AU2004243789B2 - Improvements in grass endophytes - Google Patents

Description

WO 2004/106487 PCT/NZ2004/000116 IMPROVEMENTS IN GRASS ENDOPHYTES TECHNICAL FIELD This invention relates to fungal endophytes and combinations of endophytes with grass plants. More particularly the invention relates to endophytes which form combinations with perennial, annual and hybrid ryegrasses and some other related grasses. Even more particularly the invention relates to combinations having reduced toxicity to grazing livestock as compared to cultivars of endophyte/ryegrass combinations in common use whilst still retaining resistance against pests and/or abiotic stresses.

BACKGROUND ART Fungal endophytes of the genus Neotyphodium (formerly Acremonium) infect a number of temperate climate Pooideae grasses. The Neotyphodium endophytes can produce alkaloids which are considered to confer degrees of pest and possibly disease protection upon the plants in which they naturally occur (Rowan and Latch, 1994; Blank and Gwinn, 1992). Resistance to drought conditions has also been claimed (Elberson and West, 1996). The Neotyphodium endophytes are vertically transmitted through the seed of the grasses and no natural horizontal transmission has been established (Leuchtmann, 1997).

Many of the predominating natural endophyte infections of improved grass cultivars used for pastoral agriculture production also cause significant animal disorders, for example fescue toxicoses (Stuedemann and Hoveland, 1988) and ryegrassendophyte toxicosis (Fletcher et al., 1999). These may be complex toxic reactions by animals to alkaloids produced under a range of plant growth conditions. Significant economic loss within pastoral agriculture systems can occur due to such animal toxicoses. On the other hand presence of at least some endophytes may be essential WO 2004/106487 PCT/NZ2004/000116 for the competitive persistence of the chosen grass in a pasture (Elberson and West, 1996, Fletcher and Easton, 2000).

It has also been found that grass lines can be artificially infected with selected endophytes. Axenic cultures of endophytes can be used to infect grass seedlings, grown initially under sterile conditions (Latch and Christensen, 1985), which can then be selected for desirable qualities, and multiplied for commercial use. Three significant examples of this technology have been developed by AgResearch Ltd: GREENSTONETM tetraploid hybrid ryegrass with ENDOSAFETM endophyte (Tapper and Latch, 1999, NZ Patent 233083); various perennial and hybrid ryegrasses with AR1 endophyte (Fletcher, 1999); and tall fescue cultivars with MaxQTM(Bouton et al., 2002, US Patent 6,111,170).

Ryegrass-endophyte toxicosis Perennial ryegrass infected with its common wild-type endophyte, grown for both forage and turf, frequently produces compounds of the lolitrem sub-group of indole diterpenes in concentrations in herbage sufficient to cause the serious animal disorder known as ryegrass staggers. Lolitrem B is considered the most abundant active substance and concentrations in excess of about 2 ppm of herbage dry matter may result in clinical symptoms of ryegrass staggers in grazing sheep, cattle, deer and horses.

The same ryegrass-endophyte associations also produce ergovaline and perhaps other ergot alkaloids which are believed to cause other symptoms in grazing sheep, cattle, deer and horses commonly associated with the ryegrass-endophyte toxicosis syndrome. These symptoms may include hyperthermia in warm humid conditions as WO 2004/106487 PCT/NZ2004/000116 evidenced by increased rectal temperatures and respiration rates and depressed basal prolactin levels.

These responses are likely to be elicited at ergovaline concentrations in ryegrass pastures above 0.5 ppm. Ergovaline is also believed to be responsible for the depressed growth rates associated with the toxicosis syndrome. Increased faecal moisture and faecal soiling in sheep is also associated with ryegrass-endophyte toxicosis but causes have not been ascribed to any particular toxins.

The ryegrass staggers symptoms and overall effect of lolitrems may be enhanced by the presence in herbage of other toxins such as ergovaline.

Both lolitrem B and ergovaline concentrations tend to be higher in leaf sheath and seed heads of perennial ryegrass than in the roots or leaf blade. They also undergo seasonal variation with peaks in summer to autumn.

Enhanced plant protection with reduced toxicosis Endophytes confer degrees of protection to host plants against biotic and abiotic stress. Some endophyte-derived alkaloids are known to be toxic or deterrent to insect pests. Peramine is a feeding deterrent for and lolitrem is toxic to Argentine stem weevil, (Listronotus bonariensis) (Rowan et al., 1990; Prestidge and Gallagher 1985).

Ergovaline is deterrent to black beetle (Heteronychus arator) (Ball et al., 1997).

Where these alkaloids are absent or in very low concentration in plants, infestation by such pests become a problem. Hence it can be seen from the above discussion that it is desirable to have a ryegrass that has low mammalian toxicity but which also contains deterrent and/or insecticidal compounds to help avoid insect or other pest problems.

1- 0All references, including any patents or patent applications, cited in this specification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. The discussion of the reference states what their authors assert, and the applicants reserve the right to challenge the accuracy and pertiency of the cited documents. It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that 0 10 any of these documents forms parts of the common generally knowledge in the art, in Australia or in any other country.

0It is acknowledged that the term 'comprise' may, under varying jurisdictions, be attributed with either an exclusive or an inclusive meaning. For the purpose of this specification, and unless otherwise noted, the term 'comprise' shall have an inclusive meaning ie that it will be taken to mean an inclusion of not only the listed components it directly references, but also other non-specified components or elements. This rationale will also be used when the term 'comprised' or 'comprising' is used in relation to one or more steps in a method or process.

Further aspects and advantages of the present invention will become apparent from the ensuing description which is given by way of example only.

DISCLOSURE OF INVENTION According to one aspect of the present invention there is provided a method of increasing the resistance of a grass plant to biotic stresses selected from the group consisting of: root aphid (Aploneura lentisci); mealy bug (Balanoccus poae); black beetle (Heteronychus arator); porina (Wiseana cervinata); and combinations thereof, by the step of elevating the level ofjanthitrem epoxide compounds in the host grass to levels that confer the resistance by inoculating the grass with at least one endophyte from Neotyphodium lolii species characterised by having B10 allele size of approximately 160bp-161bp and a BI 1 allele size of approximately 132bp.

According to a further aspect of the invention there is provided the use ofjanthitrem epoxide compounds in grass plants to confer resistance to biotic stresses selected from the group consisting of: root aphid (Aploneura lentisci); mealy bug (Balanoccus poae); black beetle (Heteronychus arator); porina (Wiseana cervinata); and combinations 0 thereof, to the grass by inoculating the grass with at least one endophyte strain from 0 s Neotyphodium lolii species characterised by having a B 10 allele size of approximately S160-161bp and a B Il allele size of approximately 132bp.

C In an embodiment the endophyte is selected from the group consisting of: AR37; variations in N. lolii species are exemplified by AR37; variations in N. Lolii species are 10 exemplified by AR40; and combinations thereof; 00 In the endophyte or endophyte culture described above, the compound or compounds produced by the endophyte that confers protection to the host grass is an indole compound from the class ofjanthitrem epoxides.

S In the present invention, the endophytes described above do not provide the hitherto known toxic alkaloids lolitrem B and ergovaline at levels in excess of 2 ppm lolitrem B and 0.5 ppm ergovaline. Preferably the lolitrem B and ergovaline levels are below detection levels of less than 0.1 ppm of dry matter.

The endophytes described above do however, produce sufficient levels of other substances to protect the host grass from pests or abiotic stresses (such as water deficit) or both. In particular, the endophyte-infected ryegrass produces a group of indole diterpene derivatives from the class ofjanthitrem epoxide compounds not formerly observed and identified in endophyte-infected grasses. It is the understanding of the applicant that these compounds confer protection from pest predation upon the host grass plants and the grass-dominant pasture or turf as a whole without causing toxicosis of practical significance.

Preferably, the host grass is a perennial, annual or hybrid ryegrass. Most preferably, these are selected from the species: Lolium perenne; Lolium multiflorum; Lolium x hybridum.

Preferably, the toxicosis which is avoided is ryegrass-endophyte toxicosis. Most preferably the toxicosis is caused by an ergovaline toxin or a lolitrem toxin or a combination of ergovaline and lolitrem toxins.

Preferably the endophyte culture, if used, is an axenic culture.

Preferably, the endophyte culture, if used, has the same characteristics with respect to 0 taxonomic classification, plant infectivity, alkaloid production, animal performance, and plant protection properties as the endophyte itself.

CU

In an embodiment, infection is achieved by the breeding, crossing, hybridisation, C selection, or genetic modification of grass containing the endophyte or endophyte culture.

In an embodiment the grass has enhanced root growth and more tillers in comparison to a host grass without endophyte infection.

In an embodiment the grass is a Pooideae grass.

In an embodiment the combination of an endophyte as described above and a Pooideae grass features selected from the group consisting of: enhancement of pest protection; resistance to insects; pasture persistence; and combinations thereof.

In an embodiment the combination of an endophyte as described above and a Pooideae grass has the features of enhancement of grazing animal growth, and increased animal productivity in comparison with grass infected with know endophytes capable of inducing the disorder known as ryegrass-endophyte toxicosis.

The invention consists of the foregoing and also envisages constructions of which the following are examples.

BEST MODES FOR CARRYING OUT THE INVENTION Culture conditions and description The endophytes of this invention are strains from collections of seed of perennial ryegrass originally sourced from France. Seed from many various ryegrass collections from any countries were examined for the presence of endophyte by seed squash technique. A few plants for each seed sample, where endophyte was shown to be present, were grown for a few weeks in glasshouse conditions and re-tested for endophyte presence in their leaf sheaths.

The endophytes from plants with chemotypes of interest, primarily those not producing lolitrem B or ergovaline were isolated and grown in culture according to the method of O Latch and Christensen (1985). The endophytes of this invention are held in seek stocks, o 5 a culture collection, or in cloned plants at the AgResearch Ltd site in Palmerston North, SNew Zealand. The cultures are also deposited at the Australian Government Analytical laboratories at Sydney, Australia.

All strains of endophyte of this invention can be accommodated within a single sub- S 10o grouping of the species Neotyphodium Lolii. The isolates when grown on potato 00 dextrose agar at 22 0 C are typically slow growing (radial growth approximately 0.1 0.3 mm per day) with colonies typically white and cottony, becoming fawn with age.

Conidia have not be observed.

Inoculations Axenic cultures of endophyte AR37 as an example of this invention were successfully inoculated (Latch and Christensen, 1985) into seedlings grown from surfaced sterilized seed of perennial ryegrass cultivars Lolium perenne, for example Grasslands Nui and various experimental lines, generally with a satisfactory success rate usually in excess of 5% of attempts. Similarly annual ryegrasses Lolium multiflorum, for example Grasslands Moata, and Corvette, and hybrid ryegrasses Lolium x hybridum have been successfully inoculated for further examination with the chemotype characteristics of the combinations substantially the same as for perennial ryegrasses.

Chemotype identification Basal parts of endophyte-infected tillers were freeze dried, sometimes milled, and extracted and analysed qualitatively for the presence or absence of peramine, lolitrems and ergovaline by high performance liquid chromatography (HPLC) using minor modifications of the methods of Barker et al., (1993) and Spiering et al., 2002. Some endophytes from such selections lacking both lolitrems and ergovaline were isolated, classified by culture attributes, and generally re-inoculated into seedlings of endophytefree perennial ryegrass, cultivar Grasslands Nui, as a typical improved pasture host for comparative purposes. Samples from such plants at various stages of growth were analysed in more detail for alkaloid production. Following seed multiplication two groups of endophyte-grass combinations (with and without peramine in excess of ppm) were tested in field plot trials to further determine their general agronomic qualities, persistence, and practical resistance to insect predation. Some endophytes, O not of this invention, produce peramine but not lolitrems not ergovaline and are the o 5 subject of USA Patent 6,072,107.

CU

SThe endophytes of this invention are of a class that does not produce lolitrem B (or C other closely related lolitrems of similar chromatographic and fluorescence properties) or ergovaline at detection levels of 0.1 ppm of herbage dry matter. Neither do they to normally produce peramine at a detection level of 1 ppm of herbage dry matter.

00 Identification of new alkaloids SThe endophytes of this invention produce indole diterpenes not seen before from any C 15 grass infected with endophytes. Typically 50 mg portions of ground freeze dried herbage of plants infected with these endophytes were extracted for 1 hour with I ml of dichloroethane-methanol 9:1 by volume, and the extract collected by centrifugation or filtration. The extracts were examined for the presence or absence of lolitrems by normal phase HPLC, for example with Alltima silica 150 x 4.6 mm columns (Alltech Associates, Deerfield, II) and dichloromethane-acetonitrile, 7:1 by volume at 1 ml/min using fluorescence detection (excitation 265 nm, emission 440 nm). Two fluorescent peaks were observed with the endophytes of this invention that are not characteristic of the N. lolii endophytes normally producing lolitrems. One of the peaks was less retained than lolitrem B while another peak was more retained. The same general pattern peaks was observed for extracts of herbage containing endophytes AR37 and Extracts were also analysed by reverse phase HPLC, typically with a Prodigy 150 x 4.6 mm column (Phenomenex, Torrance, CA, USA) and with a solvent mixture of typically 5.6:1 acetonitrile:aqueous ammonium acetate buffer (0.005 M) adjusted to pH 6 with acetic acid. The solvent flow rate was 1 ml/min, and eluted peaks were detected by fluorescence (excitation 265 nm, emission 440 nm or excitation 333 nm, emission 385). The order of elution was reversed and resolution enhanced in comparison to the above normal phase separation. The fluorescent peaks identified here as components I, II, III and IV had retention times 7.7, 21.5, 24.2 and 25.1 min respectively for the above typical separation conditions. The normal phase peak B corresponded to reverse phase -8- O component I while the normal phase peak A resolved into three components II, III and o s IV. The chemical identity of these components was further investigated.

SUV and fluorescence spectra of components I, II, III and IV were obtained by reverse C phase HPLC using diode array and fluorescence stopped-flow techniques (Shimadzu SPD-MIOA and RF-10A detectors) with spectral maxima as in Table 1. These data 10 compare substantially to the spectra reported for the indole diterpene class of 00 janthitrems (Gallagher, 1980; de Jesus et al., 1984) or related shearinines (Belofsky c1- 1995).

Table 1: UV absorption and fluorescence spectral peaks UV Fluorescence Component X Max nm X Em Max nm (XEx 260 nm) I 259,333 381 II 259,333 383 III 259,333 387 IV 259,333 384 HPLC with mass spectrometry (LC-MS) was performed using reverse phase chromatography with electrospray ionisation (ESI) (Shimadzu QP-8000a detector) and with variations of scan range and deflector voltage to induce and explore ion fragmentation. Table 2 lists the m/z of the indicated MH ions together with major fragment ions. The loss of a fragment of mass 58 (assigned here as a loss of Me 2

CO)

has been reported for El MS ofjanthitrem C (Penn at el., 1993) and shearinine B (Belofsky, 1995).

Table 2: Mass spectral peaks from ESI LC-MS Component ESI mass spectral peak attributions MH MH MH MH MH m/z -H 2 0 -Me -Me 2 CO -CsH 9 I 646.5 628.4 588.3 II 670.5 655.1 612.4 600.95 III 672.5 614.6 IV 714.5 656.3 The further isolation and characterisation of component I was achieved by extracting 715g of perennial ryegrass seed infected with endophyte AR37 with 3 litres of dichloromethane (DCM) at ambient temperature with stirring for 1.5 2 hr followed by a further 2 litres of DMC similarly treated. The combined extract was concentrated under reduced pressure and redissolved in hexane for a cycle of flash chromatography (Merck Silica Gel 60 0.040-0.063 mm, 170g, 85 mm with elution in 500 ml is volume steps of hexane:DCM, DCM, DMC:acetonitrile (in proportions 19:1, 9:1, 4:1 and 1:1) and acetonitrile (MeCN). The fraction eluting with DCM:MeCN was enriched with I and was evaporated to dryness (0.04g), redissolved in a small volume of DCM:MeCN and coated on to C-18 silica gel (2 this was put on top of a reverse phase silica gel flash column (Alltech octadecyl coated, 32 g, 28 mm and fractions were eluted with 70 ml volumes of MeCN:H 2 0 in steps 7:3, 4:1, 4:1, MeCN, and DCM. The second MeCN:H 2 0 4:1 fractions enriched in I was concentrated and used in two portions for flash chromatography on amino-coated silica (Analytichem Sepralyte Primary Secondary Amine, 2.1 g, 11 mm Fractions were eluted with 5 ml volumes of MeCN:H 2 0 fractions were concentrated to reduce volume, absorbed on a C-18 SPE column (2 g, 11 mm eluted with MeCN and concentrated for examination by high resolution mass spectrometry and 'H and 3

C

NMR.

The high resolution mass spectrum obtained on a VG 70-250S mass spectrometer with a DCI probe yielded characterising ions with m/z 645.3647 (M calculated for

C

39

H

5 IN0 7 645.3665) and m/z 630.3451 (calculated for C 38

H

48 N0 7 630.3431).

00 0 0 Samples of I were examined in nuclear magnetic resonance (NMR) experiments to support a proposed structure of I which is also consistent with the high resolution masses.

NMR spectra were recorded in deuterioacetone ((CD 3 2 CO) solvent on a Bruker AC400 spectrometer. Chemical shifts are reported relative to TMS. The experiments included 00 10 one-dimensional "C (100.62 MHz) and 'H (400.13 MHz) spectra together with shortrange and long range proton-proton (COSY) and proton-carbon correlation coupling S(HMBC and HMQC) spectra. Signals were assigned by comparison with published NMR data for janthitrems (de Jesus et al., 1995), supported by the correlation data.

The proposed structure may be considered an epoxide of the known janthitrem G (de Jesus et al., 1984) and hence trivially named as 11,12-epoxy-janthitrem G (Figure 1).

The structure and numbering system for I is: 37 H 22 21 0 19 18 16 14 38 I 3 13 11 29 31 2 3 12 0 \\OAc 0 2 N H 4 -11- 11 The supporting chemical shift data is in Table 3.

Table 3: NMIR chemical shifts Atom 1 3 c 'H {Atom 1 3 c 'H I 154.3 51.1 42.6 26.4 28.2 71.8 76.0 68.2 61.8 70.8 77.4 29.7 21.0 50.3 27.2 116.3 127.2 113.8 136.4 32.0 1.62, 2.60 1.80, 2.21 4.17 3.45 5.14 3.52 1.58 1.50, 1.90 2.70 2.31, 2.60 7.13 2.63, 3.06 23 24 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Acetate Me Acetate CO 49.3 74.2 72.4 119.1 140.8 133.0 103.5 140.4 16.0 18.3 69.9 26.1 26.3 22.0 30.1 31.8 29.9 20.8 170.0 2.82 5.90 7.36 1.35 1.16 1.13 1.12 1.05 1.25 1.32 1.27 2.09 12- O By comparison and analysis of the UV, fluorescence and mass spectra we propose structures for II IV: II: The 10-deacetyl 10,34-(3-methylbut-2-enyl acetal) derivative of I.

III: The 10-deacetyl-34-0-(3-methylbut-2-enyl) derivative of I.

00 IV: The 34-0-(3-methylbut-2-enyl) derivative of I.

1- Genotype characterisation of endophyte All endophytes of this invention so far tested are characterised by DNA "fingerprinting" (selected polymorphic microsatellite loci and/or AFLP techniques) as belonging to a sub-group of Neotyphodium Lolii.

Samples of about 50 mg fresh or 15 mg dry basal tiller were used for the extraction of DNA using FastDNA kit for plants (Bio 101, Vista, California) using procedures recommended with the kit. Alternatively genomic DNA was extracted from cultured endophyte (Moon et al., 1999). Microsatellite PCR amplification was performed using primer pairs labeled with fluorescent dyes, B10.1 (5'-TET) B10.2 and B 11.1 BI 1.4, as described by Moon et al., (1999). The apparent size of microsatellite PCR fluorescent-labelled products was measured relatively to with an estimated 0.3 nucleotide units by capillary electropheresis using an ABI 3100 Genetic Analyzer with POP6 polymer chemistry in 50 cm capillary arrays and Gene-Scan-400HD standards (Applied Biosystems Inc., Foster City, CA).

The apparent sizes of PCR products by this technique (adjusted by subtracting a unti where an adenine nucleotide appears to have been terminally added) are in Table 4 and show that the endophytes of this invention may be distinguished from other groups of N. Lolii endophytes of the appreant sizes of alleles. Thus the strains of this invention may be characterised by B10 allele of apparent size about 160.6 and a BI 1 allele of apparent size about 132.0. Other strains of N. Lolii and some of Epichlod festucae have been shown to generally have a single B10 allele with apparent size about 175.6 and a single variable apparent sized B 11 allele although the size 132.0 was not observed in any endophyte outside the endophytes of this invention. A single allele for each locus is typical ofN. Lolii and Epichlod festucae.

Table 4. Apparent size of B10 and Bl 1 microsatellite PCR products Source material B10 allele size B 11 allele size N. lolii strain Lpl9 175.7 180.3 N. lolii strain Lp7 175.6 188.3 AR29 lolii strain from Grasslands Nui ryegrass) 175.7 176.2 (a strain lacking lolitrem B) 175.6 240.7 ARI (a strain lacking both lolitrem B and 175.7 147.8 ergovaline) FI1 (Epichlo festucae from Festuca longifolia) 175.6 115.6 AR37 160.6 132.0 160.7 132.0 The finding of single sizes of alleles (B10 c. 160.6 and BI 1 c. 132.0) for endophytes of this invention does not preclude a possibility that closely related 0o endophytes with the same functional properties might have different alleles.

Analysis by AFLP (Griffiths et al., 1999) also confirmed that endophyte examples AR37 and AR40 of this invention are from a sub-group that can be distinguished from other N. lolii endophytes outside this sub-group by one or more polymorphic differences from within more than 200 AFLP bands observed to be polymorphic for the genus Neotyphodium.

Endophyte and growth of pasture The growth of the cultivar Grasslands Nui infected with AR37 and wild-type, and endophyte-free was assessed in a series of field trials, both grazed and mown, in four regions of New Zealand over a period of more than 3 years from 1996.

Plots infected with AR37 generally yielded more ryegrass herbage than wild-type plots.

In 11 trials sown in 1996 and 1997 annual yields measured from AR37 plots were on average 11% greater over 3 years. The greatest differences occurred from late summer through autumn.

For example, in Site 1, where conditions are favourable for good ryegrass growth (e.g.

wild-type yields 15000 kg DM/ha/year), AR37 plots yielded 6% more annual herbage (P<0.05) with the greatest yield advantages in the autumn (Table At another site, Site 2, less favourable for ryegrass growth and persistence wild-type yields 8700 kg DM/ha/year), AR37 plots had higher yields in all seasons and significantly so for 3 seasons and for the total annual yields (Table Table 5. Ryegrass yields of Grasslands Nui infected with AR37 relative to yields of Nui with wild-type endophyte (=100) for field plots at tow contrasting locations. Average of yields for 3 years for trials sown in autumn 1996.

Site Winter Spring Summer Autumn Annual Site 1 108 100 107 120* 106* Site 2 113 114* 117 123 116* Indicates value at the site is significantly different to wild-type 0.05) At Site 2, another trial shown in 1998 with Nui ryegrass and a ryegrass selection known as 'GA66' resulted in higher annual yields for AR37 plots for both ryegrasses and (P<0.05) compared with wild-type plots.

Differences in number of tillers were apparent from mid-summer to early winter, being from 22% to 64% greater for AR37 compared with wild-type (P<0.05) (Table 6).

Table 6. Grasslands Nui ryegrass tiller numbers in autumn (per metre row at Site 3, per m 2 at Site 1) Site AR37 Wild-type Endophyte-free Site 3, Area 1 1340 a 1100 b 1120 Site 3, Area 2 1680 a 1300 1030 b Site 1 7200 a 4400 4100b For each site numbers without a letter in common are significantly different (P<0.05).

Total root organic matter was examined in a trial at Site 3 after Grasslands Nui rows were occasionally mown to simulate rotational grazing. Cores, 25 mm diameter by 300 mm soil depth were assessed and the grass infected with AR37 shown to have significantly more root mass that either endophyte-free or wild-type infected grass (-i (Table 7).

C

Table 7. Root mass (grams organic dry matter per core) (-i AR37 Wild-type Endophyte-free 00 1.42 Total root organic matter 2.05 a 1.39 b 42b 1o Numbers without a letter in common are significantly different (P<0.05) Thus it was shown that infections of perennial ryegrass cultivars with AR37 results in generally superior pasture growth and potential pasture productivity especially in late summer and autumn.

Endophyte and growth of turf Perennial ryegrass is frequently used as a main component of utility turf for aesthetic and recreational purposes. An observation that Grasslands Nui cultivar infected with AR37 had persistence and green colour compared to other endophyte infections of Grasslands Nui during a dry summer season in a further site, Site 4, stimulated a small plot trial comparison of Grasslands Nui infected with either its own natural high level of wild-type endophyte or artificially infected with AR37. Trials were conducted as Site 4 and at Site 1.

The plots were managed to simulate turf growth conditions and typical turf management with regular mowing to 2 cm height when the height had grown to an estimated 3 cm. Fertiliser was applied at 30 units of nitrogen per month generally when raining and discontinued during drought periods. Water was applied only to avoid plant death from desiccation.

The measurements made included tiller density, grass production (mowing), observations on disease and pests, soil moisture and bulk density, root mass and top r- 0mass (under the mower height) and plant morphology measurements including leaf and sheath dimensions.

CU

Although there was little difference in yield above mower height there was differences in grass mass below mower height, particularly at Site 1 where the AR37 plot was about Sdouble the wild-type treatment (P<0.001).

00 i The tiller density per unit of area at both Site 1 and Site 4 was significantly greater for j AR37 plots (P<0.005). Similarly root mass was consistently higher with AR37 plots by 0about 25% or more (P<0.02) at both sites. Leaf (P<0.03) and sheath (P<0.02) widths, measured at the base of each part, were consistently less for AR37 plots measured just at Site 4. The mean tiller dry matter for AR37 was approximately 40% less than for wild-types (P<0.014) at Site 4 however the mean number of leaves per tiller was very nearly three for both endophyte plots and not significantly different.

Thus it was shown that infection with AR37 of Grasslands Nui results in a denser sward of smaller tillers when managed as a turf. These swards have increased root mass and herbage below cutting height compared to wild-type endophyte. These characteristics have high utility for improving the ground cover and lateral shear strength of turf systems.

Endophyte and pest protection The endophytes of this invention provide their host perennial ryegrass with resistance to a range of insect pests including Argentine stem weevil, black beetle, mealy bug and root aphid. In a combination of field and pot trials the degree of protection provided by the AR37 endophyte when compared with endophyte-free ryegrass is equivalent to that provided by the naturally occurring wild-type endophyte for all these pests except root aphid against which the wild-type endophyte provides little or no protection (Table 8).

For Argentine stem weevil (Listronotus bonariensis) the mode of resistance afforded by endophyte differs between AR37 and the wild-type. In AR37 adult feeding and oviposition are the same as in endophyte-free plants whereas in the wild-type defence against the weevil is mediated primarily via deterrence of the adult from feeding and oviposition by the alkaloid peramine. Observations indicate that AR37 reduces larval damage ti tillers because it is toxic to larvae. AR37 has been tested against Argentine stem weevil extensively in field and in pot trials and has consistently reduced damage by this pest to low levels where compared to damage in endophyte-free ryegrass.

AR37 also reduces black beetle (Heteronychus arator) damage by larvae in the field, mainly through deterrence of the adult. Adult black beetle damage to ryegrass tillers infected with AR37 was 17.3% whereas 46% of endophyte-free tillers were damged.

Survival of root aphid (Aploneura lentisci), mealy bug (Balanococcus poae) and porina (Wiseana cervinata) are also less on ryegrass with AR37 than on endophyte-free ryegrass.

Table 8. Examples of the effect of AR37 on different insect pests Insect Parameter AR37 Wild-type Endophyte-free Argentine stem Tillers with 13a 17a 36 b weevil farval damage Black beetle No. larvae/m 2 13.8a 13.8a 6 0 .0b Root aphid Log 0.27a 1.61 b 2.13 b (n=l)/plant Mealy bug No./10 cores 0.3a 0.6a 16.8 b Porina survival 50.2a 60.0 ab 8 9 For each insect, numbers without a letter in common are significantly different (P<0.05) Endophyte and animal performance Sheep grazing ryegrass cultvars with their wild-type endophyte in summer and autumn may exhibit one or all of the symptoms of ryegrass-endophyte toxicosis. These include reduced live weight gain, ryegrass staggers, increased rectal temperatures and respiration rates, especially in warm humid conditions, increased incidence of faecal soiling (dags) and fly strike and reduced basal prolactin levels. Using these parameters, the health and production responses of sheep grazing the same ryegrass cultivar without endophyte, with its wild-type endophyte or with AR37 endophyte in summer and autumn over 3 years were compared (Table 9).

Table 9. Mean responses (3 years) of sheep grazing ryegrass with AR37 compared to same ryegrass without endophyte or with its wild-type endophyte Endophyte- Wild- AR37 free type Live weight change (g/day) 62 -12 47 Ryegrass staggers (0-5 ascending scale) 0 2.7 1.8 Rectal Temperature (oC) 40.4 40.7 40.5 Respiration rate (breaths/minute) 85 109 Plasma prolactin (ng/ml) 208 110 210 The sheep grazing endophyte-free ryegrass exhibited none of the adverse responses typically associated with ryegrass-endophyte toxicosis. Those grazing ryegrass with AR37 had mild ryegrass staggers but the incidence and severity was significantly less than for those sheep grazing ryegrass with its wild-type endophyte. Mean live weight change was slightly lower than those grazing endophyte-free but significantly better than the negative growth rates of those grazing ryegrass with wild-type endophyte. For 0all the other parameters (rectal temperature, respiration rate and plasma prolactin levels) o 5 measured there was no significant difference between sheep grazing endophyte-free Sryegrass and those grazing ryegrass with AR37. However respiration rates and rectal temperatures were significantly higher for sheep grazing ryegrass with its wild-type endophyte than for those grazing AR37, while plasma prolactin levels were significantly lower for ryegrass with wild-type endophytes.

00 M In another replicated trial there was no evidence of ryegrass staggers in sheep grazing O endophyte-free ryegrass cultivars with AR37 whereas on the same ryegrass cultivars with wild-type endophyte the sheep had serious ryegrass staggers. Mean live weight gains in sheep grazing AR37 treatments were 130 g/day whereas those grazing the same ryegrass with its wild-type endophyte grew at only 90 g/day.

In a larger on-farm grazing trial where the ryegrass was sown with clover, responses were similar in sheep grazing AR37 treatments to those on endophyte-free treatments with no ryegrass staggers on AR37 treatments.

Aspects of the present invention have been described by way of example only and it should be appreciated that modifications and additions may be made thereto without departing from the scope thereof as defined in the appended claims.

WO 2004/106487 PCTNZ2004OOO116 0

REFERENCES:

SBall, Miles, Prestidge, R.A. 1997: Ergopeptine alkaloids and Neotyphodium lolii-mediated resistance in perennial ryegrass against Heteronychus Sarator (Coleoptera: Scarabaeidae). Journal of Economic Entomology 90: 1383-1391.

00 Barker, Davies, Lane, GA; Latch, Nott, Tapper, BA. 1993: g Effect of water deficit on alkaloid concentrations in perennial ryegrass endophyte Sassociations. In Proceedings of the Second International Symposium on Acremonium/Grass Interactions. Eds. Hume, Latch, Easton, H.S.

AgResearch, New Zealand, pp. 67-71.

o1 Belofsky, G. Gloer, Wicklow, Dowd, P.D. 1995: Antlinsectan alkaloids: shearinines A-C and a new paxilline derivative from the ascostromata of Eupenicillium shearii. Tetrahedron, 51: 14, 3959-3968.

Blank, C.A; Gwinn, K.D. 1992: Soilbome seedlings diseases of tall fescue: influence of the endophyte Acremonium coenophialum. Phytopathology 82: 1089.

Bouton, Latch, Hill, Hoveland, McCann, Watson, R.H.; Parish, Hawkins, Thompson, F.N. 2002: Re-infection of tall fescue cultivars with non-ergot alkaloid-producing endophytes. Agronomy Journal 94: 567-574.

de Jesus, Steyn, van Heerden, Vieggaar, R. 1984: Structure elucidation of the janthitrems, novel tremorgenic mycotoxins from Penicillium janthlnellum. Journal of the Chemical Society, Perkin Transactions 4, 697-701.

Elberson, West, C.P. 1996: Growth and water relations of field grown tall fescue as influenced by drought and endophyte. Grass and Forage Science 51: 333- 342.

-21- WO 2004/106487 PCT/NZ2004/0001 16 O Fletcher, L.R. 1999: "Non-toxic" endophytes in ryegrass and their effect on livestock U health and production. In Ryegrass endophyte: an essential New Zealand symbiosis.

Grassland Research and Practice Series No. 7, pp 133-139.

Fletcher, Easton, H.S. 2000: Using Endophytes for Pasture Improvement in New S 5 Zealand. In Proceedings of The Grassland Conference 2000, 4th International r- Neotyphodium/Grass Interactions Symposium. Eds. Paul, Dapprich, P.D.

UniverstAt, Paderborn, pp 149-162.

cN Fletcher, Sutherland, Fletcher, C.G. 1999: The impact of endophyte on the health and productivity of sheep grazing ryegrass-based pastures. In Ryegrass o0 endophyte: an essential New Zealand symbiosis. Grassland Research and Practice Series No. 7, pp 11-17.

Gallagher, Latch, Keogh, R.G. 1980: The janthitrems: fluorescent tremorgenic toxins produced by Penicillium janthinellum isolates from ryegrass pastures. Applied and Environmental Microbiology, 39: 1,272-273.

Griffiths, Moon, Tapper, Christensen, M. 1999: Non-radioactive

AFLP

fingerprinting for detection of genetic variation in Epichlo/Neotyphodium endophytes.

Proceedings of the 11th Australian Plant Breeding Conference.

Latch, Christensen, M.J. 1985: Artificial infection of grasses with endophytes.

Annals of Applied Biology 107:17-24.

Leuchtmann, A. 1997: Ecological diversity in Neotyphodium-infected grasses as influenced by host and fungus characteristics. In Neotyphodium/Grass Interactions, Eds. Bacon, Hill, N.S. Plenum Press, New York, pp 93-108.

Moon, Tapper, Scott, D.B. 1999: Identification of Epichloe endophytes in planta by a microsatellite-based PCR fingerprinting assay with automated analysis.

-22- WO 2004/106487 PCT/NZ2004/000116 C Applied and Environmental Microbiology 65: 1268-1279.

U

SPenn, Swift, Wigley, Mantle, Bilton, Sheppard, R. N. 1993: N Janthitrems B and C, two principal indole-diterpenoids produced by Penicillium janthinellum. Phytochemistry, 32: 6,1431-1434.

00 r- 5 Prestidge, R. Gallagher, R. T. 1985: Lolitrem B a stem weevil toxin isolated from Acremonium-infected ryegrass. Proceedings 38th New Zealand weed and pest 0 control conference: 38-40.

Rowan, D. Dymock, J. Brimble, M. A. 1990: Effect of fungal metabolite peramine and analogs on feeding and development of Argentine stem weevil (Listronotus bonariensis). Journal of Chemical Ecology 16: 1683-1695.

Rowan, Hunt, Gaynor, D.L. 1986: Peramine, a novel insect feeding deterrent from ryegrass infected with the endophyte Acremonium loliae. Joumal of the Chemical Society. Chem. Commun. 1986. 935-936.

Rowan, Latch, G.C.M. 1994: Utilization of endophyte-infected perennial ryegrasses for increased insect resistance. In Biotechnology of endophyte fungi in grasses. Eds. Bacon, C.W. White, J. CRC Press, pp 169-183.

Siegel, Latch, Bush, Fannin, Rowan, Tapper, B.A.; Bacon, Johnson, M.C. 1990: Fungal endophyte-infected grasses: alkaloid accumulation and aphid response. Journal of Chemical Ecology 16: 3301-3315.

Spiering, Davies, Tapper, Schmid, Lane, G.A. 2002: Simplified extraction of ergovaline and peramine for analysis of tissue distribution in endophyteinfected grass tillers. Journal of Agricultural and Food Chemistry, 50: 5856-5862.

Stuedemann, Hoveland. 1988: Fescue endophyte: History and impact on animal agriculture. Journal of Production Agriculture 1: 39-44.

-23- WO 2004/106487 WO 204/16487PCTINZ2004/0001 16 Tapper, Latch, G.C.M. 1999: Selection against toxin production in endophyteinfected perennial ryegrass. In Ryegrass endophyte: an essential New Zealand symbiosis. Grassland Research and Practice Series No. 7, pp 107-111.

Wilkins, Miles, Ede Gallagher, Munday, S.C. 1992: Structure elucidation of janthitrem B, a tremorgenic metabolite of Penicillium janthinellum, and relative configuration of the A and B rings of janthitrems B, E, and F. Journal of Agricultural and Food Chemistry, 40: 8, 1307-1309.

-24-

Claims (19)

1. A method of increasing the resistance of a grass plant to biotic stresses Sselected from the group consisting of: root aphid (Aploneura lentisci); mealy C bug (Balanoccus poae); black beetle (Heteronychus arator); porina (Wiseana cervinata); and combinations thereof, by the step of elevating the 0to level ofjanthitrem epoxide compounds in the host grass to levels that confer 00 the resistance by inoculating the grass with at least one endophyte from SNeotyphodium lolii species characterised by having B 10 allele size of 4 approximately 160bp-161bp and a Bl 1 allele size of approximately 132bp.

2. The method as claimed in claim 1 wherein the endophyte used produces toxic alkaloids lolitrem B and ergovaline at levels of less that 2 ppm of dry matter lolitrem B and 0.5 ppm of dry matter ergovaline.

3. The method as claimed in claim 1 or claim 2 wherein the endophyte strain produces toxic alkaloids lolitrem B and ergovaline at detection levels of less than 0.1 ppm of dry matter.

4. The method as claimed in any one of the above claims wherein the level of janthitrem epoxide compounds is increased to resistance conferring levels by inoculating the grass with: AR37, AR40, and combinations thereof; AR37 and AR40 cultures deposited on 23 May 2003 at the Australian Government Analytical Laboratries (AGAL) accession number NMO3/35819 (AR37) and NMO3/35820

5. The method as claimed in any one of the above claims wherein the endophyte strain is in the form of an endophyte culture.

6. The method as claimed in any one of the above claims wherein infection is achieved by modifying the host grass using techniques selected from the group consisting of: breeding, crossing, hybridisation, selection, genetic modification, and combinations thereof.

7. The method as claimed in any one of the above claims wherein the grass has enhanced root growth and more tillers in comparison to grass without increased levels ofjanthitrem epoxide compounds. 1- O

8. The method as claimed in any one of the above claims wherein the grass N 5 produced by the method confers the features of enhancement of grazing Sanimal growth and increased animal productivity in comparison with grass Swithout janthitren epoxide compounds.

9. The method as claimed in any one of the above claims wherein the grass is a 10 perennial, annual or hybrid ryegrass. 00 1- J

10. The method claimed in any one of claims 1 to 9 wherein the grass is a N Pooideae grass. C 1 s

11. The method as claimed in any one of the above claims wherein the grass is selected from the species: Lolium perenne, Lolium Multiflorum; Lolium x hybridum.

12. The use ofjanthitrem epoxide compounds in grass plants to confer resistance to biotic stresses selected from the group consisting of: root aphid (Aploneura lentisci); mealy bug (Balanoccus poae); black beetle (Heteronychus arator); porina (Wiseana cervinata); and combinations thereof, to the grass by inoculating the grass with at least one endophyte straine from Neotyphodium lolii species characterised by having a B10 allele size of approximately 160-161bp and a B 11 allele size of approximately 132bp.

13. The use as claimed in claim 12 wherein the endophyte strain inoculated is selected from: AR37, AR40, and combinations thereof; AR37 and cultures deposited on 23 May 2003 at the Australian Government Analytical Laboratories (AGAL) accession number NMO3/35819 (AR37) and NMO3/35820

14. The use as claimed in claim 12 or claim 13 wherein the endophyte strain is in the form of an endophyte culture.

The use as claimed in any one of claims 12 to 14 wherein inoculation is achieved by modifying the host grass using techniques selected from the group consisting of: breeding, crossing, hybridisation, selection, genetic modification, and combinations thereof. -26- O

16. The use as claimed in any one of claims 12 to 15 wherein the grass has enhanced root growth and more tillers in comparison to grass without increased levels ofjanthitrem epoxide compounds. c

17. The use as claimed in any one of claims 12 to 16 wherein the grass is a perennial, annual or hybrid ryegrass. 00

18. The use as claimed in any one of claims 12 to 17 wherein the grass is a Pooideae grass.

19. The use as claimed in any one of claims 12 to 18 wherein the grass is slected C 1is from the species: Lolium perenne; Lolium multiflorum; Lolium x hybridum.