| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. of Hort. and Crop Sci., The Ohio State Univ., Columbus, OH 43210
b The Ohio State Univ. Ext., Guernsey Co., Cambridge, OH 43725
c The Ohio State Univ. Ext., Morgan Co., McConnelsville, OH 43756
d The Ohio State Univ. Ext., Jackson Co., Jackson, OH 45640
* Corresponding author (barker.169{at}osu.edu)
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: ANOVA, analysis of variance • DM, dry matter • E-, endophyte-free ‘Jesup’ tall fescue • Nontoxic-E, ‘Jesup’ MaxQ tall fescue infected with nontoxic-endophyte (Neotyphodium coenophialum, strain AR542) • Toxic-E, volunteer tall fescue infected with toxic endophyte
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
One strategy for alleviating endophyte toxicity for livestock is to establish endophyte-free tall fescue (Ball et al., 2003). Perhaps the only disadvantage with this approach is the difficulty in maintaining endophyte-free stands, especially where previous stands have been infected with toxic endophyte. Neotyphodium coenophialum is a seed-borne fungus that is not dispersed by spores or by pollen of infected plants, and mechanisms for its dispersal are restricted to infected seed and, in rare cases, tillers. Five mechanisms for contamination of endophyte-free tall fescue stands include: (i) use of seed contaminated with toxic endophyte, (ii) plants resulting from buried seed and tillers that survive prior herbicide and cultivation practices (Hume, 1999), (iii) plants resulting from seed that have survived digestion and are dispersed in manure (Shelby and Schmidt, 1991), (iv) plants establishing from seed that is dispersed in hay or from other farm equipment, and (v) seed that is dispersed by feral animals. Efforts to control these processes have contributed to the recommendation that establishment of endophyte-free stands in previously toxic-endophyte infected fields might require a 1- to 2-yr period of tall fescue or ryegrass (Lolium perenne L.) exclusion (Hume, 1999).
Once toxic endophyte-infected seedlings become established in an endophyte-free stand ecological and physiological processes combine to give them sufficient advantage that stands are likely to become reinfested over time (Hill et al., 1998). Endophyte-infected plants have greater growth, greater seed production, and greater germination of that seed (Clay, 1987). Endophyte-infected plants have physiological mechanisms allowing better growth and survival under stressful conditions such as drought and low fertility (Burns and Chamblee, 1979; Malinowski and Belesky, 2000). Endophyte-infected plants also are less likely to be consumed by herbivorous insects (Richmond and Shetlar, 2000) and livestock (Edwards et al., 1993; Penrose et al., 2000; Cosgrove et al., 2002). In combination, these factors explain the high levels of endophyte infection that are typically found in old fescue fields (Shelby and Dalrymple, 1987) and account for the increase of endophyte levels that frequently occur in new endophyte-free seedings (Penrose et al., 2001).
One alternative management strategy might be to use a nontoxic endophyte that has negligible production of the toxic alkaloids, such as ergovaline, that are reported to cause negative effects on animals (West et al., 2001; Bouton et al., 2002). The nontoxic-endophyte may confer identical stress resistance to plants as does the toxic endophyte. Thus, the pastures established with nontoxic-endophyte tall fescue might be less susceptible to reinfestation by volunteer tall fescue and its associated toxic endophyte. The objective of this study was to test the hypothesis that nontoxic-endophyte infected (Nontoxic-E) swards are more resistant to reinfestation by volunteer tall fescue and its associated toxic endophyte (Toxic-E), than endophyte-free (E-) tall fescue swards.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
The experimental design used at South Charleston was a randomized complete block design with four replicates. The same experimental treatments were used, as described above, but excluded the unsown treatment (48 experimental units). Plots were 6 x 1.05 m. The K31 seed was obtained from Kentucky.
Defoliation Management
The Jackson site was first grazed with cattle on 7 Sep. 2001. Spring 2002 was too wet for grazing and accumulated forage was removed in a single hay harvest averaging 5810 kg DM ha–1 on 12 June 2002. During August through October 2002 plots were grazed four times, sequentially, by the same group of six cattle, 3 d per grazing. Pasture mass was measured with a calibrated rising plate meter (Ashgrove Pasture Products, Palmerston North, NZ) before each grazing and averaged 2950 kg dry matter (DM) ha–1 (maximum 3440 kg DM ha–1).
The Belle Valley site was first grazed by cattle on 28 June 2001, and again on 29 Aug. and 15 Oct. 2001. As at Jackson, spring 2002 was too wet for grazing; the experiment was mown to 30-cm height on 23 May 2002, and was subsequently grazed during 1 to 10 June and 21 to 28 June with cattle. Average pasture mass was 3730 kg DM ha–1 (maximum 5150 kg DM ha–1).
The South Charleston site was mown, with harvests to 7.5-cm height on 3 June, 17 July, 27 Aug. and 4 Nov. 2002. The pasture mass at harvest averaged 3200 kg DM ha–1.
Seed and Endophyte Testing
All seed was tested for germination with up to 100 seeds rolled within paper towel, with alternating incubator (Model 1-36VL, Percival Scientific Ltd., Boone IA) temperatures of 15°C for 16 h and 25°C for 8 h for 14 d (Anon., 2000). Those data were used to adjust sowing rates to a viable seed basis. Seed weight of each lot was determined from a random sample of 100 seed. The K31 seed used at the South Charleston site was obtained from Kentucky and inadvertently was not tested for endophyte, germination or seed size before all the seed being planted. It was assumed the germination and seed size were similar to the E- and Nontoxic-E seed.
Endophyte was determined in up to 25 seed of E-, Nontoxic-E and K31 (Ohio source) with an immunoblot antibody endophyte kit customized for seed (ENDO797-1, Agrinostics Ltd, Watkinsville GA). Since the antibody test did not distinguish between living (viable) endophyte and dead (inactive) endophyte in seed, Nontoxic-E and K31 (Ohio source) were retested with an antibody endophyte kit customized for greenhouse-grown tillers (ENDO797-2, Agrinostics Ltd, Watkinsville GA) with tillers grown from seed in a greenhouse for 6 wk. Subsequent calculations were conducted using these ‘grow-out’ results. Tests were conducted according to manufacturer's directions.
Tillers (20–25 tillers plot–1) were collected on 7, 12, and 19 Sep. 2001 at Jackson, Belle Valley, and South Charleston, respectively. Tillers from the 100% E- and 100% Nontoxic-E treatments were sampled from both the subplot and the 0.1-ha area around subplots. Since there were insufficient tall fescue plants in most unsown plots, all tillers from those plots were combined for a single analysis. Healthy tillers were taken near ground level (including roots if possible) and transported on ice until they could be stored in a refrigerator. Tests for presence of endophyte were conducted using an immunoblot antibody endophyte kit customized for field tillers (ENDO797-3, Agrinostics Ltd, GA) according to manufacturer's directions, on 20 tillers per plot. Tests were repeated on tillers collected from the field in spring (June) 2002 and autumn (3 Sept.) 2002.
Alkaloid Testing
Toxic-E and Nontoxic-E infected tall fescue plants were distinguished by a commercial ELISA kit (ENDO899-2-96, Agrinostics Ltd, GA) according to the manufacturer's directions. This kit had a monoclonal antibody that was specific to the lysergic ring structure of the ergot alkaloids and, therefore, tested for ergot alkaloid presence (Hill et al., 2002). Tests were completed on pseudostem sections collected from Nontoxic-E plots, using the same 20 tillers per plot as for endophyte testing described above.
Alkaloid tests on all Nontoxic-E plots in autumn 2001 (1080 tillers averaging 69% endophyte infection) found only one plant producing alkaloids that did not contain endophyte, which was discounted as trivial, and all subsequent alkaloid analyses were only done on plants that tested positive for endophyte. Lower alkaloid occurrence than expected was found in the analyses of September 2001, and tests were repeated on 18 Dec. 2001 with new testing kits and the same samples which had been stored at –18°C for 3 mo. Of the 561 Nontoxic-E tillers that were repeated, 533 gave the same result, 20 showed as new positives for alkaloid presence, and eight did not react that had reacted previously (many of those eight might have been due to low sample loading onto the antibody plates). Reported alkaloid data for autumn 2001 are those plants that reacted in either of the tests. The alkaloid test was repeated on samples collected 3 Sep. 2002, and which had been refrigerated for 10 d before analysis.
Calculations and Statistical Analysis
The relationship between the observed endophyte status of mixed population (Eobs) and the proportions of components (PK31 and Px, x = E- or Nontoxic-E,) and their respective endophyte status (EK31 and Ex) is given in Eq. [1].
 | [1] |
Since the K31 seed used at the South Charleston site was inadvertently not tested for endophyte, its endophyte status (EK31) was calculated by rearrangement of Eq. [1] to give Eq. [2], and using Eobs from the autumn 2001 data for the South Charleston site.
 | [2] |
For all three sites, the observed endophyte percentage was regressed on expected endophyte percentage for E- and Nontoxic-E tall fescue for each site and sampling date by Excel (Microsoft Office Excel 2003, Microsoft Corp., Redman WA). In addition, observed alkaloid infection level (percentage of 20 tillers tested per plot) in September 2001 and September 2002 was regressed on the Nontoxic-E to K31 seed planting ratio. The statistical significance of regression was tested using PROC REG of SAS (The SAS System for Windows Release 8.02, SAS Inst. Inc., Cary, NC).
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Summer and autumn 2001 rainfall was less than average at Jackson (July–September 2001 rainfall was 147 mm, 52% of average) with negligible harvestable growth during that period, but was above average at South Charleston (rainfall 354 mm, 132% of average). Winter 2001–2002 was warmer and drier than average (December 2001–February 2002 air temperature at Jackson was 2.6°C, which was 2.6°C above average, and at South Charleston was 1.5°C, which was 2.7°C above average) with relatively little precipitation (precipitation for the same period at Jackson was 155 mm, 64% of average; and at South Charleston was 171 mm, 81% of average).
Spring 2002 was extremely wet (March–April 2002 rainfall at Jackson was 402 mm, twice average; at South Charleston was 220 mm, 118% of average; and at Belle Valley was 233 mm, 135% of average) and precluded grazing of plots for fear of hoof damage from livestock or machinery. Most months during summer and autumn 2002 had above average rainfall with the exception of August (76 mm at Jackson, 83% of average; 43 mm at South Charleston, 48% of average; and 13 mm at Belle Valley, 16% of average).
Endophyte Testing
Seed
The E- tall fescue had no seed that reacted to the endophyte test kit antibody. Nontoxic-E tall fescue had 20 positive seed out of 21 tested (95.2% infected seed); however, when the test was repeated with live greenhouse-grown tillers, only 14 out of 17 tillers tested positive (82.4%). The Ohio K31 tall fescue had 18 positive seed out of 21 tested (85.7% infected seed); however, when the test was repeated with live plants, only eight out of 17 tillers tested positive (47.1%) (Table 2). Kentucky K31 seed (used at South Charleston) was not tested but was calculated in fall 2001 by Eq. [2] to have averaged 57% infected seed.
|
Sown Plots
The observed versus expected endophyte percentages in tall fescue-sown plots for the three sites for September 2001, June 2002 and September 2002, are shown in Fig. 1 to 3
, respectively (Table 3). In seven of nine observations (3 dates x 3 sites) the observed endophyte levels were similar to the proportion of endophyte-infected K31 that had been sown. For E- treatments at Jackson (June 2002 and September 2002), the observed endophyte levels were consistently above the expected endophyte levels. Nontoxic-E treatments at all three sites were equal to, or less than, the expected endophyte levels. The spring and fall 2002 observations at South Charleston had consistently lower observed endophyte levels. It was assumed that contamination from volunteer tall fescue at South Charleston in Sept. 2001 was negligible and the proportion of endophyte infection in the sown seed was calculated to be 57% by Eq. [2].
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Endophyte levels were greatest at Jackson, and by autumn 2002 had reached an average of 47.5%. Endophyte levels were intermediate at Belle Valley (39.2%) and least at South Charleston (38.2%) in autumn 2002. Reasons for those differences could not be determined with certainty but probably included both the presowing treatment and dispersal mechanisms for tall fescue seed. The plots at South Charleston had been planted in corn (Zea mays L.) for at least 20 yr before this trial; plots at Belle Valley had been planted in corn for 2 yr before this trial, and at Jackson had been in turnips (Brassica rapa L. subsp. rapa) for only 6 mo before this trial. It is likely that the population of buried seed and volunteer tall fescue plants was least at South Charleston, intermediate for Belle Valley and greatest at Jackson. Additional evidence was the unsown plots, which were abundantly populated with tall fescue at Jackson (presumably from buried seed and volunteer plants that survived pre-sowing treatments), and sparsely populated with tall fescue at Belle Valley. Although livestock grazed plots at Jackson and Belle Valley, it was unlikely these were a mechanism of transferring endophyte-infected tall fescue since they were always grazed for 2 to 3 d on tall fescue-free areas adjacent to the trials before entering the trial. Other mechanisms such as dispersal by feral animals cannot be discounted.
Endophyte levels were slightly lower in spring (37.4% in 2002) than autumn (43.8 and 41.6% in 2001 and 2002), and trends were similar between sites and treatments. Sampling in spring may underestimate endophyte levels in the northern fescue belt.
Infestation of E- Plots by Volunteer Tall Fescue
A large deviation between sown (expected) and observed endophyte levels occurred for E- at Jackson and Belle Valley. Observed endophyte levels for E- at Jackson averaged 77, 118, and 143% greater than expected in autumn 2002, spring 2002, and autumn 2002, respectively. Observed endophyte levels for E- at Belle Valley averaged 32, 70, and 39% greater than expected in autumn 2001, spring 2002, and autumn 2002, respectively. Those results were consistent with observations that endophyte-free tall fescue is susceptible to reinfestation by volunteer tall fescue and its associated toxic endophyte (Shelby et al., 1989; Thompson et al., 1989). The endophyte level of two grazed endophyte-free pastures in Tennessee increased to an average of 11% of tillers over 2 yr (Thompson et al., 1989), and in six grazed endophyte-free pastures in Alabama the endophyte level increased in three of these pastures by an average 9.5% over 4 or 5 yr (Shelby et al., 1989). Regressions of observed on expected endophyte levels for E- had average slope and intercept of 0.63 and 10.0, respectively. The intercept of 10 showed an average 10% endophyte infection in plots that should have been endophyte free. The slope being less than 1 suggested that contamination was largely uniform across all E- treatments, an observation that was consistent with contamination from unsown sources (e.g., buried seed), rather than better survival and production from the K31 seed sown with the treatment.
Without doubt, plots were infected by volunteer tall fescue, even as early as September 2001. The source of this contamination cannot easily be determined but was likely to include (i) surviving buried seed, (ii) surviving tiller remnants, (iii) dispersal by feral animals (birds, deer etc.), and (iv) hay and barnyard manure mixed with straw that was spread at Belle Valley before sowing. Observed endophyte levels at Jackson were higher than could have been accounted for by the amount of K31 included at sowing and suggested significant volunteer tall fescue contamination occurred at Jackson even as early as September 2001.
The mean contamination in 100% E- plots (8.3%) suggests an average volunteer tall fescue contamination of 12.8% (adjusting for 65% endophyte infection found in an adjacent old pasture). This was similar to the average deviation from the expected infection level averaged over all E- treatments (8.1 or 12.4% volunteer tall fescue contamination), and suggested the deviation for all E- treatments was due to volunteer tall fescue invasion rather than preferential establishment and survival of the K31 seed included in the sowings.
The mean contamination in 100% Nontoxic-endophyte plots (1.3%) suggested an average volunteer tall fescue contamination of 2.1% (adjusting for 65% endophyte infection found in an adjacent old pasture). This was similar to the average deviation from expected infection levels average over all E- treatments (1.4%), and suggested the deviation for all Nontoxic-E treatments was due to volunteer tall fescue invasion rather than preferential establishment and survival of the K31 seed included in the sowings.
Noninvasion of Nontoxic-E Plots by Volunteer Tall Fescue
Observed endophyte levels for Nontoxic-E for the three sites averaged 94.9, 77.2, and 87.9% of expected levels in autumn 2001, spring 2002, and autumn 2002, respectively. That these levels did not exceed expectations was evidence of the failure of volunteer tall fescue plants (or its associated toxic endophyte) to invade into Nontoxic-E tall fescue swards. Mechanisms accounting for lower levels of endophyte infection might include instability of the Nontoxic-E or volunteer-endophyte associations in their respective hosts, or negative bias in the antibody method. Given the randomized methods used in the laboratory, bias in the antibody method was unlikely. In general (seven of nine cases in Fig. 1–3), the 90 and 100% Nontoxic-E treatments had observed endophyte levels similar to those expected, which suggested loss of endophyte was from the K31 tall fescue population rather than Nontoxic-E plants. Further attention to stability of endophytes in their respective host plants is warranted.
The average proportion of Toxic-E tillers in the Nontoxic-E plots at Jackson, as tested by ELISA for toxic alkaloids, was consistent with the proportion of K31 seed that was sown into those plots (Fig. 4). This was further evidence of the failure of buried tall fescue seed to infect these plots. The failure to detect toxic alkaloids at levels that were expected (based on K31 included in sowing treatments) cannot be explained. Possibilities include failure of the ELISA to detect all incidences of Volunteer-E tillers, as well as loss of endophyte from the K31 sown. Although we had no positive controls, we had no reason to suspect any fault with the ELISA method since it did detect Toxic-E at expected levels at Jackson. The loss of endophyte from field plants is uncommon. Of the 42 fields reported by Shelby et al. (1989) and Thompson et al. (1989), only four showed a decrease in endophyte status, averaging 5%. The high loss of endophyte from Nontoxic-E treatments in this study (Fig. 1 to 3) cannot be explained.
Results from this trial suggest that recommendations for establishment of nontoxic-endophyte tall fescue swards could be varied from recommendations for establishing endophyte-free tall fescue. Current recommendations for new tall fescue seedings are to have a 6-mo to 1-yr exclusion of endophyte-infected tall fescue, which can be achieved by rotating to row crops or alternative forage species (Roberts and Andrae, 2004). In this study, 2 yr of corn at Belle Valley were insufficient to prevent contamination of endophyte-free plots by toxic-endophyte tall fescue. Also, at Jackson, a nontoxic-endophyte tall fescue stand established following a 6-mo turnip crop showed no measurable contamination by toxic-endophyte tall fescue after two summers. Initial evidence suggests that establishment recommendations for nontoxic-endophyte tall fescue could be relaxed from the 6-mo to 1-yr exclusion of tall fescue currently recommended for establishing endophyte-free tall fescue; however, testing in more environments and for longer periods is warranted.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication August 7, 2003.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Related articles in Crop Science:
This article has been cited by other articles:
|
|
 |
|
  B. F. Tracy and I. J. Renne Reinfestation of Endophtye-Infected Tall Fescue in Renovated Endophyte-Free Pastures under Rotational Stocking Agron. J., October 19, 2005; 97(6): 1473 - 1477. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
