studies are ongoing to address these questions.

In summary, Pseudomonas sp. strain DF41 produces thelipopeptide sclerosin that closely resembles corpeptins, mem-bers of the tolaasin class of lipopeptides, and represents anew addition to this class. The biological activity of sclerosindiffers in presenting a strong antifungal activity againstS. sclerotiorum as well as antibacterial activity against Bacil-lus species. On the other hand, sclerosin is not phytotoxic to-wards canola plants, shows no inhibition of P. infestansoomycete growth, and does not affect grazing by or passagethrough C. elegans.

AcknowledgementsWe thank Dr. K.G. Standing and Dr. W. Ens of the Depart-

ment of Physics and Astronomy for access to their massspectrometers and the software needed to analyse the spectra.V. Spicer is thanked for making adaptations to TOFMA, cre-ating pTOOL, and modifying pTOOL to include unusualamino acid residues. Financial support for this work was pro-vided through grants from the Natural Sciences and Engi-neering Research Council (NSERC) Discovery GrantsProgram (T.R. de K., A.K.C.B., W.G.D.F., and P.C.L.) andthe Canada Research Chair Program (P.C.L.).

ReferencesBallio, A., Barra, D., Bossa, F., Collina, A., Grgurina, I., Marino, G.,

et al. 1991. Syringopeptins, new phytotoxic lipodepsipeptides ofPseudomonas syringae pv. syringae. FEBS Lett. 291(1): 109–112.doi:10.1016/0014-5793(91)81115-O. PMID:1936237.

Bender, C.L., Alarcon-Chaidez, F., and Gross, D.C. 1999. Pseudo-monas syringae phytotoxins: mode of action, regulation, andbiosynthesis by peptide and polyketide synthetases. Microbiol.Mol. Biol. Rev. 63(2): 266–292. PMID:10357851.

Berry, C., Fernando, W.G., Loewen, P.C., and de Kievit, T.R. 2010.Lipopeptides are essential for Pseudomonas sp. DF41 biocontrolof Sclerotinia sclerotiorum. Biol. Control, 55(3): 211–218. doi:10.1016/j.biocontrol.2010.09.011.

Berti, A.D., Greve, N.J., Christensen, Q.H., and Thomas, M.G. 2007.Identification of a biosynthetic gene cluster and the six associatedlipopeptides involved in swarming motility of Pseudomonassyringae pv. tomato DC3000. J. Bacteriol. 189(17): 6312–6323.doi:10.1128/JB.00725-07. PMID:17601782.

Fig. 3. Sclerosin-mediated inhibition of Sclerotinia sclerotiorum as-cospore and sclerotia germination. Filter papers containing asco-spores were applied to potato dextrose agar (PDA) alone (A), PDAcontaining sclerosin extract (B), or PDA containing dried methanol(C). Plates were examined after 72 h at room temperature. Arrowspoint to ascospores that have germinated (panels A and C) or havenot germinated (panel B). Sclerotia that were untreated (D) or pre-treated with sclerosin (E) or with methanol (F) were placed ontoPDA and allowed to incubate at room temperature for 72 h.

Fig. 4. Caenorhabditis elegans larval production. Nematodes fedPseudomonas sp. strains DF41 (○), DF41-1278 (D), or the Escher-ichia coli OP50 control strain (×) were incubated at 25 °C on King’sB agar. Experiments were repeated twice.

1032 Can. J. Microbiol. Vol. 58, 2012

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

Biemann, K. 1990. Appendix 5. Nomenclature for peptide fragmentions (positive ions). Meth. Enzymol. 193: 886–887. doi:10.1016/0076-6879(90)93460-3. PMID:2074849.

Bjørnlund, L., Rønn, R., Péchy-Tarr, M., Maurhofer, M., Keel, C.,and Nybroe, O. 2009. Functional GacS in Pseudomonas DSS73prevents digestion by Caenorhabditis elegans and protects thenematode from killer flagellates. ISME J. 3(7): 770–779. doi:10.1038/ismej.2009.28. PMID:19340083.

Brenner, S. 1974. The genetics of Caenorhabditis elegans. Genetics,77(1): 71–94. PMID:4366476.

Dalla Serra, M., Fagiuoli, G., Nordera, P., Bernhart, I., Della Volpe,C., Di Giorgio, D., et al. 1999. The interaction of lipodepsipeptidetoxins from Pseudomonas syringae pv. syringae with biologicaland model membranes: a comparison of syringotoxin, syringo-mycin, and two syringopeptins. Mol. Plant–Microbe Interact.12(5): 391–400. doi:10.1094/MPMI.1999.12.5.391. PMID:10226372.

de Bruijn, I., de Kock, M.J.D., Yang, M., de Waard, P., van Beek, T.A., and Raaijmakers, J.M. 2007. Genome-based discovery,structure prediction and functional analysis of cyclic lipopeptideantibiotics in Pseudomonas species. Mol. Microbiol. 63(2): 417–428. doi:10.1111/j.1365-2958.2006.05525.x. PMID:17241198.

de Bruijn, I., de Kock, M.J., de Waard, P., van Beek, T.A., andRaaijmakers, J.M. 2008. Massetolide A biosynthesis in Pseudo-monas fluorescens. J. Bacteriol. 190(8): 2777–2789. doi:10.1128/JB.01563-07. PMID:17993540.

de Souza, J.T., de Boer, M., de Waard, P., Van Beek, T.A., andRaaijmakers, J.M. 2003. Biochemical, genetic, and zoosporicidalproperties of cyclic lipopeptide surfactants produced by Pseudo-monas fluorescens. Appl. Environ. Microbiol. 69(12): 7161–7172.doi:10.1128/AEM.69.12.7161-7172.2003. PMID:14660362.

Emanuele, M.C., Scaloni, A., Lavermicocca, P., Jacobellis, N.S.,Camoni, L., Di Giorgio, D., et al. 1998. Corpeptins, new bioactivelipodepsipeptides from cultures of Pseudomonas corrugata. FEBSLett. 433(3): 317–320. doi:10.1016/S0014-5793(98)00933-8.PMID:9744818.

Finking, R., and Marahiel, M.A. 2004. Biosynthesis of nonribosomalpeptides. Annu. Rev. Microbiol. 58(1): 453–488. doi:10.1146/annurev.micro.58.030603.123615. PMID:15487945.

Fiore, A., Mannina, L., Sobolev, A.P., Salzano, A.M., Scaloni, A.,Grgurina, I., et al. 2008. Bioactive lipopeptides of ice-nucleatingsnow bacterium Pseudomonas syringae strain 31R1. FEMSMicrobiol. Lett. 286(2): 158–165. doi:10.1111/j.1574-6968.2008.01247.x. PMID:18789127.

Forge, T.A., and MacGuidwin, A.E. 1989. Nematode autofluores-cence and its use as an indicator of viability. J. Nematol. 21(3):399–403. PMID:19287626.

Gerard, J., Lloyd, R., Barsby, T., Haden, P., Kelly, M.T., andAndersen, R.J. 1997. Massetolides A–H, antimycobacterial cyclicdepsipeptides produced by two pseudomonads isolated frommarine habitats. J. Nat. Prod. 60(3): 223–229. doi:10.1021/np9606456. PMID:9157190.

Govers, F., and Latijnhouwers, M. 2004. Late blight. In Encyclopediaof plant and crop science. Marcel Dekker, N.Y., USA. pp. 1–5.

Grgurina, I., Bensaci, M., Pocsfalvi, G., Mannina, L., Cruciani, O.,Fiore, A., et al. 2005. Novel cyclic lipodepsipeptide fromPseudomonas syringae pv. lachrymans strain 508 and syringo-peptin antimicrobial activities. Antimicrob. Agents Chemother.49(12): 5037–5045. doi:10.1128/AAC.49.12.5037-5045.2005.PMID:16304170.

Grünewald, J., and Marahiel, M.A. 2006. Chemoenzymatic andtemplate-directed synthesis of bioactive macrocyclic peptides.Microbiol. Mol. Biol. Rev. 70(1): 121–146. doi:10.1128/MMBR.70.1.121-146.2006. PMID:16524919.

Hope, I.A. 1999. C. elegans: a practical approach. Edited by B.D.Hames. Oxford University Press, Oxford.

Hutchison, M.L., and Gross, D.C. 1997. Lipopeptide phytotoxinsproduced by Pseudomonas syringae pv. syringae: comparison ofthe biosurfactant and ion channel-forming activities of syringo-peptin and syringomycin. Mol. Plant–Microb. Interact. 10: 347–354. doi:10.1094/MPMI.1997.10.3.347. PMID:9100379.

Jousset, A., Rochat, L., Scheu, S., Bonkowski, M., and Keel, C. 2010.Predator–prey chemical warfare determines the expression ofbiocontrol genes by rhizosphere-associated Pseudomonas fluor-escens. Appl. Environ. Microbiol. 76(15): 5263–5268. doi:10.1128/AEM.02941-09. PMID:20525866.

Kozlovski, V.I., Donald, L.J., Montero-Collado, V., Loboda, A.V.,Chernushevich, I.V., Ens, W., and Standing, K.G. 2005. TOF Massspectrometer with ESI source, orthogonal injection, and 16 kVacceleration voltage, for study of noncovalent complexes. InProceedings of the 53rd ASMS Conference Mass Spectrometryand Allied Topics, San Antonio, Texas.

Kuiper, I., Lagendijk, E.L., Pickford, R., Derrick, J.P., Lamers, G.E.,Thomas-Oates, J.E., et al. 2004. Characterization of twoPseudomonas putida lipopeptide biosurfactants, putisolvin I andII, which inhibit biofilm formation and break down existingbiofilms. Mol. Microbiol. 51(1): 97–113. doi:10.1046/j.1365-2958.2003.03751.x. PMID:14651614.

Lavermicocca, P., Iacobellis, N.S., Simmaco, M., and Graniti, A.1997. Biological properties and spectrum of activity of Pseudo-monas syringae pv. syringae toxins. Physiol. Mol. Plant Pathol.50(2): 129–140. doi:10.1006/pmpp.1996.0078.

Loboda, A.V., Krutchinsky, A.N., Bromirski, M., Ens, W., andStanding, K.G. 2000. A tandem quadrupole/time-of-flight massspectrometer with a matrix-assisted laser desorption/ionizationsource: design and performance. Rapid Commun. Mass Spectrom.14(12): 1047–1057. doi:10.1002/1097-0231(20000630)14:12<1047::AID-RCM990>3.0.CO;2-E. PMID:10861986.

Loper, J.E., and Lindow, S.E. 1987. Lack of evidence for in situfluorescent pigment production by Pseudomonas syringae pv.syringae on bean leaf surfaces. Phytopathology, 77(10): 1449–1454. doi:10.1094/Phyto-77-1449.

Marahiel, M.A. 1997. Protein templates for the biosynthesis ofpeptide antibiotics. Chem. Biol. 4(8): 561–567. doi:10.1016/S1074-5521(97)90242-8. PMID:9281530.

Mazzola, M., de Bruijn, I., Cohen, M.F., and Raaijmakers, J.M. 2009.Protozoan-induced regulation of cyclic lipopeptide biosynthesis isan effective predation defense mechanism for Pseudomonasfluorescens. Appl. Environ. Microbiol. 75(21): 6804–6811.doi:10.1128/AEM.01272-09. PMID:19717630.

Mott, K.A., and Takemoto, J.Y. 1989. Syringomycin, a bacterialphytotoxin, closes stomata. Plant Physiol. 90(4): 1435–1439.doi:10.1104/pp.90.4.1435. PMID:16666948.

Nielsen, T.H., Sorensen, D., Tobiasen, C., Andersen, J.B., Christo-phersen, C., Givskov, M., and Sorensen, J. 2002. Antibiotic andbiosurfactant properties of cyclic lipopeptides produced byfluorescent Pseudomonas spp. from the sugar beet rhizosphere.Appl. Environ. Microbiol. 68(7): 3416–3423. doi:10.1128/AEM.68.7.3416-3423.2002. PMID:12089023.

Nutkins, J.C., Mortishire-Smith, R.J., Packman, L.C., Brodey, C.L.,Rainey, P.B., Johnstone, K., and Williams, D.H. 1991. Structuredetermination of tolaasin, an extracellular lipodepsipeptideproduced by the mushroom pathogen, Pseudomonas tolaasiiPaine. J. Am. Chem. Soc. 113(7): 2621–2627. doi:10.1021/ja00007a040.

Purdy, L.H. 1979. Sclerotinia sclerotiorum: history, diseases andsymptomatology, host range, geographic distribution, and impact.Phytopathology, 69(8): 875–880. doi:10.1094/Phyto-69-875.

Berry et al. 1033

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.

Raaijmakers, J.M., de Bruijn, I., and de Kock, M.J. 2006. Cycliclipopeptide production by plant-associated Pseudomonas spp.:diversity, activity, biosynthesis, and regulation. Mol. Plant–Microbe Interact. 19(7): 699–710. doi:10.1094/MPMI-19-0699.PMID:16838783.

Raaijmakers, J.M., de Bruijn, I., Nybroe, O., and Ongena, M. 2010.Natural functions of lipopeptides from Bacillus and Pseudomonas:more than surfactants and antibiotics. FEMS Microbiol. Rev.34(6): 1037–1062. doi:10.1111/j.1574-6976.2010.00221.x.PMID:20412310.

Rohwer, F., Fritzemeier, K.H., Scheel, D., and Hahlbrock, K. 1987.Biochemical reactions of different tissues of potato (Solanumtuberosum) to zoospores or elicitors from Phytophthora infestans.Planta, 170(4): 556–561. doi:10.1007/BF00402991.

Savchuk, S., and Fernando, W.G.D. 2004. Effect of timing ofapplication and population dynamics on the degree of biologicalcontrol of Sclerotinia sclerotiorum by bacterial antagonists. FEMSMicrobiol. Ecol. 49(3): 379–388. doi:10.1016/j.femsec.2004.04.014. PMID:19712288.

Shaner, N.C., Campbell, R.E., Steinbach, P.A., Giepmans, B.N.,Palmer, A.E., and Tsien, R.Y. 2004. Improved monomeric red,orange and yellow fluorescent proteins derived from Discosomasp. red fluorescent protein. Nat. Biotechnol. 22(12): 1567–1572.doi:10.1038/nbt1037. PMID:15558047.

Shevchenko, A., Loboda, A., Shevchenko, A., Ens, W., Standing, K.G., Ens, W., and Standing, K.G. 2000. MALDI quadrupole time-of-flight mass spectrometry: a powerful tool for proteomic

research. Anal. Chem. 72(9): 2132–2141. doi:10.1021/ac9913659. PMID:10815976.

van de Mortel, J.E., Tran, H., Govers, F., and Raaijmakers, J.M. 2009.Cellular responses of the late blight pathogen Phytophthorainfestans to cyclic lipopeptide surfactants and their dependence onG proteins. Appl. Environ. Microbiol. 75(15): 4950–4957. doi:10.1128/AEM.00241-09. PMID:19502443.

West, S.E., Schweizer, H.P., Dall, C., Sample, A.K., and Runyen-Janecky, L.J. 1994. Construction of improved Escherichia–Pseudomonas shuttle vectors derived from pUC18/19 andsequence of the region required for their replication in Pseudo-monas aeruginosa. Gene, 148(1): 81–86. doi:10.1016/0378-1119(94)90237-2. PMID:7926843.

Yang, S.Z., Wei, D.Z., and Mu, B.Z. 2006. Determination of theamino acid sequence in a cyclic lipopeptide using M.S. with DHTmechanism. J. Biochem. Biophys. Methods, 68(1): 69–74. doi:10.1016/j.jbbm.2006.03.008. PMID:16650479.

Zhang, Y., Fernando, W.G.D., de Kievit, T.R., Berry, C., Daayf, F.,and Paulitz, T.C. 2006. Detection of antibiotic-related genes frombacterial biocontrol agents with polymerase chain reaction. Can. J.Microbiol. 52(5): 476–481. doi:10.1139/w05-152. PMID:16699573.

Zhong, H., Marcus, S.L., and Li, L. 2005. Microwave-assisted acidhydrolysis of proteins combined with liquid chromatographyMALDI M.S./M.S. for protein identification. J. Am. Soc. MassSpectrom. 16(4): 471–481. doi:10.1016/j.jasms.2004.12.017.PMID:15792716.

1034 Can. J. Microbiol. Vol. 58, 2012

Published by NRC Research Press

Can

. J. M

icro

biol

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

UN

IV M

AN

ITO

BA

on

08/2

9/12

For

pers

onal

use

onl

y.