search
for
 About Bioline  All Journals  Testimonials  Membership  News


Actinomycetes
University of Udine, Mycology Department
ISSN: 0732-0574
Vol. 5, Num. 2, 1994
Actinomycetes, Vol. 5, Part 2, 31-39, 1994 RAPID DETECTION OF CLAVIBACTER TOXICUS AND OF ITS BACTERIOPHAGE RESPONSIBLE FOR ANNUAL RYEGRASS T0XICITY IN AUSTRALIA AND THE EFFECT OF SELECTED HERBICIDES ON TOXIN PRODUCTION

D.I. KURTBOKE

Murdoch University, School of Biological and Environmental Sciences, Perth, W.A. 6150, Australia

Code Number: AC94009
File Sizes:      
     Text: 27K
     Graphics:  Photo (Jpg) -  133K
     
ABSTRACT. Annual ryegrass toxicity bacterium, Clavibacter toxicus, was isolated from toxic ryegrass seeds and livestock feed by exposing the material to bacteriophages affecting plant pathogenic coryneform and saprophytic bacteria. Susceptibility to phages provided a selective means of reducing the cell numbers on isolation plates and hence facilitated the detection and isolation of C.toxicus. Bacteriophages specific to C.toxicus were also isolated from toxic ryegrass seeds and livestock feed. Plaque morphology, host range, and particle morphology of the phage isolated are described. Various concentrations (20-400 ug/ml) of herbicides, commonly used in Western Australia, Simazine (registered trademark), Glean (registered trademark), Hoegrass (registered trademark), Trifluralin (registered trademark), and Sertin (registered trademark), were tested to determine their effect on the conversion rate of nontoxigenic C.toxicus strains into toxigenic derivatives by two different bacteriophages isolated from the toxic ryegrass seeds and livestock feed. The number of toxigenic strains increased when herbicide concentrations of 200 and 400 ug/ml were applied.

Annual Ryegrass Toxicity (ARGT) often leads to fatal poisoning of livestock grazing on ryegrass pasture infected with Clavibacter toxicus (McKay and Ophel, 1993). As well as animal loss, ARGT causes other direct and indirect costs to agricultural productivity in Australia. These include those associated with the control of ryegrass, loss of grazing potential, disruption of livestock management, herbicide resistance, and reduced marketability of produce such as hay, feed grain and grain due to contamination (McKay and Ophel, 1993).

The bacterium is very difficult to isolate and purify despite being present in high numbers in toxic galls. It grows very slowly on artificial media and is easily overgrown by saprophytic bacteria (McKay and Ophel, 1993). It has also been reported that, following the isolation of the bacterium, toxin production declined rapidly after subculturing of fresh isolates from field material (McKay and Ophel, 1993). Recently Ophel et al., (1993) reported that C. toxicus regularly produced toxin in vitro when infected by a specific bacteriophage and, following phage infection, colony morphology of the C. toxicus changed, toxin producing colonies became extremely sticky and had a consistency resembling melted cheese. Although the mechanism by which the bacteriophage affects toxin production is still unknown, Ophel et al. (1993) concluded that bacteriophage presence was correlated with toxin production and toxin-producing strains of C.toxicus were in a phage-carrier state rather than in a true lysogenic state.

Although serological (Riley, 1987), alloenzyme electrophoretic (Riley et al., 1988) and adhesion (Riley and McKay, 1990) techniques are currently available for the detection of the bacterium, they have not proved to be definitive in its identification (Riley and Gooden, 1991). Therefore, rapid techniques are required for the detection, isolation and identification of the bacterium such as phage typing which is currently used to facilitate the rapid identification of bacterial strains suspected of being toxigenic (Riley and Gooden, 1991).

Herbicides can affect the lytic activity of phage and interfere with lysogenisation (Toure and Stenz, 1977). Roslycky (1982) showed that Paraquat concentrations from 20 to 400 ug/ml gradually reduced the adsorption frequency and average burst size in phage-host systems although the product had no effect on virus attachment or on the length of the latent period. Annual ryegrass is ubiquitous throughout the cropping zones of Southern and Western Australia and herbicides are widely used for postemergent control of weeds (Riley, 1992; McKay and Ophel, 1993). However, the effect of herbicides on annual ryegrass toxicity bacterium and its bacteriophage have received little attention in Australia .

This report describes a rapid technique for the isolation of C.toxicus. Such a technique may prove useful in routine isolation and identification of the species associated with the ARGT. It also examines the detection of specific bacteriophages for C. toxicus in grain and whether they can be used for the rapid identification of C.toxicus like the other specific phage reported by Riley and Gooden (1991). In addition the effects of selected herbicides on the conversion rate of a non-toxigenic ARGT bacterium into a toxigenic strain by specific bacteriophages are investigated.

MATERIALS AND METHODS

Isolation and identification of C.toxicus

Toxic ryegrass seeds (Lolium rigidum Gaudin) and livestock feed for the isolation of C.toxicus were obtained from the Western Australian Department of Agriculture (WADA). These materials were exposed to high titre (10^l2 pfu/ml) stock phage suspension as described by Kurtboke et al. (1992). The activity spectra of the phage used in the stock phage suspension are given in Table 1. The polyvalent phage did not affect C. toxicus. Phage previously isolated for Bacillus and Pseudomonas spp. (Kurtboke et al., 1993a) were also added to the stock suspension (10^l2 pfu/ml) with the aim of reducing the numbers of as many different types of bacteria as possible.

523M medium was used for the isolation of C. toxicus (Riley and Ophel, 1992a), with cycloheximide (50 ug/ml) to reduce fungal growth (Williams and Davies, 1965). The plates were then inoculated with 0.3 ml of the selected dilution of the test samples treated with and without phage and dried in a laminar flow cabinet for 30 min (Vickers and Williams, 1987). The plates (6 replicates for each treatment) were then incubated at 25øC for 10 dd and the resulting colonies on the plates with and without phage were counted. Results were analysed using the Student's 't' test.

All colonies resembling C. toxicus were transferred to 523M medium (Riley and Ophel, 1992a) from the plates treated with and without phage. Tentative identification of the isolates was carried out using morphological criteria (Schaad, 1980; Vidaver, 1980) and biochemical (Riley and Ophel, 1992a). Phage typing was also carried out to confirm the identity of the isolates using C.toxicus specific phages. Isolates resistant to the C.toxicus phages were further tested for their susceptibility to phages utilising plant pathogenic corvneform bacteria (Table 1).

-------------------------------------------------------------------

     HOST                         STRAIN CODES**     PHAGES

                                                     AR*     C1*    
-------------------------------------------------------------------

Arthrobacter ilicis           PDDCC 2607-69     PH     
-
Clavibacter iranicus          PDDCC 3496-73     -      PH     
C. michiganensis   
subsp. insidiosus             NCPPB 1109        -       +     
C. michiganensis 
subsp. michiganensis          NCPPB 2979        -       +                    
C. michiganensis 
subsp. nebraskensis           NCPPB 2581        -       +     
C.rathayi                     NCPPB 797         -       +     
                              NCPPB 2980        -       +    

C.tritici                     PDDCC 2624        -       +     
                              NCPPB 255         -       +      
                              NCPPB 1953        -       +    
C.toxicus                     CS2               -       -     
                              CS14 (NCPPB 3552) -       -      
                              CS29              -       -
                              CS30              -       -    
                              CS31              -       -
                              CS34              -       -
Curtobacterium flaccumfaciens 
subsp. betae                  PDDCC 2594-69     -       -     
C.flaccumfaciens 
subsp. oortii                 NCPPB 2113        -       -     
C.flaccumfaciens 
subsp.poinsettiae         NCPPB 854     -       -     
Rhodococcus fascians          NCPPB 3067        -       -     
-------------------------------------------------------------------

Table 1 contd. PHAG STRAIN CODES C2* C3* C4* CTl CT2 CU* RH ----------------------------------------------------- PDDCC 2607-69 - - - - - - - PDDCC 34496-73 + + + - - - - NCPPB 1109 + + + - - - - NCPPB 2979 PH + + - - - - NCPPB 2581 + + + - - - - NCPPB 797 + PH + - - - - NCPPB 2980 + + + - - - - PDDCC 2624 + + PH - - - - NCPPB 255 + + + - - - - NCPPB 1953 + + + - - - - CS2 - - - + + - - CS14

(NCPPB 35-52) - - - PH PH - - CS29 - - - + + - - CS30 - - - + + - - CS31 - - - + + - - CS34 - - - + + - - PDDCC 2594-69 - - - - - PH - NCPPB 2113 - - - - - + - NCPPB 854 - - - - - + - NCPPB 3067 - - - - - - PH

----------------------------------------------------------------

Table 1. Strains and phages used in the study. (*: phage used to prepare the stock suspension, **: CS: Western Australian Department of Agrlculture South Perth, Australia (Riley, 1987, Riley and Ophel, 1992), NCPPB: National Collection of Plant Pathogenic Bacteria, Plant Pathology Laboratory, Hatching Green, Harpenden, Hertfordshire, UK; PDDCC: Plant Disease Division Culture Collection, Plant Disease, D.S.I.R., Auckland, New Zealand;

PH: Propagation host; +: confluent lysis or congruent plaques; -: no reaction).

Phage isolation and partial characterisation .

Phages were isolated to C.toxicus and purified following the methodology described by Bradley et al. (1961) from toxic ryegrass seeds and livestock feed. Phages utilising other plant pathogenic coryneform bacteria (Table 1) were also isolated from an organic mulch. Characteristics of the mulch were previously described (Kurtboke et al., 1993b).

Host ranges of the phages were detected using 20 strains of plant pathogenic coryneform actinomycetes and 11 strains of C.toxicus obtained from WADA (Table 1). Details of the strains were previously described (Riley, 1987; Riley and Ophel, 1992a).

Particle morphology was studied with a transmission electron microscope (JEOL-2000 FX II) operated at 80 kV.

Effects of selected herbicides on conversion rates of non-toxigenic C.toxicus isolates into toxigenic strains by bacteriophage.

Three tentatively identified non-toxigenic C.toxicus isolates from toxic ryegrass seeds were used to test the effect of herbicides on the conversion of the strains into toxigenic ones. Broth cultures of 523M medium (100 ml) were inoculated with one of the isolates (10^8 cfu/ml) and incubated at 26 C for 2 dd as described by Ophel et al. (1993). After 2 dd of incubation a mixture of a phage and herbicide was added to the growing cultures. The phages were either phiCT1 or phiCT2 (10^8 pfu/ml) and the herbicides included Simazinet (registered trademark), (Ciba-Geigy Australia Ltd.), Glean (registered trademark), (Dupont Australia Ltd.), Hoegrass (registered trademark), Trifluralin (registered trademark), (Hoechst Australia Ltd.), Sertin (registered trademark), (Schering Australia Pty. Ltd.). Herbicide concentrations ranged from 20 to 400 ulg/ml (Roslycky, 1982). Controls contained the phage but no herbicides. Cultures were further incubated for 5 dd at 25 C and aliquots (200 ul) of the Iysates ( 10^9 pfu/ml) were plated onto 523M agar and incubated for 14 dd (Ophel et al., 1993). Colonies were grouped into types according to the morphological appearances as described by Ophel et al. (1993). Three replicates were used for each treatment and results were analysed using the Student's 't' test. The toxicity of C.toxicus isolates was tested before and after herbicide and phage applications using the rapid technique described by Riley and Ophel (1992b).

RESULTS

Detection, isolation and identification of C.toxicus from the plates treated with and without phage.

On plates without phage, colonies resembling Curtobacterium flaccumfaciens, Clavibacter michiganensis (subspp. insidiosus, michiganensis, nebraskensis and sepedonieus) and Bacillus spp. formed confluent growth and the enumeration and isolation of C.toxicus was difficult. On the other hand, as a result of the reduction in the number of the bacteria masking the growth of C. toxicus on the plates treated with phage, detection of the bacterium was much easier (Table 2). Most of the tentatively identified isolates from phage treated plates were susceptible to the phage specific to C. toxicus and biochemical tests also confirmed that these isolates were C. toxicus (Table 3).

Partial characterisation of C.toxicus phage.

Phages phiCT1 and phiCT2 isolated to C.toxicus formed small circular clear plaques 1 mm in diameter on the host species .

-------------------------------------------------------------------

                  Ryegrass Seeds        Livestock Feed

Treatment cfu/plate cfu/g dry wt cfu/plate cfu/g dry wt

-------------------------------------------------------------------

Control 4.2+-1.20 0.14x10^4 5.9+-1.12 0.2x10^4

(38.86+-1.50) (1.4x10^4) (33 9+-1.53) (1.13x10^4)

Phage 15.3+-1.33 0.51x10^4 11.08+-1.60 0.37x10^4

(11.2+1-.60) (0.4x10^4) (13.8+-1.45) (0.46x104)

-------------------------------------------------------------------

Table 2. number of colony forming units (cfu) of C.toxicus and of other bacteria (in brackets) in isolation plates treated or not with phage. (The 't test is significant at P<0.001 for the reduction of bacteria and at P<0.01 (ryegrass) and P<0.02 (livestock feed) for the increase of C.toxicus on isolation plates treated with phage.

-------------------------------------------------------------------
                     Ryegrass Seeds          Livestock
Feed

Organisms Control Phage Control Phage Clavibacter toxicus 5 25 3 18 Clavibacter spp. 11 3 11 3 Arthrobacter spp. 2 - - - Curtobacterium spp. 1 - 1 - Rhodococcus spp. 2 - - - Other bacteria 9 3 15 9 Total No. of isolates 30 30 30 30 -------------------------------------------------------------------

Table 3. Distribution of isolates from plates with or without phage (control) according to biochemical Riley and Ophel, 1992a) and phage typing criteria. (-: none isolated).

-------------------------------------------------------------------

Negatively stained particles of phages phiCT1, and phiCT2 belonged to Siphoviridae (B1) morphotype (Francki et al., 1991) with icosahedral heads 52 and 50 nm in diameter respectively. Rigid tails of the phages were 6.35 and 6.25 nm wide and 145 and 1.40 nm long respectively (Fig. 1). Phages phiCTl and phiCT2 were species specific and only lysed C.toxicus strains (Table 1).

Effects of selected herbicides on the conversion rate . of non-toxigenic C.toxicus isolates into toxigenic ones in the presence of the bacteriophage.

Colony morphology.

Colony morphology designated as Type 2 fitted the description of Ophel et al. (1993) and had a glassy appearance. The colonies characterised by this morphology grew very slowly (10-15 dd) and were resistant to phages phiCT1 and phiCT2. They were extremely sticky and resembled melted cheese. Colonies that grew 5-7 dd after the Iysate was plated and showed normal morphology were designated as Type 3 following the description of Ophel et al. (1993). These colonies were resistant to species specific phages phiCT1 and phiCT2.

Figure 1. Morphology of phages phiCTl (a) and phiCT2 (b). Bar = 50 nm.

Toxin production.

On the basis of treatments with and without herbicides, all Type 2 isolates were found to be toxic. However, herbicide treated host-phage systems yielded greater numbers of toxic Type 2 colonies in comparison with the non-herbicide treated host-phage systems. This increase was observed only when high concentrations of herbicide (200-400 ug/ml) were applied (Table 4). Increase in colony forming units (cfu) of toxic colonies of C.toxicus was observed with each type of the herbicide and was independent of the type of the herbicide used (Table 4).

DISCUSSION

Although there are studies characterising the population of C. toxicus in grain, knowledge of the ecology of this bacterium in such substrates is still poor. This is due to the dominance of other saprophytic bacteria on isolation plates which overgrow and inhibit the development of slow growing C.toxicus (McKay and Ophel, 1993). The method of preincubating samples with phage utilising unwanted bacterial taxa on isolation plates successfully reduced the numbers of bacteria masking the development of slow growing C.toxicus and

hence facilitated the detection and isolation of the bacterium. The methodology described here may be used for the rapid isolation and culturing of the ARGT bacterium once it is detected in natural substrates using other techniques such as Randomly Amplified Polymorphic DNA - Polymerase Chain Reaction RAPD-PCR). Use of this methodology would increase our knowledge of the ecology of the ARGT bacterium and its phage in toxic fields, grain and livestock feed. In addition it would provide information on the host-phage interactions, essential for further exploitation of phage susceptibility for selective isolation purposes (Kurtboke and Williams, 1991).

-------------------------------------------------------------------

                                     PHAGE

Herbicide     Concentration
                (ug/ml)          phiCTl               phiCT2

                            C.T.2      C.T.3     C.T.2     C.T.3
-------------------------------------------------------------------
Control                       2         30         3        32

Simazine (R)     20           2         28         4        36
                 50           2         29         3        33
                100           2         37         2        32
                200           4         32         4        33
                400           8*        24         5        34

Glean (R)        20           2         24         5        36
                 50           3         36         4        39
                100           3         28         5        30
                200           4         33         4        33
                400           4         27         4        26

Hoegrass(R)      20           2         27         5        36
                 50           2         33         5        37
                100           2         28         4        36
                200           3         34         4        33
                400           4         36         5        40
                                                               
Trifluralin(R)   20           3         33         3        34
                 50           3         27         2        32
                100           3         30         3        38
                200           4         33         7        33
                400           6*        32         8*       36

Sertin (R)       20           2         32         3        35
                 50           2         31         2        31
                100           2         34         4        36
                200           3         36         4        34
                400           4         35         6        31
-------------------------------------------------------------------
Table 4. Number (cfu/plate) of colony types (C.T) 2 and 3 after plating Iysates from the host phage systems treated with different herbicides. Grouping according to colony morphology (Ophel et al., 1993).(R) = registered trademark.

-------------------------------------------------------------------

The importance of phage typing for the identification of plant pathogenic bacteria has already been stressed by Gross and Vidaver (1979). The species specific phages isolated from two different sources in this study facilitated the rapid tentative identification of the ARGT bacterium. The isolated phages showed great similarities in their morphology and resembled the phage isolated by Riley and Gooden (1991) and Ophel et al. (1993). The relative frequency of morphological phage types may have significance in the detection of phage responsible for the toxin production by C.toxicus when it is infected with a bacteriophage since McKay and Ophel (1993) noted that the phage isolates to C.toxicus were indistinguishable on the basis of DNA restriction patterns.

Although further chemical tests are required to confirm the production of the toxin, such as high-performance liquid chromatography (Cockrum and Edgar, 1985), results obtained with the semiquantitative method of Riley and Ophel (1992b) indicated that herbicide application at high concentrations can increase the numbers of toxic colonies and the frequency of toxigenic conversion.

The capability of a temperate phage to lysogenise a sensitive host is controlled by three factors: the genetic composition of the phage, the genetic composition of the host, and the environment (Gold, 1959). During the first minutes after infection with a temperate phage lysis or lysogenisation take place. This depends on many conditions such as cation concentration in the environment, changes in temperature and multiplicity of infection (Gold, 1959). These factors can also affect a carrier-type lysogeny (Gold, 1959). However, a thorough study on the environmental parameters and their effect on the carrier-type lysogenicity of C.toxicus phage has not so far been conducted in Australia. Further studies in the Southern and Western Australian grain belt would provide more information on the effects of herbicides and of other environmental parameters on the toxigenic conversion of C.toxicus.

ACKNOWLEDGEMENTS. Parts of this study were conducted in collaboration with the WADA, Division of Animal Industries. I thank Dr.S.S. Sutherland (WADA) for her cooperation and for supplying materials and type strains used in this study. I am also grateful to Assoc.Prof.K.Sivasithamparam at the University of Western Australia (UWA) and Dr.R.Gilmour (WADA Plant Industries) for their support and for allowing me to use their facilities. Mr.S.Parry (UWA) assisted with the preparation of the electron micrographs.

Bradley, S.G., D.L.Anderson & L.A.Jones (1961). Phylogeny of actinomycetes as revealed by susceptibility to actinophage. Deu.lnd.Microbiol., 2: 223-237

Cockrum, P A. & J. Edgar (1985). Rapid estimation of corynetoxins in bacterial galls from ryegrass (Lolium rigidum Gaud.) by high-performance liquid chromatography. Aust. J. Agric. Res. 36: 35-41

Francki, R. I. B., C. M. Fauquet, D. L. Knudson & F.Brown (1991). Classification and nomenclature of viruses. V Report of the International Committee on Taxonomy of Viruses. Arch. Virol.

Suppl. 2: 161-166

Gold, W. (1959). Effects of the medium and its composition on the activities of actinophage for Streptomyces griseus. Ann. N. Y. Acad. Sci., 81: 99510 16

Gross, D. & A. Vidaver (1979). A selective medium for isolation of Corynebacterium nebraskense from soil and plant parts. Phytopathology, 69: 82-87

Kurtboke, D. I. & S. T. Williams (1991). Use of polyvalent phage for selective isolation purposescurrent problems. Actinomycetes, 2: 31-34

Kurtboke, D. I., C-F.Chen & S. T. Williams (1992). Use of polyvalent phage for reduction of streptomycetes on soil dilution plates. J. Appl. Bacteriol., 72: 103-111

Kurtboke, D. I., N. E. Murphy & K. Sivasithamparam (1993a). Use of bacteriophage for the selective isolation of thermophilic actinomycetes from composted eucalyptus bark. Can. J. Microbiol., 39: 46-51

Kurtboke, D. I., C. R. Wilson & K. Sivasithamparam (1993b). Occurrence of Actinomadura phage in organic mulches used for avocado plantations in Western Australia. Can.J.Microbiol.. 39: 389-394

McKay, A. C. & K. M.Ophel (1993). Toxigenic Clavibacter/Anguina associations infecting grass

Copyright 1994 C. E. T. A.


The following images related to this document are available:

Photo images

[ac94007a.jpg]

Line drawing images

[ac94007a.gif]
Home Faq Resources Email Bioline
© Bioline International, 1989 - 2024, Site last up-dated on 01-Sep-2022.
Site created and maintained by the Reference Center on Environmental Information, CRIA, Brazil
System hosted by the Google Cloud Platform, GCP, Brazil