| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Ottawa-Carleton Institute of Chemistry, Carleton University, Ottawa, Ontario, K1S 5B6
Gregory W. Adams
JD Irving Ltd. Sussex Tree Nursery, 181 Aiton Road, Sussex, New Brunswick, E4G 2V5
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Previous studies conducted in growth chambers had demonstrated the successful experimental inoculation of white spruce seedlings with anti-insectan toxin producing needle endophytes. Needles colonized with the rugulosin-producing endophyte 5WS22E1 (DAOM 229536) contained rugulosin in concentrations that impaired spruce bud-worm growth. Here we report experimental inoculations conducted under nursery conditions. To improve the reliability of detecting successful colonization, a polyclonal assay was developed for the target endophyte 5WS22E1. It was able to reliably detect the fungus in 500 ng subsamples of colonized needles. Seventeen months post-inoculation, 330 seedlings from 1235 inoculated were colonized. A random selection of 113 colonized seedlings was analyzed for rugulosin. Needles of most (90%) contained detectable concentrations of rugulosin. The range and distribution of the rugulosin concentrations was similar to that found in earlier tests done in growth chambers.
Key words: anti-insectan compounds, fungal endophytes, rugulosin, spruce budworm, white spruce
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, Clay and Holah 1999). Conifer needles are also infected by systemic fungal endophytes that may fulfill several ecological roles (Carroll 1988, Ganley et al 2004). We have been interested in their possible role in limiting conifer needle herbivory. Over a period of approximately 10 y, a collection of white spruce endophytes was obtained according to Wilson et al (1994). Strains producing anti-insectan toxins were identified together with the chemical structures of their toxins from sites in New Brunswick, Canada (Findlay et al 2003).
In previous studies of some of these strains, seedlings were inoculated with a needle to inject fungal cells into the stem. After 3 mo in growth chambers, colonization was determined by culturing surface-disinfested needles. A small percentage (2–3%) was colonized by, among others, a rugulosin-producing endophyte (5WS22E1). The colonized needles contained rugulosin (as determined by TLC and HPLC analysis) and larvae eating such needles did not gain as much weight as those eating uncolonized needles (Miller et al 2002). This demonstrated that the presence of the endophytes and their toxin(s) would affect herbivory under experimental conditions. Rugulosin has been shown to be produced not only by conifer endophytes but a number of species including P. islandicum (Turner et al 1983).
Assays using antibody-based methods for the detection of grass endophytes have been developed for several endophyte species both with microplate assays and tissue immunoblot methods (Gwinn et al 1991, Johnson et al 1982, Reddick and Collins 1998). Compared to competing methods such as PCR, these have a greater potential for field use based on simple protocols. The purpose of this paper is to report: (i) the development of a polyclonal antibody for the rugulosin-producing endophyte 5WS22E1; (ii) the inoculation of 1235 seedlings and subsequent growth outdoors under commercial nursery conditions (iii) analysis of these needle samples using the culture method previously employed and the antibody method and (iv) HPLC analysis for the presence of rugulosin in ~10% of the positive samples.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
. Control-pollinated full-sib families of white spruce were produced for inoculation. The families were chosen to provide a diverse set of genotypes. Nine families originating from unrelated white spruce parents were used. The parents were selected from the J.D. Irving Ltd. genetic improvement program and originated within a range of 45N to 47.5N latitude and 65W to 70W longitude in New Brunswick and Maine. The minimum distance between trees selected in the forest was 200 m. Seeds were taken from frozen storage at –5 C, planted in plastic seedling containers in a media containing 3:1 peat moss/vermiculite and placed in a greenhouse. Fertilization via irrigation water started 1 mo after sowing (10-52-10 for 2 wk followed by 8-20-30 for 1 wk followed by 20-8-20 at 100 µg/L and increasing to a maximum of 125 µg/L).
5WS22E1 (DAOM 229536) cultures were grown on 9 cm plates containing 2% malt-extract agar (Difco) at 25 C for 8 wk. After the incubation period, 5 mL sterile water was poured on agar surface, which was rubbed gently with a sterile bent glass rod. The resulting suspension was taken up with a sterile pipette, macerated and diluted with sterile water to deliver an average of three fungal hyphal fragments per drop (6 µL) from a sterile 1 mL syringe with a 0.45 mm needle (B-D No. 309597) as determined by counting with a hemocytometer. Wound inoculation of 1235 seedlings was performed in a laminar flow hood by injecting 6 µL into the unlignified tissue of the stem typically 10 mm away from the terminal shoot (Miller et al 2002). This was done at the Sussex Tree Nursery on 16 Apr, 2001. This is located at 45°43'N, 65°31'W; elevation 21.30 m. Mean annual temperature is 5.8 C ( Jan –8.5 C, Jul 19.0 C) with average precipitation of 245 cm snow and 915 mm rain.
The trees grew in trays for 6 mo in a greenhouse, at which point they were planted into pots and left in a shaded area with irrigation until sampling in mid-Sep 2002.
ELISA development.— – Cells of 5WS22E1 were grown on two types of liquid media and on irradiated, young uninoculated white spruce needles. The needles were irradiated with 25kGy (MDS Nordion, Montreal, PQ) and 200 mg was placed in a sterile glass Petri dish containing a filter paper followed by the addition of 1 mL sterile water. After 24 h, the culture was inoculated with a small piece of culture taken from the leading edge of a 2% malt-extract agar plate. The first liquid medium used was a glucose/sucrose mineral salts medium (1 g/L KH2PO4, 1 g/L KNO3, 0.5 g/L MgSO4·7H2O, 0.5 g/L KCl, 0.2 g/L glucose, 0.2 g/L sucrose) and the second, 2% malt-extract. An aliquot (50 mL) of each medium was dispensed into 250 mL Erlenmeyer flasks, respectively and autoclaved. An agar culture was macerated in 100 mL sterile water under aseptic conditions. An aliquot (2.5 mL) was used to inoculate the flasks. All cultures were incubated at 18 C ca. 3 mo. At the end of the incubation period, the cells growing on the needles were carefully scraped off with a scalpel and freeze dried. Cells from the liquid cultures were filtered and washed several times with sterile distilled water and freeze dried.
Polyclonal antibody production was performed in goats at Cedarlane® Laboratories Limited, Hornby, Ontario. This laboratory meets the requirements of the Canadian Council on Animal Care. Freeze-dried cells from each medium were ground up and each diluted in sterile PBS to a concentration of 20 mg/mL for the antigen solution. A total of 0.5 mL of this solution was emulsified with 0.5 mL of complete Freund’s adjuvant (Brenntag Biosector, Denmark) for the primary immunization. A total of 0.5 mL of incomplete adjuvant was used for the subsequent boosts. A pre-immune sample was obtained from the jugular vein of each goat with a needle and vacutainer before the primary immunization. Each goat then was injected with a 21-gauge needle intra-muscularly in the hind quarter at four sites with 0.25 mL of the emulsified antigen solution per injection site. After 28 d the goat received its first boost as described above, its second boost at day 53 and a test bleed was taken at day 66.
The antibodies produced from the three goats were tested to determine their avidity and cross reactivity with powdered, freeze-dried young uninoculated spruce needle cells, as well as cells of the most common needle phylloplane fungi isolated from these needles (Alternaria alternata, Phoma herbarum, Cladosporium cladosporioides and Aspergillus fumigatus; Miller et al 1985, Miller et al 2002), and other white spruce conifer endophytes: 5WS11I1 (DAOM 229535; vermiculin producer) and 5WS331L1 (a rugulosin-producer). In addition a number of balsam fir endophytes were tested (data not reported here because they could not occur in white spruce needles). In the case of the needle endophytes, cells were produced on irradiated needles as above. The phylloplane species tested were grown in shake culture with a maltose, yeast extract, peptone medium (Miller and Mackenzie 2000), filtered, washed and freeze dried as above. This method was shown to be suitable for antigen production by such fungi in unrelated studies. Cells were ground to a fine powder in small mortar and carefully weighed. Suspensions of known concentration were made in TBS (0.8 g/L NaCl, 0.2 g/L KCl, 1.89 g/L Tris-HCl, and 1.57 g/L Tris base) in vials and vortexed.
Avidity and cross reactivity experiments were conducted on sera from the goats treated with the various immunogens in a similar fashion, first optimizing cell additions/serum dilutions, and then conducting cross-reactivity experiments. As needed aliquots of cells were diluted in 0.1 M carbonate buffer pH 9.6 (Sigma) coating buffer to defined concentrations and pipetted into 96 well Nunc brand microplates. The plates were covered with an acetate-sealing sheet and placed on a rotary shaker for 4 h at room temperature. The plate was removed, turned upside down and shaken to remove all of the coating solution in the sink. A total of 200 µL of Blotto (10 g of nonfat dry milk/L of TBS) was added to each well, covered and placed in a refrigerator at 5 C overnight. The plate was removed and washed with a Molecular Devices Skan Washer 400 to remove the Blotto solution. The washing solution used was TTBS (0.5 mL of tween-20/L of TBS) with a washing program of three cycles of soaking, washing and rinsing. Various dilutions of goat serum in Blotto were made from which 100 µL was added to the microplate wells. The plate was covered and placed on a rotary shaker 1 h at room temperature; it was removed and washed with TTBS as described above. A total of 100 µL of antigoat IgG-horseradish peroxidase conjugate (Sigma) diluted 5000 times in Blotto was added to each microplate well. The plate was covered and incubated at room temperature 1 h. The plate was washed for the final time and substrate was added. A total of 100 µL of TMB (Tetra-methly benzidine, Sigma) was added to each well. The plate was covered and incubated at room temperature 30 min. The reaction was stopped with 50 µL of 0.5 M sulfuric acid. The plate was read immediately at 450 nm with subtraction of 630 nm on a Molecular Devices Spectra Max 340PC reader.
The polyclonal antibody produced with 5WS22E1 cells grown on the defined medium had low avidity and was not studied further. The antibody from the 2% MEA medium had acceptable avidity but unacceptable cross-reactivity. The polyclonal antibody to the cells cultured on irradiated needles was used in all further studies. A 4000-fold dilution from the latter serum with a 5000 dilution of the secondary antibody was determined to be optimal for tests with 5WS22E1 cells, allowing a preliminary estimate of the sensitivity of the assay to be made. Tests with this and the other conifer endophytes were done with cell weights of 15–240 ng over a fivefold range in antibody concentration. Using a serum dilution of 4000, response to 15, 30, 60, 120 and 240 ng cells of the phyloplane species and 60, 100 and 500 ng freeze dried white spruce needles was determined. Powdered freeze-dried white spruce needles (500 ng/well) were spiked with additions of 5WS22E1 cells over the above range.
Needle analyses for 5WS22E1: plating method.— –
At the time of sampling, average tree height was 12.9 cm. Needles from each tree were carefully removed radiating out from the inoculation point, placed in sterile plastic bags and immediately frozen for transport to the laboratory. Each bag was taken from the freezer and ca. 20 needles removed. Each needle was surfaced-disinfested by dipping in 70% ethanol for 1 min, rinsing in sterile distilled water 1 min and blotted dry on sterile tissue. This was placed in a sterile Petri dish and cut into two segments and the needle half that was attached to the stem plated on 2% malt-extract agar. Plates were incubated at 18 C for 6 wk and were inspected regularly by microscopy for 5WS22E1 growth (Miller et al 2002).
ELISA.— – The remaining needles were freeze-dried. Approximately 20 needles from each frozen needle sample were removed, ground to a fine powder in a vial with a Spex-Certiprep grinder-mixer (model 5100) and 10 mg weighted out. One mL of TBS (0.8 g/L NaCl, 0.2 g/L KCl, 1.89 g/L Tris-HCl, and 1.57 g/L Tris base) was added to each vial and placed on the vortex until completely mixed (ca. 1 min). The samples were assigned codes unrelated to the tree codes and randomized to ensure that samples from individual trees were analyzed across many plates. All were diluted in 0.1 M carbonate buffer pH 9.6 (Sigma) coating buffer to concentrations of 100 and 500 ng of needles per 100 µL well and pipetted in duplicate onto 96-well micro-plate. The remaining steps in the analysis were as described above. In each trial 60 ng of 5WS22E1 cells were used as positive control to assess the performance of the assay; relative standard deviation of the net value of 25 representative experiments was 8.7%. Unless an acceptable positive control result was obtained, the results from individual plates were redone. Samples with high absorbance values in both 100 ng and 500 ng tests were rejected as indicating dilution problems. Results were scored as positive when absorbance of the 500 ng sample was greater than the lowest absorbance value above 1 (1.000) plus the mean absorbance value of 30 ng of the target endophyte on that plate.
Chemical analyses.— – Rugulosin (purity > 95%) used for standards was a gift from Dr John A. Findlay, University of New Brunswick. The presence of rugulosin was to be determined in a subsample of 113 randomly selected trees of the 330 trees determined to be endophyte positive by antibody. A 100 mg sample of freeze-dried needles was ground to a fine power as describe above and extracted with 10 mL of ice-cold petroleum ether by stirring for 45 min under low light. The flask was kept on ice during the extracting and was covered with aluminium foil to prevent degradation by light. The suspension was filtered by suction and discarded. The needles were returned to the flask and extracted with 10 mL of chloroform for 45 min as above for petroleum ether. The new suspension was filtered by suction and retained while the needles were discarded. The chloroform extract was washed with 10 mL of 5% NaHCO3 in a separator funnel. This first chloroform layer was discarded, the pH was acidified to pH 3 with 1 N HCl, and a new 10 mL of chloroform was added to the funnel and extracted. The chloroform was removed and dried in an amber vial under a gentle stream of nitrogen.
The dried extracts were redissolved in 50 µL of acetonitrile, 10 µL was removed and injected into an 1100 series Agilent Technologies HPLC-DAD, with a Synergi Max RP 80A, 250 x 4.6 column (Phenomonex) and a gradient method adapted from Frisvad (1987). The gradient started at 90% water with 0.05% TFA and 10% acetonitrile and changed to 10% water with 0.05% TFA and 90% acetonitrile over the 20 min run. Samples were analyzed at 389 nm, the maximum UV/VIS absorption for rugulosin and peak identity was confirmed by full spectrum data from the diode array detector. The limit of quantification was 150 ng/g freeze dried powdered material; recoveries from spiked needles averaged 75%. Statistical analyses were done with SYSTAT v. 10.2 (Point Richmond, California).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
, results of triplicate experiments presented). Relative cross-reactivity to 15, 30, 60, 120 and 240 ng cells of the phyloplane species was moderate (8%). Over the same range, there was slightly greater cross-reactivity to the cells of the two white spruce endophytes tested (~15%), one of which produced rugulosin. Even at 240 ng cells, the response was below the 1 absorbance unit threshold used. The response of the polyclonal to the above range of the nontarget fungal cells was not linear across a range of antibody concentrations. The response to 60, 100 and 500 ng of white spruce cells again across a range of antibody concentrations was moderate (~6%) and nonlinear. A comparison of the response to 60 ng of nontarget cells to the target endophyte is provided (FIG. 1); average relative standard deviation of replicates included in these data was 6.3%.
|
, results of triplicate experiments presented).
|
). As before the colonies of these taxa typically arose from the sides of the needles rather than the cut ends.
When the same samples were analyzed by the antibody method, 330 or 27% were positive. All samples where the fungus was seen in culture were positive by the antibody assay. Of the 113 samples tested for rugulosin by HPLC from the 330 antibody positive needles, 101 (90%) were positive at the limit of quantification. The range of concentrations was 0.15–24.8 µg/g needle. The distribution of values (assuming half the detection limit for the nondetects) is shown in FIG. 3). The Geometric Mean needle rugulosin concentration was 1.02 µg/g. Mean frozen weight of 100 representative needles was 2.6 mg/needle. Freeze dried weight was 1.08 mg/needle.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). It was known that the endophyte grew and colonized the seedling slowly. The 3 mo incubation might not have been sufficient to permit the seedling to be colonized to a detectable level. In addition it was not clear in what direction the fungus would spread from the point of colonization. These latter factors made it impossible to know whether the needles harvested for the test could reasonably be expected to be colonized at all.
The polyclonal assay developed for the target endophyte 5WS22E1 was of comparable sensitivity to similar assays for grass endophytes (Gwinn et al 1991, Johnson et al 1982, Reddick and Collins 1998). This was achieved despite the greater difficulties of the conifer needle matrix compared to grass leaves. Cuttings of the latter can be placed directly in microplates whereas the tough, hydrophobic conifer needles must be ground to expose fungal cells to the antibody. Cross-reactivity to the potentially competing fungi was acceptable (FIG. 1
). Epiphytic biomass is typically low in young needles (Carroll 1979) and is comprised mostly of fungi (Swisher and Carroll 1980). Based on an extrapolation of our data on their data (Swisher and Carroll 1980 TABLE II), the fungal epiphytic biomass on these 19 mo old needles would have < 3 µg/g. This means the presence of such fungi in our needle samples had no effect on the antibody response in these assays. In addition relatively large amounts of powdered white spruce needles compared to fungal cells did not affect the reliable detection of 30–60 ng cells of target endophyte (FIG. 2).
Nonendophyte fungi grew from nearly all the surface-disinfested needles of these nursery-grown seedlings. This was in contrast to the situation from the younger, growth chamber-grown seedlings where this was observed previously in half the needles (Miller et al 2002). The presence of these faster growing fungi made it much more difficult to recognize the presence of the target endophyte. When the same needle samples were tested by the antibody method, the rate of colonization was eight times higher or 27%. The present data cannot completely resolve the reason for this difference, but it is clear that a major factor was the insensitivity of culture method. It also is likely that endophyte spread and biomass were greater than we could detect and this will be the subject of further study.
Most (90%) of the needles shown to be endophyte positive by the antibody method contained rugulosin concentrations above the detection limit. This provides additional evidence of the reliability of the antibody method. The mean (1 µg/g), range and distribution of rugulosin concentrations in these needles (FIG. 3) were similar to that found in growth chamber-grown seedlings (Miller et al 2002). The analytical method in the present study included examination of the full scan UV spectrum of the rugulosin peak. This provides additional confirmation of the presence of this compound compared to the previous HPLC UV and TLC analyses (Miller et al 2002).
The analyses were done with replicate 500 ng sub-samples obtained from a 20 needle sample. The conservative assumption used in the scoring of the 330 positive seedlings was that they contained the limit of quantification. Using this assumption, each needle would contain 60 µg endophyte biomass/g needle or ~6%. For comparison Swisher and Carroll (1980) report that 1–4 y old Douglas-fir needles have ~10 µg epiphyte biomass/g needle. This comprised mainly fungi but including algae and bacteria in older needles. This measurement enables another comparison to be made: the amount of rugulosin per weight of fungal cells.
Several studies have been made of the production of mycotoxins in living plants with culture and ergosterol to assess fungal biomass. A representative example is a study of deoxynivalenol in experimentally inoculated preharvest corn with corresponding measurements of ergosterol and viable fungi among other data (Miller et al 1983). With the ergosterol-fungal biomass conversion discussed in Gessner and Newell (2002), it is possible to estimate that in planta dexoynivalenol concentration corresponded to ~3% of the fungal biomass. With the mean rugulosin concentration, the ratio was ~2%.
In summary the polyclonal assay developed for the target endophyte 5WS22E1 reliably detected the fungus in 500 ng subsamples of colonized needles. Nineteen mo post-inoculation, rates of colonization detected were much higher than that documented from earlier chamber studies. Analysis showed most colonized needles (90%) contained detectable concentrations of the 5WS22E1 anti-insectan compound rugulosin. The range and distribution of needle rugulosin concentrations was similar to that found in earlier tests done in growth chambers.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Corresponding author. Tel: 613-520-2600 X 1053, 613-520-3749. E-mail: david_miller{at}carleton.ca
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
———, Holah J. 1999. Fungal endophyte symbiosis and plant diversity in successional fields. Science 285:1742–1744.
Carroll GC. 1979. Needle microepiphytes in a Douglas-fir canopy: biomass and distribution patterns. Can J Bot 57:1000–1007.[CrossRef]
———. 1988. Fungal endophytes in stems and leaves: from latent pathogen to mutualistic symbiont. Ecology 69: 2–9.[CrossRef]
Findlay JA, Li G, Miller JD, Womilouju TO. 2003. Insect toxins from spruce endophytes. Can J Chem 81:284–292.[CrossRef]
Ganley RJ, Brunsfeld SJ, Newcombe G. 2004. A community of unknown, endophytic fungi in Western White pine. Proceedings of the National Academy of Sciences of the United States of America 101:10107–10112.
Gessner MO, Newell SY. 2002. Biomass, growth rate, and production of filamentous fungi in plant litter. Manual of Environmental Microbiology. 2nd ed. p 390–408.
Gwinn KD, Collins-Shephard HM, Reddick BB. 1991. Tissue print-immunoblot, an accurate method for the detection of Acremonium coenophialum in tall fescue. Phytopathology 81:747–748.[CrossRef]
Miller JD, Mackenzie S. 2000. Secondary metabolites of Fusarium venenatum strains with deletions in the Tri5 gene encoding trichodiene synthetase. Mycologia 92: 764–771.[CrossRef]
———, ———, Foto M, Adams GW, Findlay JA. 2002. Needles of white spruce inoculated with rugulosin-producing endophytes contain rugulosin reducing spruce budworm growth rate. Mycol Res 106:471–479.[CrossRef]
———, Strongman D, Whitney NJ. 1985. Observations on fungi associated with spruce budworm infested balsam fir needles. Can J Forest Res 15:896–901.[CrossRef]
———, Young JC, Trenholm HL. 1983. Fusarium toxins in field corn. I. Parameters associated with fungal growth and production of deoxynivaneol and other mycotoxins. Can J Bot 61:3080–3087.[CrossRef]
Reddick BB, Collins MH. 1988. An improved method for detection of Acremonium coenophialum in tall fescue plants. Phytopathology 78:418–420.[CrossRef]
Swisher R, Carroll GC. 1990. Fluorescein diacetate hydrolysis as an estimator of microbial biomass on coniferous needle surfaces. Microbial Ecology 6:217–226.[CrossRef]
Turner WB, Aldridge DC. 1983. Fungal Metabolites II. Academic Press Inc. New York. 152 p.
Wilson R, Wheatcroft R, Miller JD, Whitney NJ. 1994. Genetic diversity among natural populations of endophytic Lophodermium pinastri from Pinus resinosa. Mycol. Res 98(7):740–744.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
