CA2132705A1 - Biological control of post-harvest diseases - Google Patents

Abstract

Novel bacterial isolates are provided for the biological control of post-harvest rot causative organisms, particularly by antagonism to fungal growth. Particularly the present invention provides novel isolates of bacteria of species Pseudomonas fluorescens, Serratia liquefaciens, Serratia plymuthica, Bacillus subtilis, Bacillus pumilis) and Bacillus polymyxa which are particularly effective in inhibiting the growth of organisms of the post-harvest disease causing fungi Botrytis cinerea and Alternaria brassicicola). Further provided are antibiotics derived from the Bacillus species.

Description

~ W O 93/186~4 PCT/GB93/00604 J~ 0~

The present invention relates to novel bacterial isolates and to their use as agents, or source of agents, antagonistic to growth of rot causing organisms. Particularly the present invention provides novel isolates of b~cteria of species Pseudomonas fluore~ens, S~a~i~
lisscf~icn~, serrs~llLLly-uthica~ Bacillus subtilis, Bacillu~ ~umilis and ~acillus Dolvmvxa which are particularly effective in inhibiting the growth of organisms of the post-harvest disease causing fungi ~otrvtis cinerea and ~ . Further provided are antibiotics derived from the ~ species.

Post-harvest losses during storage of plant produce are caused, inter alia, by water loss, leaf senescence, regrowth and rotting, the latter particularly caused by f~ngal and bacterial pathogens. Such post -harvest losses can be OEeatly reduced by storage at low temperatures (eg.1-20C) and high relative humidity (eg.95%)(Robinson et al.
1975). Under these storage conditions ~. cinerea is the prevalent fungus found on stored cabbages and the main reason for losses (Geeson 1978, Brown et al. 1975); it be~ng an opportunistic pathogen of a wide range of leafy vegetables attacking weakened, wounded or senescent leaf tissues and is also known to attack various fruit (see patent US 5041384). Healthy leaf tissues hsve, however, been described as being highly resistant to ~sL~YSi~ attack (Newhook 1951).
Initial resistance of the exposed outer leaves is also likely to be the reason for the finding that 32~Y$15 rot of cabbages usually starts after 2-3 months of cold storage and is often confined to the outer, dried out, senescent cabbage leaves (Wale 1980).

To prevent fungal spoilage it is common practice in many countries to spray cabbages with systemic fungicides in the field snd to dip harvested cabbage hesdæ in fungicide solutions prior to storage ~Brown et al. 1975). Since the oncogenic nature of many of the most commonly used fungicides is increasingly recognized and because the persistence of most fungicides is incre~sed by the low storsge temperatures the postharvest use of fungicides is of growing concern.

W O 93/lX654 ' PCT/GB93/006~4 -~
'~

2 ';' Additionally, resistance to the fungicides ~sed has been reported (Spotts & Cervantes 1986) and suppression of the main spoila~e organism ~ _siD~a by fungicides such as benomyl has been shown to result in incressed populstion of ~ ls~l- which causes a more penetrating rot of cabbage heads than ~. cinerea (Wale & Epton 1979).

Control of post-harvest fungal pathogens by bacteria and yeast antagonists such es ~ e d~9~
Pseudomonas sYrinFae, ~ b ~ - h~ and Debarvo~vces han~enii has been described for a variety of stored vegetables and fruit including apple, apricot, cherry, Cit~lS, grape, nectarine, peach, pear, pepper, persimmon. plum, potato and tomato (see Wilson & Wisniewski, 1989, for a recent review). However, the majority of studies have only investigated the antagonistic effects at temperatures of around 20C, but not at temperatures of between 1 and 10C which are used for co~mercial storage of most fruit and vegetables. Additionally very few publications have tested the persistence of antagonists on plant surfaces or compared microbial , -populations on stored pl~nt organs with the antagonistic microflora.

The present invention provide~ novel bacterial isolates of species which have the property of inhibiting the growth o~ fungal species on post-harvest products; these being species of ~s~lnmans~. Serrati~
and ~acillux. Most advantageously, the Pseu~omcns~ and Serratia isolates of the present invention have the property of being able to inhibit fungal growth at cold storage temperatures, eg. below 10C, typically from 0 to 4C. Furthermore, the ~s~LLlD~ isolates have the ~roperty of expressing an antibiotically active fraction also provided by the invention which itself is capable of isolation in pure or semipure form and being us~d to inhibit growth o~ such fungal species.

,It should be realised that the isolates of the present invention have been selected from a background population of many thousands of such strains which are not effective. However, given access to the deposited isolates referred to above, and the data and protocols W O 93/18654 ~ b ~ PCT~GB93/00604 relating thereto below~ it will be recognised by those skilled in the art that further related strains with similar antagonistic qualities will be isolatable and thus the scope of the present invention covers such related strains in so far as then might be recognised by comparison with the strains pr~vided and the data and protocols herein.

Samples of the novel bacterial isolates of the present invention have been deposited under the terms of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the purpose of patent procedures at the UK National Collection of Industrial and Marine Bacteria, 23 St Machar Drive, Aberdeen, AB2 lRY, UK, under the following accession numbers:

~e~ignation Accession No. ~ t~_Q~ osit~
Pseudomonas fluorescens CL42 NCIM3 40495 24.03.92 '' CL66 NCIMB 40490 23.03.92 '' '' CL82 NCIMB 40497 31-03-97 5r~c~ Ym~bi~ CL43 NCIMB 40493 23.03.92 Serr~is~lic~acien~ CL80 NCIMB 4~492 23.03.97 ~Csl~l "_~b~ CL27 NCIMB 40491 23.03.92 ~ li CL45 NCIMB 40489 23.03.92 The present invention provides isolated bacteria having the identifying characteristics of an isolate selected from the group consisting of NCIMB 40489, NCIMB 40490, NClMB 40491, NCIMB 40492, NCIMB 40493, NCIMB 40494, NCIMB 40495 and NCIMB 40497. -Such isoistes share the characteristic of inhibiting growth of fungi of specie~ ~
brassi~i~Qla and/or ~L_ cinere~ on cabbage tissue:agar:water mixtures.

A particul~rly preferred aspect of the present invention provides a bacterial -olate having the identifying characteristics of an isolate selected from the group consisting of NCIMB 40490, NCIMB 40492, NCIMB
40493, NCIMB 40495 and NCIMB 40497, said isolate being cspable of inhibiting growth of A. bra&sisi~ol~ and/or B. cinerea on cabbage leaves at temperatures of 20C or less, more preferably at 10C or less, most preferably at 4C or les~. More preferably the isolate is of species selected from the ~roup consisting of Ps~udQmQn~ ' fl~n~n~. ~9~ ho-A and Serrati~ liauefacien~.
.
A further preferred aspect of the pre~ent invention provides a -bacterial isolate having the identifying characteristics of an isolate selected from the group comprising NCIMB 40489 ~nd NCïMB 40491, said isolate being capable of inhibiting growth of ~_ _b~ ol~ and/or B. cine~i~ on cabbage leaves at temperatures of about 20~C. More preferably the isolate is capable of production of an antibiotic fraction when cultured upon a broth of homogenised cabbage leaves of -concentration 50g or more per litre of water, said antibiotic being .
capable of inhibiting growth of said A. brassici~ola and/or .
~in~ia on cabbage leaves at temperatures of about 20C.

A further aspect of the invention provides aqueous suspensions of bacteria of one or more of the bacterial isolates of the invention, preferred suspensions being of sufficient cfu/ml to inhibit the fung growth referred to above on fruit and/or vegetables that have been dipped therein, and further to biologically pure cultures of such ~-bacteria at higher cfu/ml for preparing suspensions for such dipping. `~

The present invention thus further provides a process for the prevention of fungal disease in post-harvest vegetables and/or fruit products comprising dipping said products into suspension of bacteria as described above. Preferably the suspension will contain from between 104 and 109, more preferably between 105 and 108 cfu/ml.
Using the non-Bacillus group cold ~zor~ge is a preferred option.
.
A still further aspect of the invention provides antibiotics obtainable by culturing isolated bacteria of the type of NCIMB 40489 and/or NCIMB 40491 in a medium comprising homogenised cabbage leaves and water in a ratio o~ 5g to lOOg/litre. Further provided are compositions charscterised in that they have such antibiotics as active ingredients.

t~ f.~ t ~ ~

The isolates, their efficacy against fungal growth, methods of selecting similar such isolates ~nd methods for their use in achieving this in industrial application will now be exemplified, by way of illustration only by referral to the following Figures and Examples.

Firures:

Figure 1. Inhibition zone diameter (mm) around bacterial colonies on cabbage agar conSaining different concentrations of cabbage tissue seeded with A. brassicicola and ~ sin~a- - CL42;~ =CL66; ~ CL82.

F~gure 2. Mean inhibition zone diameter (mm) (vertical bars =
standard deviation) around bacterial colonies on cabbage agar containing different concentrations of cabbag~ tissue seeded with b~sssi~icQ1~ and B. cinerea (mean of 3 determinations per isolate).
= Serratia ~lvmu~hica (CL43),~ - ~erratia_liauefacien~ CL57: isolates 58, 59, 61, 62 and 80 showed similar results.

Figure 3. isolate CL27 growth, sporulation and production of antagonistic activity against A. brassici~nl~ and ~.
~ing~a in cabbage broth ~A: C85, B: C810~ batch culture. Antifungal activity wax ~ssayed in Sens-acute plates containing fungal seeded CA5. z cfu ~ = spores; e activity against B. cinerea ~ activity against ~

Figure 4. Bacillus Dumilus isolate CL45 growth, sporulation and production of antagonistic activity against ~ and E._c1n~r~a in cabbage~broth (A: CB5~ B: CB10) batch culture.
i~ ' Antifungal activity was assayed in Sens-acute plates containing fungal seeded CA5 ~ s cfu ~ =spores ~ activity against ~ cinerea;
=activity against A. brassicicsl~.

Figure 5. _~ lll _ b~lll- isolate CL27 growth, sporulation and production of antagonistic activity against A. brassicicola and ~
in nutrient broth (A: NB, B: NBIMg, C: NBllx g~ucose) batch W O 93/186$4 ~ Q ~ PCT/GB93/00604 !

~ :' culture. Antifungal activity was assayed in Sens-acute plates containing fungal seeded CA5. =cfu; O =spores; ~ =activity against -sin~a; O =activity against A. bras~icicol~.

Figure 6. isolAte CL45 growth, sporulation And production of activity against A. bras~i~icoLa and ~. cinerea in nutrient broth ~A: NB, B: NB~Mg) batch culture. Antifungal activity was assayed in Sens-acute plates containing fungal seeded CA5. O
=cfu; O =spores, ~ =activity against ~. cinereA; ~ ctivity against A.
brassicicol~ ~
:, Figure 7. ~ isolate CL27 growthc sporulation and -production of activity against A. brassi~ics~Land B~ cinere~ in defined medium (A: DM, B: DMl1% glucose) batch culture. Antifungal activity assayed in Sens-acute plates containing fungal seeded CA5.
=cfu; O =spores; -activity against B. cinerea; O =activity against A. brsssicicola. ~:

Figure 8. Effect of medium pH on the activity of antiblotics from 3ssills5 isolate GL27 and CL45 batch cultured extracted 2 days (A&C~
or 4 days (8&D) after inoculation. In A, B, ~nd D inhibition of the pathogen was determined on Sens-acute Micro-detection plates containing ~I_sin~C~ seeded CA5 medium, in C on PVA. Batch culture media: =CB10; o =CB5; -N8; 0 =NB~Mg.

Figure 9. Effect of nutrient type and concentration in the assay medium (homogeni2ed cabbsge tissue or potato dextrose broth solidified wi~h 10 g/l agar) on the size of inhibition zones formed. Media was seeded with ~, ciner~a spores. The crude extract used was taken after

3 days from B. subtilis isolate CL27 CB5 batch cultures. Different concentrations of antibiotics were obtained by diluting the crude extract (c~nc. 100%) with sterile CB5. Assay media used: ~ =50Og/ml;
O =lOOmg/ml; ~ =200mg/ml homogenized cabbage tissue; A -1/4 strength PDB; ~1/2 strength PDB; -full strength PDB(Potato dextrose broth).

W 0 93/1~654 ~ 5 ~ PCT/GB93/00604 EXAMPLE l._Isolation of anta~onis~ n~ g,~ lga leaf tissues.
A lOxlO cm square was aseptically cut from the outer~ost cabbage leaf of a head and placed in a plastic bag with 25 ml of sterile Ringer's solution. The tissue was then placed in a stomacher for lO min to remove surface organisms. Appropriate dilution steps were plated onto nutrient agar, cabbage agar (homogenized cabbage leaf tissue either lOg [CAl], 20g [CA2], 50g [CA5] or lOOg [CAlO]) and lOg agar per litre water), cabbage agar seeded with Botr~ti~ spores (as described below), Sabourough agar (Oxoid) and potato dextrose agar (Oxoid). Antagonists were detected by the inhibition zones formed either when leaf washing~
were directly plated on Botrvtis-seeded agar or after replica-plating of colonies onto Botrvtis-seeded agar.

IdentificaL~l~of the general~m~ a~ cabba~e leav~
The number of colonies of filamentous fungi was counted directly on ~ isolation plates. TQ estimste the number of yeasts, fluorescent : pseudomonads and Enterobacteriaceae, 140 colonies were randomly picked from isolation plates and identified. Phase contrast microscopy was used So identify yeasts and to examine bacteria for motility. Since all bacteria found were Gram-negative the isolstes showing motility and fluorescence on King's medium B were classified as fluorescent pseudomonads and those showing anaerobic growth (Hugh & Leifson 1953) as Enterobacteriaceae.. Other bacterial isolates were not identified.
:: :
The number of Bacillus spores was estimated from the number of colonies developing after 20 fold concentration of the leaf washings by centrifugation at 3600 rev/min and heating (800C for ~0 min).

Gram's stain (Jensen's modification), shape, motility (under phase contrast microscopy), oxidase test (Kovacs 1956), heating (20 min at 80C) and the oxidation/fermentation test (Hugh ~ Leifson 1953) were used for initial identification. Oxidase-negative, acidifyin~
Gram-negative rods (Enterobacteriaceae) were further tested with API
20E, oxidase-positive, non-acidifying Gram-negative rods with API 20NE

4 PCI /GB93/00604 _ ~ ~
~ "~
~ ~ f O ~

and heat resistant Gram-positive rods (~sil1~ spp.) with API 20E and 50C B test strips (API-BioMerieux, UK.~. Results were analysed by the API computer identification software and identification of ~ia liau~f~i~n~ (Grimes & Hennerty) was confirmed by testing acid production from raffinose, manolate, lactose and adonitol using the methods described by Brenner (1984).

aration of fgn8~l_in~culum.
All fungHl stock cultures were grown on cabbage agar 5 (CA5; 50 g homogenized cabbag~ tissue, 10 g agar, 1 litre distilled water3 at 4C in the dark. Malt extract ~gar (MA) was used for spore production. e~ Y~ was grown for 5 days at 20C in the dark followed by 3 days under W-light to induce sporulation and Alternaria ~rassicicola was grown for 14 days at 20C prior to use in experiments. Spores were harves~ed by pouring 10 ml of sterile Ringer's solution onto a fungal plate and suspending spores using a bacteriological loop. Spore suspensions were filtered through a , double layer of muslin and adjusted to a concentration of 4X106 spores/ml using a haemacytometer.
. .

Inocula were prepared by suspending bacterial growth from a 24 hr NA
culture in 1/4 strength Ringer's solution. The turbidity was then adjusted to an absorbance of 1.0 at 625 nm using a ULTRO SPEC 4051 spectrophotometer (LKB 8IOCHROM Ltd., Cambridge, UK.) giving a bacterial concentration of 7X108 cfu/ml for ~seudomo~ e~cs and 5X108 cfu/ml for ~erratia isolates (the number of cfu was obtained by plating dilution steps of the suspension onto NA). Lower inoculum levels were prepared by appropriate dilution in 1/4 strength Ringer`s solution.

'In vitro' ~ss~Y~for antifun~al activitv.
Twenty ml of liquid CA5 was cooled to 40~ 5~ L~ spores are very sensitive to heating) and inoculated with 0.2 ml of a fungal suspension containing 4x105 spores/ml. After the agar had set plates W O 93/186~4 PCT/GB93/00604 a 9 "

were dried in a la~inar flow cabinet for 90 min and then inoculated immediately with a loopful of a bacterial suspension containing 2X108 cfu/m. When testing the antibioti~ activity in crude extracts (cell free culture filtrates) from batch cultures of ~a~illu5 spp. fungal seeded medium was poured into sens-acùte Micro detection trays (Intl.
Patent Application No PCT/GB90/01067; Proteus Molecular Design Ltd., Marple, Cheshire, UK); After the agar had set 4 mm wells were cut into the agar and filled with 30 ~1 of the crude extract. Inhibition zone diameters around colonies or wells were measured 2 days after inoculation of bacteria or cell free ex~r~c~s of ~otrvti~ snd 4 days after inoculation of Alternaria seeded media.

'In vivo' (l~af disk) assav~ an~ifuD~ tivitv, Lesf disks (15 mm diameter) were cut from internal leaves of cabbages (the outermost 5 leaves were discarded) using a cork~orer. Disks were transferred to a 25 ml universal bottle and heated for 20 min at 50C.
This treatment was found necessary to break the natural resistance of non-senescent internal cabbage leaf tissue to enable infection by ~
Sin~s and A. brassicicola spores. Disks were left to cool and then suspended in a bacterial suspension containing 7x108, 7x107 or 7X106 cfu/ml for p. fluorescens isolates and 5x108, 5x107 or 5X106 cfu/ml for ~ ia isolates. This resulted in approximately 107, 106 or 105 cfu/disk, when the num~ of c~u on leaf disks w~.s determined immediately after inocuLation m is determinat ;s w~s carried out by wa~hing (10 min at speed 6 US~llg a Gallenkamp ~, ~t shaker) disks in 2 ml sterile 1/4 strength Ringer's ~olution to r~-~ove bacteria ~rom the plant surfsc~s ~nd plating at appropriate dilutions on nutrient agar. ,;

After inoculation with bacteria disks were transferred to a tissue culture assay plate (8.5x13 cm) with 24 wells of 15 mm dia~eter. A
non-heated disc was placed under the inoculated heated disk to mimic the situation in a stored cabbage head where susceptible, senescent outer leaves overlay resistant metabolically active inner leaves.
Assay plates were then placed open in a laminar flow cabinet until the disk surfaces were dry. Afterwards 2.5 ~1 of fung~1 suspension W O 93/186~4 PCT/GB93~00604 ~
-.
;J

contaîning 103, 104 or 105 spores was plaoed in the centre of the disk. Assay plates were placed over water in a seed tray (15x20 cm) and the tray sealed with plastic film. The humidity in trays was ~-measured using a Vaisala 100 humidity probe (V~isala OY, PL26, Helsinki, Finland) and found to be between 97 and 99% throughout the experiment. Trays were incubated at 40C in darkness for 10 weeks.

Persisten~e o~ reinQculated anta~onist~ on leaf ~urfaces. `
To deter~ine the persistence of reinoculated antagonistic bacteria on leaf surfaces, non-hea~ed leaf disks (1.5 cm diameter) were dipped in a bacterial suspension and incubated in assay plates as described above. Prior to the removal of surface microorganisms a lxl cm square was cut from the disk in order to avoid sampling bacteria on the cut edges of the disk. The square was suspended in 2 ml of Ringer's solution and shaken for 10 min using a Gallenkamp wrist shaker at speed 6 to remove surface bacteria. The number of antagonistic bacteria in the leaf washing was determined on the basis of their ~n `
vitro antagonism against A. brassicicola by plating leaf washings on CA5 seeded with A. brassicicola and counting of colonies with inhibition zones.

G~neral and anta~onifitic microflsra found on Brassica s~.
When the number of viable microorganisms on the cabbage leaf surface was determined, leaves were found to hsrbour approximately 103-104 cfu/cm2 (Table 1). or these between 1 and 15% were yeasts, 65-80%
were fluorescent pseudomonads, 5-20% Enterobacteriaceae. Less than 1%
of surface microorganisms were filamentous fungi and Bacillus spp.
could only be isolated by the selective enrichment procedure described above. The number of viable ~acillus spores was between 1 and 20 cfu/20 cm2. Numbers of surface microorganisms were lower on internal leaves of cabbages Table 6 shows the numbers found on the 6th leaf.

W O 93/18~54 PCT/GB93J00604 f~ 5 Table 1. Numbers of viable microorganisms (cfu)/cm2 on Dutch White and January Glory cabbage leaves.

Dutch White January Glory Isolation medium nmean STD n mean STD
. _ ___ .
CA5 4 lx104 5x103 5 6x103 2x103 NA 7 8x103 3x103 4 5x103 2x103 SA 7 5x103 2x103 4 4x103 5x103 _ _ . _ n = number of leaves examined STD = standard deviation Sixty one pe~cent of antagonistic bscterial strains isolated from ~r~5sisa spp. were oxidase positive fluorescent pseudomonads, 30 %
were ~2cillY5 and 9~ were ~~gsia species. The different groups are described in separate sections below.

Fluorescent pseudomonads appear to be a major part of the natural flora of cabbage leaves since they were repeatedly isolated in large numbers. All antagonistic fluoresc~t pseudomonads isolated belonged to the species psçudomonas_f1~2~Sc5D~. All strains were isolated from conventionally and organically grown 'Dutch White' or 'January Glory' cabbages, broccoli or Brussels sprouts using a variety of different isolation media (Table 2). There were two main groups of antagonistic pseudomonads, those producing fluoresc~nt pigments on nutrient agar (OEoup 1: CL6,7,9,10,11,12,15,16,18,21,29,39,40) and those which did not (group 2: CL42,66,74,82,13,14,17,22,23,30,31-37,39), _~

1~:~ ' , Group 1 pseudomonads produced small zones of inhibitiQn (<4 mm) on W O 93/l8654 PCT/GB93/~0604 ~
.., ..

d ~ ;J ~ ~J J
12 .

. cinerea and A. brassicicola seeded CA5 ~gar, while Group 2 included: a. strains CL31-CL37 and CL49 which caused rotting of cabbage leaves (not tested further for 'in_Yis~' antagonism) and b.
all strains (CL42, 49, 66. 74, 82 and 50) which produced large inhibition zones (Table 2). Inhibition zones formed by di~ferent strains of group 1 pseudomonads at different temperatures and nutrient concentrations in the medium were generally small (1-4 mm, individual results not shown). However, when inhibition zones of strains from group 2 were compared, strain specific patterns were found (Table 2).
The size of inhibition zones on c~bbage agar decreased with decreasing concentrations of nutrients in the medium (between 5 and 100 g/l of cabbage tissue) for deposited strains CL42 ~nd CL82. However, strain CL66 showed increasing zones of inhibition with decreasing concentration of nutrients on cabbage agar see~ed with Alterna~ia but decreasing zones of inhibition in Botry~i~ seeded cabb~ge agar.
Significantly the inhibition zones at 4C were larger at all nut~ient concentrations with both fungi around CL82 colonies and around CL42 colonies on ~15~n~ia seeded agar than at 10C.

Table 2. In vitro inhibition of Altern~ia bra~:icicola and Botrvtis ~in~ by non-pathogenic group 2 E~eYl~mQn~ spp. isolated from ~aE~is~ spp. on CA5 (means from 3 determinations) . ,, ._ . _ _ _ _ _ _ Inhibition zone diameter (mm) . Alt~rn~ Q~
_ ~ _ Isolate Isolation code Origin medium 20C 10C 4C 20C 10C 4C
, _ _ CL74 DWLo NA 5 5 ., .

. W C Q3/l8654 PCT/GB93/00604 , . . .

Tsble 2 continued:
DWL Dutch white cabbage leaves DWLo Dutch white cabbage leaves (organically grown) JGL January Glory cabbage leaves ' yivo inhibition ~SY~lD~nsE antagonists also suppressed infection and disease development on cabbage tissue when cabbage leaf disk assay was used to detect ' m vivo' antagonism. P.~rl~nre~ isolates CL42, CL66 and CL82:showed best~control of both fungi on cabbage disks - preventing development of disease symptoms on disks when dipped in a suspension of 7X106 cfu/ml; CL73 only gave intermediate control. ~.

Table 3. Suppression of ~Lla~s~ a and ~ 1 Is~
by ~s~gd2ign~s:s~pp. on~cabbage:leaf disks (each experiment was x3) : , ~ ~, ;' ~ lr~ ~ ':
Bscteria inoculum - inoculum ~Bacterlal ~ inoculum No. of spores/disk No. of spores/disk ~i601àte~ (c~fu/ml) 104 103102 104 103 1o2 ;~ ; , CL42,66,82 7X108 ~ ~ ~ ~

6 1 ~V ~ J
-, . ~: .: : ~ , : :
inhibition - no inhibition;of fungal growth V vàriable~:results ND not determined End of experiment (6 weeks after control discs showed fungal growth) ExamDle ~: Serratia SD~.
Nine percent of antagonistic bacterial stràins isolated~from ~gssl~ spp. were Sass9~1a 8pp. (Enterobacteriaceae), eight s;trains being ~ la ll~"~si~n~ and one strain being . A}l strains were isolated from conventionally ;,: , ~ ;
- ~ ~
,,, , -: ..
, - ~, W 0 93/18654 PCT/GBg3/00604 ~ .
i~. 7,'.,', ' ~ 3 14 ' -and organically grown 'Dutch White' or 'January Glory' cabbages using a variety of different isolation media (Table 4). No antagonistic~ :
Enterobacteria~eae could be isolated from Brussel's sprout or broccoli.

Serratia strains grew well and showed 'in vitrQ' antagonism on CA5 medium at temper~tures of between 4 and 20C (Table 4). Inhibition zones on Botrvtis seeded CA5 were larger at 20 and 10C but smaller at 4C than on Al~ernaria seeded CA5. It should be noted that Botr~tis grew faster at all temperatures (developing visible growth on the medium) than Alternaria. Inhibition zones of S. lics~d~i~ns strains CL57, 58, 59, 61, 62, 73 and 80 were found to have a similar size, but smaller than those of S. ~lvmuthica strain CL43.

Table 4. vitro inhibition of Alternaria bras~icicQla and BotrYti~
~in9~ by Serratia spp. isolated from Brassica on CA5 (mean of 3).
_ _ . _ _ .
Inhibition zone diameter (mm) ~:
, _ _ . _ ~1S~SD~ Botrytis .:
: Isola~e Isolation code Origin medium 20C 10C 4C 20C 10C 4C
. __ ~h~ .' . _ _ _ ::~
~ .' CL59 DWL CA5 9 14 5 i3 18 1 CL62 JGL SA 8 14 6 16 10 0 .
CL73 DWLo CA5 6 CL80 DWLo CA5 10 14 6 15 10 `

_ _ See Table 2 for key WO 93/18654 PCr/GB93/00604 S ~ ~

The size of inhibition zone of all S. liauefacien~ strains were found to be similar at all nutrient concentrstions (between 5 and 100 g/l of cabbage tissue), except for inhibition zones formed around CL43 colonies at 10C with 10 g/l of cabbage tissue which are larger ~Figure 2). Inhibition zones of S. ~lvmuthica strain CL43, however, were generally larger and CL43 was the only strain exhi~iting `
significant antagonism at low nutrient concentrations ~5 and 10 g/l;
Figure 2).

:Table 5. Suppression of ~ ~ Yli~ e~a and AlL~ ic~
by ~s~ spp. on cabbage leaf disks at 4C (each experiment x3) :.
:''', . ~ ~otrvtis cinerea Alternaria brassic1cola inoculum inoculum ' ..
Bacterial: ~ :No. of spores/disk No. of spores/disk : isloate inoculum (cfu/ml) 10~ 103 102 104 103 102 : ~ _ ~ : ~ ,:

CL435xlOB ~ ~ ~ ~ ~ I
5x107 ~ V ~ , + + ,

5~ ~ V

CL80i 5x108~' : . ~ ~ ~ ~ , 5x107 V ~ ~ ~ ~ ~ .
: .5xI06 _ V ~ +
.
: : See Table 3 for key.
Experiment ends a~ 6 weeks after control disks showed fungal growth , , .

. W O ~3/18654 PCTtGB93/00604 . .

The size of inhibition zone of all 5 ~ strains were found to be similar at all nutrient concentrations (between 5 and 100 g/1 of cabbage tissue), except for inhibition zones formed around CL43 colonies at 10C with 10 g/l of cabbage tissue which are larger (Figure 2). Inhibition zones of ~ lYm~hi~a strain CL43, however, were generally larger and CL43 was the only strain exhibiting significant antagonism at low nutrient concentrations (5 ~nd 10 g/1;
Figure 2).

Table 5. Suppression of Botrv~i~ cinere~ and ~l~e na i-_~- 5519i 1 by Serratia spp. on cabbage leaf disks at 4C (each experiment x3) _ _ _ .
Botrvtis ~ine~a Alternari~ brassicicola inoculum inoculum Bacterial No. of spores/disk No. of spores~disk isloate inoculum (cfù/ml~ 104 103 102 104 103 102 ~ _ ~ . . ,' ~e~ra~i~ plvmuthica CL43 5x10 + ~ ~ + ~ +
5x107 V , + . + ~ +
5X106 .. _ , V + + '~' _ _ _ ~ 5 CL80 5X108 ~ + ~ , + +
5x107 V ~ + ~ ~ +
5X106 _ V + ~ ~ +
_ _ _ See Table 3 for key.
Experiment ends at 6 weeks after control disks showed fungal growth W O 93/186~4 PCT/GB93/00604 : ;~,;.

In vivo inhi~ition.
Serratia antagonists also suppressed infection and disease development on cabbage tissue when the cabbage leaf disk assay was used to detect 'in Yiv~' antagonism. When cabbage di~ks were dipped in a bacterial suspension with 2x109 (results not s~own), 2x108, 2x107 or 2X106 cfu deterioration of the tissue and the development of v~sible growth of a9s~Y5~ was inhibited to different extents, depending on :-bacterial species used and the fungal inoculum applied (Table 53. ;

Persistence of Pseudomona~ Se~ra~ia on cab~a~e leaves.
When the persistence of reinoculated antagonists was studied ~ ~
~luorescens and Serratia spp. were able to persist in large numbers ~:;
on cabbage leaf surfaces at low temperatures ~Table 6).

Table 6. Persistence of antagonistic Pseudomnnas fluorescens and ~-Serratia ~lvmuthica on cabbage leaf disks after inoculation of disks by dipping into a b~cterial suspension containing 7x107 cfu/ml for -~5~g~n~QD~5 and 5x107 cfu/ml for S~rr~a (mean is of 2 determinations).
_ _ . .`.
Total No. of cfu/cm2: (% of cfu showing antagonism :
against ~ 191 ~1 ) __ _ __ . ~............... _ , . ''~
Days after Non-inoculated P.fluoresc~ ~ S . D1VmU~hiGa inoculation control CL42 l L82 CL43 . . _ . _ .
0 7X102 (0) 5X104(>99%) 4X104(>99X) 3X104(>99%) :
14 7X102 (0) 2X104 ( >990 9X103 (98%) 5X104 ( >~90 28 . 1X103 (0.2*) 4X103(~7%) 6X103(95%) 2X104(>99%)48 3X103 (0) 5X103 (97%) 2X103 (80%) 2X104 ( >9~%) . _ ' ~

. _ * one antagonistic colony was observed W O 93/18654 PCT/~B93J00604 ,.. , .. : .
Q ~ ! :

xam~l~ 4: Bacillus SD~. General characteristics:

Thirty percent of all bacteria showing antagonism against ~otrvtis cin~ were found to be ~ssi11Y~ spp. Fourteen Bacillus isolates were identified and found to be B. su~tilis (8 isolate~
~milus (3 isolates) or B. ~olvmvxa (3 isolates).

Antibiotic DroductiQn in vitro (semiau~n~ iy~l_ ~S111~5 isolates isolated from ~ sEi~ spp. did not grow at 4 and 10C (except for Bacillus strains CL47, 52, 64 and 67 which produced very poor growth after 4 weeks al 10 and 6 weeks at 40C). At 20C
most strains produced large inhibition zones on CA5 medium (Table 7).
However, ~ci11Y~ did not grow or failed to produce large inhibition zones at lower nutrient concentrations in the medium (Table 7j.

:
Cabbage broth (50 g tCB5] or 100 g tCB10] of homogenized cabbage leaves, l litre distilled H20), nutrient broth (NB; Oxoid!, nutrien~
broth + 0.005g/1 MgS04 (NB+Mg), nutrient broth + 1~ glucos~ ~NB+1%
glucose), a defined medium (DM: 2~ NH4Cl, 6g Na2HP0~ ~- KH2i~04 3g NaCl, 0.05g MgS0~ 7H20, 5 mg L-methionine, 10 mg glyc ~ 2.5 mg L-aspartic acid, 10 g agar, 1 litre distilled H20) an 1 + 1X glucose were used as growth media. 100 ml of medium was tra- rred to 250 ml Erlenmeyer conical flasks and autoclsved and inoculate~ with bacteria from a 3 da~ old nutrient agar culture of the ~2~ a ætrains. Amino acids were filter sterilised and added to the medium ~fter autoclaving. Four ml of medium was sampled asept~cally at various times after inoculation. The number of cfu was determined by plating of dilution steps (1/4 strength Ringer's solùtion) of the sampled med~um on nutrient agar. The number of ~n~ll9a spores was determined by mixing 0.1 ml of absolute alcohol wlth 0.1 ml of the culture medium in wells of a microwell plates (0.3 ml wells; Nunc Ltd., DK) for 20 min (to kill off vegetative cells) before plating o~ dilution steps on r.utrient agar.

w ~ ( 0 '~

Crude antibiotic extracts from the various media were made by centrifuging the liquid for 30 min at 3000 g using a Mistral 3000i centrifuge (Fisons, UK.). The supernatant was filtered through a 0.2 ~
~m cellulose nitrate filter (Sartorius, Gottingen, Germany) to obtain `
cell free filtrates. To obtain lower concentrations of antibiotics, these crude extracts were then diluted with cabbage broth. 30 ul of the crude extract or the different dilutions were plsced in 4 mm diameter wells in CA5 plates seeded with ~ insr~ or A.
5i~isQl~. When the effect of pH on the activity of antibiotics in crude extracts was determined the pH of the assay medium was adjusted --tc values of between 3 and 9 with HCl and NaOH prior to placing the crude extracts into the wells.
`:
:'' Inhibition zones were also formed around cell free culture filtrates Of ~gSillg~ spp. g--own for 5 days in cabbage broth (CB5) or nutrient broth (NB) as growth media (Table 8). Cell free cabbage agar ~-filtrates of ~esillls isolate CL27 and CL41 batch cultures had a very high activity against Botrvtis cinerea and lower activity against `
D~Lia kL~5aisioQle (Table 8). Nutrient broth filtrates only had activity against when Petri dishes were used to perform assays, but showed similar activity against B. cinerea and ~_ ~$es3lsisQle when assays were performed in the more sensitive Sens`-acute Microdetection plates; see Hampson et al.

Hampson et al 1992a gives a detailed description of the differences between Sensacute Micro-detection plates and Petri-dishes. Nutrient broth and cabbage broth (NB5) filtrates from strain CL45 were only active against`~, ciner~a (Table 8). ~ -The production of antibiotics in batch culture after different ;~
incubation times and on different liquid media (NB, NB~Mn, NB~1%
glucose, CA5, CA10, DM and DM~1% glucose) was studied. Culture ~
filtrates of CL27 and CL45 again developed very strong activity `

!

(inhibition zones of around 40 mm) against B. cinerea when grown in cabbage broth (Fig. 3 & 4), but developed much lower activity (inhibition zones <15mm) when grown in nutrient broth (Fig 5 & 6) or defined media ~Fig 7.). In cabbage broth (CA5 & CA10) both CL27 and CL45 developed anti~80trvtis activlty after 2 days, approximately 24 hrs before ~he first newly formed ~nsi1lu~ spore~ could be detected.
On NB and DM antibiotic act~vity also developed after Z days but an increase in the number of spores could only be detected after 9 days~
Addition of 1% glucose inhibited ~sill~ spore formation but not the development of antifungal activity in NB and DM (Fig. 5 & 7).

As before, isolate CL27 but not CL45 developed activity against Alternaria~h~isL~i~l~. Activity was highest in cell free cul~ure filtrates from NB~Mg snd CB10 batch cultures ~20 mm inhibition zones) and similar in all other media (inh~bition zoneC of between 15 and 10 mm) (Fig 3 to 7). Similar to the activity again~t Botrv~
anti-A1bcr~ activity preceded spore formation by 24 h in CA5 and CA10 and 7 days in NB and DM while addition of 1% glucose did not stop the development of antifungal activity (Fig. 3 to 6).

The activity of antifungal antibiotics formed in li~uid batch culture of ~ l11L~ ~bLili. (measured as the inhibition zone diameter on funæal seeded CA5 medium) was affected by the pH (Fig. 8) and the concentration and type of nutrients in the assay medium (Fig. 9).

~H ef~ç~ Antibiotics produced in cabbage br~th (CB5 & CBlO) after 2 and 4 days and by CL45 in NB+Mg after 2 days showed the highest activity (formed the largest inhibition zones) against Bp~rvtis between pH 5.6 and 6Ø Activity decreased slightly when the pH was reduced to 3.1 and to very low values when the pH was increased to values above 6.0 (Fig. 8). Antibiotics produced by CL27 on NB+Mg had s~milar activity between pH 5.6 and 8.9. but decreased activity when the pH was lower than 5.6. In NB activity could only be detected if th~ pH was above 6.0 and increased slightly with increasing pH. When W O g3/186~4 PCT/GB93/00604 . 7 ~

antibiotic activity was assayed in ~Q~Y~is-seeded potato dextrose agar crude extracts from nutrient broth and CB5 showed no activity, ;~
extracts from CA10 showed lower activity at pH values below 6 but similar activity at higher pH and extracts from nu~rient broth with manganese showed increased activity thsn on ~otrv~ seeded cabbage agar used (Fig. 8C). ~-Nutrient levels: With increasing concentration of nutrients, in the form of cabbage tissue or potato extract and dextrose in the assay medium, the activity of antibiotics &lso decreased (Fig. 9). On ~`
Potato dextrose agar no, or only very small inhibition zones were ~-~
formed (Fig. 9).

Antibiotics formed were found to be very stable at 4C since no decrease in activity could be detected in the orude cell free culture filtrates of strains CL27, CL41 and CL45 stored sterile for 4 months.
Filtrates were assayed at monthly intervals during storage at 4C.

.
".
~ :.
Bacteria were grown for 7 days in batch culture of either cabbage broth (lOg homogenized cabbage leaf tissue) or nutrient broth (~0.005g Mn per litre). Cell free culture extracts were prepared as described abov~). 0.04ml of the cell free culture filtrate was placed on Silica gel 60 TLC plates (Art.5271; Merck) and separated using the solve~t system described by Swinburne et al (Trans. Br. Mycol. Soc. fi~
(2), 211-217) ie. Butan-l-ol: acetic acid: H20; 3:1;1. Plates were dried in a laminar flow cabinet then an oven for 5hrs at 70C to remove solvent and acetic acid residues.

Fungal suspensions were made in either nutrient broth or cabbsge broth (5 g/l homogenized cabbage leaves). The dried TLC plates were sprayed with spore suspensions of either B. ~ineria or ~ 5 ~ l~go- and W O 93/l8654 PCT/GB93tO0604 ~ él3~
!

incubated for 3 days or 4 days (depending on experiment) at 20C in darkness to allow fungal growth on the plates. Inhibition ~ones were visualised according to the methods of Lund & Lyon (J. Chromatography lQQ 192-196 (1975)) and used to calculate Rf values for the antibiotics produced.

~a~ihin¢i~ Rf values CL27s Anti-Botrytis antibiotic produced in cabbage broth: 0.39 Assumed non peptide ~negative TDM): activity at pH <5.6.
CL27b Anti-Botrytis/Alternaria antibiotic produced in cabbage 0.56 broth: Assumed peptide (positive TDM): activity at pH ~5.6 CL27c Anti-Alternaria antibiotic produced in nutrient broth 0.61 Assumed peptide nature (positive TDM): activity at pH >5.6.
CL45a Anti-Botrytis antibiotic produced in cabbage broth. 0.39 Assumed non peptide tnegative TDM): activity a~ pH <5.6.

TDM-4,4'-tetramethyldiaminodiphenylmethAne reagent: positive = blue /green. CL27 antibiotics are formed in Cabbage Broth (CB5), Nutrient Broth and Nutrient broth with manganese: CL45 antibio~ics only Cabbage broth (CB5) Activity spectrum of antibiotics (plate asssys as described above).
TablP 7 _ _ _ Fungus Strain CL27a CL27b CL45a ~9~ STl,Pl9 + + +
Trichocladium SDD. H7 + +

Zveorrhyn~h~_ so~. P23 ~ + ~
p~nicillium 8DD . P5,I4 - + -~:'4~LL~ . Pl - , _ Alternaria soo. H9 - +
_ W O 93/18654 ~ iO ~ PCT/GB93/00604 . _ Table 8 Dia.inhibition zones (mm) in fungal seeded medi~ around well containing crude extracts of CL27 And CL45 br~ths: includes 3mm well ~ ~:
Genera Strain ~mll ~ h~ acillus Dumili~
CA5/2d CA5/4d NB/4d CA5~4d NB/4d d=day .
. . . _ ~s~crçill~ P1 0 8 8 0 o ~:-Hg 0 4 4 0 0 .
AL~crns:1a OUT3 17 17 0 0 ~.
Botrvti~ P19 27 15 10 34 C1 21 14 9 24 0 ~
MP3 30 19 12 28 0 `~.
MP5 12 8 12 26 0 ;

Debarvo~Yces P27 0 0 6 0 0 Esnisilliu~ P5 0 0 8 0 0 . I4 0 0 11 0 0 ~hQm~ P13 0 5 6 5 o ~Lshnsl~di~ H7 6 5 14 0 0 TricoohYton C3 0. 0 11 0 l~i~hnsQc~J~ P29 31 12 0 46 0 ~y~orrhvnchus P23 0 23 20 0 Bhs~n¢s~sl~ G4/1 0 0 0 0 O Yeast ~L~9~9cgsss~ G4/6 0 0 8 0 0 Yeast ~ .
When TLC plates on which cell free culture filtrates from CB5 medium inoculated with either CL27 or Cl45 were over layered with Botrvtis ~in~ea seeded medium large inhibition zones t30-40~m) developed at an rf of 0.38 to 0.40. When filtrates from C85 medium inoculated with strain CL27 were examined an additional small band ~3mm) without fungal growth could be detected at an rf-value of 0.56 (not present when CB5 i~oculated C45 was tested). When filtrates from NB and NB~Mn W O 93/18654 PCT/G~93/00604 .

jrJ i inoculated with CL45 were overlayed in this manner~ only a sm~ll band of inhibition with an rf-value of O.56 could be detected (there were no zones of inhibition when NB and NB~Mn medium inoculated with CL-45 were examined).

When the overlaying w~s carried out against ~lL9~ b~
seeded medium, 2 inhibition zones were detected (rf 0.56 and 0.61) from NB, NB+Mn and CB5 CL27 growth media inoculsSions. Thus it is seen that CL2j produces at least three antibiotics; nne active against ~. cinerea (r~ 0.39) only produced on cabbage extract and active at pH less thsn 6; one weakly active against ~.__si~¢~ and active against A~ ~rassicocQl~ (rf 0.56) produced on all media tested and active above pH 5.6; and a third active only again~t ~
brassicocola (rf 6.1) again produced on all media tested. l~è similar rf values for the CL45 and CL27 antibiotics suggest that they are the same, but activlty patterns suggest the oppo~ite.

Table 9. Ln~ inhibition (Petri dlshes) o~ E~ L~AJ~ by BBS111Y~ spp. isolated from a~Bssls~ 6pp.; mean 3 determinations.
_ _ _ _ , Inhibition zone diameter (mm) at 20C
on ~QI~Y~i8_seeded cabb~ge medium . . _ .

Isolate Isolation Concentration of nutrients (g/l) code Origin Medium 100 50 10 5 . ~ _ _ subtilis CLoo48 .. SA 22 CL0052 .. PDA 21 CL0055 DWLo SA 29 CLoo56 .. .. 34 CLoo63 .. .. 28 CL0071 .. PDA 24 W O 93/186~4 PCT/GB93/00604 ~ ;

. .

Table 9 continued.
i ' '-:
B.~umilis 100 50 10 5 (g/l) CLoo45 DWL CA5 26 34 CLoo64 DWLo SA 10 CLoO69 PDA 21 B. ~olvmvxa CLoo47 DWL SA 10 CLoo67 DWLo SA 19 CLoo68 PDA 18 - ~:
_ _ ~ _ .:
- no visible bacterial growth 8RL Broccoli lesve~ ~SL Brussel's sprout Table 10. Inhibition zones (mm on Petri dishes) formed by cell free culture filtra~es of ~Dsillg~ spp. of lawns of ~ L1~ :
and ~. ciner~ (mean is of 2 determinations; ~ 11Y~ spp. were grown for 5 days in the medium) _ ~.

Bacterial Nutrient broth Cabbage broth strain Alternaria Botrytis Alternaria Botrytis cLwoa 6 O 2 20' j CL0019 8 0 2 30~
CL0020 6 0 2 8+

CLoo26 6 0 4 22 12* 10* 20* 31*
CLoo28 6 0 4 10 . 13* 11* 20* 3*
CLoo45 O O O 35 o* 9* o* 45*
- - -~ Sens-acute (Proteus) trays; + = some fungal growth W O 93/186~4 PCT/GB93/00604 ~ .
invention.
All bacterial strains isolated from ~a~i~a spp. showing yitro antagonism against ~otrvtis cin8r~a we ~ found to be Pseudomonas ~l~QLe~can~ (group 1), ~sillls or Se~-~tia species. Fluorescent pseudomonads and Enterobacteriaceae have been found to be the major inhabitants of cabbage leaves while ~sillsE spp. were only found in very low numbers on Brassi~g leaves. Fluorescent pseudomonads and ~ucill~ spp. have previously been described as potential biocontrol agents against post-harvest fungal diseases of fruit and vegetables, whereas Serratis nlvmuthica and 5g~sl~ have not previously been shown to have such activity (see Wilson & Wisniewski 1989 for a recent review on biocontrol of post-harvest fungal diseases of fruit and vegetables).

Pseudomonas fluorescens Fluorescent pseudomonsd strains CL42. CL66 and CL82 of all bacterial strains tested showed the highest in vivo activity on leaf disks at 4C, inhibiting Botrvtis infection with all fungal inocula tested.
Pseudomonads such as p. ceDacia. ~. Dutida. and P. svrin~e have been described as post-harvest biocontrol agents for a variety of stored fruit and vegetables and their mode of actian has been described as being either by antibiotic production or by nutrient competition (Colyer & Mount 1984, James & Gutterson 1986, Gurusiddaiah et al. 1986, Janisiewicz 1987, Janisiewicz & Roitman 1988). It is important to note that E.__sY~ine~e and ~ _ 5~Z~e can be pathogenic to plants and that some pseudomonsds including ~ __5cse51sal have been described as opportunistic human pathogen (Agrios 1988, Bergan 1981).
~,, ; ! ' j ~ , ~
Some of the antsgonistic pseudomonads isolated by the present inventors were slso found to be psthogenic to some vegetables eg.
cabbages. Antagonistic pseudomonads should therefore be identiPied to species level to avoid using possible human pathogens an~ tested for effect on plsnts before use as postharvest control agents. Because the antagonistic strains isolated were able to persist on cabbage W O 93/186~4 PCT~GB93/006~4 le~ves they qualify as ideal poten~ial candidates for biocontrol during storage of ~assi~a spp..

Serratia s~p.
Serrati~ spp. isolated from cabbage leaves grew well and showed strong in vitrQ antagonism at low temperatures. However, only S~
Ply~9¢hi~ strain CL42 showed high in_YiYQ antagonism against ~ing~ on leaf disks. All S. liauefacien~ strP ns showed intermediate to poor control 'i~ viY~'. Se~ra~ia spp. have not previously been identified as fungal antagonists. ~B~ia spp. have not been described as plant pa~hogens but S. mar~ has been described as causing human nosocomial infection ~Brenner 1984). Such organisms should therefore only be used once they have been shown to be non-pathogenic to man. Enterobacteriaceae form a major part of the -~
natural flora of cabbage leaves and because the antagonistic strains isolated were able to persist on ~abbsge leaves they have potential a~
biocontrol agents, eg. particularly during storage of ~assica spp..

Bacillus s~.
~aGillus spp. are considered to be unsuitable as potential biocontrol agents for cold-storage since they did not grow or grew very poorly at low temperatures. This has not been reported before because most workers describing the use of ~sillua species for post-harvest biocontrol of fungal pathogens of fruit and vegetables have only tested 'in vi~ro' and 'in vivo' antagonism at temperatures of around 200C (Pusey & Wilson 1984, Singh & Deverall 1984, Utkhede & Sholberg 1986, Guel &er ~ al~ 1988). Their poor growth at low temperatures could, hcwever, be the explanation for vPriable results described in the field and in pilot tests for commercial application of ~sill~
sLbeilis based biocontrol agents (Baker ~_al_ 1985; Pusey ~_al_ 1986 & 1988). ~g&111~5 spp. are known to produce peptide antibiotics with antifungal activity during spore formation (see K~tz & Demain 1977 for a review) and peptide antibiotics have been identified as the active conpounds in the use of ~acill~ subt;l;s as a biocontrol.agent for post-harvest fungal diseases (McKeen ~t al. 1986, Gueldner ~_al~
1988).

W O 93/186~4 PCT/GB93/00604 i ~; . ..

Production of such antibiotics by fermentation and their use as 'bio-pesticides' is the preferred use of such ~cill~a if no cold resistant antagonistic Bacillus spp. is isolable. However the antibiotics and the organisms might also be used to prolong the shelf-life when fruit and vegetables are removed from the cold-store.
The poor activity of crude antibiotic extracts when asfiayed in potato dextrose agar (PDA) indicates that the antibiotics faund are novel, since PDA has been used by most other authors to select ~cillus spp.
which produce anti-fungal antibiotics (Katz & Demain 1977; Pusey &
Wilson 1984; Singh ~ Deveral 1984; McKeen et ~1_ 1986; Utkhede Sholberg 1986; Wilson & Wisniewski 1989).

Details of all the strains of the invention are provided in Tables ll and 12.

~1cu I~L 10: In vivo activitv of all strains against B.cinerea on vounF mlant lqaf surfaces.
CL82, CL80 and CL27 were tested for ability to control B. ciner~ on leaf surfaces in vivo using a young plant assay in high humidity propagation chambers. Only CL27 showed significant control, similar to that achieved by fungicides. Application of cell free culture filtrates of the CL27 broth gave similar control to that provided by ~he cell suspension, thus indicating an antibiosis effect. This result shows that CL27 and its derivatives can be used as bio-control agents applied before harvest. In order to determine the effectiveness of such an application after harvest, persistence studies were carried out.

~sLllhe microplants were planted into 3 x 3 cm co~post plugs and sprayed at weekly intervals with fungal spores and antagonists in a humidity of 86% to 95~. For the first 6 weeks all treatments gave similar success, but after 10 weeks only fungicide and treatment using CL27 or its broth gave significant control. CL27 returned better control than fungicide in these tests.

W O 93/186~4 PCT/GB93/00604 ~ . .

ExamDl~ Ef~ec~ of Serr~tia and P~es~o~ e~ 93~Q~d ~roduce.
Dutch white cabbages (var. Morgan) were grown and treated with insecticide and herbicides in the normal fashion, but not fungicides.
After harvest cabbages were dipped in either fungicide suspension (1 g/l Rovral and 2 g/l Ridomil) or into bacterial suspensions of 107 to 108 c~u/ml. Approx. 14 kg cabbage (10-12 heads) were transferred into plastics trays and sealed in plastics to avoid cross-con -tamination; 5 replicate trays were used for each treatment. Trial~
were set up in a cold store at 4-6C and at 1-3C with humidity measured a~ 7 day intervals and found to be 92 to 96% in all cases.
In one of the trials 106 spores/ml B. cinerea were applied after the fungicides or bacteria; 5 ml spore suspension sprayed onto each tray.

Cabbages were scored as follows: 0 = no visible growth; 1 = l-10X, 2 = 11-25%, 3 = 26-~0%, 4 = 51-75%, 5 = 76-90%, 6 = 91-99%, 7 = 100%
of surface covered by fungal grow~h and measured at 6 weekly intervals. When 80X coverage occurred in the untreated controls (when trimming occurs in commercial storage) destructive assessment was made by reference to weight loss on removal of infected leaves. Heads were stored again for 10 to 14 weeks and then trimmed after 42 weeks.
Trial at 1-3C did not allow destructive testing so extrapolation of leaf infected ares with trimming loss was carried out.

Use of ~L80 and CL82 gave control similar to that of fungicides in all cases, CL82 giving better inhibition in added ~ _CiL~C~
treatment than A. brassicicola; the opposite of in vitro results.
CL82 persisted on cabbage in higher numbers than other antagonists and thus is la preferred example of the invention. Of the other strains, CL43 was the most efficacious in cold store.

W 0 93J18~4 PCT/GB93/00604 .:
~ J

. . . ~ .
Tahle 11. Characteristics of antagonistic PseudomQna~ and Serratia spp. .. , _ ___ _ _ _ _ fluorescens Plym~thicaliauefacien Characteristics (group 1) ,_. . _ . _._ __ .
Gram stain - _ _Hugh L Leifson -Motility +
Fluorescence on KB-agar + _ Oxidase + _ Beta-galactosidase - ~ +
Arginine dihydrolase ~* _ _ Lysine decarboxylase ND _ +
Ornithine decarboxylase ND _ ~
Citrate utilisation ND + +
H2S production ND _ _ Urease - _ _ Esculin hydrolysis -} ND ND .
Tryptophan de~minase ND _ _ Indole produ~tion - _ _ Acetoin production ND + ~
Gelatine hydro.lysis + + +
N03-reduction Growth on:
Glucose + ~& +&
Mannitol I +
Inositol ND +*
Sorbitol ND + +
Rhamnose ND _ _ Sucrose ND I& +&
~ . _ . _ W O 93/18654 PCT/GB93/00604 ~ , , ~ ~J~ S ~ a 3 `~

Contd.
. . . _ _ _ :.;
MelibioseND ~ ~
AmygdalinND + + ~;
Arabinose + + +
.Mannose + ND ND
N-Acetylglu~osamine + ND MD .
Maltose ~ ND ND .
Gluconate ~ ND ND ~
Caprate + ND ND :
Adipate - ND ND
Malate + ND ND
Citrate + ND ND :~
Phenyl-acetate - ND ND :~
Raffinose3 ND ND ~ :~
Manolate3 ND ND _ :~.
Lactose3 ND ND _ .
Adonitol3 ND ND _ . _. .. _ _ _ _ _ & positive after 6 h 1 isol~tes CL74, positive -~
positive after 48 h 2 SC¢~ spp. were tested for I positive test result acidification from carbohydrates - negative test result 3 additional tests (Brenner 1984) ND not deter~ined .

. W O 93/186~4 PCT/GB93/00604 J

31 . I

Table 12: Characteristics of antagonistic ~acillus subtilis ~illss Du~i~ and Bacill~sL~gLymvxa spp.
_ . _ _ _ Cill~Q ~slll~ Bacillu~
~ubtili~ P~mil~ polvmYxa CL27,41,48,52 CL45,64,69 CL47,67.68 Characteris~ics55e56~63~71 _ _ _ . _ _ _ Gram stain + +
Motility +
Heat resistant spores + + ~
Beta-galactosidase + l +
Arginine dihydrolase - _ _ _ysine decarboxylase - _ _ Ornithine decarboxylase - _ _ Citrate utilisation Vl Vl _ H2S production - _ _ Urease - _ .
Tryptophan deaminase - _ _ Indole production - _ _ Acetoin production ~ , .
Gelatine hydrolysis + + ~
NO3-reduction + + +
~-Acid from:
Glycerol + + + .
Erythritol - _ _ D-Arabinose - _ _ L-Arabinose + + ~+g Ribose + + +g .
D-Xylose + + +g L-Xylose - _ _ ~-Adonitol - . _ ~-Methyl-xyloside - _ +
.' _ _ _ ~ 32 Contd.
~ ~ _ Galactose V2 + +g :
D-Glucose + ~ +g D-Fructose + + +g D-Mannose + + +g L-Sorbose - _ _ .' Rhamnose - _ _ Dulcitol - _ _ Inositol + _ _ Mannitol + + +g .
Sorbitol ~ _ _ alpha-Methyl-D-mannoside - V3 +
~lpha-Methyl-D-~lucoside + _ +
N-Acetyl ~lucosamine - ~ _ A~ygdAline + ~ +g Arbutine + + l Esculine ~ + +
Salicine ~ + +g Cellobiose + + +g Acid from: .
MP1 tose + V4 +g Lacto~e - V5 +g Melibiose + V6 +g Sucro~e + ~ ~g Trehalose ~ + +g Inuline + _ +g Melezitose - _ _ D-Raffinose ~ V7 +g Amidon + _ +g Glycogene + _ +g Xylitol - _ _ .
~-Gentiobiose V8 + +

. W O 93/18654 PCT/~B93/00604 h ~

Contd.

D-Turanose V9 V9 +
D-Lyxose - _ _ ,.
D-Tagatose - ~ _ -.
D-Fucose - _ . _ L-Fucose - _ _ D-Arabitol - _ _ L-Arabitol - _ _ Gluconate -P _ + :~
2-ceso-gluconate - _ _p :
5-ceto-gluconate -positive test result - negative test result -.
V variable test result 1 CL27,45,52,56,63 negativei CL41,48,55,64,69,71 positive 2 CL27,41,48,52,55,56,71 ~egative: C163 positive 3 CL45 negative; ;
CL64,69 positive :~.
4 CL64,69 negative; CL45 positive .
5 CL45,69 negative; CL64 positive :

6 CL45 negative; CL64,69 positive

7 CL45 negative; CL64,69 positive

8 CL41 negative; CL27,48,52,55,56,63,71 positive 9 :`
CL27,41,48,63,64,69 negative; CL45,52,55,56,71 positive `'.

~.

:, ~ 3 3 34 References Agrios, G.N. (1988) Plant Pathologv. New York: Acadcmic Press.
Baker, C.J et al (1983) el~n~_ni5~s9~_ 69, 770-772-Bergan, T. (1981)in The ~rokarvotes Vol I. ed. Starr et al, pp. 666-701.
Brenner D J (1981)in The Prokarvotes Vol II. ed Starr et al ppllO3-1128.
Brenner, D.J. (1984) Bergev's Manual of Svstematic B~cteriology (9th Edition) (Ed. by Krieg, N.R. & Holt, J.G.), pp. 408-420.
Williams and Wilkins, Baltimore.
Brown, A etal (lg75) Proc 8th.Brit.Insecticide/FunFicide Conf 1,339-346.
Colyer, P.D & Mount, M.S. (1984) ~lane Disease. 68, 703-706.
Dickinson, C.H. & Preece (1976~ Microbiology of aerial plant surfaces. Academic Press, London.
Geeson, J.D. (1978) Grower. 89,27-31.
Gueldner, R.C et al (1988) J, Agricultural and Food Chem. 36~366-370.
Gurusiddaish, S.et Al (1986) Antimicrob. Ag.and ~he~otherapv, 29,488-495.
Hampson, S.B. et al (1991) Howell, C.R.et al ,(1988) ~h~l~9~eh~ Y~ 78, 1075-1078.
James, D.W. et al (1986) A~pl.and Environmental Microbiol. 52, 1183-1189 Janisiewicz, W.J. (1987) Phv~n~athologY. 77, 481-485.
J~nisiewicz, W.J. et al (1988) ~hYtoDatholog~, 78, 1697-1700.
Katz, E. et al (1977) Bacteriological B~yie~E. 41, 449-474.
Leifert, C. et al (1992) Letters of ADolied MicrobiQlogy McKeen, C.D.,et al (1986) EbYIC9-~h~l9 ~_ 7~. 136-139-Newhook, F.J. (1951) Annals of Microbiologv. 38, 169-184.
Pusey, P.L.et al (1984) ~l~n~_nis~&e. 68, 753-56-Pusey, P.L.et al (1986) Plant~Disease. 70, 587-590.
Pusey, P.L.et al (1988) Plant nisease. 72, 622-626.
Robinson, D.H.et al (1975) Ann~ f An~l~ed ~i9lQ8Y_ 81,399 408.
Singh, V.et al (1984) ~ 83,487-490.

Spotts, R.A. et al (1986) ~lnn~_niasssg. 70,106-108.
Ut~hede, R.S.et al (1986) 5~ C~_ 32, 963-967.
Wale, S.J. (1980) The post-harvest pathology of Dutch white cabbage in refrigerated storage. Ph.D. Thesis, Manchester University.
Wilson, C.L et al (1989) Annual Review of Phvtoea~hology . 27,425-441.
Wisniewski, M. (1989) Canadian Journal of aQ~nY~ 67, 2317-2323.

Claims (25)

1. Isolated bacteria having the identifying characteristics of an isolate selected from the group consisting of NCIMB 40489, NCIMB
40490, NCIMB 40491, NCIMB 40492. NCIMB 40493, NCIMB 40494, NCIMB 40495 and NCIMB 40497 including the characteristic of being capable of inhibiting growth of fungi of species Alternaria brassicicola and/or Botrytis cinerea on homogenised cabbage tissue:agar:water mixtures.

2. Isolated bacteria as claimed in claim 1 having the identifying characteristics of an isolate selected from the group consisting of NCIMB 40490, NCIMB 40492. NCIMB 40493, NCIMB 40495 and NCIMB 40497 said isolate being capable of inhibiting growth of Alternaria brassicicola and/or Botrytis cinerea on cabbage leaves at temperatures below 20°C.

3. Isolated bacteria as claimed in claim 2 being capable of inhibiting growth of Alternaria brassicicola and/or Botrytis cinerea on cabbage leaves at temperatures of 10°C or less.

4. Isolated bacteria as claimed in claim 2 being capable of inhibiting growth of Alternaria brassicicola and/or Botrytis cinerea on cabbage leaves at temperatures of 4°C or less.

5. Isolated bacteria as claimed in any one of claims 1 to 4 wherein the bacteria is selected from the group consisting of Pseudomonas fluorescens, Serratia plymuthica and Serratia liquifaciens

6. Isolated bacteria as claimed in claim 1 having the identifying characteristics of an isolate selected from the group consisting of NCIMB 40489 and NCIMB 40491, said isolate being capable of inhibiting growth of Alternaria brassicicola and/or Botrytis cinerea on cabbage leaves at temperatures of about 20°C.

7. Isolated bacteria as claimed in claim 6 wherein the bacteria is capable of production of an antibiotic fraction when cultured upon a broth of homogenised cabbage leaves of concentration 5g to 100g per litre of water, said antibiotic being capable of inhibiting growth of said Alternaria brassicicola and/or Botrytis cinerea on cabbage leaves at temperatures of about 20°C.

8. An aqueous suspension comprising one or more of the isolates of claims 1 to 7.

9. A suspension as claimed in claim 8 wherein the cfu/ml value of said bacteria is sufficient to inhibit growth of Alternaria brassicicola and/or Botrytis cinerea on fruit and/or vegetables that have been dipped therein or sprayed therewith.

10. A suspension as claimed in claim 9 wherein the cfu/ml of said bacteria is between 104 and 109.

11. A suspension as claimed in claim 9 wherein the cfu/ml of said bacteria is between 105 and 108.

12. A biologically pure culture of any one of the isolated bacteria of claims 1 to 7.

13. A process for the prevention of fungal disease in post-harvest vegetables and/or fruit products comprising dipping said produce into suspension of bacteria as claimed in any one of claims 9 to 11 or spraying them with such suspension.

14. A process as claimed in claim 13 further comprising the step of cold storing said produce.

15. An antibiotic obtainable by culturing isolated bacteria as claimed in claim 7 in a medium comprising homogenised cabbage leaves and water in a concentration of 5g to 100g leaves per litre.

16. An antibiotic as claimed in claim 15 wherein the culture medium comprises 10g to 50g/litre of homogenised cabbage leaves.

17. An antibiotic as claimed in claim 15 wherein the product culture medium is filtered to remove cells, the filtrate is dewatered and the residue is separated using a silica gel column or plate with a butan-1-ol:acetic acid:water; 3:1:1 eluant.

18. An antibiotic as claimed in claim 17 having an Rf value of about 0.39 using a silica gel plate and butan-1-ol:acetic acid:water; 3:1:1.

19. An antibiotic as claimed in claim 17 having an Rf value of about 0.56 using a silica gel plate and butan-1-ol:acetic acid:water; 3:1:1.

20. An antibiotic obtainable by culturing an isolate having the identifying characteristics of NCIMB 40491 in nutrient broth, characterised in that it has an Rf value of about 0.61 using a silica gel plate and butan-1-ol:acetic acid:water; 3:1;1.

21. An antibiotic obtainable by a method as claimed in any one of claims 15 to 21 as described in any one of Examples 5, 6. 7 or 8 herein.

22. An antibiotic composition characterised in that it has cell free antibiotic as claimed in any one of claims 15 to 21 as active ingredient.

23. Use of a bacteria as claimed in any one of claims 1 to 7 as a biocontrol agent for the inhibition of post-harvest disease.

24. Use of an antibiotic or antibiotic composition as claimed in any one of claims 15 to 23 as a biocontrol agent for the inhibition of post-harvest disease.

25. Vegetable or fruit produce characterised in that it has been treated with an isolate, suspension, antibiotic or composition as claimed in any one of claims 1 to 22.