AU2017251752B2 - Novel organisms (3) - Google Patents

Abstract

The present invention relates to new methods of selecting and breeding organisms, in particular organisms which exhibit symbiotic behaviour with symbionts, and to new 5 organisms and symbiota developed thereby.

Description

Field of the Invention

The present invention relates to new methods of selecting and breeding organisms, in particular organisms which exhibit symbiotic behaviour with symbionts such as 0 fungal endophytes or epiphytes or bacterial microbiome in plants, and such as rumen microbiome in ruminant livestock, and to new organisms and symbiota developed thereby.

Background of the Invention

The phenotype of many species of livestock, crops and pastures depends on the interaction between the genotype of the individual and the genotype of a symbiont. Important plants, including forage grasses, legumes, trees, shrubs, and vines are commonly found in association with endophytes including fungi, bacteria, viruses and microbes. Similarly, important animals, including cattle, sheep, pigs, goats, etc. have 20 microbiomes present in their gut and rumen.

Both beneficial and detrimental horticultural, agronomic and veterinary properties result from such associations, including improved tolerance to water and nutrient stress and resistance to insect pests.

For example, ryegrass plants can show improved drought tolerance and persistency if a fungal endophyte of the correct genotype colonises the plant. Similarly, in grasses, insect resistance may be provided by specific metabolites produced by the endophyte, in particular loline alkaloids and peramine. Other metabolites produced 30 by the fungal endophyte, for example lolitrems and ergot alkaloids, may be toxic to

-32017251752 25 Oct 2017 grazing animals and reduce herbivore feeding.

Considerable variation is known to exist in the metabolite profile of symbionts. For example, fungal endophyte strains that lack either or both of the animal toxins have 5 been introduced into commercial ryegrass varieties.

Bacterial microbiomes also influence plant performance (e.g. in ryegrass, wheat).

In animals, the microorganisms present in the gut are responsible for digestion of an 0 animal’s feed. Ruminants host an extremely large and complex array of microrgansisms in their rumen, and it is this microbial community which allow them to convert low quality forage into high quality protein and lipids for human consumption in the form of meat and milk. During this process methane, a potent greenhouse gas is produced. Rumen microbes-bovine symbiota may be important, for example, in 5 improving feed conversion efficiency and reducing methane production. In ruminants, successful digestion of poor quality feed may depend on having a particular rumen microbiome profile. If it could be demonstrated that regions of the host genome are associated with differences in rumen microbial profile, then this could be exploited to breed cattle with lower methane emissions and improved feed conversion efficiency.

:0

Molecular genetic markers such as simple sequence repeat (SSR) markers have been developed as diagnostic tests to distinguish between symbiont taxa and detect genetic variation within taxa. For example, the markers may be used to discriminate symbiont strains with different toxin profiles.

However, there remains a need for methods of identifying, isolating and/or characterising organisms which exhibit symbiotic behaviour with symbionts. Difficulties in artificially breeding of these symbiota limit their usefulness. For example, many of the novel endophytes known to be beneficial to pasture-based 30 agriculture exhibit low inoculation frequencies and are less stable in elite germplasm.

Moreover, in traditional breeding techniques, for example in forage grasses such as

2017251752 25 Oct 2017

-4 perennial ryegrass and tall fescue, grass varieties are bred using classic crossbreeding techniques and grass genotypes are selected for their superior characteristics, after monitoring their performance over a period of multiple years. The selected grass genotypes that form the experimental variety are then inoculated 5 with a single endophyte and the resulting grass-endophyte associations are evaluated for any favourable characteristics such as insect resistance. The individual experimental synthetic varieties deploying a single endophyte in them are then evaluated for agronomic performance and resulting animal performance by grazing animals over a period of years. This evaluation process may reveal that the single 0 endophyte being deployed in the different experimental synthetic varieties may not show vegetative and/or intergenerational stability in some of these varieties or the desired alkaloid profile conferred by the single endophyte may vary between different synthetic varieties failing to confer appropriate levels of insect resistance or causing animal toxicoses. It would be a significant development in the art if this time5 consuming process could be accelerated or otherwise improved.

It is accordingly an object of the present invention to overcome, or at least alleviate, one or more of the difficulties or deficiencies associated with the prior art.

Summary of the Invention

Accordingly, in a first aspect of the present invention, there is provided a method for producing an improved organism including (i) providing a library of germplasm of an organism; and a library of symbionts;

(ii) inoculating the germplasm library with one or more symbionts selected from the symbiont library to generate symbiota;

(iii) breeding, selecting, screening and/or evaluating the inoculated germplasm or symbiota for desired symbiota characteristic(s); and (iv) subsequently identifying, culturing or otherwise using the symbiota exhibiting desired characteristic(s) to produce the improved organism.

-52017251752 25 Oct 2017

By ‘symbiotum’ (or plural symbiota) is meant a supra-organism representing an association of organism with symbiont. For example, symbiota may be inoculated plant germplasm.

By ‘library’ is meant a resource, such as a collection of symbionts.

Applicants have established that it is possible to deploy multiple symbionts in multiple organisms and to select for improved symbiotic compatibility and performance early in the breeding process. That is symbiont-organism genotype combinations are bred 0 and screened for desired characteristics including improved symbiota compatibility and performance ab initio. This may be contrasted with prior art techniques in which an organism host, e.g. a grass variety, would be bred and selected over a period of time after which symbiont inoculation would occur with a single symbiont late in the varietal development.

As used in this application, the term organism refers to a eukaryotic multicellular organism, including without limitation animals and plants. In particular embodiments, the animal may be a mammal or an avian. Of particular significance are mammalian animals of agricultural significance, such as cattle, sheep, pigs, goats and the like 0 that harbour microbiomes in the gut or rumen. In particular embodiments, the organism is a plant, including without limitation monocots and dicots. Specific types of plants include, without limitation perennial grasses, legumes, decorative or fruit bearing trees, vines, shrubs and bushes, herbs, and decorative or edible flowers.

In a preferred embodiment of this aspect of the present invention, the organism may be a plant or animal and the symbiont(s) may be capable of forming symbiotic associations with the plant or animal. Preferably, the organism is a plant establishing symbiotic associations with endophytes.

By ‘symbiont’ is meant a microorganism that is associated with a multicellular organism.

-62017251752 07 Oct 2019

By ‘associated with’ is meant that the symbiont lives on, in or in close proximity to the organism. For example, the symbiont may be a microorganism that lives within the body or cells of another organism or on or closely associated with the surface of the organism, for example a biofilm in close association with a plant. In the case of a 5 plant the symbiont may be enodophytic, for example living within the internal tissues of the plant, or it may be epiphytic, for example growing externally on the plant.

The symbiont may be selected from one or more of fungi, viruses, bacteria and other microbes. For example, it may be an endophyte, such as a fungal or bacterial 0 endophyte, an epiphyte, such as a fungal or bacterial epiphyte, a bacterial microbiome, or a mycorrhiza such as an arbuscular mycorrhiza.

Accordingly, in a preferred embodiment of the present invention, there is provided a method as described above wherein the organism is a plant, the method including:

(i) providing a library of plant germplasm; and a library of endophytes and/or epiphytes and/or plant-associated microbiomes;

(ii) inoculating the plant germplasm library with one or more endophytes and/or epiphytes and/or plant-associated microbiomes selected from the library to generate symbiota;

(iii) breeding, selecting, screening and/or evaluating the symbiota for desired symbiota characteristic(s); and (iv) identifying, culturing or otherwise using the symbiota exhibiting desired characteristic(s) to produce the improved plant.

In a preferred embodiment, the method may be used to develop improved symbiota varieties exhibiting desired characteristic(s).

In a preferred embodiment, the present invention provides a method for producing an 30 improved plant, said method including:

(i) providing a library of germplasm of an organism; and a library of symbionts;

-6a 2017251752 07 Oct 2019 (ii) inoculating the plant germplasm library with one or more symbionts selected from the symbiont library to generate symbiota;

(iii) selecting, screening and/or evaluating the inoculated plant germplasm library or the symbiota for desired symbiota characteristic(s); and (iv) subsequently identifying, culturing or otherwise using the symbiota exhibiting desired characteristic(s) to produce the improved plant, wherein:

said symbiont(s) are endophytes and/or epiphytes and/or plant-associated microbiomes;

said library of germplasm includes 100s to 1000s of genotypes;

said library of symbionts includes 10s to 100s of genotypes; and said step (iii) includes a step of genetic analysis and/or metabolic analysis; and said plant germplasm is present as an embryo; and said inoculation step includes coating said embryo with a symbiont-containing coating layer to form an artificial seed.

In a further aspect of the present invention there is provided a method of genomic selection of organism-symbiont associations, said method including (i) providing a library of nucleic acid samples from said organism or said organism-symbiont association;

-72017251752 25 Oct 2017 (ii) analysing said samples using metagenomics to obtain a genetic profile for each sample;

(iii) selecting organisms or organism-symbiont associations with a desired genetic and/or metabolic profile.

In a preferred embodiment, the method may include the preliminary steps of:

(i) providing an organism-symbiont association;

(ii) subjecting said organism-symbiont association to one or more environmental conditions; and (iii) preparing a library of nucleic acid samples from each of the environmentally treated organism-symbiont associations.

In a preferred embodiment of this aspect of the present invention, the organism may be a plant or animal and the symbiont may be a bacterial microbiome.

In a preferred embodiment, the method may be used to develop improved symbiota varieties exhibiting desired characteristic(s).

The plant may be a grass, preferably a perennial grass, legume, vine, shrub, tree, 0 herb, flower, shrub or bush. The method according to this aspect of the present invention is particularly applicable to grasses and legumes.

The endophyte may be a fungal or bacterial endophyte. The epiphyte may be a fungal or bacterial epiphyte. In a preferred embodiment, the library may be a 25 bacterial microbiome.

In a preferred embodiment, the symbiont may be a fungal or bacterial endophyte, a fungal or bacterial epiphyte, or a bacterial microbiome.

By ‘analysing said samples by metagenomics’ is meant analysing genetic material recovered from the organism or organism-symbiont association, preferably directly from the organism or organism-symbiont association, so that largely unbiased

-82017251752 25 Oct 2017 samples of a significant proportion, preferably substantially all, genes from a significant proportion, preferably substantially all, of the members of the sampled association are analysed.

This analysis may involve detecting the presence or absence of polymorphic markers, such as simple sequence repeats (SSRs) or mating-type gene markers.

Alternatively, or in addition, this analysis may involve sequencing genomic and/or mitochondrial DNA and/or ribosomal RNA and performing sequence comparisons to 0 databases of known nucleic acid sequences, for example 16S rRNA sequences which are indicative of bacterial genes.

The one or more environmental conditions to which said organism-symbiont association may be subjected include, but are not limited to, different nutritional 5 conditions, such as nutrient stresses, eg. low nitrogen or phosphorus; and different light, temperature or water conditions, such as low light, cold stress or water stress, respectively.

In a preferred embodiment of the present invention, the symbiont(s) may include a 0 genetic variation, to enhance stability of the symbiotum, abiotic stress tolerance (e.g.

water stress) of the symbiotum, biotic stress tolerance (e.g. disease resistance) of the symbiotum, nutrient use efficiency (e.g. phosphorus use efficiency, nitrogen use efficiency) of the symbiotum; and for example, for symbiont trait introgression in animals such as ruminant livestock species to enhance feed conversion efficiency of 25 the symbiotum (i.e. ruminant livestock organism with its rumen microbiome) or to mitigate methanogenesis from the symbiotum.

The genetic variation may be introduced utilizing any standard techniques, e.g. via one or more of random mutagenesis, di/poly-ploidisation, targeted mutagenesis;

cisgenesis; transgenesis; intragenesis.

-92017251752 25 Oct 2017

The genetic analysis may be conducted as described above. The seedlings may for example be screened for bacterial 16S rRNA sequences.

The symbiota may be of any suitable form, including inoculated embryos, plant 5 seeds, germinating seeds, seedlings, plantlets, plants, etc.

In a preferred form the nucleic acid samples may be extracted from the leaves of seedlings, more preferably from the epicotyl, hypocotyl or similar embryonic shoot of the seedlings. In grasses, the DNA/RNA may be extracted from tillers. In another 0 preferred form the nucleic acid samples may be extracted from root samples.

Alternatively, the nucleic acid samples may be extracted from whole germinating seeds or seedlings.

Preferably the nucleic acid samples are RNA samples, which may then be used to 5 construct cDNA libraries.

The method according to this aspect of the present invention may further include subjecting the selected symbiota populations to phenotyping for assessment of symbiota performance and/or maintenance of desired characteristics; and selecting 0 symbiota for generating a synthetic symbiota variety, for example by polycrossing.

For example, the selected symbiota variety may be subjected to an symbiont identification assay, followed by polycrossing to generate a next generation seed. Optionally, the above steps may be repeated to confirm symbiota stability, desired 25 characteristics, symbiont identity and/or symbiont incidence in the next generation, for example the next seed generation.

Accordingly, in a further aspect of the present invention, there is provided improved symbiota including one or more organisms containing one or more symbionts 30 produced utilising the method described above.

-10 2017251752 25 Oct 2017

Accordingly, in a further aspect of the present invention, there is provided an improved organism exhibiting symbiota with an symbiont and produced utilising the method described above.

The improved organism may be a plant or animal.

Where the organism is a plant, the plant may be a grass, tree, flower, herb, shrub or bush, vine or legume, or a product thereof.

The plant material may be in the form of a seed, seedling, embryo or the like.

The method steps described above may be repeated to develop later generations of symbiota seeds, plants or animals.

In a further aspect, the present invention provides a plant, plant seed or other plant part derived from a seed or plant of the present invention and stably infected with and symbiont.

Preferably, the plant cell, plant, plant seed or other plant part is a grass, more 0 preferably a forage, turf or bioenergy grass, such as those of the genera Lolium and Festuca, including L. perenne and L. arundinaceum, and of the genera Brachiaria and Urochloa, including B. brizantha, B. decumbens, B. humidicola and U. mosambicensis.

By ‘plant cell’ is meant any self-propagating cell bounded by a semi-permeable membrane and containing plastid. Such a cell also required a cell wall if further propagation is desired. Plant cell, as used herein includes, without limitation, seeds suspension cultures, embryos, meristematic regions, callus tissue, leaves, roots, shoots, gametophytes, sporophytes, pollen and microspores.

A large scale endophyte discovery program has been undertaken to establish the library of fungal endophyte strains. A collection of perennial ryegrass and tall fescue

-11 2017251752 25 Oct 2017 accessions has been established.

The endophytes selected to inoculate the plant germplasm may be selected utilising the techniques described in an Australian provisional patent application filed 1 June 5 2012 entitled “Novel Endophytes”, the entire disclosure of which is incorporated herein by reference. The novel endophytes described therein are particularly preferred.

Genetic analysis of endophytes in these accessions has led to the identification of a 0 number of novel endophyte strains. These novel endophyte strains are genetically distinct from known endophyte strains. Metabolic profiling may be undertaken to determine the toxin profile of these strains.

Specific detection of endophytes in planta with SSR markers may be used to confirm 5 the presence and identity of endophyte strains artificially inoculated into, for example, grass plants, varieties and cultivars.

In the screening step (iii) according to the method of this aspect of the present invention, the inoculated germplasm may be screed by genetic analysis and/or 0 metabolic profiling. For example, techniques of genetic analysis described in the Australian provisional patent application entitled “Novel Endophytes” may be used.

Alternatively, or in addition, the inoculated germplasm may be subjected to genetic analysis (genetically characterised) to demonstrate genetic distinction from known 25 symbiont-genotype symbiota and to confirm the identity of symbiont strains artificially inoculated into, for example, grass plants, varieties and cultivars.

By ‘genetic analysis’ is meant analysing the nuclear and/or mitochondrial DNA of the symbiont.

This analysis may involve detecting the presence or absence of polymorphic markers, such as simple sequence repeats (SSRs) or mating-type markers. SSRs,

-12 2017251752 25 Oct 2017 also called microsatellites, are based on a 1-7 nucleotide core element, more typically a 1-4 nucleotide core element, that is tandemly repeated. The SSR array is embedded in complex flanking DNA sequences. Microsatellites are thought to arise due to the property of replication slippage, in which the DNA polymerase enzyme 5 pauses and briefly slips in terms of its template, so that short adjacent sequences are repeated. Some sequence motifs are more slip-prone than others, giving rise to variations in the relative numbers of SSR loci based on different motif types. Once duplicated, the SSR array may further expand (or contract) due to further slippage and/or unequal sister chromatid exchange. The total number of SSR sites is high, 0 such that in principle such loci are capable of providing tags for any linked gene.

SSRs are highly polymorphic due to variation in repeat number and are codominantly inherited. Their detection is based on the polymerase chain reaction (PCR), requiring only small amounts of DNA and suitable for automation. They are 5 ubiquitous in eukaryotic genomes, including fungal and plant genomes, and have been found to occur every 21 to 65 kb in plant genomes. Consequently, SSRs are ideal markers for a broad range of applications such as genetic diversity analysis, genotypic identification, genome mapping, trait mapping and marker-assisted selection.

:0

Known SSR markers which may be used to investigate endophyte diversity in perennial ryegrass are described in van Zijll de Jong et al (2003).

Alternatively, or in addition, the genetic analysis may involve sequencing genomic 25 and/or mitochondrial DNA and performing sequence comparisons to assess genetic variation between symbionts.

The inoculated germplasm or organism-symbiont association may be subjected to metabolic analysis to identify the presence of desired metabolic traits.

By ‘metabolic analysis’ is meant analysing metabolites, in particular toxins, produced by the symbionts. Preferably, this is done by generation of inoculated plants for each of the symbionts and measurement of eg. toxin levels, resistance to pests and/or

-132017251752 25 Oct 2017 diseases, or tolerance to water and/or nutrient stress in planta. More preferably, this is done by generation of isogenically inoculated plants for each of the endophytes and measurement of toxin levels in planta.

By a ‘desired genetic and metabolic profile’ is meant that the symbiont possesses genetic and/or metabolic characteristics that result in a beneficial phenotype in an organism harbouring, or otherwise associated with, the symbiont.

Such beneficial properties include improved tolerance to water and/or nutrient stress, improved resistance to pests and/or diseases, enhanced biotic stress tolerance, enhanced drought tolerance, enhanced water use efficiency, enhanced tolerance to extremes of temperature, reduced toxicity, enhanced nutrient availability and enhanced vigour in, for example, a plant with which the symbiont is associated, relative to a control plant lacking the symbiont or containing a control symbiont such as standard toxic (ST) endophyte.

Such beneficial properties also include reduced toxicity of the associated plant to grazing animals.

For example, tolerance to water and/or nutrient stress may be increased by at least approximately 5%, more preferably at least approximately 10%, more preferably at least approximately 25%, more preferably at least approximately 50%, more preferably at least approximately 100%, relative to a control symbiont such as ST endophyte or to a no symbiont control plant. Preferably, tolerance to water and/or 25 nutrient stress may be increased by between approximately 5% and approximately 50%, more preferably between approximately 10% and approximately 25%, relative to a control symbiont such as ST endophyte or to a no symbiont control plant.

For example, plant resistance to pests and/or diseases may be increased by at least 30 approximately 5%, more preferably at least approximately 10%, more preferably at least approximately 25%, more preferably at least approximately 50%, more preferably at least approximately 100%, relative to a control plant. Preferably, plant

-14 2017251752 25 Oct 2017 resistance to diseases and/or pests may be increased by between approximately 5% and approximately 50%, more preferably between approximately 10% and approximately 25%, relative to a control plant.

For example, water use efficiency and/or plant vigour may be increased by at least approximately 5%, more preferably at least approximately 10%, more preferably at least approximately 25%, more preferably at least approximately 50%, more preferably at least approximately 100%, relative to a control symbiont such as ST endophyte or to a no symbiont control plant. Preferably, tolerance to water and/or 0 nutrient stress may be increased by between approximately 5% and approximately 50%, more preferably between approximately 10% and approximately 25%, relative to a control symbiont such as ST endophyte or to a no symbiont control plant.

For example, toxicity may be reduced by at least approximately 5%, more preferably at least approximately 10%, more preferably at least approximately 25%, more preferably at least approximately 50%, more preferably at least approximately 100%, relative to a control symbiont such as ST endophyte or a no symbiont control plant. Preferably, toxicity may be reduced by between approximately 5% and approximately 100%, more preferably between approximately 50% and approximately 100% relative to a control symbiont such as ST endophyte or a no symbiont control plant.

In a preferred embodiment toxicity may be reduced to a negligible amount or substantially zero toxicity.

In a preferred embodiment, the endophyte may exhibit a desired toxin profile.

Preferably the endophyte is isolated from a fescue species, preferably tall fescue.

Preferably, the endophyte is of the genus Neotyphodium, more preferably it is from a species selected from the group consisting of N uncinatum, N coenophialum and N 30 lolii, most preferably N coenophialum. The endophyte may also be from the genus Epichloe, including E typhina, E baconii and E festucae. The endophyte may also be of the non-Epichloe out-group. The endophyte may also be from a species selected

-152017251752 25 Oct 2017 from the group consisting of FaTG-3 and FaTG-3 like, and FaTG-2 and FaTG-2 like.

The endophyte may also be from the genus Acremonium, including A. implicatum and endophytes from Brachiaria-Urochloa grasses as described in Australian patent 5 application No. 2011902393 entitled “Fungi and associated methods”, to the present applicant, the entire disclosure of which is incorporated herein by reference.

By a ‘desired toxin profile’ is meant that the symbiont produces significantly less toxic compounds and/or significantly more beneficial compounds than a control organism. 0 For example in the case of plants, the endophyte may produce significantly less toxic alkaloids, such as ergovaline, compared with a plant inoculated with a control endophyte such as standard toxic (ST) endophyte; and/or significantly more alkaloids conferring beneficial properties such as improved tolerance to water and/or nutrient stress and improved resistance to pests and/or diseases in the plant with which the 5 endophyte is associated, such as peramine, N-formylloline, N-acetylloline and norloline, again when compared with a plant inoculated with a control endophyte such as ST or with a no endophyte control plant.

In a particularly preferred embodiment, the endophyte may be selected from the 0 group consisting of E1, NEA10, NEA11 and NEA12, which were deposited at The National Measurement Institute on 5 January 2010 with accession numbers V10/000001, V10/000002, V10/000003 and V10/000004, respectively, and are described in International patent application PCT/AU2011/000020, the entire disclosure of which is incorporated herein by reference.

In a particularly preferred embodiment, the endophyte may be selected from the group consisting of NEA16, NEA17, NEA18, NEA19, NEA20, NEA21 and NEA23, which were deposited at The National Measurement Institute on 3 April 2012 with accession numbers V12/001413, V12/001414, V12/001415, V12/001416, 30 V12/001417, V12/001418 and V12/001419, respectively, and are described in an

Australian patent application filed 1 June 2012 entitled ‘Novel endophytes’, the entire disclosure of which is incorporated herein by reference.

- 162017251752 25 Oct 2017

In a particularly preferred embodiment, the endophyte may be selected from the group consisting of Acremonium 1.1.A (1.1 A), 3.3.A (3.3A), 5.1.B (5.1 B), 9.2.A (9.2A) and 12.1.A (12.1A), which were deposited at The National Measurement Institute on 5 15 June 2011 with accession numbers V11/011370, V11/011371, V11/011372,

V11/011373, and V11/011374, respectively, which are described in Australian patent application No. 2011902393 entitled “Fungi and associated methods”, the entire disclosure of which is incorporated herein by reference.

Such endophytes may have a desired toxin profile as hereinbefore described.

In a preferred embodiment of the present invention, the symbiont(s) may include a genetic variation, for example, for endophyte trait introgression in plants such as grasses to enhance vegetative stability of the symbiotum, intergenerational stability of 5 the symbiotum, abiotic stress tolerance (e.g. water stress) of the symbiotum, biotic stress tolerance (e.g. disease resistance) of the symbiotum, nutrient use efficiency (e.g. phosphorus use efficiency, nitrogen use efficiency) of the symbiotum; and for example, for symbiont trait introgression in animals such as ruminant livestock species to enhance feed conversion efficiency of the symbiotum (i.e. ruminant 0 livestock organism with its rumen microbiome) or to mitigate methanogenesis from the symbiotum.

The genetic variation may be introduced utilizing any standard techniques, e.g. via one or more of random mutagenesis, di/poly-ploidisation, targeted mutagenesis;

cisgenesis; transgenesis; intragenesis.

In a preferred embodiment, the endophyte(s) may be endophyte variants as described in an Australian patent application filed 1 June 2012 entitled “Designer Endophytes”, the entire disclosure of which is incorporated herein by reference.

In a particularly preferred embodiment, the endophyte may be selected from the group consisting of an endophyte variant selected from the group consisting of

-172017251752 25 Oct 2017

NEA12dh5, NEA12dh6, NEA12dh13, NEA12dh14, and NEA12dh17, which were deposited at The National Measurement Institute on 3 April 2012 with accession numbers V12/001408, V12/001409, V12/001410, V12/001411 and V12/001412, respectively.

Such endophytes may have a desired toxin profile as hereinbefore described.

Preferably, the organism is inoculated with the symbiont by a method selected from the group consisting of infection, breeding, crossing, hybridization and combinations 0 thereof.

In one embodiment, the plant may be inoculated by isogenic inoculation. This has the advantage that phenotypic effects of endophytes may be assessed in the absence of host-specific genetic effects. More particularly, multiple inoculations of endophytes 5 may be made in plant germplasm, and plantlets regenerated in culture before transfer to soil or other growth medium.

In another embodiment, a library of plant germplasm may be inoculated with multiple endophytes. This has the advantage of enabling favourable host-endophyte 0 associations to be identified.

The identification of an endophyte of the opposite mating-type that is highly compatible and stable in planta provides a means for molecular breeding of endophytes for perennial ryegrass. Preferably the plant may be infected by hyper25 inoculation.

Hyphal fusion between endophyte strains of the opposite mating-type provides a means for delivery of favourable traits into the host plant, preferably via hyperinoculation. Such strains are preferably selected from the group including an 30 endophyte strain that exhibits the favourable characteristics of high inoculation frequency and high compatibility with a wide range of germplasm, preferably elite perennial ryegrass and/or tall fescue host germplasm and an endophyte that exhibits

-182017251752 25 Oct 2017 a low inoculation frequency and low compatibility, but has a highly favourable alkaloid toxin profile.

The endophyte-infected plants may be cultured by known techniques. The person 5 skilled in the art can readily determine appropriate culture conditions depending on the plant to be cultured.

The screening step (iii) may include analysing plant metabolites. The metabolites may be analysed by known techniques such as chromatographic techniques or mass 0 spectrometry, for example LCMS or HPLC. In a particularly preferred embodiment, endophyte-infected plants may be analysed by reverse phase liquid chromatography mass spectrometry (LCMS). This reverse phase method may allow analysis of specific metabolites (including lolines, peramine, ergovaline, lolitrem, and janthitrems, such as janthitrem I, janthitrem G and janthitem F) in one LCMS chromatographic run 5 from a single endophyte-infected plant extract.

In a particularly preferred embodiment, the endophytes may be selected from the group consisting of consisting of NEA2, NEA3, NEA6, NEA10, NEA11, NEA12, E1, NEA17, NEA21, NEA23, NEA18, NEA19, NEA16, NEA20, NEA12dh5, NEA12dh6, 0 NEA12dh13, NEA12dh14, NEA12dh17, NEA12-DsRed and IRM1-35.

In another particularly preferred embodiment, LCMS including EIC (extracted ion chromatogram) analysis may allow detection of the alkaloid metabolites from small quantities of endophyte-infected plant material. Metabolite identity may be confirmed 25 by comparison of retention time with that of pure toxins or extracts of endophyteinfected plants with a known toxin profile analysed under substantially the same conditions and/or by comparison of mass fragmentation patterns, for example generated.

As stated above, in the method according to this aspect of the present invention, the organism selected may be a plant or animal, preferably a plant. Where the organism

-192017251752 25 Oct 2017 is a plant, the plant germplasm may be present as an embryo and the embryo may be treated to form an artificial seed.

Accordingly, in a preferred embodiment there is provided a method for preparing artificial seeds which method includes:

providing a source of plant seeds; subjecting the seed(s) to a surface-sterilisation step; isolating seed embryo(s) from the surface-sterilised seed(s); and coating the embryo(s) with a coating to form artificial seed(s).

o

The artificial seeds may be prepared utilising the techniques described in Australian provisional patent applications filed 1 June 2012 and 7 September 2012 entitled “Method for large scale generation of symbiota”, the entire disclosures of which are incorporated herein by reference.

The seeds may be from any suitable plant. The plant may be a grass, preferably a perennial grass, legume, vine, shrub, tree, herb, flower, shrub or bush. The method according to this aspect of the present invention is particularly applicable to grasses and legumes.

:0

The seeds may be surface-sterilised by any suitable technique. Preferably the seeds are sterilised by treating them with an acid such as hydrochloric acid and bleach, such as sodium hypochlorite. Preferably the acid and bleach treatments are performed sequentially. The acid treatment may be for a period of from 1 hour to 24 25 hours, preferably overnight. The bleach treatment may be for a period of 5 minutes to 1 hour, preferably approximately 20 minutes. The bleach treatment may be performed twice on successive days, with the seeds being washed after each treatment, for example using sterile distilled water, and stored at approximately 4 to 30°C, preferably approximately 24°C.

Embryos may be isolated from the treated seeds by techniques known to those skilled in the art.

-20 2017251752 25 Oct 2017

In a preferred embodiment, the embryos may be treated to create one or more points of entry for the symbiont, eg. endophyte. For example, the embryo may be punctured or its surface ortherwise damaged, for example by scratching or etching, to 5 facilitate entry of the symbiont. In a particularly preferred embodiment, a hypodermic needle or similar may be used to create singly or multiple puncture holes in the surface of the embryo.

The coating may be of any suitable type to encapsulate the embryo, including 0 alginate, agar, polyco 2133, carboxy methyl cellulose, carrageenan, gelrite, guargum, sodium pectate, tragacanth gum and the like. In a preferred embodiment the coating is alginate, more particularly calcium alginate.

In a preferred embodiment, the embryos may be mixed with the coating and drops of 5 coating containing individual embryos placed in a polymerising solution such as calcium chloride solution, preferably while stirring, to form artificial seeds. Artificial seeds may be collected following approximately 1-60 minutes stirring, preferably after approximately 15 minutes stirring.

In a preferred embodiment the embryos may be inoculated with a symbiont such as a fungal endophyte prior to coating. In a preferred form, the embryos may be directly inoculated with endophyte mycelium.

Alternatively, in a particularly preferred embodiment, isolated embryos may be coated 25 with a symbiont-containing coating layer, such as a fungal endophyte-containing coating layer.

In this embodiment, the inoculation step may include: providing a source of seed embryos;

inoculating the embryos with one or more symbionts such as fungal endophytes; and coating the inoculated embryo(s) with a coating to form artificial seed(s).

-21 2017251752 25 Oct 2017

Alternatively, the inoculation step may include: providing a source of seed embryos; and coating the embryos with a coating containing symbionts such as fungal 5 endophytes to form artificial seed(s).

In a preferred embodiment the seeds may be double coated. Preferably the second coating layer is alginate, more preferably calcium alginate, even more preferably coloured calcium alginate. In a preferred embodiment, the artificial seeds with the 0 first coating layer may be air dried prior to coating with the second layer.

In a preferred embodiment, the method may further include coating the artificial seeds with a second coating layer, said second coating layer preferably containing added nutrients suitable for sustaining the embryo and/or symbiont.

Alternatively, the second coating layer may not contain added nutrients, this nutrient deprived layer being designed to eg. reduce endophyte out-growth during germination and restrict endophyte growth in close proximity to the embryo.

In a preferred embodiment the method may further include growing the artificial seeds to form plantlets or seedlings; and screening the plantlets or seedlings for symbiont presence such as fungal endophyte presence.

The step of growing the artificial seeds may be undertaken using any suitable growth medium. A germination medium such as MS (Murashige and Skoog), modified MS or MS + BAP (6-benzylamino purine) is particularly preferred.

The genetic analysis may be conducted as described above. The seedlings may for 30 example be screened for symbiont-specific, eg. endophyte-specific simple sequence repeats (SSRs).

-22 2017251752 25 Oct 2017

Alternatively, or in addition, the seedlings may be screened for the presence of favourable symbiota via molecular phenotyping. The molecular phenotyping may be performed utilising the methods described in an Australian provisional patent application filed 1 June 2012 entitled “Molecular phenotyping method”, the entire 5 disclosure of which is incorporated herein by reference.

In this method seedlings may be screened for the presence of favourable symbiota via molecular phenotyping. The seedlings may, for example, be assessed for improved alkaloid production and/or improved water soluble carbohydrate: protein 0 ratio. Such techniques may utilise an enzymatic assay, colorimetric assay, SSR markers and/or metabolomics analysis. Such analyses may be semi- or substantially fully automated.

Thus the method may include screening symbiota for the presence of desirable 5 characteristics, said method including molecular phenotyping a population of symbiota.

In a preferred embodiment, the method may include assessing the population of symbiota for alkaloid production and/or water soluble carbohydrate (WSC):protein 0 ratio. Preferably this assessment is done using one or more methods selected from the group consisting of enzymatic assays, colorimetric assays, SSR markers and metabolomic analysis.

In a preferred embodiment, assessment of alkaloid production includes measurement 25 of alkaloid profile and/or content in the population. Preferred alkaloids include peramine, lolitrem B and ergovaline. In a preferred embodiment, alkaloids may be inferred by SSR markers and detected by metabolomic analysis, more preferably a combination of SSR marker and metabolomic analysis are used.

In another preferred embodiment, WSC:protein ratio may be assessed. WSC may be quantified using an enzymatic assay. In a preferred embodiment, individual

-232017251752 25 Oct 2017 concentrations for sucrose, glucose, fructose and fructans may be determined. Protein may be quantified using a colorimetric assay.

In a particularly preferred embodiment, protein may be quantified by a method 5 including:

extracting proteins from the symbiota using an alkali, such as NaOH, preferably a weak NaOH solution;

quantification of proteins using a colorimetric assay, such as a Bradford assay. 0 Detection may be carried out, for example, using a plate reader.

The symbiota may be of any suitable form, including inoculated embryos, plant seeds, germinating seeds, seedlings, plantlets, plants, etc.

Preferably the seeds are derived from endophyte-infected plants i.e. plant/endophyte symbiota.

In the method according to this aspect of the present invention, the screening step (iii) may include screening artificial seeds by accelerated ageing, which is described 0 in an Australian patent application filed 1 June 2012 entitled ‘Method for selection of stable symbiota’, the entire disclosure of which is incorporated herein by reference.

Accordingly, the present invention provides a method of assessing the compatibility and/or stability of a plant/endophyte symbiota, said method including:

providing a source of seeds including symbiont such as fungal endophyte inoculated plant embryos;

screening the seeds and/or their offspring for compatibility and/or stability of the plant/symbiont association (i.e. symbiota) such as plant-fungal endophyte symbiota by applying accelerated ageing thereto.

In the accelerated ageing procedure, the seeds, or their offspring, may be subjected to deteriorative conditions, preferably by means of high temperature and/or increased

-24 2017251752 25 Oct 2017 moisture content. In a particularly preferred embodiment the seeds may be exposed to an environment of high relative humidity. For example, the seeds may be exposed to temperatures of approximately -20 to 50°C, preferably 10 to 45°C, more preferably 15 to 40°C, even more preferably 25 to 40°C and/or to humidity levels of 5 approximately 60% to 100%, preferably 80% to 100% for periods of e.g.

approximately 1 to 30 days, preferably 2 to 10 days, more preferably 4 to 7 days.

Accelerated ageing reduces endophyte viability i.e. it allows counter-selection of unstable associations and permits the ranking of symbiota based on their stability.

o

Preferably the method includes the further step of subjecting the selected symbiota populations to a rapid symbiont, such as fungal, endophyte viability assay.

Accordingly, the method of the present invention may further include assessing the 5 compatibility and/or stability of a plant/symbiont association (i.e. symbiotum) such as plant-fungal endophyte symbiota including providing a source of seeds including symbiont, eg. fungal endophyte inoculated plant embryos;

screening the seeds and/or their offspring for compatibility and/or stability of 0 the plant/symbiont association (i.e. symbiotum) such as plant-fungal endophyte symbiota by applying accelerated ageing thereto; and subjecting the selected symbiota populations to a rapid symbiont such as fungal endophyte viability assay.

The viability assay step according to this aspect of the present invention may include: culturing the seeds to generate plantlets, seedlings or germinating seeds; extracting DNA and/or RNA from the plantlets, seedlings or germinating seeds; and subjecting the extracted DNA and/or RNA to an assay for in planta expressed 30 symbiont-specific gene(s) such as fungal endophyte-specific gene(s).

Preferably the seeds are derived from symbiont-inoculated plants such as fungal

-252017251752 25 Oct 2017 endophyte-inoculated plants.

Preferably the seeds are artificial seeds, as hereinbefore described.

The rapid endophyte viability assay is described in an Australian patent application 5 filed 1 June 2012 entitled ‘Method for rapid endophyte viability assessment’, the entire disclosure of which is incorporated herein by reference.

Preferably the seeds are cultured for a relatively short period of time, so that a rapid assessment of symbiont viability such as fungal endophyte viability may be obtained.

Preferably the seeds are cultured for approximately 1 to 10 days, more preferably 3 to 10 days, more preferably 3 to 7 days, more preferably 3 to 5 days.

Applicants have found that endophyte specific genes are expressed in this time frame, enabling early in planta endophyte viability assessment.

In a preferred form the DNA/RNA may be extracted from the leaves of seedlings, more preferably from the epicotyl, hypocotyl or similar embryonic shoot of the seedlings. In grasses, the DNA/RNA may be extracted from tillers. In another preferred form the DNA/RNA may be extracted from whole germinating seeds.

:0

Preferably the RNA and DNA may be co-extracted, preferably in a single step. Preferably, the DNA/RNA may be extracted from 1 to 10 day-old, preferably 3 to 10 day old, more preferably 3 to 7 day old, more preferably 3 to 5 day-old epicotyls, hypocotyls or similar embryonic shoots of seedlings, in order to accelerate the 25 process.

The assay may be an assay used to amplify and simultaneously quantify a targeted DNA/RNA molecule in the extracted DNA/RNA. Preferably the assay is a quantitative real-time polymerase chain reaction (Q-PCR/qRT-PCR) assay, or kinetic 30 polymerase chain reaction (KPCR) assay. In a particularly preferred form, the assay may be a TaqMan or similar assay.

-262017251752 25 Oct 2017

The endophyte specific genes may be of any suitable type. Preferably it is only, mainly or highly expressed in planta. Genes encoding the proteins 7490, 8263, 0005 and 2232 are particularly preferred.

Primers are designed for amplification of the targeted gene(s) by methods known to those skilled in the art.

The seeds may be from any suitable plant. The plant may be a grass, preferably a perennial grass, legume, vine, shrub, tree, herb, flower, shrub or bush. The method 0 according to this aspect of the present invention is particularly applicable to grasses and legumes.

Preferably the seeds are derived from symbiont eg. endophyte-infected plants eg. plant/endophyte symbiota.

Preferably the seeds are artificial seeds.

The method according to this aspect of the present invention may further include subjecting the selected symbiota populations to phenotyping for assessment of 0 symbiota performance and/or maintenance of desired characteristics; and selecting symbiota for generating a synthetic symbiota variety, for example by polycrossing.

For example, the selected symbiota variety may be subjected to an symbiont identification assay, followed by polycrossing to generate a next generation seed. 25 Optionally, the above steps may be repeated to confirm symbiota stability, desired characteristics, symbiont, eg. Fungal endophyte identity and/or symbiont, eg. Fungal endophyte incidence in the next generation, for example the next seed generation.

Accordingly, in a further aspect of the present invention, there is provided improved 30 symbiota including one or more organisms containing one or more symbionts produced utilising the method described above.

-272017251752 25 Oct 2017

Accordingly, in a further aspect of the present invention, there is provided an improved organism exhibiting symbiota with an symbiont and produced utilising the method described above.

The improved organism may be a plant or animal.

Where the organism is a plant, the plant may be a grass, tree, flower, herb, shrub or bush, vine or legume, or a product thereof.

The plant material may be in the form of a seed, seedling, embryo or the like.

The method steps described above may be repeated to develop later generations of symbiota seeds, plants or animals.

In a further aspect, the present invention provides a plant, plant seed or other plant part derived from a seed or plant of the present invention and stably infected with a symbiont, eg. an endophyte.

Preferably, the plant cell, plant, plant seed or other plant part is a grass, more 0 preferably a forage, turf or bioenergy grass, such as those of the genera Lolium and Festuca, including L. perenne and L. arundinaceum, and of the genera Brachiaria and Urochloa, including B. brizantha, B. decumbens, B. humidicola and U. mosambicensis.

By ‘plant cell’ is meant any self-propagating cell bounded by a semi-permeable membrane and containing plastid. Such a cell also required a cell wall if further propagation is desired. Plant cell, as used herein includes, without limitation, seeds suspension cultures, embryos, meristematic regions, callus tissue, leaves, roots, shoots, gametophytes, sporophytes, pollen and microspores.

In an alternative embodiment of the present invention the organism may be an animal, preferably selected from cattle, sheep, goats, deer or the like.

-282017251752 25 Oct 2017

Accordingly, in a preferred embodiment of the present invention, there is provided a method as described above wherein the organism is an animal, the method including:

(i) providing a library of animal germplasm; and a library of symbionts;

(ii) inoculating the animal germplasm library with one or more symbionts selected from the symbiont library to generate symbiota;

(iii) breeding, selecting, screening and/or evaluating the symbiota for desired symbiota characteristic(s); and (iv) identifying, culturing or otherwise using the symbiota exhibiting desired characteristic(s) to produce the improved animal.

In a further aspect of the present invention, there is provided a method of mapping regions of a host organism’s genome that affect rumen microbial profile in the organism, said method including (i) providing a library of polymorphisms of the host genome;

(ii) identifying the effect on rumen microbiome profile of the polymorphisms by carrying out an association study; and (iii) identifying one or more regions of the host genome that affect rumen microbial profile in the organism.

In a preferred embodiment, the organism may be a cow.

In a further aspect, the present invention includes identifying and/or cloning nucleic 25 acids including genes encoding polypeptides or transcription factors from the genomes of the symbionts.

Methods for identifying and/or cloning nucleic acids encoding such genes are known to those skilled in the art and include creating nucleic acid libraries, such as cDNA or 30 genomic libraries, and screening such libraries, for example using probes for genes of the desired type; or mutating the genome of the symbiont of the present invention, for example using chemical or transposon mutagenesis, identifying changes in the

-292017251752 25 Oct 2017 production of polypeptides or transcription factors of interest, and thus identifying genes encoding such polypeptides or transcription factors.

Thus, in a further aspect of the present invention, there is provided a substantially purified or isolated nucleic acid encoding a polypeptide or transcription factor from the genome of an symbiont of the present invention.

By ‘nucleic acid’ is meant a chain of nucleotides capable of carrying genetic information. The term generally refers to genes or functionally active fragments or 0 variants thereof and or other sequences in the genome of the organism that influence its phenotype. The term ‘nucleic acid’ includes DNA (such as cDNA or genomic DNA) and RNA (such as mRNA or microRNA) that is single- or double-stranded, optionally containing synthetic, non-natural or altered nucleotide bases, synthetic nucleic acids and combinations thereof.

By a ‘nucleic acid encoding a polypeptide or transcription factor’ is meant a nucleic acid encoding an enzyme or transcription factor normally present in an symbiont of the present invention.

The present invention encompasses functionally active fragments and variants of the nucleic acids of the present invention. By ‘functionally active’ in relation to the nucleic acid is meant that the fragment or variant (such as an analogue, derivative or mutant) is capable of manipulating the function of the encoded polypeptide, for example by being translated into an enzyme or transcription factor that is able to catalyse or 25 regulate a step involved in the relevant pathway, or otherwise regulate the pathway in the symbiont. Such variants include naturally occurring allelic variants and nonnaturally occurring variants. Additions, deletions, substitutions and derivatizations of one or more of the nucleotides are contemplated so long as the modifications do not result in loss of functional activity of the fragment or variant. Preferably the 30 functionally active fragment or variant has at least approximately 80% identity to the relevant part of the above mentioned sequence to which the fragment or variant corresponds, more preferably at least approximately 90% identity, even more

-30 2017251752 25 Oct 2017 preferably at least approximately 95% identity, most preferably at least approximately 98% identity. Such functionally active variants and fragments include, for example, those having conservative nucleic acid changes.

Preferably the fragment has a size of at least 20 nucleotides, more preferably at least 50 nucleotides, more preferably at least 100 nucleotides.

Preferably, said fragments are able to produce the same activity as the original gene when expressed. Preferably, said fragments that maintain conserved regions within 0 consensus sequences.

Preferably said variants are variants of the specified sequences that provide either conserved substitution, or limited modifications in consensus sequences to a level, for example, of no more than approximately 5%, more preferably no more than 1%.

For example, fragments and variants of a sequence encoding X may include a wild type sequence from species Z that encodes X, a fragment of a wild type sequence wherein the fragment encodes X, and that retains conserved regions within consensus sequences from species Z, and variants of the wild type sequence or 0 fragments which encode X activity and have only conservative substitutions, a variant

X’ that encodes X activity and in which sequence differs only by substitutions found in one or more contributing sequences used in formulating the consensus sequence, or a variant X that encodes X activity in which the variant has not more than approximately 95% amino acid variation, more preferably not more than 25 approximately 99% amino acid variation from the wild type sequence or fragment.

By ‘conservative nucleic acid changes’ or ‘conserved substitution’ is meant nucleic acid substitutions that result in conservation of the amino acid in the encoded protein, due to the degeneracy of the genetic code. Such functionally active variants and 30 fragments also include, for example, those having nucleic acid changes which result in conservative amino acid substitutions of one or more residues in the corresponding amino acid sequence.

-31 2017251752 25 Oct 2017

By ‘conservative amino acid substitutions’ is meant the substitution of an amino acid by another one of the same class, the classes being as follows:

Nonpolar: Ala, Val, Leu, lie, Pro, Met, Phe, Trp

Uncharged polar: Gly, Ser, Thr, Cys, Tyr, Asn, Gln

Acidic: Asp, Glu

Basic: Lys, Arg, His

Other conservative amino acid substitutions may also be made as follows:

Aromatic: Phe, Tyr, His

Proton Donor: Asn, Gln, Lys, Arg, His, Trp

Proton Acceptor: Glu, Asp, Thr, Ser, Tyr, Asn, Gln

In a further aspect of the present invention, there is provided a genetic construct 5 including a nucleic acid according to the present invention.

By ‘genetic construct’ is meant a recombinant nucleic acid molecule.

In a preferred embodiment, the genetic construct according to the present invention 0 may be a vector.

By a ‘vector’ is meant a genetic construct used to transfer genetic material to a target cell.

The vector may be of any suitable type and may be viral or non-viral. The vector may be an expression vector. Such vectors include chromosomal, non-chromosomal and synthetic nucleic acid sequences, eg. derivatives of plant viruses; bacterial plasmids; derivatives of the Ti plasmid from Agrobacterium tumefaciens; derivatives of the Ri plasmid from Agrobacterium rhizogenes; phage DNA; yeast artificial chromosomes;

bacterial artificial chromosomes; binary bacterial artificial chromosomes; vectors derived from combinations of plasmids and phage DNA. However, any other vector may be used as long as it is replicable or integrative or viable in the target cell.

-32 2017251752 25 Oct 2017

In a preferred embodiment of this aspect of the invention, the genetic construct may further include a promoter and a terminator; said promoter, gene and terminator being operatively linked.

By a ‘promoter’ is meant a nucleic acid sequence sufficient to direct transcription of an operatively linked nucleic acid sequence.

By Operatively linked’ is meant that the nucleic acid(s) and a regulatory sequence, 0 such as a promoter, are linked in such a way as to permit expression of said nucleic acid under appropriate conditions, for example when appropriate molecules such as transcriptional activator proteins are bound to the regulatory sequence. Preferably an operatively linked promoter is upstream of the associated nucleic acid.

By ‘upstream’ is meant in the 3’->5’ direction along the nucleic acid.

The promoter and terminator may be of any suitable type and may be endogenous to the target cell or may be exogenous, provided that they are functional in the target cell.

:0

A variety of terminators which may be employed in the genetic constructs of the present invention are also well known to those skilled in the art. The terminator may be from the same gene as the promoter sequence or a different gene. Particularly suitable terminators are polyadenylation signals, such as the (CaMV)35S polyA and 25 other terminators from the nopaline synthase (nos) and the octopine synthase (ocs) genes.

The genetic construct, in addition to the promoter, the gene and the terminator, may include further elements necessary for expression of the nucleic acid, in different 30 combinations, for example vector backbone, origin of replication (ori), multiple cloning sites, spacer sequences, enhancers, introns, antibiotic resistance genes and other selectable marker genes [such as the neomycin phosphotransferase (nptll) gene, the

-332017251752 25 Oct 2017 hygromycin phosphotransferase (hph) gene, the phosphinothricin acetyltransferase (bar or pat) gene], and reporter genes [such as beta-glucuronidase (GUS) gene (gusA) and the green fluorescent protein (GFP) gene (gfp)]. The genetic construct may also contain a ribosome binding site for translation initiation. The genetic 5 construct may also include appropriate sequences for amplifying expression.

Those skilled in the art will appreciate that the various components of the genetic construct are operably linked, so as to result in expression of said nucleic acid. Techniques for operably linking the components of the genetic construct of the 0 present invention are well known to those skilled in the art. Such techniques include the use of linkers, such as synthetic linkers, for example including one or more restriction enzyme sites.

Preferably, the genetic construct is substantially purified or isolated.

By ‘substantially purified’ is meant that the genetic construct is free of the genes, which, in the naturally-occurring genome of the organism from which the nucleic acid or promoter of the invention is derived, flank the nucleic acid or promoter. The term therefore includes, for example, a genetic construct which is incorporated into a 0 vector; into an autonomously replicating plasmid or virus; or into the genomic DNA of a prokaryote or eukaryote; or which exists as a separate molecule (eg. a cDNA or a genomic or cDNA fragment produced by PCR or restriction endonuclease digestion) independent of other sequences. It also includes a genetic construct which is part of a hybrid gene encoding additional polypeptide sequence.

Preferably, the substantially purified genetic construct is at least approximately 90% pure, more preferably at least approximately 95% pure, even more preferably at least approximately 98% pure.

The term “isolated” means that the material is removed from its original environment (eg. the natural environment if it is naturally occurring). For example, a naturally occurring nucleic acid present in a living plant is not isolated, but the same nucleic

-34 2017251752 25 Oct 2017 acid separated from some or all of the coexisting materials in the natural system, is isolated. Such nucleic acids could be part of a vector and/or such nucleic acids could be part of a composition, and still be isolated in that such a vector or composition is not part of its natural environment.

As an alternative to use of a selectable marker gene to provide a phenotypic trait for selection of transformed host cells, the presence of the genetic construct in transformed cells may be determined by other techniques well known in the art, such as PCR (polymerase chain reaction), Southern blot hybridisation analysis, histochemical assays (e.g. GUS assays), thin layer chromatography (TLC), northern and western blot hybridisation analyses.

The genetic constructs of the present invention may be introduced into organisms, for example plants, animals, microorganisms or fungi by any suitable technique.

Techniques for incorporating the genetic constructs of the present invention into plant cells or fungal cells (for example by transduction, transfection, transformation or gene targeting) are well known to those skilled in the art. Such techniques include Agrobacterium-mediated introduction, Rhizobium-mediated introduction, electroporation to tissues, cells and protoplasts, protoplast fusion, injection into reproductive organs, injection into immature embryos and high velocity projectile introduction to cells, tissues, calli, immature and mature embryos, biolistic transformation, Whiskers transformation, and combinations thereof. The choice of technique will depend largely on the type of plant or fungus to be transformed, and may be readily determined by an appropriately skilled person. For transformation of 25 protoplasts, PEG-mediated transformation is particularly preferred. For transformation of fungi PEG-mediated transformation and electroporation of protoplasts and Agrobacterium-mediated transformation of hyphal explants are particularly preferred.

Cells incorporating the genetic constructs of the present invention may be selected, as described below, and then cultured in an appropriate medium to regenerate transformed plants or fungi, using techniques well known in the art. The culture

-352017251752 25 Oct 2017 conditions, such as temperature, pH and the like, will be apparent to the person skilled in the art. The resulting plants or fungi may be reproduced, either sexually or asexually, using methods well known in the art, to produce successive generations of transformed plants or fungi.

As used herein, except where the context requires otherwise, the term comprise and variations of the term, such as comprising, comprises and comprised, are not intended to exclude further additives, components, integers or steps.

Reference to any prior art in the specification is not, and should not be taken as, an acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in Australia or any other jurisdiction or that this prior art could reasonably be expected to be ascertained, understood and regarded as relevant by a person skilled in the art.

Detailed Description of the Embodiments

In the figures:

Figure 1 shows genotypic analysis of endophyte content in accessions from a 0 targeted fescue germplasm collection.

Figure 2 shows genetic diversity analysis of tall fescue endophytes.

Figure 3 shows diversity analysis of host and endophyte.

Figure 4 shows selection of fescue-endophyte combinations for metabolic profiling, endophyte isolation and isogenic inoculation.

Figure 5 shows selection of fescue-endophyte combinations for metabolic profiling, 30 endophyte isolation and isogenic inoculation.

Figure 6 shows a desired toxin profile of tall fescue endophytes.

-362017251752 25 Oct 2017

Figure 7 shows a metabolic profile analysis.

Figure 8 shows endophytes selected for semi-quantitative analysis of metabolites.

Figures 9 and 10 show metabolomics analyses of fescue endophytes.

Figure 11 shows a semi-quantitative analysis of metabolic profile under temperature/water stress.

o

Figure 12 shows endophytes selected for isogenic inoculation.

Figure 13 shows SSR-based genotyping of isolated endophytes cultures prior to isogenic inoculation.

Figure 14 shows endophyte vegetative stability in tall fescue and perennial ryegrass host genotypes (stability at 12 months post inoculation).

Figure 15 shows endophytes selected for isogenic inoculation.

:0

Figures 16-19 show metabolic profiling of isogenic tall fescue-endophyte associations.

Figure 20 shows anti-fungal bioassays of fescue endophytes. Column 1 25 Colletotrichum graminicola, Column 2 Drechslera brizae, Column 3 Rhizoctonia cerealis.

Figure 21 shows sequencing of selected novel fescue endophytes.

Figure 22 shows peramine biosynthetic pathway.

Figures 23 A-C show presence of perA gene within non-Epichloe out-group

-372017251752 25 Oct 2017 endophytes (Fig 23A NEA17; Fig 23B NEA18; Fig 23C NEA19).

Figure 24 shows ergovaline biosynthetic pathway.

Figure 25 shows genes in the eas gene cluster.

Figures 26 A-D show presence of dmaW gene for ergovaline biosynthesis in endophyte strains (Fig 26A NEA17; Fig 26B NEA16; Fig 26C AR542; Fig 26D NEA20).

Figures 27 A-D show presence of eas gene cluster for ergovaline biosynthesis. Fig 27A FaTG-2 NEA17 (287819); Fig 27B non-Epichloe out-group NEA18 (FEtc6-75); Fig 27C FATG-3 NEA21 (231557); Fig 27D N. coenophialum NEA16 (FEtc7-342).

Figure 28 shows the Lolitrem B biosynthetic pathway.

Figure 29 shows genes in the Lolitrem B biosynthetic gene cluster.

Figures 30 A-D show presence of Lolitrem B biosynthetic gene cluster 1 (ItmG, ItmM 0 and ItmK) in endophyte strains. Fig 30A FaTG-2 NEA17 (287819); Fig 30B nonEpichloe out-group NEA18 (FEtc6-75); Fig 30C FATG-3 NEA21 (231557); Fig 30D N. coenophialum NEA16 (FEtc7-342).

Figures 31 A-D show presence of Lolitrem B biosynthetic gene cluster 2 (ItmB, ItmQ, 25 ItmP, ItmF and ItmC) in endophyte strains. Fig 31A FaTG-2 NEA17 (287819); Fig 31B non-Epichloe out-group NEA18 (FEtc6-75); Fig 31C FATG-3 NEA21 (231557); Fig 31D N. coenophialum NEA16 (FEtc7-342).

Figures 32 A-D show presence of Lolitrem B biosynthetic gene cluster 3 (ItmE and 30 ItmJ) in endophyte strains. Fig 32A FaTG-2 NEA17 (287819); Fig 32B non-Epichloe out-group NEA18 (FEtc6-75); Fig 32C FATG-3 NEA21 (231557); Fig 32D N. coenophialum NEA16 (FEtc7-342).

-382017251752 25 Oct 2017

Figure 33 shows the loline biosynthetic pathway.

Figure 34 shows the loline biosynthetic gene cluster.

Figures 35 A-D show presence of Loline biosynthetic gene cluster in endophyte strains. Fig 35A FaTG-2 NEA17 (287819); Fig 35B non-Epichloe out-group NEA18 (FEtc6-75); Fig 35C FATG-3 NEA21 (231557); Fig 35D N. coenophialum NEA16 (FEtc7-342).

Figures 36 A-F show alkaloid biosynthetic gene analysis for endophyte strain NEA23 (269850). Fig 36A Presence of loline gene cluster; Fig 36B Presence of peramine gene; Fig 36C Analysis of Lolitrem gene cluster 01; Fig 36D Analysis of Lolitrem gene clusters 02 and 03; Fig 36E Analysis of dmaW gene for ergovaline production;

Fig 36F Analysis of eas gene cluster for ergovaline production.

Figure 37 shows genotypic analysis of NEA23 and NEA21.

Figure 38 shows genotypic analysis of NEA16 and NEA20.

:0

Figure 39 shows the desired toxin profile of perennial ryegrass endophytes.

Figure 40 shows genotypic analysis of endophyte content in accessions from a targeted ryegrass germplasm collection.

Figure 41 shows analysis of endophyte content in accessions from a targeted ryegrass germplasm collection. Grey circles - candidates; Black circles - controls.

Figure 42 shows an isogenic host plant genotype cultivar ‘likeness’ determination

Figure 43 shows an isogenic host plant genotype cultivar ‘likeness’ determination.

-392017251752 25 Oct 2017

Figure 44 shows development of a host-panel molecular genetic test for QC of hostendophyte associations.

Figure 45 shows inoculation of candidate endophytes into meristem cultures of 5 diverse ryegrass host panel.

Figure 46 shows selected endophytes.

Figure 47 shows a detailed characterisation of the known knowns and their 0 precursors.

Figure 48 shows a host-panel identification test.

Figures 49 and 50 show in-depth metabolic profiling of novel isogenic host5 endophyte associations.

Figure 51 shows LEPJ: the known knowns.

Figure 52 shows accurate mass of LEPJ.

:0

Figure 53 shows identification of LEPJ. LCMS analysis of one replication of associations of Bro08_ST_1 displaying extracted ion chromatogram. (A) positive ion extraction (B) peramine; (C) ergovaline; (D) lolitrem B (E) Janthitrem (F) Negative ion extraction.

Figure 54 shows identification of LEPJ. LCMS analysis of one replication of associations of Tolosa_NEA12 displaying extracted ion chromatogram. (A) positive ion extraction (B) peramine; (C) ergovaline; (D) lolitrem B (E) Janthitrem (F) Negative ion extraction.

-40 2017251752 25 Oct 2017

Figure 55 shows quantification of LEPJ. LCquant analysis of alkaloid content according to the retention time and spectrum (A) peramine; (B) ergovaline; (C) lolitrem B; (D) Janthitrem.

Figure 56 shows a comparison of semi-quantitative alkaloid (LEPJ) profiles of E⁺ versus E' host plants. Y error bars denote standard errors; Letters above each column indicate significant differences between different hosts harbouring same endophyte; * indicate significant differences between the same host harbouring different endophytes; **Ratio values are relative to Bronsyn ST.

o

Figure 57 shows a comparison of semi-quantitative alkaloid profiles for diverse endophytes in an isogenic host (lmp04). The Y error bars denote standard errors; Indicate significant differences between the same host harbouring different endophytes.

Figure 58 shows an assessment of stability of alkaloid (LEPJ) profile of selected endophytes across different isogenic hosts. Y error bars denote standard errors; Letters above each column indicate significant differences between different hosts harbouring same endophyte. * indicate significant differences between the same host 0 harbouring different endophytes; **Ratio values are relative to Impact.

Figure 59 shows lolitrem B biosynthesis.

Figure 60 shows ergovaline biosynthesis.

Figure 61 shows an analysis of the lolitrem B biosynthetic pathway.

Figure 62 shows the paspalinine - A tremorgen pathway.

Figure 63 shows a pathway analysis: paspalinine - A tremorgen.

Figure 64 shows the presence of lolitrem and paspalinine in symbiota.

-41 2017251752 25 Oct 2017

Figure 65 shows that Peramine is present in planta in trace amounts in absence of a peramine-producing endophyte.

Figure 66 shows the procedure used to assess changes to host phenotype mediated by the endophyte. 1 isogenic host: Impact (lmp04); 5 host-endophyte associations: NEA10, NEA11, NEA12, ST, E’; 6 clonal replicates.

Figure 67 shows endophyte effect on symbiotum performance. A. Tiller number; B. 0 Shoot fresh weight; C. Root fresh weight; D. Root dry weight

Mid grey - 0.5 mM ΝΟβ' Dark grey - 2.5 mM ΝΟβ' Light grey - 10.0 mM ΝΟβ'

Figure 68 shows plants before the T1 harvest.

Figure 69 shows performance as measured by tiller number.

Figure 70 shows sequence analysis using the 454 sequencing platform.

Figure 71 shows sequence analysis using the Illumina sequencing platform.

:0

Figure 72 shows a summary of sequenced perennial ryegrass endophyte genomes. Total Yield = 84.8 gigabases (GAII: 7.4Gb, HiSeq: 77.4Gb); Average yield on HiSeq = 2.2Gb per strain; Average yield on GAII = 373Mb per strain.

Figure 73 shows a summary of sequenced perennial ryegrass endophyte genomes. Total number of reads = 436,920,622 pairs; Average number of paired reads on HiSeq = 12,844,656 per strain; Average number of paired reads on GAII = 1,863,852 per strain.

Figures 74-78 show Illumina reads v 454 contigs.

Figure 79 shows 454 FLX v 454 Titanium v Illumina GAII coverage display.

-42 2017251752 25 Oct 2017

Figure 80 shows 454 FLX v 454 Titanium v Illumina GAI I v Hi Seq coverage display.

Figure 81 shows Homo polymer correction: graphical view.

Figure 82 shows Homo polymer correction: base view.

Figure 83 shows Pan-genome SNP calling: graphical view, read level.

Figure 84 shows 77 independent sequencing runs, ordered by similarity.

Figure 85 shows that independent sequencing runs can also display strain specific deletions.

Figure 86 shows that independent sequencing runs can predict amino acid changes from SNP data.

Figure 87 shows that contig number decreases with rounds of assembly refinement.

Figure 88 shows that N50 increases with rounds of assembly refinement.

Figure 89 shows that assembly refinement decreases the number of scaffolds, and single ‘unscaffolded’ contigs.

Figure 90 shows that average number of bases per contig increases with assembly refinement.

Figure 91 shows the structures of Lolitrem B, Erogvaline and Peramine, with desirable toxin profiles indicated.

Figure 92 shows in vitro bioassays to assess antifungal activity of Neotyphodium endophytes.

-432017251752 25 Oct 2017

Figure 93 shows a detached leaf assay to assess resistance to crown rust (Puccinia coronata f. sp. Lolii) of perennial ryegrass plants with and without Neotyphodium endophytes.

Figure 94 shows glasshouse and field trial screens for drought tolerance and water use efficiency of perennial ryegrass plants with and without Neotyphodium endophytes.

Figure 95 shows the steps involved in cell division.

Figure 96 shows experimental work flow for chromosome doubling of endophyte cells.

Figure 97 shows flow cytometry calibrations for DNA content assessment in Neotyphodium endophyte strains. Peaks indicate relative nuclear DNA content.

Figure 98 shows flow cytometry analysis of NEA12^dh Neotyphodium endophyte strains.

:0

Figure 99 shows analysis of growth rate in culture after 8 weeks of NEA12^dh Neotyphodium endophyte strains compared to control endophyte strains.

Figure 100 shows analysis of growth rate in culture over 5 weeks of NEA12^dh 25 Neotyphodium endophyte strains compared to control endophyte strains.

Figure 101 shows antifungal bioassays of NEA12^dh Neotyphodium endophyte strains.

Figure 102 shows antifungal bioassays of NEA12^dh Neotyphodium endophyte strains.

Figure 103 shows analysis of genome survey sequencing read depth of colchicinetreated Neotyphodium endophyte strains.

-44 2017251752 25 Oct 2017

Figure 104 shows analysis of genome survey sequencing reads mapping to NEA12 genome survey sequence assembly.

Figure 105 shows experimental work flow for X-ray mutagenesis.

Figure 106 shows the indole-diterpene biosynthetic pathway of Neotyphodium endophytes.

Figure 107 shows in vitro growth of X-ray irradiated Neotyphodium endophyte strains.

Figure 108 shows Itm gene clusters of Neotyphodium endophytes.

Figure 109 shows determination of genome sequence variation in X-ray irradiated 5 Neotyphodium endophyte strains.

Figure 110 shows single nucleotide polymorphisms (SNPs) in genome sequences of X-ray irradiated Neotyphodium endophyte strains.

Figure 111 shows small insertions/deletions (INDELs) in genome sequences of X-ray irradiated Neotyphodium endophyte strains.

Figure 112 shows deletions in genome sequences of X-ray irradiated Neotyphodium endophyte strains.

Figures 113 shows numbers of SNPs in genic regions of genome sequences of X-ray irradiated Neotyphodium endophyte strains.

Figure 114 shows numbers of INDELs in genic regions of genome sequences of X30 ray irradiated Neotyphodium endophyte strains.

Figure 115 shows the spectrum of genome sequence changes (deletions) in genome

-452017251752 25 Oct 2017 sequences of X-ray irradiated Neotyphodium endophyte strains.

Figures 116 shows mutagenesis index of X-ray irradiated strains based on number of genome sequence changes observed in genome sequences of X-ray irradiated 5 Neotyphodium endophyte strains.

Figure 117 shows metabolic profiling of NEA12^dh Neotyphodium endophyte strains.

Figure 118 shows metabolic profiling of X-ray irradiated Neotyphodium endophyte 0 strains.

Figure 119 shows artificial seeds generated through Ca-alginate coating of perennial ryegrass embryos using a coating with Ca-alginate matrix without added nutrients.

Figure 120 shows Ca-alginate coating of perennial ryegrass embryos into artificial seeds using coating with coloured Ca-alginate matrix. Artificial seeds of perennial ryegrass coloured with Queen Green (90610); a) air-dried artificial seeds; b) artificial seeds plated on germination medium. Artificial seeds of perennial ryegrass coloured with Queen Pink (92330); c) air-dried artificial seeds; d) artificial seeds plated on 0 germination medium.

Figure 121 shows Ca-alginate coating of perennial ryegrass embryos into artificial seeds using coating with multiple Ca-alginate matrix layers, a) Artificial seeds of perennial ryegrass coated with first coating (non-coloured) Ca-alginate layer (layer A) 25 with added nutrients, b) Artificial seeds of perennial ryegrass coated with two (first layer A; non-coloured plus second layer B; Queen Green-coloured) Ca-alginate layers with added nutrients; c) double-coated artificial seeds placed on germination medium.

Figure 122 shows Ca-alginate coating of perennial ryegrass embryos into artificial seeds using coating with multiple Ca-alginate matrix layers, a) - c) Cross-sections of artificial seeds of perennial ryegrass coated with first coating (non-coloured) Ca-462017251752 25 Oct 2017 alginate layer (layer A) and second coating with Queen-Pink or Queen-Green coloured Ca-alginate layer (layer B). d) - e) Cross-sections of artificial seeds of perennial ryegrass coated with first coating (non-coloured) Ca-alginate layer (layer A) and second coating with Queen-Green coloured Ca-alginate layer (layer B).

Figure 123 shows germination of seeds, embryos and artificial seeds of perennial ryegrass cv. Bronsyn E- (endophyte free, 2668 seed batch), a) Original seeds:1% germination frequency on filter paper; b) Surface-sterilized seeds: 10% germination frequency on filter paper; c) Isolated embryos: 48% germination frequency on 0 germination medium; d) Artificial seeds (with germination medium): 40% germination frequency on MS medium.

Figure 124 shows Germination of seeds, embryos and artificial seeds of perennial ryegrass cv. Bronsyn E+ (endophyte plus, 2667 seed batch), a) Original seeds:10% 5 germination frequency on filter paper; b) Surface-sterilized seeds: 30% germination frequency on filter paper; c) Isolated embryos: 90% germination frequency on germination medium; d) Artificial seeds (with germination medium): 81% germination frequency on MS medium.

Figure 125 shows germination of artificial seeds and development of artificial-seed derived seedlings in perennial ryegrass.

Figure 126 shows freshly isolated seed-derived embryos of perennial ryegrass individually placed in wells of a) 96-well and b) endophyte mycelium suspension 25 added to individual wells and allowed to partly air-dry under laminar flow prior to c) production of artificial seeds coated with Ca-alginate layer.

Figure 127 shows artificial seeds produced by method 1.

Figure 128 shows germinating artificial seeds produced by method 1.

Figure 129 shows artificial seeds produced by method 2.

-472017251752 25 Oct 2017

Figure 130 shows artificial seeds produced by method 2 with endophyte outgrowth.

Figure 131 shows artificial seeds produced by method 3.

Figure 132 shows artificial seeds produced by method 3 with endophyte outgrowth.

Figure 133 shows germinating artificial seeds produced by method 3.

Figure 134 shows an outline of accelerated ageing.

Figure 135 shows seed germination rate of perennial ryegrass cv. Bronsyn with different endophytes after accelerated ageing treatment of seed followed by seed storage.

Figure 136 shows endophyte viability in seed from perennial ryegrass cv. Bronsyn with different endophytes after accelerated ageing treatment of seed followed by seed storage.

Figure 137 shows seed germination rates of symbiota representing different endophytes in different host genetic backgrounds after accelerated ageing treatment of seed.

Figure 138 shows endophyte viability of different endophytes in different host genetic 25 backgrounds after accelerated ageing treatment of seed.

Figure 139 shows endophyte viability of different endophytes in different host genetic backgrounds after accelerated ageing treatment of seed.

Figure 140 shows primers designed forTaqMan assay.

-482017251752 25 Oct 2017

Figure 141 shows TaqMan Primer Functionality for the Endophyte-Specific Gene LtmJ; RNA and DNA labelled.

Figure 142 shows TaqMan Primer Functionality for the Endophyte-Specific Gene 5 7490; RNA and DNA labelled.

Figure 143 shows TaqMan Primer Functionality for the Endophyte-Specific Gene 8263; RNA and DNA labelled.

Figure 144 shows TaqMan Primer Functionality for the Endophyte-Specific Gene 0005; RNA and DNA labelled.

Figure 145 shows TaqMan Primer Functionality for the Endophyte-Specific Gene 2232; RNA and DNA labelled.

Figure 146 shows TaqMan assay control - no template.

Figure 147 shows TaqMan assay control - Plant GAPDH. RNA and DNA labelled.

Figure 148 shows Detection in Co-Extracted DNA/RNA Pooled Samples for the Endophyte-Specific Gene 7490; RNA and DNA labelled.

Figure 149 shows Detection in Co-Extracted DNA/RNA Pooled Samples for the Endophyte-Specific Gene 0005; RNA and DNA labelled.

Figure 150 shows a control (no template) for Detection in Co-Extracted DNA/RNA Pooled Samples; RNA and DNA labelled.

Figure 151 shows detection in co-extracted DNA/RNA samples from individual 10 30 day old epicotyls of the endophyte-specific gene 2232; RNA and DNA labelled.

-492017251752 25 Oct 2017

Figure 152 shows detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls of the endophyte-specific gene 7490; RNA and DNA labelled.

Figure 153 shows detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls of the endophyte-specific gene 0005; RNA and DNA labelled.

Figure 154 shows a control (no template) for detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls; RNA and DNA labelled.

Figure 155 shows Detection in Co-Extracted DNA/RNA from 3, 5 and 7 Day-Old Pooled Epicotyls of the Endophyte Specific Gene 2232. Circles = Day 7 E+ Pool; Squares = Day 5 E+ Pool; Triangels = Day 3 E+ Pool; Diamonds = Day 7 E- Pool.

Figure 156 shows Detection in Co-Extracted DNA/RNA from 3, 5 and 7 Day-Old

Pooled Epicotyls of the Endophyte Specific Gene 7490. Circles = Day 7 E+ Pool;

Squares = Day 5 E+ Pool; Triangles = Day 3 E+ Pool; Diamonds = Day 7 E- Pool.

Figure 157 shows Detection in Co-Extracted DNA/RNA from 3, 5 and 7 Day-Old Pooled Epicotyls of the Endophyte Specific Gene 0005. Circles = Day 7 E+ Pool;

Squares = Day 5 E+ Pool; Triangles = Day 3 E+ Pool; Diamonds = Day 7 E- Pool.

Figure 158 shows metabolic profiling of population of symbiota to enable ab initio breeding and selection.

Figure 159 shows the top 20 peramine producing plants.

Figure 160 shows molecular phenotyping of a population of symbiota to enable ab initio breeding and selection.

Figure 161 shows the production of glucose-6-phosphate from glucose.

-50 2017251752 25 Oct 2017

Figure 162 shows the presence of water soluble carbohydrates in a population of symbiota.

Figure 163 shows quantification of proteins.

Figure 164 shows genetic diversity and relatedness between germplasm.

Figure 165 shows genetic diversity and relatedness between germplasm.

Figure 166 shows genotyping by sequencing.

Figure 167 shows procedure for genotyping by sequencing.

Figure 168 shows a generic scheme for a current commercial ryegrass breeding 5 program.

Figure 169 shows a schematic representation of relationships between requirements for numbers of individuals to be genotyped and phenotyped for genomic selection studies, and values of heritability for specific traits.

:0

Figure 170 shows a scheme for genomic selection implementation in a ryegrass breeding program.

Figure 171 shows the characterization of the microbiome in perennial ryegrass (Lolium perenne). Vertical axis shows % mapped sequences from cDNA libraries of perennial ryegrass-microbiome symbiota grown under full nutrient conditions foilwing mapping to non-plant database sequences in database 1. Horizontal axis shows samples analysed: L = leaf; R = root; FR = free of endophyte; ST = standard toxic endophyte.

-51 2017251752 25 Oct 2017

Figure 172 shows a 16S analysis of microbione in perennial ryegrass (Lolium perenne). The figure shows a taxonomic distribution of the microbes represented in the Lolium perenne microbiome.

Figure 173 shows leaf and root microbiomes in perennial ryegrass (Lolium perenne). The graph plots the sum by bacterial class of sequence read counts from leaves and roots, respectively.

Figure 174 shows leaf and root microbiomes in perennial ryegrass (Lolium perenne). 0 The figure shows a sample tree produced by hierarchical clustering of sample sets.

Figures 175 and 176 show leaf and root micobiomes in perennial ryegrass (Lolium perenne). The figures show two of eleven clusters containing more than 10 genes: Fig 175: Cluster 5 - low NH₄ induced; Fig 176: Cluster 3 - low N repressed.

Figure 177 shows leaf and root microbiomes in perennial ryegrass (Lolium perenne). The figures shows reads mapping to sequences annotated as Azospirillum species.

Figure 178 shows leaf and root micorbiomes in Antarctic hairgrass (Deschampsia 0 antarctica). The figure shows a taxonomic distribution of the microbes represented in the Deschampsia antarctica leaf and root microbiome.

Figure 179 shows leaf and root microbiomes in Antarctic hairgrass (Deschampsia antarctica). The figure shows a heat map displaying differences in bacterial species 25 predominance in shoots and roots.

Figure 180 shows root microbiomes in Antarctic hairgrass (Deschampsia antarctica). The figure shows a zoomed region of the heat map displaying differences in bacterial species predominance in response to differing nutritional status in root samples.

-52 2017251752 25 Oct 2017

Figure 181 shows leaf microbiomes in Antarctic hairgrass (Deschampsia antarctica). The figure shows a zoomed region of the heat map displaying differences in bacterial species predominance in response to differing nutritional status in leaf samples.

Figure 182 shows leaf and root microbiomes in Antarctic hairgrass (Deschampsia antarctica). The figure shows a zoomed region of the heat map displaying Azospirillum species clustered together.

Figure 183 shows endophyte suspensions at different dilution rates.

o

Figure 184 shows a genome wide association study for assessing effect of 624930 polymorphisms in the host (15 dairy cows) on their rumen microbiome profiles. Results for five contigs are shown. Grey indicates odd numbered chromosomes, black even numbered chromosomes.

The invention will now be described with reference to the following non-limiting examples.

Example 1 - A new paradigm in forage grass breeding :0

The Old Paradigm:

Breeding and selection of grass only followed by single endophyte inoculation and symbiota evaluation leading to synthetic grass varieties deploying a single unselected endophyte.

The New Paradigm:

Ab initio breeding and selection of grass-endophyte symbiota followed by symbiota evaluation leading to synthetic symbiota varieties deploying multiple endophytes.

For Application in Lolium/Festuca and Brachiaria/Urochloa.

Next Generation Ab Initio Molecular Breeding of Grass-Endophyte Symbiota.

2017251752 25 Oct 2017

Organism	Endophyte	Symbiota characteristic
Lolium perenne (perennial ryegrass)	Neotyphodium lolii	Reduced toxicity to grazing animals
Lolium arundinaceaum (tall fescue)	Neotyphodium coenophialum	Enhanced persistence
Brachiaria species	Acremonium implicatum	Biotic stress tolerance

-54 2017251752 25 Oct 2017

New Principles and Approaches

Extend Concept of Synthetic Varieties to Both Partners of the Symbiotum i.e. Grass Host and Endophyte

Deploy multiple endophyte and grass genotypes in populations selected for optimal 5 symbiota compatibility and performance

Capture Ab Initio Plant Genotype X Endophyte Genotype Effects

Breed and select ab initio symbiota for optimal symbiota compatibility and 0 performance rather than breed and select grass host only followed by endophyte inoculation and symbiota evaluation

Capture and Exploit Broader Endophyte Genotype Effects on Symbiota Performance

Exploit significant endophyte genotype effects on symbiota performance well beyond pest resistance (and reduced animal toxicosis)

Capture and Exploit Global Novel Endophyte Genetic Diversity [Step 1]

Deploy multiple novel endophytes ab initio rather than single branded endophytes late in varietal development process thus exploiting wide range of endophyte genotypic diversity [i.e. ‘endophyte germplasm collection’]

Generate and Exploit De Novo Endophyte Genetic Diversity [Step 2]

Generate de novo genetic variation [i.e. random mutagenesis (e.g. ionizing radiation); di-/poly-ploidisation (e.g. colchicine treatment); targeted mutagenesis (e.g. ExZactDelete and ExZact-Edit); cisgenesis; transgenesis; intragenesis; integrative genome editing (e.g. ExZact-Add)] in novel endophytes for enhanced endophyte trait 30 introgression in large-scale establishment of grass-endophyte symbiota [i.e. ‘breeding designer endophytes’]

-552017251752 25 Oct 2017

Capture Broad Genotypic Grass and Endophyte Diversity in Symbiota Populations for Ab Initio Symbiota Breeding [Step 3]

Establish large-scale symbiota populations involving 10s to 100s of endophyte 5 genotypes in 100s to 1000s of grass genotypes for ab initio breeding and selection of symbiota

Deploy method for large-scale endophyte inoculation based on artificial seeds generated from isolated grass embryos directly inoculated with different endophytes 0 followed by coating in alginate layers or coated by endophyte-containing alginate layer [i.e. ‘symbiota artificial seeds’]

For Brachiaria/Urochloa grass genotypes inoculate single endophytes and also coinoculate multiple endophytes in (particularly apomictic) grass host genotypes

Undertake Ab Initio Symbiota Breeding and Selection [Steps 4-8]

Select large-scale symbiota populations involving 10s to 100s of endophyte genotypes in 100s to 1000s of grass genotypes for compatibility and stability by applying accelerated ageing to symbiota artificial seeds and/or their offsprings as 0 selection tool [Step 4]

Select symbiota populations from Step 4 for viability by applying rapid endophyte viability assay [Step 5]

Subject symbiota populations from Step 5 to comprehensive, multi-year phenotyping for symbiota performance and alkaloid production [Step 6]

Select SynO symbiota from Step 6 for poly-crossing to generate synthetic symbiota experimental varieties [Step 7] followed by determination of endophyte incidence and 30 identity [Step 8]

-562017251752 25 Oct 2017

Next generation ab initio molecular breeding of grass-endophyte symbiota endophyte workflow

Endophyte Germplasm

Step 1: Discovery of global diversity in novel endophytes

Step 2: De novo generation of novel endophyte diversity

Establishing symbiota

Step 3: Large scale generation of grass-endophyte symbiota o

Selecting symbiota for stability, desired alkaloid profile and performance

Step 4: Selection tools for symbiota stability

Step 5: Rapid endophyte viability assay

Step 6: Multi-year phenotyping (incl molecular) of symbiota

Step 7: Selection of SynO symbiota parents for poly-crossing to generate experimental synthetic varieties for comprehensive evaluation

Step 8: Endophyte ID assay

Poly-crossing SynO symbiota parents

Step 9: Poly-crossing selected symbiota parents (SynO) with known endophyte identity and incidence and recovery of Syn1 seed

Confirming symbiota stability, alkaloid profile and endophyte identity and incidence

Step 10: Selection tool for symbiota stability

Step 11: Rapid endophyte viability assay

Step 12: Molecular phenotyping of symbiota*

Step 13: Endophyte ID assay*

Evaluating Syn1 symbiota

Step 14: Evaluation of symbiota (Syn1) experimental synthetic variety for agronomic performance

-572017251752 25 Oct 2017 *Assessment of alkaloid profile and endophyte identity and incidence on pooled samples at Syn1 generation

Poly-crossing Syn1 symbiota plants

Step 15: Poly-crossing advance of symbiota (Syn1) with known endophyte identity and incidence and recovery of Syn2 seed

Step 16: Selection tool for symbiota stability

Step 17: Rapid endophyte viability assay

Step 18: Molecular phenotyping of symbiota*

Step 19: Endophyte ID assay*

Evaluating Syn2 symbiota

Step 20: Evaluation of symbiota (Syn2) experimental synthetic variety for agronomic and animal performance *Assessment of alkaloid profile and endophyte identity and incidence on pooled samples at Syn2 generation

Example 2 - Tall fescue endophyte discovery

The objectives of this work on discovery and characterization of endophytes in tall fescue (Lolium arundinaceum) were:

1. Identification and characterisation of novel tall fescue endophytes for evaluation in germplasm.

2. Development and evaluation of optimised associations between novel endophytes and elite germplasm.

The endophyte discovery was based on screening 568 accessions to identify endophyte positive plants followed by genotyping 210 endophytes to identify novel endophytes in tall fescue.

-582017251752 25 Oct 2017

The characterisation in planta of novel endophytes from tall fescue was based on the following steps:

• Meristem cultures for tall fescue cultivars established for isogenic host panel • Endogenous metabolic profiles determined for 48 samples · Isolation of 38 endophytes was undertaken • Inoculation of 15-20 endophytes into isogenic host panel was undertaken • Isogenic host-endophyte associations were characterised

Genotypic analysis of endophyte content in accessions from a targeted fescue 0 germplasm collection

Initially, 472 accessions from 30 countries were tested for endophyte incidence; with 2 replicates of 6-10 seeds in each bulk per accession used in the analysis and endophyte incidence assessed with 6 SSRs.

New accessions were included in the analysis from the under-represented geographic origins; with a total of 568 accessions from 40 countries tested for endophyte incidence.

	Number of geographic origins	Percentage positive accessions
	FEtc collection	GRIN collection	FEtc collection	GRIN collection
Incidence assessment 01	7	23	96%	30%
Incidence assessment 02	-	10	-	45%

Table 1: Genotypic analysis of endophyte content in accessions from a targeted fescue germplasm collection

-592017251752 25 Oct 2017

Genotypic analysis of endophyte content in accessions from a targeted fescue germplasm collection is shown in Table 1. 233 endophyte positive accessions (41%) were detected. The geographical origins are represented in the endophyte incidence assessment.

A genetic diversity analysis of tall fescue endophytes is shown in Figure 2. A selected set of 210 accessions were used to assess genetic diversity of tall fescue endophytes. Genetic diversity was assessed with 38 SSR markers. Six different taxa were detected. The majority were N. coenophialum. Twenty were FaTG-2.

Six were putative FaTG-3. Thirteen were FaTG-3 like.

Diversity of host and endophyte is shown in Figure 3.

Selection of fescue-endophyte combinations for metabolic profiling, endophyte isolation and isogenic inoculation is shown in Figure 4. 52 accessions were initially selected for metabolic profiling and endophyte isolation. Endophyte presence was consistently detected in 25 accessions (red). An additional 48 accessions from underrepresented clusters were established in the glasshouse and screened for endophyte presence. 20 accessions were endophyte positive (blue) and have been selected for 0 further analysis.

Selection of fescue-endophyte combinations for metabolic profiling, endophyte isolation and isogenic inoculation is shown in Figure 5. Initial selections are shown in red. Additional selections are shown in blue.

The desired toxin profile of tall fescue endophytes is shown in Figure 6.

Example 3 - Metabolic profiling

The experimental design used for semi-quantitative metabolic profile analysis of tall 30 fescue-endophyte associations for the detection of alkaloid production in the endogenous host background is described below.

-602017251752 25 Oct 2017

Experimental design for semi-quantitative analysis of metabolites

Trim and re-pot the plants

0_

16h L: 21 C

8h D: 16°C, 2 Weeks

Trim 5 cm above ground ; Notes:!

¹I ii ! Light intensity: 620mmol m-1s-1;

¹I

Water supply: 50 ml/day!

'!

I Maintain 4 copies per accession in general ;

Lp«xi

Harvest 10 tillers (2-4 weeks age) o

16h L: 21 C

I

8h D: 16°C, ² Weeks

JL

Trim 5 cm above ground o A metabolic profile analysis for detection of ergovaline and peramine is shown in Figure 7.

Endophytes selected for semi-quantitative analysis of metabolites are shown in Figure 8.

Metabolic profile analysis for the detection of alkaloid production of different fescue endophytes

A metabolic analysis of tall fescue-endophyte associations for the detection of alkaloid production including loline, loline formate, peramine, ergovaline and lolitrem

B in the endogenous host background is shown in Figure 9. The alkaloid profile (i.e. lolines, peramine, ergovaline and lolitrem B) of tall fescue-endophyte associations in

-61 2017251752 25 Oct 2017 the endogenous host background for a range of endophyte strains belonging to different endophyte species is shown in Table 2.

Tall fescue accession details	Alkaloid profile
Tall fescue accession	Endophyte strain	Endophyte species	Lo lines	Peramine	Ergovaline*	Lolitrem B
BE9301	E34	N. coenophialum	+	+	L +	-
8PC	NEA13	N. coenophialum	n.d	+	+	n.d
FEtc7-180	NEA14	N. coenophialum	+	+	H +	-
FEtc7-58	NEA15	N. coenophialum	+	+	M +	-
FEtc7-342	NEA16	N. coenophialum	+	+	-	-
FEtc7-343	NEA20	N. coenophialum	+	+	-	-
234746	NEA22	N. coenophialum	+	+	M +	-
FEtc6-83	NEA24	N. coenophialum	+	+	H +	-
FEtc7-289	NEA25	N. coenophialum	+	-	H +	-
FEtc6-68	NEA26	N. coenophialum	+	+	+	-
FEtc6-85	NEA27	N. coenophialum	n.d	+	+	n.d
FEtc6-87	NEA28	N. coenophialum	n.d	+	+	n.d

-62 2017251752 25 Oct 2017

FEtc7-127	NEA29	N. coenophialum	+	+	+	-
FEtc6-128	NEA30	N. coenophialum	+	+	+	-
FEtc6-129	NEA31	N. coenophialum	+	+	+	-
287819	NEA17	FaTG-2	-	+	M +	-
231557	NEA21	FaTG-2	+	+	-	-
269850	NEA23	FaTG-3	+	+	-	-
231553	NEA19	Out group 1	-	-	-	-
FEtc6-75	NEA18	Out group 1	-	-	-	-
ST	ST	N. lolii	-	+	+	+
AR542*	AR542	N. coenophialum	+	+	-	-
KY31*	KY31	N. coenophialum	+	+	+	-
E77*	E77	N. coenophialum	+	+	+	-

Table 2 Alkaloid profile (i.e. lolines, peramine, ergovaline and lolitrem B) of tall fescue-endophyte associations in the endogenous host background for a range of endophyte strains belonging to different endophyte species (* Published data; nd = not determined).

Further metabolic analysis of the fescue endophytes is shown in Figure 10.

Example 4 - Semi-quantitative Analysis of Metabolic Profile under Temperature/Water Stress

In addition to the metabolic analysis of tall fescue-endophyte associations grown under standard conditions, for the detection of alkaloid production conferred by the endopohytes in the endogenous host background (Figures 7 - 10), a semi-632017251752 25 Oct 2017 quantitative analysis of metabolic profiles of tall fescue-endophyte associations grown under high temperature and water stress conditions was undertaken. Corresponding tall fescue-endophyte associations were grown under 16h Light and 30°C; 18h Dark and 20°C, and then sampled for alkaloid profile analysis as described 5 below:

• Harvest (control) -+ freeze dry -+ 50 mg pseudostem material -+ 80% methanol extraction -+ LCMS analysis • Recovery and water stress • Second harvest (stress) -+ freeze dry -+ SSR confirm all of the plant material o again.

This was performed in a controlled (growth chamber) environment simulating summer conditions, with light watering as required. Nine copies per accession were planted in general potting mix. A Randomized Complete Block with subsampling was 5 used.

Figure 11 shows a semi-quantitative analysis of metabolic profile of tall fescueendophyte associations grown under high temperature and water stress conditions.

Example 5- In planta isogenic inoculation in tall fescue with novel endophytes Summary:

A total of 36 fescue endophytes have been isolated from a range of fescue accessions from different geographic origin as described in Table 3, and found to 25 belong to different taxa as follows: 19 of them being N. coenophialum; 5 of them being FaTG-2; 3 of them being Outgroup; 3 of them being FaTG-3; 3 of them being FaTG-3 like; and 3 of them being N. uncinatum.

-64 2017251752 25 Oct 2017

Fescue Accession	Endophyte Strain	Origin	Cluster	Taxon	Fescue Accession	Endophyte Strain	Origin	Cluster	Taxon
1 8PC	8PC		C01.1	N. coenophialum	23 231557	NEA21	Morocco	C09	Fa TG-2
2 BE9301	E34		C01.1	N. coenophialum	24 287819	NEA17	Spain	C09	Fa TG-2
3 E77	E77		C01.2	N. coenophialum	25 598834		Morocco	C09	Fa TG-2
4 FEtc6-62		Catalunya (Spain) 4	C01.2	N. coenophialum	26 231559		Morocco	C09	Fa TG-2
5 FEtc6-68	NEA26	Catalunya (Spain) 14	C01.2	N. coenophialum	27 598852		Morocco	C09	Fa TG-2
6 FEtc7-127	NEA29	Aragon (Spain)14	C01.2	N. coenophialum	28 598934		Italy	C10	Outgroup
7 FEtc7-289	NEA25	Aragon (Spain)14	C01.2	N. coenophialum	29 231553	NEA19	Algeria	C10	Outgroup
8 FEtc7-58	NEA15	Aragon (Spain) 1	C01.2	N. coenophialum	30 FEtc6-75	NEA18	Sardegna (NW Italy) 5	C10	Outgroup
9 234746	NEA22	Spain	C01.2	N. coenophialum	31 269850	NEA23	Tunisia	C12	Fa TG-3
10 632582		Italy	C02.1	N. coenophialum	32 610918		Tunisia	C12	Fa TG-3
11 Kentucky 31	KY31		C02.1	N. coenophialum	33 610919		Tunisia	C12	Fa TG-3
12 FEtc6-128	NEA30	Pyrenees 13	C02.2	N. coenophialum	34 598829		Morocco	C13	Fa TG-3 like
13 FEtc6-129	NEA31	Pyrenees 17	C02.2	N. coenophialum	35 598863		Morocco	C13	Fa TG-3 like
14 FEtc7-180	NEA14	PaySardegna (Basque (FraniC02.2	N. coenophialum	36 598870		Morocco	C13	Fa TG-3 like
15 440364		Kazakhstan	C03	N. coenophialum	37 M311046		Russion Federation	C14	N. uncinatum
16 619005		China	C03	N. coenophialum	38 M595026		United Kingdom	C14	N. uncinatum
17 FEtc6-83	NEA24	Corsica (France)7	C04	N. coenophialum	39 M611046		Russion Federation	C14	N. uncinatum
18 FEtc6-85	NEA27	Corsica (France) 15	C04	N. coenophialum
19 FEtc6-87	NEA28	Corsica (France) 17	C04	N. coenophialum
20 AR542	AR542	Morocco	C05	N. coenophialum
21 FEtc7-342	NEA16	Gaurda (Portugal)	C06	N. coenophialum
22 FEtc7-343	NEA20	Gaurda (Portugal)	C06	N. coenophialum

Table 3 - Isolation of fungal cultures from selected fescue-endophyte combinations

Establishment of Meristem Cultures for Diverse Fescue Host Panel for In Planta Inoculation of Fescue Endophytes

Table 4 shows selected tall fescue and perennial ryegrass cultivars used to identify representative plant genotypes included in the diverse host panel for in planta 10 inoculation of fescue endophytes. All the selected plant genotypes have a high regeneration frequency of >80%.

-652017251752 25 Oct 2017

Cultivar	Genotype code	Species	Characteristics
Bariane	BARI 27	L. arundinaceum	Soft leaved, later maturing, highly palatable
Dovey	DOV 24	L. arundinaceum	High yielding, fast establishing
Quantum	QUAN 17	L. arundinaceum	Soft leaved with improved rust resistance
Jesup	JES01	L. arundinaceum	Cool season perennial forage
Bronsyn	BRO 08	L. perenne	Standard perennial ryegrass forage type

Table 4 - Selected tall fescue and perennial ryegrass cultivars used to identify representative plant genotypes included in the diverse host panel for in planta inoculation of fescue endophytes

Isolated fungal endophytes from endophyte-containing fescue accessions selected for in planta isogenic inoculation into the diverse host panel are shown in Figure 12. Figure 13 shows SSR-based genotyping of isolated endophyte cultures prior to in planta isogenic inoculation to confirm their identity.

Results from the SSR genotyping indicating the allele number and sizes for different

SSR markers for the different fescue endophyte strains are shown in Table 5.

-662017251752 25 Oct 2017

Endophyte Strain ID

Tall Fescue Accession ID

NCESTA1DH04 (FAM) NLESTA1TA10 (FAM) NCESTA1HA02 (HEX) NCESTA1CC10 (HEX)

		Allele 1	Allele 2	Allele 3	Allele 1	Allele 2	Allele 3 Allele 1	Allele 2	Allele 3	Allele 1	Allele 2	Allele 3
AR542	-	212	218	227	165	175	322	327	330	198	201	211
E34	BE 9301	212	218	224	165	175	322	329	330	198	201	211
E77	-	212	218	224	165	175	308	322	330	197	201	211
NEA13	8PC	212	218	224	165	175	322	330		197	200	210
NEA14	FEtc7-180	215	218	229	165	175	322	329	330	198	201
NEA15	FEtc7-58	212	218	224	165	175	322	329	330	197	201	211
NEA16	FEtc7-342	215	227		165	175	309	322	330	198	201	211
NEA17	287819	215	221	227	171	175	322			201	203
NEA18	FEtc6-75	218	227		171	175	304	322		201
NEA19	231553	221	227		171	175	304	325		201

Table 5 - Presence of alleles in endophyte strains

Results from the in planta isogenic inoculation into the diverse host panel of selected isolated fungal endophytes from endophyte-containing fescue accessions are shown in Table 6. Data on number of inoculations tested, number of successful inoculations and % of successful inoculations are provided in Table 6 to illustrate the inoculation ability of tall fescue endophytes in tall fescue and perennial ryegrass hosts.

A. Number of inoculations tested

	E77	E34	NEA13	NEA15	NEA14	AR542	NEA16	NEA17	NEA18	NEA19
	E77	BE9301	8PC	Fetc7-58	FEtc7-180	AR542	FEtc7-342	287819	FEtc6-75	231553	Total
BAR/ 27	23	25	30	34	38	38	24	32	40	27	311
BRO 08	39	31	24	27	35	36	30	33	48	22	325
DOV24	10	14	Nl	Nl	Nl	17	8	18	14	16	97
JESS 01	23	23	39	27	20	36	33	17	28	14	260
QUAN 17	8	31	20	15	17	21	18	16	15	8	169
Total	103	124	113	103	110	148	113	116	145	87	1162

B. Number of successful inoculations

	E77	E34	NEA13	NEA15	NEA14	AR542	NEA16	NEA17	NEA18	NEA19
	E77	BE9301	8PC	Fetc7-58	FEtc7-180	AR542	FEtc7-342	287819	FEtc6-75	231553	Total
BARI 27	3	3	4	0	1	11	3	17	18	2	62
BRO 08	0	0	2	0	2	0	0	4	2	5	15
DOV24	3	0	Nl	Nl	Nl	1	0	1	4	0	9
JESS 01	7	0	5	0	7	10	3	2	1	2	37
QUAN 17	3	0	1	0	0	0	0	6	5	3	18
Total	16	3	12	0	10	22	6	30	30	12	141

C. Percent of successful inoculations

	E77	E34	NEA13	NEA15	NEA14	AR542	NEA16	NEA17	NEA18	NEA19
	E77	BE9301	8PC	Fetc7-58	FEtc7-180	AR542	FEtc7-342	287819	FEtc6-75	231553	Total
BARI 27	13.0	12.0	13.3	0.0	2.6	28.9	12.5	53.1	45.0	7.4	18.8
BRO 08	0.0	0.0	8.3	0.0	5.7	0.0	0.0	12.1	4.2	22.7	5.3
DOV24	30.0	0.0	Nl	Nl	Nl	5.9	0.0	5.6	28.6	0.0	10.0
JESS 01	30.4	0.0	12.8	0.0	35.0	27.8	9.1	11.8	3.6	14.3	14.5
QUAN 17	37.5	0.0	5.0	0.0	0.0	0.0	0.0	37.5	33.3	37.5	15.1
Total	22.2	2.4	9.9	0.0	10.8	12.5	4.3	24.0	22.9	16.4	12.7
Cluster	1	1	1	1	2	3	3	7	8	8

Outgroup 1

Species coenophialum

FaTG-2

W.

Nl

Not inoculated

Table

- Inoculation

Ability of Tall Fescue

Endophytes in

Tall Fescue and

Perennial Ryegrass Hosts

-672017251752 25 Oct 2017

Example 6 - Endophyte Vegetative Stability in Tall Fescue and Perennial Ryegrass Host Genotypes

Following in planta isogenic inoculation with a range of selected isolated endophytes from fescue accessions, the endophyte vegetative stability of these endophytes in the different tall fescue and perennial host genotypes (i.e. BRO 08, BARI 27, DOV 24) was assessed, showing that:

• Several tall fescue endophytes (e.g. NEA17, NEA18, NEA19) were stable in perennial ryegrass (BRO08).

· BARI27 formed stable associations with all endophytes except for NEA15.

• NEA15 failed to form stable associations with any of host genotypes tested.

• DOV24 formed few stable associations.

The stability of these associations of novel tall fescue endophytes inoculated in different tall fescue and perennial ryegrass genotypes from the diverse host panel was assessed 12 months post-inoculation. Corresponding results are shown in Table

7.

-682017251752 25 Oct 2017

Plant Genotyp e	E7 7	E34	NEA1 3	NEA1 5	NEA1 4	AR54 2	NEA1 6	NEA1 7	NEA1 8	NEA1 9
E7 7	BE930 1	8PC	Fetc7- 58	FEtc7- 180	AR54 2	FEtc7- 342	28781 9	FEtc6- 75	23155 3
BARI 27	1/2	2/2	1/4	NA	1/1	7/7	1/1	1/2	8/10	1/1
BRO 08	NA	NA	0/1	NA	0/2	NA	NA	5/5	2/2	3/5
DOV 24	1/2	NA	Nl	Nl	Nl	0/1	NA	2/2	2/4	NA
JESS 01	5/5	NA	4/6	NA	5/6	5/10	2/3	0/1	0/1	3/3
QUAN 17	2/3	NA	0/1	NA	NA	NA	NA	3/6	3/5	1/2

Table 7 - Stability of associations of novel tall fescue endophytes (e.g. NEA13, NEA14, NEA15, NEA16, NEA17, etc.) inoculated in different tall fescue and perennial ryegrass genotypes (BARI 27, BRO 08, DOV 24, JESS 01 and QUAN 5 17) from the diverse host panel assessed 12 months post-inoculation. NA - not applicable, Nl - not inoculated, number of stable association/number of associations

Figure 14 shows stability at 12 months post inoculation of selected endophytes in tall 10 fescue and perennial ryegrass host genotypes from the diverse host panel.

The range of novel fescue endophytes selected for in planta isogenic inoculation is shown in Figure 15.

1. Table 8 shows additional novel tall fescue endophytes (e.g. NEA20, NEA21,

NEA22, etc.) selected for in planta isogenic inoculations in tall fescue genotypes (i.e. BARI 27, JESS 01 and QUAN 17) from the diverse host panel, based on the following selection criteria Produces little or no ergovaline

2. Produces no lolitrem B

-692017251752 25 Oct 2017

3. Produces lolines and/or peramine

	NEA2 0	NEA2 1	NEA2 2	NEA2 3	NEA2 4	NEA2 7	NEA3 0
FEtc7- 343	231557	234746	269850	FEtc6- 83	FEtc6- 85	FEtc6- 128
Neo	FaTG-3	Neo	FaTG-3	Neo	Neo	Neo
Lol/-/P/-	L0I/-/P/-	Lol/E/P/-	L0I/-/P/-	Lol/E/P/-	7/E/P/?	7/E/P/?
BARI 27	28	30	30	TBI	30	25	30
JESS 01	23	20	20	TBI	20	20	30
QUA N 17	30	30	40	TBI	30	35	25

Table 8 - Additional novel tall fescue endophytes (e.g. NEA20, NEA21, NEA22, etc.) selected for in planta isogenic inoculations in tall fescue genotypes (i.e. 5 BARI 27, JESS 01 and QUAN 17) from the diverse host panel. Neo = N.

coenophialum; ? = alkaloid profile not tested; TBI = To Be Inoculated.

Example 6 - Metabolic profiling of endophyte-tall fescue associations established following in planta isogenic inoculations of novel tall fescue 10 endophytes in tall fescue genotypes from the diverse host panel

Metabolic profiling of endophyte-tall fescue associations established following in planta isogenic inoculations of novel tall fescue endophytes in tall fescue genotypes from the diverse host panel is shown in Figures 16, 18 and 19. These figures:

· Compare semi-quantitative alkaloid profiles of selected endophytes across different isogenic hosts

-702017251752 25 Oct 2017 • Compare semi-quantitative alkaloid profiles for diverse endophytes in an isogenic host • Compare semi-quantitative alkaloid profiles of tall fescue and perennial ryegrass endophytes in the perennial ryegrass genotype Bro08

Figure 17 shows the presence of peramine and ergovaline in endophyte-tall fescue associations established following in planta isogenic inoculations of novel tall fescue endophytes in tall fescue genotypes from the diverse host panel.

Table 9 shows metabolic profiling of endophyte-tall fescue associations established following in planta isogenic inoculations of novel tall fescue endophytes in tall fescue genotypes from the diverse host panel. Confirmed endophyte positive (E+) plants were split to 5 replicates and regularly trimmed to promote tillering. Four months later E+ plants were re-potted in 12 replicates.

One month later E+ plants were re-potted if less than 9 positive copies were available at the time. Endophyte status was tested using SSR markers after each re-potting.

			Endophyte genotype
Host genotypes νεαιθ	NEA17	8PC	AR542 E34	E77 NEA18	NEA14	NEA16	NEA15
231553	287819		BE9301	FEtc6-75	Fetc7-180	Fetc7-342	Fetc7-58

Bariane (Bari27)

2/5

2/5

3/3

11/11 3/3

10/11 5/5

10/10 1/4

8/12 NA

9/14

5/5

12/12 3/4

5/12

1/4

1/6 16/25

Dovey (DOV 24)

NA

2/5

8/8 NA

NA

NA

3/5

6/12

3/5

3/12 NA

NA

NA

Jessup (JessOI)

2/4

4/8

NA

3/3

12/12 4/4

12/12 NA

2/3

8/11

NA

2/4

7/19

2/3

12/12 NA

Quantum (Quart 7)

2/5

8/7

4/5

12/12 NA

NA

4/5

12/12 2/4

5/12 NA

NA

NA

Bronsyn (Bro08)

9/9

10/11 5/5

11/12 1/9

0/8

NA

NA

3/4

7/7 0/5

NA

NA

Table 9 - Endophyte-tall fescue associations established following in planta 20 isogenic inoculations of novel tall fescue endophytes in tall fescue genotypes from the diverse host panel used for metabolic profiling.

A range of endophyte-tall fescue associations established following in planta isogenic inoculations of novel tall fescue endophytes in tall fescue genotypes from 25 the diverse host panel were selected for metabolic profiling (Table 9). In total, 29 isogenic host-endophyte associations were subject to LCMS analysis, following the experimental design described below.

-71 2017251752 25 Oct 2017

Experimental design • Trim and re-pot plants • 16h L, 30°C; 18h D, 20°C · Harvest (control) —» freeze dry -*· 50mg pseudostem material —> 80% methanol extraction -> LCMS analysis • Recovery and water stress • Second harvest (stressed) freeze dry —> 50mg pseudostem material 80% methanol extraction -> LCMS analysis.

o

This was performed in a controlled (growth chamber) environment simulating summer conditions, with light watering as required. Nine copies per accession were planted in general potting mix. A Randomized Complete Block with subsampling was used.

Example 8 - Bio-protective properties of fescue endophytes

Three fungal pathogens (i.e. Colletrotrichum graminicola, Drechslera brizae and

Rhizoctonia cerealis) - causing a range of fungal diseases and infecting a range 0 of different plant hosts - were included in antifungal bioassays used to analyse the potential anti-fungal activities of isolated fescue endophytes. Figure 20 shows results from anti-fungal bioassays of isolated fescue endophytes. Results of antifungal bioassays are also shown in Table 10. A range of endophytes were found to have high (H) and medium (M) antifungal activity (Table 10).

-722017251752 25 Oct 2017

Tall Fescue endophytes Antifungal activity against

1	Strain ID 440364	Accession	Taxon N. coenophialum	Colletotrichum graminicola Drechslera brizae Rhizoctonia cerealis
H	H	H
2	AR542	AR542	N. coenophialum	M	H	H
3	E34	BE9301	N. coenophialum	M	M	H
4	NEA13	8 PC	N. coenophialum	M	H	H
5	NEA14	FEtc7-180	N. coenophialum	M	M	H
6	NEA15	FEtc7-58	N. coenophialum	M	H	H
7	NEA16	FEtc7-342	N. coenophialum	M	H	H
8	NEA22	234746	N. coenophialum	H	M	M
9	NEA27	FEtc6-85	N. coenophialum	L	M	L
10	NEA30	FEtc6-128	N. coenophialum	M	H	H
11	E1		Non-M lolii	L	L	M
12	NEA18	FEtc6-75	Outgroup 1	M	H	H
13	598852		FaTG-2	M	H	H
14	610918		FaTG-3	M	H	H
15	NEA21	231557	FaTG-3	M	H	M
16	598829		FaTG-3 like	M	L	M

Antifungal activity: Low, Medium, High

Table 10 - Anti-fungal bioassays of isolated novel fescue endophytes

Example 9 - Genome survey sequencing of novel tall fescue endophytes

A range of novel tall fescue endophtyes were subjected to genome survey sequencing (GSS).

Figure 21 shows a strategy for GSS of selected novel fescue endophytes. The alkaloid profiles of novel fescue endophytes subjected to GSS analysis are shown in 10 Table 11.

-732017251752 25 Oct 2017

Tall fescue accession details	Alkaloid profile in Endogenous Host
Endophyte strain	Accession No/isolated ID	Endophyte species	Lolines	Peramine	Ergovaline	Lolitrem B
E34	BE9301	N. coenophialum
NEA13	8PC	N. coenophialum
NEA14	FEtc7-180	N. coenophialum
NEA15	FEtc7-58	N. coenophialum
NEA16	FEtc7-342	N. coenophialum
NEA20	FEtc7-343	N. coenophialum
NEA22	234746	N. coenophialum
NEA24	FEtc6-83	N. coenophialum
NEA17	287819	FaTG-2
NEA21	231557	FaTG-3
NEA23	269850	FaTG-3
NEA19	231553	non- Epichloe out-group
NEA18	FEtc6-75	non- Epichloe out-group
AR542*	AR542*	N. coenophialum
E77*	E77*	N. coenophialum
598852	598852	FaTG-2
AR501*	AR501*	FaTG-3
598829	598829	FaTG-3 like
E81	E81	N. uncinatum
9340	9340	E. typhina
9707	9707	E. baconii

Table 11 - Alkaloid profiles of sequenced endophytes. Green: alkaloid present, Yellow: Alkaloid absent, Grey: alkaloid profile not determined * Profiles are taken from published data

Figure 22 shows the peramine biosynthetic pathway. PerA encodes a single multifunctional enzyme that catalyses all the biosynthetic steps. GenBank accession Number: AB205145. The presence of the perA gene in non-Epichloe out-group

-74 2017251752 25 Oct 2017 endophytes is shown in Figure 23.

Figure 24 shows the ergovaline biosynthetic pathway. Genes in the eas gene cluster which are involved in ergovaline biosynthesis are shown in Figure 25 and Table 12.

The dmaW gene encodes DMAT synthase enzyme, which catalyzes the first committed step in ergovaline biosynthesis. Presence of the dmaW gene in novel fescue endophytes is shown in Figure 26 and presence of the eas gene cluster in novel fescue endophytes is shown in Figure 27.

Gene Cluster	Gene	GenBank Accession No
	dmaW	AY259838
	easA	EF125025
	easE	EF125025
eas gene cluster	easF	EF125025
	easG	EF125025
	easH	EF125025
	IpsA	AF368420
	IpsB	EF125025

Table 12 - Genes in the eas cluster

Figure 28 shows the Lolitrem B biosynthetic pathway. Genes in the gene cluster which are involved in Lolitrem B biosynthesis are shown in Figure 29 and Table 13.

Presence of gene cluster 1 (ItmG, ItmM and ItmK) in endophytes is shown in Figure 15 30, presence of gene cluster 2 (ItmB, ItmQ, ItmP, ItmF and ItmC) is shown in Figure and presence of gene cluster 3 (ItmE and ItmJ) is shown in Figure 32.

-752017251752 25 Oct 2017

Gene Cluster	Gene	GenBank Accession No
	ItmG	AY742903
gene cluster 01	ItmM	AY742903
	ItmK	AY742903
	ItmB	DQ443465
	ItmQ	DQ443465
gene cluster 02	ItmP	DQ443465
	ItmF	DQ443465
	ItmC	DQ443465
gene cluster 03	ItmJ	DQ443465
ItmE	DQ443465

Table 13 - Genes in the gene cluster involved in Lolitrem B biosynthesis

Figure 33 shows the Loline biosynthetic pathway. Genes in the gene cluster which are involved in Loline biosynthesis are shown in Figure 34 and Table 14. Presence 5 of Loline biosynthetic gene cluster in novel fescue endophytes is shown in Figure 35.

Gene Cluster	Gene	GenBank Accession No
	lolF	EF012269
	lolC	EF012269
	lolD	EF012269
	lolO	EF012269
LOL gene cluster	lolA	EF012269
	lolU	EF012269
	lolP	EF012269
	lolT	EF012269
	lolE	EF012269

Table 14 - Genes in the Loline biosynthetic gene cluster

Figure 36 and shows an alkaloid biosynthetic gene analysis for endophyte strain

NEA23. Tables 15 and 16 show alkaloid biosynthetic gene analyses for various endophyte strains. Table 15 shows results from the assessment of alkaloid

-762017251752 25 Oct 2017 biosynthetic gene presence/absence for different endophytes by mapping genome survey sequence reads corresponding to the different alkaloid biosynthetic genes/gene clusters.

-772017251752 25 Oct 2017

I

-792017251752 25 Oct 2017

Table 16 shows results from the assessment of alkaloid biosynthetic gene presence/absence for different endophytes by mapping genome survey sequence reads corresponding to the different alkaloid biosynthetic genes/gene clusters as 5 well as corresponding alkaloid profile observed for corresponding tall fescueendophyte associations.

-802017251752 25 Oct 2017

Tall fescue accession details	Alkaloid profile and Gene presence
Endophyte strain	Accession No/isolated ID	Endophyte species	Lolines	Peramine	Ergovaline*	Lolitrem B
E34	BE9301	N. coenophialum	G+	G+	G+	PG+
8PC	8PC	N. coenophialum	G+	G+	G+	PG+
NEA14	FEtc7-180	N. coenophialum	G+	G+	G+	PG+
NEA15	FEtc7-58	N. coenophialum	G+	G+	G+	PG+
NEA16	FEtc7-342	N. coenophialum	G+	G+	G-	PG+
NEA20	FEtc7-343	N. coenophialum	G+	G+	G-	PG+
NEA22	234746	N. coenophialum	G+	G+	G+	PG+
NEA24	FEtc6-83	N. coenophialum
NEA17	287819	FaTG-2	G-	G+	G+	G+
NEA21	231557	FaTG-3	G+	G+	G-	G-
NEA23	269850	FaTG-3	G+	G+	G-	G-
NEA19	231553	non- Epichloe out-group	G-	G+	G-	G-
NEA18	FEtc6-75	non- Epichloe out-group	G-	G+	G-	G-
AR542*	AR542*	N. coenophialum	G+	G+	G-	PG+
E77*	E77*	N. coenophialum
598852	598852	FaTG-2
AR501*	AR501*	FaTG-3	G+	G+	G-	G-
598829	598829	FaTG-3 like
E81	E81	N. uncinatum
9340	9340	E. typhina
9707	9707	E. baconii	G-	PG+	G-	G-

Table 16 - Alkaloid biosynthetic gene analysis. Green: alkaloid present, Yellow: Alkaloid absent, Grey: alkaloid profile not determined, *Profiles are taken from published data, G+ = gene/gene cluster present, G- = gene/gene cluster absent, PG+ = gene/gene cluster partially present

-81 2017251752 25 Oct 2017

Table 17 shows novel fescue endophytes (NEA16, NEA18, NEA19, NEA20, NEA21 and NEA23) with favourable toxin profiles.

Tall fescue accession	Taxon	Alkaloid profile (Lol/P/E/L)	Antifungal
NEA21 (231557)	FaTG-3	+/+/-/-	High
NEA23 (269850)	FaTG-3	+/+/-/-	To be tested
AR501*	FaTG-3	+/+/-/-	-
NEA18 (FEtc6- 75)	Non-Epichloe Outgroup	-/-/-/-	High
NEA19 (231553)	Non-Epichloe Outgroup	-/-/-/-	To be tested
NEA16 (FEtc7- 342)	N. coenophialum	+/+/-/-	High
NEA20 (FEtc7- 343)	N. coenophialum	+/+/-/-	To be tested
AR542*	N. coenophialum	+/+/-/-	High

Table 17 - Novel fescue endophytes (NEA16, NEA18, NEA19, NEA20, NEA21 and NEA23) with favourable toxin profiles and antifungal activities observed in bioassays. * Control commercial endophyte

A genotypic analysis of the novel fescue endophytes NEA23 and NEA21 is shown in Figure 37.

-82 2017251752 25 Oct 2017

Example 10 - Perennial Ryegrass Endophyte Discovery and Characterisation

The objectives were:

1. Identification and characterisation of novel perennial ryegrass endophytes for evaluation in germplasm.

2. Development and evaluation of optimised associations between novel endophytes and elite germplasm.

Experimental strategies

1. Identification and characterisation of novel perennial ryegrass endophytes for 0 evaluation in proprietary germplasm

Activity: Establishment of a proprietary ‘library’ of novel endophyte strains • Targeted germplasm collection • Genotypic analysis of germplasm samples • Metabolomic analysis of germplasm samples · Fungal culture isolation and inoculation • Metabolomic analysis of inoculated plants • Assessment of endophyte stability

2. Development and evaluation of optimised associations between novel endophytes 0 and elite germplasm

Activity: Delivery of novel endophytes into the germplasm improvement process • Large-scale inoculation of selected endophytes • Genetic analysis of selected host-endophyte associations • Phenotypic analysis of selected host-endophyte associations

The desired toxin profile of perennial ryegrass endophytes is shown in Figure 6.

The outcome requirements for endophyte stability are as follows: Intergenerational stability • Maximal < 5% loss per generation • Ideally 2-3% loss per generation

Seed storage stability • 3 years in seed

-832017251752 25 Oct 2017

Establishment of ryegrass germplasm collection for endophyte discovery identification and establishment of 244 accessions.

Figure 40 shows genotypic analysis of endophyte content in accessions from a targeted ryegrass germplasm collection.

Figure 41 shows analysis of endophyte content in accessions from a targeted ryegrass germplasm collection. Genetically unique; toxin production within required 0 limits based on genotypic prediction.

Isolation of fungal cultures from selected ryegrass-endophyte associations • In vitro cultures of candidate endophytes established and genotypes confirmed • Long-term cryopreservation of endophyte cultures established

-84 2017251752 25 Oct 2017

Species	No. Isolates	Examples
N. Ioiii	57	ST, NEA3, NEA2, 38 novel
N. coenophialum	37	KY31, AR542 and 11 other AR strains, E34 and 5 other RBG strains, 19 novel
LpTG-2	6	AR6, NEA4, NEA11
Non-N. ioiii	6	NEA12, E1
FaTG-2	6	8907 and 5 novel
FaTG-3	8	AR501 and 4 other AR strains, 3 novel
Other fescue endophytes	6	3 novel FaTG3-like, 3 Outgroup
N. uncinatum	4	E81 and 3 novel
Total	130

Table 18 - Isolation of fungal cultures from selected ryegrass-endophyte associations

Example 11 - Isogenic inoculation of perennial ryegrass endophytes

Cultivar	Species	Characteristics
Barsandra	L. perenne	Turf type
Barsintra	L. perenne	Tetrapioid forage type
Bealey	L. perenne	Tetrapioid forage type
Bronsyn	L. perenne	Standard forage type
Impact	L. perenne	Late flowering, dense tillering forage type
Meridian	L. perenne	Early flowering forage type
Tolosa	L. perenne	Distinct forage type
Barfest	Festulolium

Table 19 - Establishment of meristem cultures for a diverse perennial ryegrass host panel

-852017251752 25 Oct 2017

Figure 42 shows an isogenic host plant genotype cultivar ‘likeness’ determination • Bronsyn and Tolosa (96 individuals) were assessed for genetic diversity using 58 perennial ryegrass SSR markers • Bronsyn and Tolosa cultivars easily discriminated

Figure 43 shows an isogenic host plant genotype cultivar ‘likeness’ determination.

Impact, Bealey and Barsandra genotypes were assessed for genetic diversity using 56 perennial ryegrass derived SSR markers.

Figure 44 shows development of a host-panel molecular genetic test for QC of hostendophyte associations.

• A panel of 5 perennial ryegrass SSR markers has been identified to distinguish the different host-panel genotypes • Full QC capabilities for isogenic host plant as well as endophyte

Figure 45 shows inoculation of candidate endophytes into meristem cultures of diverse ryegrass host panel.

Regeneration of Mature Isogenic Ryegrass- Endophyte Associations · Quantitative score used to assess endophyte inoculation frequency • 3 diagnostic SSR markers are used to determine endophyte presence and identity and samples are scored on a scale of 0-3 • Quantitative score is also used for seed purity testing and endophyte presence and identity testing

-862017251752 25 Oct 2017

Quantitative score	Aiieies present and of correct size for given SSR loci
3	Endophyte present
2	Endophyte present
1	Endophyte absent
0	Endophyte absent

Table 20 - Quantitative score used to assess endophyte inoculation frequency • Average inoculation frequency of 15% across all endophyte-host combinations

Inoculation success rates

	15285	NEA12	15441	15931	AR1	NEA11	F2	NEA10	STWT	Average
Bea02	6.7	15.0	16.7	4.8	0.0	66.7	0.0	0.0	32.0	15.8
Bro08	8.3	5.0	21.1	0.0	5.0	93.3	0.0	0.0	52.0	20.5
Fest02	4.8	0.0	33.3	0.0	0.0	31.8	0.0	0.0	17.4	9.7
lmp04	0.0	19.0	3.8	0.0	0.0	81.6	0.0	9.7	45.7	17.8
Mer05	NA	4.5	16.0	0.0	4.0	62.5	0.0	8.7	8.0	11.5
San02	0.0	25.0	40.0	8.7	0.0	43.6	0.0	0.0	34.4	16.9
Sin04	0.0	4.3	44.4	0.0	0.0	45.5	0.0	0.0	55.6	16.6
Tol03	9.1	11.1	5.9	0.0	0.0	42.9	3.8	0.0	27.3	11.1
Average	3.6	10.5	22.7	1.7	1.1	58.5	0.5	2.3	34.0	15.3

Table 21 - Percent of successful inoculations

	15285	NEA12	15441	15931	AR1	NEA11	F2	NEA10	STWT	Total
Bea02	1	3	4	1	0	12	0	0	8	29
Bro08	2	1	4	0	1	14	0	0	13	35
Fest02	1	0	8	0	0	7	0	0	4	20
lmp04	0	4	1	0	0	40	0	3	16	64
Mer05	NA	1	4	0	1	5	0	2	2	15
San02	0	6	10	2	0	17	0	0	11	46
Sin04	0	1	4	0	0	5	0	0	10	20
Tol03	1	2	1	0	0	3	1	0	3	11
Total	5	18	36	3	2	103	1	5	67	240
Table 22	- Number of successful inoculations
	15285	NEA12	15441	15931	AR1	NEA11	F2	NEA10	STWT	Total

	Bea02	15	20	24	21	12	18	22	24	25	181
	Bro08	24	20	19	20	20	15	25	19	25	187
	Fest02	21	19	24	18	19	22	20	9	23	175
	lmp04	25	21	26	25	25	49	44	31	35	281
	Mer05	0	22	25	24	25	8	23	23	25	175
	San02	24	24	25	23	24	39	49	47	32	287
	Sin04	7	23	9	18	13	11	24	19	18	142
	Tol03	11	18	17	15	19	7	26	17	11	141
10	Total	127	167	169	164	157	169	233	189	194	1569

Table 23 - Total number of inoculations tested

-872017251752 25 Oct 2017

Figure 46 shows the endophytes selected.

Endophyte vegetative stability in priority ryegrass host panel genotypes • Plants were re-sampled 6-12 months (top line) after initial inoculation and again after 2 years and 3 years (bottom line) after initial inoculation • NEA11 endophyte highly stable across all host plants • NEA12 exhibits varying degrees of stability

Plant genotype	Endophyte genotype
ST	NEA10	NEA11	NEA12
Impact	9/10	2/3	12/12	1/4
(lmp04)	3/3	2/2	3/3	1/1
Barsandra	4/6		7/7	2/4
(San02)	2/3	NA	3/3	1/2
Tolosa	1/2		3/3	2/2
(Tol03)	1/1	NA	2/3	2/2
Bealey	3/3		9/9	0/2
(Bea02)	2/3	NA	3/3	0/1
Bronsyn	3/6		9/9	1/1
(Bro08)	2/2	NA	4/4	0/1

o Table 24 - Endophyte vegetative stability in priority ryegrass host panel genotypes

-882017251752 25 Oct 2017

Perennial Ryegrass Isogenic Inoculations

Endophytes NEA10 and NEA12 are recalcitrant to inoculation, with low frequencies of successful inoculation

NEA10 is only compatible with Impact and Bronsyn

NEA12 is only compatible with Impact, Barsandra and Tolosa

A· Number of inoculations performed

	ST	NEA10	NEA11	NEA12	Total
Bea02		40		30	70
Bro08		80		75	155
lmp04		90		50	140
San02		80		50	130
Tol03		80		40	120
Total	0	370	0	245	615

B. Number of inoculations tested

	ST	NEA10	NEA11	NEA12	Total
Bea02		31		21	52
Bro08		59		50	109
lmp04		60		21	81
San 02		64		31	95
Tol03		32		27	59
Total		246		150	396

C. Number of successful inoculations

	ST	NEA10	NEA11	NEA12	Total
Bea02		0		1	1
Bro08		1		0	1
lmp04		1		2	3
San 02		0		1	1
Tol03		0		2	2
Total		2		6	8

D. Percent of successful inoculations

	ST	NEA10	NEA11	NEA12	Total
Bea02		0		1	1.0
Bro08		1.7		0	1.7
lmp04		1.7		9.5	11.2
San 02		0		3.2	3.2
Tol03		0		7.4	7.4
Total		3.4		21.2	24.5

Table 25 - Perennial Ryegrass Isogenic Inoculations

-892017251752 25 Oct 2017

Stable associations previously formed Unstable association

Endophytes NEA3, NEA2 and E34 have not been successfully inoculated

E1 is broadly compatible

A. Number of inoculations performed

	NEA3	NEA2	E1	E34	Total
Bea02	30	30	55	26	141
Bro08	80	80	30	Nl	190
lmp04	80	80	30	15	205
San02	80	80	55	30	245
Tol03	80	80	80	40	280
Total	350	350	250	111	1061

B. Number of inoculations tested

	NEA3	NEA2	E1	E34	Total
Bea02	17	18	19	26	80
Bro08	39	43	18	Nl	100
lmp04	71	58	12	12	153
San02	51	55	8	19	133
Tol03	37	41	17	26	121
Total	215	215	74	83	587

C. Number of successful inoculations

	NEA3	NEA2	E1	E34	Total
Bea02	0	0	9	0	9
Bro08	0	0	13	Nl	13
lmp04	0	0	10	0	10
San02	0	0	6	0	6
Tol03	0	0	10	0	10
Total	0	0	48	0	48

D. Percent of successful inoculations

	NEA3	NEA2	E1	E34	Total
Bea02	0	0	47.4	0	11.8
Bro08	0	0	72.2	Nl	24.1
lmp04	0	0	83.3	0	20.8
San02	0	0	75.0	0	18.8
Tol03	0	0	58.8	0	14.7
Total	0	0	64.9	0	16.2

Nl not inoculated

Table 26 - Perennial Ryegrass Isogenic Inoculations

-902017251752 25 Oct 2017

Vegetative stability of E1 endophyte in priority ryegrass host panel genotypes

Plant Genotype	Endophyte E1	No. of dead plants after resampling
Impact (lmp04)	10/12 1/5	0
Barsandra (San02)	6/8 5/5	0
Tolosa (Tol03)	10/17 7/8	5
Bealey (Bea02)	9/19 8/8	2
Bronsyn (Bro08)	13/18 0/8	0

Table 27 - Vegetative stability of E1 endophyte in priority ryegrass host panel genotypes. Plants were re-sampled 6-12 months after initial inoculation (bottom 5 line).

Summary of Endophyte Stability in Isogenic Host Associations • Host and endophyte specific effects were observed

- Identified stable host-endophyte associations for novel endophytes, establishing designer associations

e.g. Impact-NEA10; Tolosa - NEA12; Barsandra-E1 • Significant genotype x genotype (G x G) effects highlight the importance of breeding for symbiota

- Bronsyn and Impact exhibited higher inoculation success rates - better range of compatibility

- E1 exhibits high rates of successful inoculation and has intermediate compatibility

- NEA11, NEA10 and ST endophytes are highly stable over time

-91 2017251752 25 Oct 2017

- Vegetative stability issues with NEA12 • Emphasis on maintaining and characterising highly valuable associations

Example 12 - Metabolic profiling of isogenic host - endophyte associations

Endophyte	Endogenous Toxin Profile	Confirmed Toxin Profile	Origin	Species
ST	L/E/P	Y (L7E/P)		N. lolii
AR1	-/-/P	N.A		N. lolii
15285	La/-/P	Y (L7-/P)	Portugal	N. lolii
15441	-/-/-	Nc	Italy	N. lolii
15931	-/E/P	N.A	Former Soviet Union	N. lolii
01030 44 F2	-/E/n,db	N.A	Spain	N. lolii
NEA10	-/E/n.d	Y (-/E/P)	Spain	N. lolii
NEA11	-/E/n.d	Y (-/E/P)	France	LpTG-2
NEA12	-/-/-	Y (-/-/-)	France	Non-N.lolii

^a Negligible levels of lolitrem B ^b Peramine not measured in Agriseeds samples ^c Differential response: Bealey, Meridian and Source plant Barsandra and Barfest L/E/P. Not selected for further analysis

N.A No stable isogenic host-endophyte associations formed

Table 28 - Metabolic profiling of isogenic host - endophyte associations

In-depth metabolic profiling of novel isogenic host-endophyte associations

Figure 47 shows a detailed characterisation of the known knowns and their precursors.

Procedure:

1. Establish methodologies for semi-quantitative analysis of known alkaloids (LEPJ)

2. Compare semi-quantitative alkaloid (LEPJ) profiles of E⁺ versus E‘ host plants

-922017251752 25 Oct 2017

3. Compare semi-quantitative alkaloid profiles for diverse endophytes in an isogenic host

4. Assess stability of alkaloid (LEPJ) profile of selected endophytes across different isogenic hosts

Host genotypes were confirmed using the host-panel identification test (Figure 48). Ten replicates of each host-endophyte association were tested.

• Two confirmed endophyte positive plants split into 4 replicates • Plants regularly trimmed to promote tillering · 2 years later (first column of table 29) - Plants re-potted to 10 replicates • 2 months later (second column of table 29) - Endophyte positive plants repotted to 15 replicates • Endophyte status was tested using SSR markers

Host genotype Endophyte genotype

ST NEA10 NEA11 NEA12 E1

Impact (lmp04)	6/10	7/15 10/10 15/15	8/8	7/7	9/10	11/15 10/10 10/15
Basandra (San02)	7/10	10/15	5/10	13/15	5/10	14/22
Tolosa (Tol03)	6/10	11/15	9/10	11/15	8/10	11/15
Bealey (Bea02)	6/10	11/15	6/10	15/15
Bronsyn (Bro08)	9/10	11/15	8/10	11/15

Table 29 - Endophyte status

Figures 49 and 50 show in-depth metabolic profiling of novel isogenic hostendophyte associations.

• Ten replicates per host-endophyte association (20*10 = 200 samples).

· Complete randomised block design • Two organs (pseudostem and leaves) were harvested following 6 weeks growth under controlled conditions • After 6 weeks regrowth there was a second harvest • Water supply: 50 ml /day · 14 hours light (620 mmolm ²s¹ light intensity) /21 °C • 10 hours dark/16 °C

-932017251752 25 Oct 2017

Outcome:

To determine the pattern of metabolic variation between endophytes, host plants and organ types.

Sample preparation —> LC/MS analysis —> LEPJ alkaloid profiles

QQQ analysis Alkaloid pathway analysis

Alkaloid isolation and quantitative characterisation

Current Study:

Freeze dried pseudostem material

50mg of freeze dried powder

Extracted twice in 1ml of methanokwater (80:20, v:v)

Figure 51 shows LEPJ: the known knowns.

Figure 52 shows accurate mass of LEPJ.

RT(min)	m/z	Accurate mass	MS2	Identity
3	248(H+)	248.15022	206,231,175	Peramine
4.7	534(H+)	534.27002	223,268	Ergovaline
10.3	686(H+)	686.40369	628,238	Lolitrem B
10.3	646(H+)	646.37238	280, 222, 588.33	Janthitrem

Table 30 - accurate mass of LEPJ

Figures 53 and 54 show identification of LEPJ.

Figure 55 shows quantification of LEPJ.

Figure 56 shows a comparison of semi-quantitative alkaloid (LEPJ) profiles of E⁺ versus E' host plants.

Figure 57 shows a comparison of semi-quantitative alkaloid profiles for diverse endophytes in an isogenic host (lmp04).

-94 2017251752 25 Oct 2017

Figure 58 shows an assessment of stability of alkaloid (LEPJ) profile of selected endophytes across different isogenic hosts.

• Significant genotype x genotype (G x G) effects highlight the importance of breeding for symbiota • Specific LEPJ alkaloid production varies between different endophytes inoculated into the same isogenic host • e.g. NEA11 inoculated into Impact (lmp04) produces more peramine and less ergovaline than NEA10 inoculated into the same host genotype • LEPJ alkaloid production varies between different host plants harbouring the same endophyte indicating host genotype effects • e.g. ergovaline was not detected in Bealey (Bea02) and Barsandra (San02) inoculated with ST and NEA11 and was detected in relatively high quantities in Bronsyn (Bro08) inoculated with the same endophytes

Assessment of an endophyte that matches the specifications and is stable across germplasm pools.

Genic content can be used to predict presence/absence of lolitrem B

Semi-quantitative LC-MS can be used to assess relative amounts of LEPJ across germplasm pools • NEA12 forms stable associations with some germplasm backgrounds, produces J across the hosts assessed, does not produce LEP and exhibits broad spectrum antifungal activity • NEA11 is a highly compatible endophyte that produces relatively less P, and more E than ST across a range of host backgrounds, but no LJ • E1 is a highly compatible endophyte that does not produce LEPJ in lmp04, but does produce J in Bro08 • NEA10 forms stable associations within a limited host background and produces more E and less P than NEA11 (and by inference ST)

-952017251752 25 Oct 2017

					N. lolii			LpTG-2	non-NJolii	E festucae
Gene	GenBank Accession No	ST	Lp19	AR1	NEA3	E9	G4	NEA10 NEA11	NEA12 E1	E2368
Metabolite production in	L	L		—	L	L	_ —	TBD

planta

Lolitrems	ItmB	00443465
	ttmQ	00443465
	ftmP	00443465
	ItmJ	DO443465
	ttmE	00443465
	ItmF	00443465
	ItrnC	00443465
	ItmG	AY742903
	ItmM	AY742903
	ItmK	AY742903
Metabolite production in planta Peramine	perA	AB205145
Metabolite production in planta Ergot Alkaloids	dmaW	AY259837
	Ip&A	AF368420
	IpsB	EF125025
	easA	EF125025
	easE	EF125025
	easF	EF125025
	easG	EF125025
	easH	EF125025
Metabolite production in planta Loline Alkaloids	IdF	AY723749
	lotC	AY723749
	IdD	AY723749
	IdO	AY723749
	IdA	AY723749
	IdU	AY723749
	IdP	AY723749
	IdT	AY723749
	IdE	AY723749

TBD

Table 31 - Semi-quantitative LC-MS used to assess relative amounts of LEPJ across germplasm pools

Pathway analysis • Lolitrem B and ergovaline biosynthetic pathways are known in Λ/. lolii • Perform pathway analysis to determine if gene deletions result in changes in production of precursors of known knowns (LPEJ) and direct flux to synthesis of other known toxins

Figure 59 shows lolitrem B biosynthesis.

2017251752 25 Oct 2017 <o σ>

2017251752 25 Oct 2017

CH

re o

Φ o

o o

re o re

o		to	to	IO	to	to
CM	CM	CM	CM	CM	CM	CM
X	s	S5	s	s	8	s?
	CM	CM	CM	CM	CM	CM
co
	u_	LL	LL	LL	U.	U-
	LU	LU	LU	LU	LU	LU

co co

Φ

Table 34 - pathway analysis Lolitrem B m CM

CM θ

Φ a:

σ> σ>

V)

Φ to Φ

Φ £ e $ QC ¢0

Κί φ g

O) .2

I s> is

LO

CM O CXI

LO co

CO co co co

O

Z

CO in

Ο)

Φ (Λ

Φ (0

CO

CM

CO

CM

CM

CO

CM CM

CD

LO

CO CD

LO LO 04

O z

LO

LO LO 04

O

Φ (0

LO

LO o

co

CO Rl

OJ

LO LO 04

ρ)

111

O

LO

ID O LO z CD

CM

CO CM CM

LO

CM

CO

Rl

E aj

0)

Φ in

Φ (D

CO

LO

LO «J φ

(0 co CM 04

CM

LO CO lOCLCLCLQ^CLCLCLCLQiiXCLQi C4ZZZZZZZZZZZZ φ

Φ in

Φ

JS? φ φ ΦΦΦ Φ TJ Φ TJ TJ Φ Φ

O

CO Φ S

O <n ergonine N.P. C30H38N5O5 548.287295 nicergoline N.P. C24H27BrN3O3 484.123578 ______________ergometrine__________N.P.___________________________________________C19H24N3O2 326.186852

Table 35 - pathway analysis ergot alkaloids (Ν

Φ g e

I uc φ

E

CO

E

I co

E

I (S

LO

II

Ϊ5 .?

co o CN

CN

CN e <u tu a co S£

CT- ID CO hCN vCN

CN

CO CN CN

CO LO

CN CN

CN

CO

CO (ft <A £

Φ to ,φ

CN

CN CN

CO CN

CO CN LO

LO

LO co'

CO

LO

LO

LO co

CN

CO co LO CN

CN O LO z o

CN

CN co O LO z o

O) O

CO

XT co

CN

O

CD σ>

(NJ co LO co co co

LO co co CO LO co

	co	LO	Tt
o	O	o	o
z	z	z	z
CN ID	co χί	s?
T	T	T	ΣΙΣ
σ>			pK,
co	co	CO	CO
O	O	O	o

CN

CO

O CN

CN CO O LO

Φ c

E (Ό

Φ

LO CO co co co

LO o

CN

CN (D

CN CN CN CN do' 00 LO do' CN (0

CO LO co LO

CO

LO (N LO

LO

CN

LO <0 CO

CO

CD xf

CD

CO

CN

LO co

CD CO

LO

LO ω E ω

4-> Ξ <-> c

RS σ c CO

Φ c

E co

Φ Q.

(0 co

E φ

«5 ω ω ra c ra co co Q.

co co

-101 2017251752 25 Oct 2017

Figure 61 shows an analysis of the lolitrem B biosynthetic pathway.

Figure 62 shows the paspalinine - A tremorgen pathway.

Figure 63 shows a pathway analysis: paspalinine - A tremorgen.

Figure 64 shows the presence of lolitrem and paspalinine in symbiota.

Endophyte Genotypes

Lolitrem 686(10.30)	ST	NEA10 NEA11 NEA12 E-	E1
Barsandra	+	NA	---	NA
Bealey	+	NA	NA	NA
Bronsyn	+	NA	NA	-
Impact	NA	-	_ _ _	-
Tolosa	+	NA	---	-

Table 37 - presence of lolitrem in symbiota

Paspalinine 434(10.17)	ST	Endophyte Genotypes
NEA10	NEA11	NEA12 E-	E1
Barsandra	+	NA	-	-	NA
Bealey	+	NA	+	NA	NA
Bronsyn	+	NA	-	NA	-
Impact	NA	+	-	-	-
Tolosa	+	NA	+	-	-

Table 38 - presence of paspalinine in symbiota

2017251752 25 Oct 2017

Table 39 - in-depth metabolic profiling of novel isogenic host-endophyte associations

-1032017251752 25 Oct 2017

Figure 65 shows that Peramine is present in planta in trace amounts in absence of a peramine-producing endophyte.

Example 13 - Endophyte effect on symbiotum performance

Assessment of endophyte effect on symbiotum performance in isogenic host background is done through measurements of:

• tiller number • shoot weight · root weight • root length • rootshoot ratio

Figure 66 shows the procedure used to assess changes to host phenotype mediated 5 by the endophyte.

Figure 67 shows endophyte effect on symbiotum performance.

ANOVA • Significant endophyte effects on all measured traits o · No significant combined endophyte/nitrate effects • No effect of nitrate on endophyte vegetative stability

F probability	Endophyte	<0.001	<0.001	<0.001	<0.001
Nitrate	<0.001	0.005	0.492	0.846
Endophyte/ Nitrate	0.065	0.208	0.905	0.847

LSDq q5	Endophyte	4.115	1.001	3.945	0.599
Nitrate	3.360	1.036	5.107	0.560
Endophyte/ Nitrate	7.099	1.829	7.786	1.065

Table 40 - Assessment of endophyte effect on symbiotum performance in isogenic host background through measurements of: Column 3: tiller number; 25 Column 4 Shoot Fresh Weight (g); Column 5: Root Length (cm); Column 6: Root Fresh Weight (g).

-104 2017251752 25 Oct 2017

Eight replicates per isogenic association were assessed for phenotypic performance

Plants were grown in sand pots for 4 weeks to establish

Treatments were applied at week 4: Drought and waterlogging

Harvest 1 at 8 weeks

Harvest 2 at 12 weeks

Figure 68 shows plants before the T1 harvest.

Figure 69 shows performance as measured by tiller number.

o

There were Significant genotype x genotype (G x G) effects • Significant variation between host plants • Significant effects of endophyte on performance

This highlights the importance of ab initio breeding and selection of symbiota.

Example 14 - Genome survey sequencing of endophytes

Progress in Performance of Different Sequencing Platforms

Roche 454 · 400-500bp reads • 1 million reads per run • 12x coverage

Illumina GAIIx • 150bp paired end reads · 8 lanes per flow cell • 20 million paired end reads per lane • 12 samples per lane • 10x coverage per sample

Illumina HiSeq 2000 · 100bp paired end reads • 8 lanes per flow cell • 2 flow cells per run

-1052017251752 25 Oct 2017 • 250 million paired end reads per lane • 24 samples per lane • 20-30x coverage per sample

Pangenome Analysis of Neotyphodium Endophytes

Sequence Analysis Using the 454 Sequencing Platform

Ten N. lolii and non-/V. lolii genomes sequenced

A reference genome for N. lolii generated

Nuclear architecture, including gene content, characterised

Intraspecific polymorphisms identified, highlighting the importance of sequencing different N. lolii genomes

Relationships between metabolic diversity and gene loss-gain for some toxins identified

Comparison with the related sexual form revealed gene loss from both lineages as a 5 source of phenotypic diversity

Comparison of the mitochondrial genome sequence with another member of the Clavicipitaceae revealed it is highly conserved, with differences in genome size due to indels in intergenic and intronic regions

Figure 70 shows sequence analysis using the 454 sequencing platform.

Sequence Analysis Using the Illumina Sequencing Platform perennial ryegrass endophyte strains sequenced:

• 16/V. lolii · 3 LpTG-2 • 4 non-/V. lolii

Reference genome construction - ST

Representatives of diversity of perennial ryegrass endophytes for pan-genome analysis

Current commercial endophytes

NEA2, NEA3, NEA4, AR1

Future (potentially) commercial endophytes

-1062017251752 25 Oct 2017

NEA10, NEA11, NEA12, E1

Within cluster analysis of genetic diversity

Endophytes from distinct geographical origins • ST (Grasslands Samson) - NA₆ (Morocco) and C9 (Spain)

Endophytes from the same geographical origin • NEA12 (France)-15310 and 15311

Figure 71 shows sequence analysis using the Illumina sequencing platform.

Figure 72 shows a summary of sequenced perennial ryegrass endophyte genomes.

Figure 73 shows a summary of sequenced perennial ryegrass endophyte genomes.

Key Research Activities · Incorporation of GAI I and HiSeq data to assemble and map a diverse range of endophyte strains and genes • Completion of the reference genome sequence assembly for ST endophyte • High resolution genome sequence analysis of ‘identical’ strains • Pan-genome sequence analysis of endophyte strains representing the diversity of perennial ryegrass endophytes

N. lolii Standard Toxic Reference Genome Refinement and Pangenome Analysis • Until recently all endophyte genome sequencing was performed using the

Roche 454 platform • There is now additional sequence data available from an independent platform (Illumina GAI I & HiSeq2000) for a large number of endophyte strains and species • This is useful as the two methods use fundamentally different sequencing chemistry hence one can be used to complement the other • N. lolii Standard Toxic (ST) is the reference strain for which we have the most 454 sequence data

-1072017251752 25 Oct 2017 • What gains are there from using both sets of data for this endophyte?

Figures 74-78 show Illumina reads v 454 contigs. Figure 74 shows that in general the Illumina data confirms 454 data. Figure 75 shows that the Illumina data can 5 correct for 454 homo-polymer error. Figure 76 shows that Illumina data from other strains identifies strain specific SNPs. Figure 77 shows that Illumina data from other species identifies haplotypes of duplicated/triplicated genes e.g FEtc7-58. Figure 78 shows ‘triplicate’ genes e.g. FEtc7-58.

Graphical Interactive Viewing and Analysis: Read Level

Figure 79 shows 454 FLX v 454 Titanium v Illumina GAI I coverage display.

Figure 80 shows 454 FLX v 454 Titanium v Illumina GAI I v Hi Seq coverage display.

Figure 81 shows Homo polymer correction: graphical view

Figure 82 shows Homo polymer correction: base view

Figure 83 shows Pan-genome SNP calling: graphical view, read level

Figure 84 shows 77 independent sequencing runs, ordered by similarity. These can display common SNPs.

Figure 85 shows that independent sequencing runs can also display strain specific deletions.

Figure 86 shows that independent sequencing runs can predict amino acid changes from SNP data.

-1082017251752 25 Oct 2017

Use of Illumina Data to Refine the 454 Sequence Assembly

Method • Generate an assembly using Newbler (Roche assembly software) and refine using information from Illumina sequence data (mate pair, read depth etc) and

Newbler assembly data (links between contigs, highly similar contigs) • Some Definitions:

=> Contig Continuous unambiguous assembled sequence => Scaffold Sequence containing more than one Contig whose connections and orientations relative to each other are known, but 0 which are connected by ambiguous sequence => N50 Length of contig in which 50% of the total assembled length is contained in contigs longer that that length

Figure 87 shows that contig number decreases with rounds of assembly refinement.

Figure 88 shows that N50 increases with rounds of assembly refinement.

Figure 89 shows that assembly refinement decreases the number of scaffolds, and single ‘unscaffolded’ contigs.

:0

Figure 90 shows that average number of bases per contig increases with assembly refinement.

Current Status of Reference Genome Sequence Assembly of ST Endophyte => 2,494 contigs, totalling 33,804,495 bp => 1,496 of these contigs in 449 scaffolds, largest 395,697 bp, total 25,953,091 bp => 998 single unscaffolded contigs, containing 7,851,404 bp => Augustus gene prediction program predicts 10,821 protein encoding genes

-1092017251752 25 Oct 2017 =^> The ST reference genome provides a basis for generating the pangenome of other N. lolii and related fungi.

Review of Candidate Endophyte Status

Previous candidate evaluation process:

- Identified E+ accessions (77)

- Identified novel, genetically diverse endophyte genotypes

- Identified 42 primary endophyte candidates subjected to metabolic profiling in endogenous host background

- Eliminated lolitrem B-producing endophyte candidates (29)

- Eliminated duplicate endophyte genotypes (5)

- Identified 8 primary endophyte candidates for advancement

5	Endophyte Identifier	Species	Origin	Endogenous Toxin Profile	Confirmed Isolated Toxin Profile
	15285	N. lolii	Portugal	La/-/P	J L/-/P
	15441	N. lolii	Italy	-/-/-	Z L/E/P
	15656	N. lolii	Turkey	-/-/-	xb
	15931	N. lolii	Former Soviet Union	-/E/P	Z NDC
	F2	N. lolii	Spain	-/E/ND	Z NDd
	NEA10	N. lolii	Spain	-/E/ND	Z -/E/P
	NEA11	LpT G-2	France	-/E/ND	Z -/E/P
:0	NEA12	Non-M. lolii	France	-/-/-	Z -/-/-
	a = negligible levels in endogenous background
	b = unable to isolate from original source plant
	c = inoculated into isogenic panel but unstable
	d= recalcitrant to inoculation into isogenic panel
		Candidates with suitable profile advanced in program
		Candidates with suitable profile not advanced due to stability issues
25		Candidates with apparent unsuitable profiles

Table 41 - Review of Candidate Endophyte Status

Potential to consider 15931 and F2 endophytes in new paradigm of large-scale establishment of grass-endophyte symbiota for ab initio selection.

-110 2017251752 25 Oct 2017

Example 15 - Overview of generation of novel designer Neotyphodium endophyte variant strains through mutagenesis

The objective of this work was to create novel variants of the perennial ryegrass endophyte, Neotyphodium lolii, through induced polyploidisation and mutagenesis, with desirable properties such as enhanced bioactivities (e.g. antifungal acitivity), and/or altered plant colonization ability and stability of grass host - endophyte variant associations (e.g. altered in vitro growth), and/or altered growth performance (e.g.

enhanced plant vigour, enhanced drought tolerance, enhanced water use efficiency) of corresponding grass host - endophyte variant associations. These grass host endophyte variant associations are referred to as novel ‘designer’ grass-endophyte associations.

Experimental strategies for the generation and characterisation of novel designer Neotyphodium endophyte variant strains through mutagenesis

The experimental activities thus included:

1. Establishment of phenotypic screens for novel ‘designer’ grass-endophyte associations such as:

• Enhanced biotic stress tolerance • Enhanced drought tolerance and enhanced water use efficiency • Enhanced plant vigour

2. Targeted generation (i.e. polyploidisation and X-ray mutagenesis) and characterisation (i.e. antifungal bioassays, in vitro growth rate, genome survey sequencing [GSS]) of novel ‘designer’ endophytes

3. Breeding of ‘designer’ grass endophyte associations • Delivery of designer endophytes into grass (e.g. perennial tyegrass) germplasm development process.

-111 2017251752 25 Oct 2017

Establishment of phenotypic screens for novel ‘designer’ grass-endophyte associations

Assessment of enhanced biotic stress tolerance using NEA12 is shown in Figures 92 and 93. Figure 92 shows in vitro bioassays to assess antifungal activity of Neotyphodium endophytes. Figure 93 shows a detached leaf assay to assess resistance to crown rust (Puccinia coronata f.sp. lolii).

Assessment of enhanced drought tolerance and enhanced water use efficiency is shown in Figure 94. This involved glasshouse and field trial screens for drought tolerance, survival and recovery, regrowth after drought, metabolic profiling and multiple trait dissection (based on assessments and measurements associated with plant morphology, plant physiology, plant biochemistry).

Generation of designer N lolii genotypes by polyploidisation

This involved creation of novel variation in Neotyphodium endophytes without the use of transgenic technology. Colchicine has been widely and successfully used for chromosome doubling in plants, e.g. perennial ryegrass. It inhibits chromosome segregation during mitosis inducing autopolyploidisation (chromosome doubling; see Figure 95). This enables the generation of novel endophytes through induced chromosome doubling and may be applicable to the production of artificial polyploid endophytes.

The experimental work flow for chromosome doubling is shown in Figure 96.

Flow cytometry calibrations to assess DNA content in Neotyphodium endophytes are shown in Figure 97. Peaks indicate relative nuclear DNA content.

Flow cytometry analysis of NEA12^dh strains is shown in Figure 98 and Table 42.

-1122017251752 25 Oct 2017

1. ST is highly stable, broadly compatible. 2. NEA12 produces JANTHITREM only.

3. AR1 produces PERAMINE only.

Endophyte	Colchicine treatment (%)	# of colonies	# colonies analysed
N.lolii ST	0.2	12	12
N.lolii NEA12	0.1	60	2
N.lolii NEA12	0.2	60	18
N.lolii AR1	0.1	60	0
N.lolii AR1	0.2	60	0

Table 42: Colchicine treated endophyte strains (ST, NEA12 and AR1 endophyte strains) subjected to colchicine treatments (at different colchicine concentrations in %) leading to the recovery of endophyte colonies (# of colonies) used for flow cytometry analysis

Analysis of in vitro growth of NEA12^dh Neotyphodium variant endophyte 0 strains

Analysis of growth rate of NEA12^dh Neotyphodium variant endophyte strains in in vitro culture after 8 weeks is shown in Figure 99. In an initial screen, analysis of variance identified two NEA12^dh Neotyphodium variant endophyte strains (NEA12^dh17 5 and NEA12^dh4) showing significantly different in vitro growth rate to the control

NEA12 endophyte:

NEA12^dh17 grows significantly faster (p<0.01**)

NEA12^dh4 grows significantly slower (p<0.05*)

Analysis of growth rate of NEA12^dh Neotyphodium variant endophyte strains in in vitro culture over 5 weeks is shown in Figure 100. In a validation screen, Student’s ttests identified two NEA12^dh Neotyphodium variant endophyte strains (NEA12^dh17 25 and NEA12^dh15) showing significantly different in vitro growth rate to the control

NEA12 endophyte:

NEA12^dh17 grows significantly faster (p<0.01**)

-1132017251752 25 Oct 2017

NEA12^dh15 grows significantly slower (p<0.01**)

Antifungal bioassays of NEA12^dh Neotyphodium variant endophyte strains

A list of fungal pathogens (causing a range of fungal diseases and infecting a range of different plant hosts) that were included in antifungal bioassays used to analyse NEA12^dh Neotyphodium variant endophyte strains to assess their spectrum of antifungal activities is shown in Table 43.

o

-114 2017251752 25 Oct 2017

Fungus	Disease	Hosts
Alternaria alternata	leaf spot, rot, blight	Numerous (dead plant materials)
Bipolaris portulacae	Damping-off	Asteraceae (daisies), Portulacaceae (purslane)
Botrytis cinerea	Stem rot, mould, seedling wilt	Many dicots, few monocots
Colletotrichum graminicola	Leaf spot, stalk rot	Poaceae (especially Zea mays)
Drechslera brizae	Leaf blight	Poaceae (Briza spp.)
Phoma sorghina	Spot (leaf, glume, seed), Root rot, Dying-off	Poaceae (grasses)
Rhizoctonia cerealis	Spot (wheat) Yellow patch (turfgrass)	Poaceae (grasses)
Trichoderma harzianum	Green mould, Parasite of other fugni	Many dicots, few monocots, Fungi
Table 43: Fungal pathogens (causing a range ol	fungal diseases and infecting

a range of different plant hosts) included in antifungal bioassays to analyse NEA12^dh Neotyphodium variant endophyte strains to assess their spectrum of antifungal activities

Antifungal bioassays of NEA12^dh Neotyphodium variant endophyte strains are shown in Figures 101 and 102. Twenty NEA12^dh strains were screened for changes in antifungal activity. Four NEA12^dh strains (i.e. dh5, dh6, dh13 and dh14) were identified as having greater antifungal activity compared to NEA12.

Genome survey sequencing and sequence analysis of NEA12^dh Neotyphodium variant endophyte strains

-1152017251752 25 Oct 2017

NEA12^dh Neotyphodium variant endophyte strains with enhanced antifungal activity, showing faster in vitro growth rate and higher DNA content were subjected to genome survey sequencing (GSS). Sequence data was generated for 10 NEA12^dh 5 strains and control NEA12 strain (highlighted in blue on Table 44).

Endophyte	Antifungal	Growth
NEA12	Std	Std
NEA12dh1	Std	Std
NEA12dh2	Std	Std
NEA12dh3	Std	Std
NEA12dh4	Std	Slower
NEA12dh5	Higher	Std
NEA12dh6	Higher	Std
NEA12dh7	Std	Std
NEA12dh8	Std	Std
NEA12dh9	Std	Std
NEA12dh10	Std	Std
NEA12dh11	Std	Std
NEA12dh12	Std	Std
NEA12dh13	Higher	Std
NEA12dh14	Higher	Std
NEA12dh15	Std	Slower
NEA12dh16	Std	Std
NEA12dh17	Std	Faster
NEA12dh18	Std	Std
NEA12dh19	Std	Std

NEA12dh20 Std Std

Table 44: List of NEA12^dh Neotyphodium variant endophyte strains showing different antifungal activity [higher than control or equal to control (standard, Std)] and different in vitro growth [slower than control, faster than conrol or 10 equal to control (standard, Std)] compared to control NEA12 strain

Genome survey sequencing (GSS) data obtained for NEA12^dh Neotyphodium variant endophyte strains derived from colchicine treated NEA12 control strain (highlighted in blue on Table 44) were analysed as follows:

· De-novo assembly of the GSS data from NEA12 control strain - to act as a reference genome sequence for the analysis of the NEA12^dh Neotyphodium variant endophyte strains

-1162017251752 25 Oct 2017 • Map the GSS data sequence reads from the NEA12^dh Neotyphodium variant endophyte strains to the NEA12 reference genome sequence • Identify potentially duplicated regions, i.e. regions with higher than expected sequence coverage · Identify gene sequences that may have been duplicated

Analysis of GSS read depth of NEA12^dh Neotyphodium variant endophyte strains is shown in Figure 103. Analysis of sequence contigs that appeared to have higher than expected read depth indicates that no major duplication event has occurred 0 (excepting whole genome events). The patterns of read depth across these contigs are not identical between strains. This suggests there are differences between the NEA12^dh Neotyphodium variant endophyte strains and the control NEA12 strain.

Analysis of GSS sequence assemblies for the NEA12^dh Neotyphodium variant 5 endophyte strains and the control NEA12 strain is shown in Table 45.

Strain	# contigs	N50	Max contig	# bases
NEA12	143202	28621	181461	32734984
NEA12dh5	305031	29444	191191	30994592
NEA12dh17	274394	37802	209957	30777017
NEA12dh18	282692	30717	177813	30889903
Table 45: Analysis of GSS sequence assemb	ies for the NEA12	h

Neotyphodium variant endophyte strains and the control NEA12 strain

Independent de novo sequence assemblies were performed using parameters identical to those used in assembling the genome sequence for the control NEA12 endophyte strain. Differences in sequence assembly statistics may indicate genomic differences between strains. GSS data obtained for the NEA12^dh Neotyphodium variant endophyte strains and used in the sequence assemblies reveal fewer bases incorporated into the sequence assembly and produce more sequence contigs.

-1172017251752 25 Oct 2017

Increased numbers of smaller sequence contigs may be caused by transposon movement/replication.

Analysis of sequence reads mapping to the NEA12 genome sequence assembly is shown in Figure 104. While we do not wish to be restricted by theory, if the genomes were the same no difference in the number of sequence reads mapping to the reference genome sequence would be expected. NEA12^dh Neotyphodium variant endophyte strains range from 35-70% sequence reads mapping to NEA12 sequence contigs > 5kb in size. There are differences between the genome sequences of the 0 NEA12^dh Neotyphodium variant endophyte strains and the control NEA12 strain.

Summary of results on generation and characterisation of novel designer Neotyphodium variant endophyte strains through colchicine treatment based mutagenesis

Sequence read depth changes were analysed in NEA12^dh Neotyphodium variant endophyte strains compared with the control NEA12 strain. Whilst no large partial genome sequence duplication events were detected, the occurrence of full genome duplication events in the NEA12^dh Neotyphodium variant endophyte strains cannot be 0 excluded based on the GSS sequence analysis.

De novo sequence assemblies were independently performed on GSS data obtained from the NEA12^dh Neotyphodium variant endophyte strains. Differences in sequence assembly statistics indicate that genomic changes were caused by the colchicine25 treatment in the NEA12^dh Neotyphodium variant endophyte strains. The number of sequence reads from NEA12^dh Neotyphodium variant endophyte strains mapping to the NEA12 reference genome sequence varies between strains. All GSS data analyses performed on the NEA12^dh Neotyphodium variant endophyte strains indicate genomic differences.

In summary, the following novel designer endophytes were generated by colchicine treatment of NEA12 endophytes:

-1182017251752 25 Oct 2017 • Four NEA12^dh Neotyphodium variant endophyte strains (dh5, dh6, dh13 and dh14) with enhanced bioprotective properties (i.e. antifungal bioactivities);

• One NEA12^dh Neotyphodium variant endophyte strain (dh17) with higher in vitro growth rate than control NEA12 strain (i.e. potentially with enhanced stability/host colonization ability);

• Ten NEA12^dh Neotyphodium variant endophyte strains (including dh5, dh6, dh13, dh14 and dh17) and control NEA12 strain subjected to genome survey sequencing; and • Five NEA12^dh Neotyphodium variant endophyte strains (including dh5, dh13 and dh 17) selected and subjected to isogenic inoculation in planta.

In planta isogenic inoculation in perennial ryegrass with NEA12^dh Neotyphodium variant endophyte strains

The following NEA12^dh Neotyphodium variant endophyte strains and control NEA12 strain were used for in planta isogenic inoculation in perennial ryegrass:NEA12

NEA12dh5 High antifungal activity

NEA12dh13 High antifungal activity

NEA12dh4 Slow-growing

NEA12dh15 Slow-growing

NEA12dh17 Fast-growing

Plant Genotype	NEA12 dh4	NEA12 dh5	NEA12 dh13	NEA12 dh15	NEA12 dh17	NEA12
IMP04	30 (1/19)	30 (0/14)	30 (0/21)	30	32 (2/19)	30 (0/6)
TOL03	25	30	30	20	30	20

Table 46: Isogenic inoculation of perennial ryegrass genotypes (IMP04 and

TOL03) with NEA12dh Neotyphodium variant endophyte strains. Numbers indicate number of perennial ryegrass plants of the two genotypes subjected to isogenic inoculation with the different NEA12dh Neotyphodium variant endophyte

-1192017251752 25 Oct 2017 strains.

Generation of designer N lolii genotypes by X-ray mutagenesis

The generation of designer Neotyphodium endophytes genotypes by X-ray mutagenesis offers the opportunity to create novel endophyte variant strains with enhanced properties, such as enhanced stability in grass hosts, broader host compatibility as well as improved toxin profiles e.g. following elimination of the production of the detrimental alkaloid lolitrem B in the highly stable and broadly compatible ST endophyte.

o

Such an novel designer endophyte would be advantageous over existing commercial endophytes, such as AR1 and AR37, as it would be highly stable and broadly compatible and with optimal toxin profile.

Figure 105 shows an experimental work flow for X-ray mutagenesis of endophyte strains.

Figure 106 shows the indole-diterpene biosynthetic pathway. Lolitrem B is the major toxin that causes ryegrass staggers, a disease of grazing animals. Ten genes in 3 0 gene clusters are required for lolitrem biosynthesis. We focused initial analysis on 3 Ltm genes, one from each gene cluster. Optimised multiplex PCR analysis was designed and implemented.

Screening of X-ray irradiated N. lolii strains

In a preliminary primary screen >5,000 colonies of X-ray irradiated N. lolii established as an initial resource of novel variation of N. lolii endoophytes induced through X-ray mutagenesis and representing a mutagenised N. lolii endophyte strain collection - of were screened by multiplex PCR analysis for the presence of targeted Ltm genes leading to a preliminary identification of -140 putative lolitrem B gene cluster PCR-negative colonies (-2.5% of 5,000 colonies screened). In a secondary screen high quality DNA was extracted (140 liquid cultures) and PCR analysis conducted. This identified 2 putative deletion mutants for one of the lolitrem B genes

-1202017251752 25 Oct 2017 (Itm J).

Dose (Gy)

Gy (1 irradation)

Gy (1 irradation)

Colony

139-6

145-15

Table 47: Putative X-ray irradiation-induced Itm gene deletion mutants of N.

lolii derived from irradiation with 30 Gy dose. The colony number represents the unique identifier of the putative X-ray irradiation-induced Itm gene deletion mutant (i.e. 139-6 and 145-15). Black represents PCR-negative result for respective Itm gene analysis, grey represents PCR-positive result for respective Itm gene analysis.

o Antifungal bioassays of designer X-ray irradiated N. lolii variant strains

There were eight X-ray irradiated N. lolii variant strains (i.e. X-ray mutagenesis derived variant strains 1-35, 4-7, 7-22, 7-47, 123-20, 124-6, 139-6, 144-16 and 14515) and one control N. lolii strain (i.e. ST endophyte strain).

Five fungal pathogens (causing a range of fungal diseases and infecting a range of different plant hosts) were included in antifungal bioassays used to analyse the X-ray irradiated N. lolii variant strains, as follows:

Bipolaris portulacae

Colletotrichum graminicola

Drechslera brizae

Phoma sorghina

Rhizoctonia cerealis

No significant difference in antifungal activities of X-ray irradiated N. lolii variant strains tested was observed compared to the spectrum of antifungal activities observed for the control ST endophyte strain.

-121 In vitro growth of designer X-ray irradiated N. lolii variant strains

Results from the analysis of in vitro growth rate of designer X-ray irradiated N. lolii variant strains are shown in Figure 107, with a statistical analysis of in vitro growth 5 undertaken at week 5 for the X-irradiated N. lolii variant strains compared to the control ST strain, revealing significant differences in in vitro growth rates as follows: p< 0.05* (for X-irradiated N. lolii variant strain 139-6) p<0.01** (for all other mutants)

Genome survey sequencing of designer X-ray irradiated N. lolii variant strains

Eight X-ray irradiated N. lolii ST variant strains and corresponding control ST strain were subjected to genome survey sequencing (GSS), leading to 46-fold to 79-fold genome sequence coverage for the different strains as shown in Table 48.

Strain	Description	Coverage
ST	ST	23x
139-6	ST irradiated	61x
145-15	ST irradiated	52x
144-16	ST irradiated	46x
1_35	ST irradiated	79x
4_7	ST irradiated	46x
7_22	ST irradiated	53x
7_47	ST irradiated	38x
123-20	ST irradiated	54x
124-6	ST irradiated	75x

Table 48: Genome sequence coverage obtained in genome survey sequencing for for 8 X-ray irradiated N. lolii ST variant strains and corresponding control

ST strain

Detecting genome sequence variation in designer X-ray irradiated N. lolii variant strains

-122 2017251752 25 Oct 2017

Results from the analysis to detect genome sequence variation in X-ray irradiated N. lolii variant strains are shown in Figure 109. Corresponding results on the detection of single nucleotide polymorphisms (SNPs) are shown in Figure 110 and results on the detection of small insertions/deletions (INDELs) are shown in Figure 111.

Differences in sequence read depth and pair insert size in X-ray irradiated N. lolii variant deletion mutant strains are shown in Figure 112.

Results on sequence analysis for Ltm gene clusters are shown in Figure 108. No deletions, large or small, were found in the coding or regulatory sequences of ltm 0 gene clusters. No SNPs, insertions or translocations were found in the coding or regulatory sequences of ltm gene clusters.

Spectrum of genome sequence changes detected in the X-ray irradiated N. lolii variant strains

Figure 113 shows numbers of SNPs detected in genic regions of X-ray irradiated N. lolii variant deletion mutant strains. There are large differences in the number of SNPs detected in the X-ray irradiated N. lolii variant deletion mutant strains and compared to the control ST strain. All X-ray irradiated N. lolii variant deletion mutant 0 strains have over double the number of SNPs per Mb across genic regions compared to the control ST strain. X-ray irradiated N. lolii variant deletion mutant strains have on average 6 SNPs per Mb, where the control ST strain has 2 SNPs per Mb.

Figure 114 shows numbers of INDELs in genic regions of X-ray irradiated N. lolii 25 variant deletion mutant strains. All X-ray irradiated N. lolii variant deletion mutant strains contain more indels in genic regions than the control ST strain. The difference in indel numbers between the X-ray irradiated N. lolii variant deletion mutant strains and the control ST strain is on average 134 indels per Mb. When grouped by irradiation treatment (i.e. irradiation dose applied and number of repeat 30 irradiations) there appears to be a peak in number of indels at 10Gy*2 treatment, consistent with the results obtained in the SNP detection analysis.

-1232017251752 25 Oct 2017

Figure 115 shows the spectrum of genome sequence changes in the form of deletions detected in X-ray irradiated N. lolii variant deletion mutant strains.

Table 49 shows examples of some of these genome sequence deletions detected in X-ray irradiated N. lolii variant deletion mutant strains.

Strain	Radiation Treatment	Deletion
123_20	30Gy*2	Contig00915 (268bp)
124_6	30Gy*2	Partial duplication
139_6	30Gy	Partial duplication
144—16	30Gy
145_15	30Gy	Partial duplication
1_35	10Gy	Contig00831 (3.6kb)
4_7	10Gy
722	10Gy*2
7_47	10Gy*2	Contig01131 (0.6kb), contig01082 (4.2kb), contig02985 (1kb), contig02725 (83bp), contig01095 (130bp)

Table 49: Deletions detected in genome sequences of X-ray irradiated N. lolii variant deletion mutant strains. Bold indicates deletions confirmed by changes in sequence read coverage. The remainder are potential transposon deletions.

The X-ray irradiated N. lolii variant deletion mutant strain # 7_47, which was generated following two X-irradiation treatments at 10 Gy dose (10Gy*2) of N. lolii ST endophyte, had the greatest number of large deletions.

-124 2017251752 25 Oct 2017

Annotation of deleted sequences in the genomes of X-ray irradiated N. lolii variant deletion mutant strains

X-Ray Irradiated N. /o///Variant Mutant Strain 1_35:

For the X-ray irradiated N. lolii variant mutant strain 1_35 the following deleted sequences in ST454Contig00831 contig with a ~ 4,400-8,000 bp length was detected, with this genome sequence region containing the following two predicted genes:

ST454contig00831_AUGUSTUS_gene_3318:6018 (847 letters)

1) ref |XP_386347.11 hypothetical protein FG06171.1 [Gibberella 660x0.0 gb|EAW12630.11 DUF500 domain protein [Aspergillus NRRL 1]; 253 x 9e-66, and ST454contig00831_AUGUSTUS_gene_3958:4728 (183 letters); and

2) gb|EAW13545.11 2,3-cyclic-nucleotide 2-phosphodiesterase [Aspergillus 32 x 2.4

X-Ray Irradiated N. lolii Variant Mutant Strain 7_47:

For the X-ray irradiated N. lolii variant mutant strain 7_47 the following deleted sequences in ST454Contig01082, ST454Contig01131 and ST454Contig02985, with these genome sequence regions containing no predicted genes:

Query= ST454contig01082 length=9120 numreads=287 gb|AAA21442.11 putative pol polyprotein [Magnaporthe grisea] 145 1e-32 Query= ST454contig02985 length=2414 numreads=99 gb|AAA21442.11 putative pol polyprotein [Magnaporthe grisea] 92 2e-17

Mutagenesis index of X-ray irradiated N. lolii variant deletion mutant strains

Figure 116 shows SNPs and Indels per Mb in genic regions of X-ray irradiated N. lolii variant deletion mutant strains derived from X-ray irradiation of N. lolii at different levels of irradiation. Strain 1_35 has a 3.6 kb deletion; Strain 7_47 has 3 deletions (4.2 kb, 1 kb, 0.6 kb in lenght). Strain 124_6 has a partial duplication. Strains 139_6 30 and 145_15 have partial duplications.

Given that ST endophyte has approximately 443.5 genes per Mb, using 10Gy*2

-1252017251752 25 Oct 2017 treatment, the expected rate of SNP/INDEL occurrence is 0.33 per gene in the genome.

Summary

X-ray irradiated N. lolii variant deletion mutant strains were analysed for many types of genome sequence variation i.e. deletions, SNPs, INDELs, inversions and translocations. SNPs, INDELs, deletions and duplications were identified in the genome survey sequences of X-ray irradiated N. lolii variant deletion mutant strains. There was an apparent peak in number of SNPs and INDELs in X-ray irradiated N.

lolii variant deletion mutant strains recovered from administering 10Gy*2 X-ray irradiation treatment to N. lolii ST endophyte. The X-ray irradiated N. lolii variant deletion mutant strain 7_47 had 3 large deletions. It was demonstrated that this mutagenesis method based on X-ray irradiation can be used to create novel designer Neotyphodium endophyte strains, and enabled:

• 5,000 X-ray irradiated N. lolii variant endophyte strains derived from X-ray irradiation of ST N. lolii endophyte were screened;

• 140 putative X-ray irradiated N. lolii variant endophyte mutant strains were identified:

· 9 X-ray irradiated N. lolii variant endophyte mutant strains were subjected to antifungal bioassays;

• 9 X-ray X-ray irradiated N. lolii variant endophyte mutant strains were subjected to in vitro growth assays;

• 9 X-ray irradiated N. lolii variant endophyte mutant strains were subjected to genome survey sequencing;

• 2 X-ray irradiated N. lolii variant endophyte mutant strains with gene deletions (1_35 and 7_47) were identified; and • 3 X-ray irradiated N. lolii variant endophyte mutant strains with gene duplications (124_6, 139_6 and 145_15) were identified.

-1262017251752 25 Oct 2017 ln planta isogenic inoculation in perennial ryegrass with X-ray irradiated N. lolii variant endophyte mutant strains

Plant Genotype	ST-IRM 139-6	ST-IRM 145-15	ST-IRM 144-16	ST-IRM 1-35	ST-IRM 7-47	ST
IMP04	3 0	2 5	3 0	3 0	3 0	2 5
TOLO 3	2 5	0	2 5	3 0	3 0	2 0

Table 50: Isogenic inoculation of perennial ryegrass genotypes (IMP04 and

TOL03) with X-ray irradiated N. lolii variant endophyte mutant strains. Numbers indicate number of perennial ryegrass plants of the two genotypes subjected to isogenic inoculation with the different X-ray irradiated N. lolii variant endophyte mutant strains (i.e. ST-IRM 139-6, ST-IRM 145-15, ST-IRM 144-16, ST-IRM 1-35 and ST-IRM 7-47) and control ST endophyte strain.

Metabolic profiling of colchicine treatment-derived NEA12dh and X-ray irradiation-derived Neotyphodium variant endophyte strains

Results from metabolic profiling of colchicine treatment derived NEA12dh endophyte variant strains is shown in Figure 117.

Results from metabolic profiling of X-ray irradiation treatment derived N. lolii ST endophyte variant strains is shown in Figure 118.

The following endophytes were grown on PDB for 3 weeks:

Control N. lolii ST endophyte strain

X-ray irradiation treatment derived N. lolii ST endophyte variant strain 4-7

X-ray irradiation treatment derived N. lolii ST endophyte variant strain 139-6

X-ray irradiation treatment derived N. lolii ST endophyte variant strain 144-16

X-ray irradiation treatment derived N. lolii ST endophyte variant strain 145-15 and subjected to metabolic profiling using LCMS on corresponding

-1272017251752 25 Oct 2017

1. Liquid filtrate

2. Mycelial extract

The X-ray irradiation treatment derived N. lolii ST endophyte variant strains could be 5 readily distinguished from control N. lolii ST strain using mycelia extracts or filtrates alone.

Example 16 - Method for Large-Scale Generation of Grass-Endophyte Symbiota (artificial seeds) o

The objective of the work was to develop an efficient, robust and low-cost method for large-scale production of grass endophyte symbiota. The method should be:

a) applicable to inoculation of 10s - 100s of endophyte in 100s - 1000s of grass genotypes;

b) applicable to perennial ryegrass, tall fescue and Brachiaria; and

c) applicable to inoculation of novel and designer endophytes with de novo generated genetic variation [i.e. induced mutagenesis (ionizing radiation, colchicine), targeted mutagenesis, transgenesis, cisgenesis, intragenesis, etc.].

The method should further enable next-generation ab initio molecular breeding, selection and evaluation of grass-endophyte symbiota [rather than breeding and selection of grass host followed by endophyte inoculation and symbiota evaluation only],

The experimental strategies - and corresponding experimental steps - implemented include:

1. Large-scale perennial ryegrass seed-derived embryo isolation and artificial seed production

A. Develop an efficient, low-cost, large-scale seed surface-sterilization method;

B. Develop an efficient, low-cost, large-scale seed-derived embryo isolation method;

C. Develop an efficient, low-cost, large-scale artificial seed production method;

D. Test germination frequency and germination stages of artificial seeds;

-1282017251752 25 Oct 2017

E. Assess endophyte presence in seedlings derived from artificial seeds generated with embryos isolated from endophyte-plus seeds;

2. Large-scale endophyte inoculation into perennial ryegrass artificial seeds

F. Develop an efficient, low-cost, large-scale endophyte inoculation method for artificial seeds [based on seed-derived embryo inoculation with endophyte mycelium followed by artificial seed production including double/multiple coating (inner layer plus endophyte, outer layer as ‘pseudo-aleurone/endosperm’) of artificial seeds]; and E. Assess endophyte presence in seedlings derived from artificial seeds generated with embryos isolated from endophyte-minus seeds inoculated with novel endophytes.

Large-Scale Perennial Ryegrass Seed-Derived Embryo Isolation and Artificial Seed Production

Seed Surface Sterilization Method

The seed surface sterilization method implemented includes the following steps:

Day 1: seeds were soaked in 10% sulphuric acid O/N

Day 2: treated with 10% Domestos for 20 min and stored at 24C after wash with 0 distilled sterile water

Day 3: treated with 10% Domestos for 20 min and stored at 24C after wash with distilled sterile water, followed by embryo isolation [see B) below].

Four independent experiments were conducted with 200 seeds each.

No bacterial or fungal contamination was observed.

Embryo Isolation Method

Based on the successful seed sterilization method [see A) above], 1,000 ryegrass seed-derived embryos can be isolated by one person within 4 hours.

Artificial Seed Production Method

Ca-alginate Coating of Perennial Ryegrass Embryos into Artificial Seeds

i) Coating with Ca-alginate matrix without added nutrients

-1292017251752 25 Oct 2017

For the Ca-alginate coating of perennial ryegrass embryos into artificial seeds using a coating with Ca-alginate matrix without added nutrients, the following steps were undertaken:

• Embryos were freshly isolated and mixed with 3% sodium alginate solution · Alginate drops were placed into 50mM calcium chloride solution while stirring at 60 rpm. Each drop contains one embryo.

• Artificial seeds were collected after 15 min stirring and washed with sufficient distilled sterile water

Artificial seeds were placed on germination medium MS or MS + 1mg/L BAP.

Figure 119 shows artificial seeds generated through Ca-alginate coating of perennial ryegrass embryos using a coating with Ca-alginate matrix without added nutrients.

ii) Coating with Ca-alginate matrix with added nutrients

For the Ca-alginate coating of perennial ryegrass embryos into artificial seeds using a coating with Ca-alginate matrix with added nutrients, the following steps were undertaken:

• Embryos were freshly isolated and mixed with 3% sodium alginate in modified

MS medium consisting of MS (without CaCI2) + 750 mg/L glutamine + 5μΜ

CuSO₄+ 1.95 g/L MES • Alginate drops (containing individual embryos) were placed in 50mM calcium chloride solution while stirring at 60 rpm.

• Each drop contains a single seed-derived isolated embryo.

· Artificial seeds were collected after 15 min stirring and thoroughly washed with distilled sterile water medium.

• Artificial seeds were placed on MS medium plates for germination.

iii) Coating with coloured Ca-alginate matrix

For the Ca-alginate coating of perennial ryegrass embryos into artificial seeds using a coating with coloured Ca-alginate matrix with added nutrients, the following steps were undertaken:

-130 2017251752 25 Oct 2017

Embryos were freshly isolated and mixed with 3% sodium alginate in modified MS medium consisting of MS (without CaCI2) + 750 mg/L glutamine + 5μΜ CuSO₄ + 1.95 g/L MES

Different food dyes [i.e. 10 pL/ml Queen Green (90610) or Queen Pink (92330)] were added to the sodium alginate coating solution to colour coating matrix thus establishing basis to demonstrate potential for multi-layer coating.

Alginate drops (containing individual embryos) were placed in 50mM calcium chloride 0 solution while stirring at 60 rpm.

Each drop contains a single seed-derived isolated embryo.

Artificial seeds were collected after 15 min stirring and thoroughly washed with 5 distilled sterile water medium.

Artificial seeds were placed on MS medium plates for germination.

Figure 120 shows Ca-alginate coating of perennial ryegrass embryos into artificial 0 seeds using coating with coloured Ca-alginate matrix.

iv) Coating with multiple Ca-alginate matrix layers

For the Ca-alginate coating of perennial ryegrass embryos into artificial seeds using a coating with multiple Ca-alginate matrix layers, the following steps were undertaken:

· Embryos were freshly isolated and mixed with 3% sodium alginate in modified

MS medium [consisting of MS (without CaCh) + 750 mg/L glutamine + 5μΜ CUSO4 + 1.95 g/L MES] as the first coating layer (layer A) to make artificial seeds.

• Alginate drops (containing individual embryos) were placed in 50mM calcium chloride solution while stirring at 60 rpm. Each drop contains a single seedderived isolated embryo.

• Artificial seeds coated with layer A were collected after 15 min stirring and

-131 2017251752 25 Oct 2017 thoroughly washed with distilled sterile water medium. The average diameter of the artificial seed freshly coated with layer A is 4 mm. Artificial seeds coated with layer A were placed in Petri dish and allowed to air-dry for 1 - 2 hours in a laminar flow cabinet. The diameter of the air-dried artificial seed coated with 5 layerAis2mm.

• Air-dried artificial seeds coated with layer A were mixed with 3% sodium alginate in modified MS medium [consisting of MS (without CaCI2) + 750 mg/L glutamine + 5μΜ CuSCU + 1.95 g/L MES] coloured with food dye dye [i.e. 10 pL/ml Queen Green (90610)] as the second coating layer (layer B) to make 0 double-coated artificial seeds; following the same procedure.

Figure 121 shows Ca-alginate coating of perennial ryegrass embryos into artificial seeds using coating with multiple Ca-alginate matrix layers.

Figure 122 shows Ca-alginate coating of perennial ryegrass embryos into artificial seeds using coating with multiple Ca-alginate matrix layers. Freshly isolated seedderived embryos of perennial ryegrass are individually placed in wells of a) 96-well or

b) 384-well plates. With the aid of a disposable syringe sodium alginate solution is added to the individual wells and single embryos in alginate solutions are loaded in 0 the syringe. With the aid of the syringe individual embryos coated with alginate solution are dropped into polymerising CaCI₂ solution under agitation for production of artificial seeds. The use of 96-well plate is preferred over the 384 well plate for production of artificial seeds of perennial ryegrass.

Assessing Germination Frequency of Artificial Seeds

In order to assess germination frequency of artificial seeds, the following steps were undertaken:

Germination of seeds, embryos and artificial seeds of perennial ryegrass 30 cv. Bronsyn E' (endophyte free, 2668 seed batch)

Seed germination frequency was comparatively assessed for (Figure 123):

a) Original seeds:1% germination frequency on filter paper

-132 2017251752 25 Oct 2017

b) Surface-sterilized seeds: 10% germination frequency on filter paper

c) Isolated embryos: 48% germination frequency on germination medium

d) Artificial seeds (with germination medium): 40% germination frequency on MS medium

Germination of seeds, embryos and artificial seeds of perennial ryegrass cv. Bronsyn E+ (endophyte plus, 2667 seed batch)

Seed germination frequency was comparatively assessed for (Figure 124):

a) Original seeds:10% germination frequency on filter paper

b) Surface-sterilized seeds: 30% germination frequency on filter paper

c) Isolated embryos: 90% germination frequency on germination medium

d) Artificial seeds (with germination medium): 81% germination frequency on MS medium

Figure 125 shows germination of artificial seeds and development of artificial-seed derived seedlings in perennial ryegrass.

Assessing Endophyte Presence in Seedlings Derived from Artificial Seeds

In order to assess endophyte presence in seedlings derived from artificial seeds, the 0 following experiments were undertaken:

Endophyte presence in seedlings derived from seeds and artificial seeds of perennial ryegrass seed cv. Bronsyn E+(endophyte plus, 2667 seed batch) Twenty seedlings of Bronsyn E⁺ (2667) seeds germinated on filter paper were 25 transferred to soil.

Twenty five seedlings from germinated artificial seeds generated with Bronsyn E plus (2667) seed-derived embryos were transferred to soil. The embryos in artificial seeds were sterilized using 10% H₂SO₄ overnight treatment.

Following 6 week grow-out of seedlings derived from seeds and artificial seeds, endophyte presence was assessed based on endophyte-specific

-1332017251752 25 Oct 2017

SSR test.

Twenty seedlings of Bronsyn E plus (2667; containing ST endophyte) seeds germinated on filter paper were transferred to soil, leading to 13 of 19 seedlings (68%) testing positive for ST endophyte presence in the endophyte-specific SSR test.

Twenty five seedlings from germinated artificial seeds generated with Bronsyn E plus (2667) seed-derived embryos were transferred to soil. The embryos in artificial seeds were sterilized using 10% H2SO4 overnight treatment, leading to 19 of 23 seedlings (83%) testing positive for ST endophyte in the endophyte-specific SSR test, clearly 0 indicating that the methods for seed surface sterilization, large-scale embryo isolation, and artificial seed production with Ca-alginate coating do not negatively affect viability of a resident endophyte.

Large-Scale Inoculation of Endophytes in Perennial Ryegrass Artificial Seeds

Different methods for the large-scale inoculation of endophtyes in perennial ryegrass artificial seeds were developed, with examples of methods 1 to 3 described below:

Inoculation of Isolated Seed-derived Embryos with Endophyte Mycelium and Production of Endophyte-infected Artificial Seeds in Perennial Ryegrass

Freshly isolated seed-derived embryos of perennial ryegrass are individually placed in wells of a) 96-well and b) endophyte mycelium suspension added to individual wells and allowed to partly air-dry under laminar flow prior to c) production of artificial seeds coated with Ca-alginate layer (Figure 126).

Method 1: Direct Inoculation of Isolated Embryos with Endophyte Suspension Prior to Ca-Alginate Coating

Method 1, inoculation of isolated seed-derived embryos with endophyte mycelium and production of endophyte-infected artificial seeds in perennial ryegrass, is based on direct inoculation of isolated embryos with endophyte suspension prior to Ca30 alginate coating as follows:

Freshly isolated embryos of perennial ryegrass are incubated with endophyte

-134 2017251752 25 Oct 2017 suspension (1/16 dilution) for 30 mins at RT in individual wells of 96-well plates.

Inoculation suspension is removed from well and inoculated embryos are allowed to partly air-dry on filter paper disks.

Artificial seeds are produced (Figure 127) with endophyte-inoculated embryos with 3% sodium alginate-containing modified MS growth medium [MS (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO4⁺1.95 g MES+ 1mg/l BAP],

Artificial seeds are allowed to germinate on MS medium for germination.

Freshly isolated embryos of perennial ryegrass are directly inoculated with endophyte suspension (1/8 dilution), partly air-dried and then coated with Ca-alginate in individual wells of 96-well plates.

Artificial seeds from perennial ryegrass directly inoculated with endophyte and then coated with Ca-alginate layer are able to germinate on MS germination medium (Figure 128).

Method 2: Direct Coating of Isolated Embryos with Endophyte-Containing CaAlginate Layer

Method 2, inoculation of isolated seed-derived embryos with endophyte mycelium and production of endophyte-infected artificial seeds in perennial ryegrass, is based on direct coating of isolated embryos with endophyte-containing Ca-alginate layer as 25 follows:

Embryos of perennial ryegrass are freshly isolated in endophyte suspension (1/16 dilution) in individual wells of 96-well plates.

Two-fold concentration sodium alginate (6%) modified MS medium [MS (without

CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO₄+1.95 g MES+ 1mg/l BAP] is added to 30 the individual wells to coat embryos with an endophyte-containing alginate layer

Artificial seeds are produced with endophyte-layer coated embryos (Figure 129).

Artificial seeds are allowed to germinate on MS medium for germination

-1352017251752 25 Oct 2017

Embryos of perennial ryegrass are freshly isolated and coated with endophyte suspension (1/8 or 1/16 dilutions) with Ca-alginate then added to generate an endophyte-containing alginate layer coating the embryos in individual wells of 96-well plates.

Following culture, endophyte out-growth is observed from the endophyte-containing alginate layer used to coat the isolated embryos of perennial ryegrass (irrespectively of endophyte suspension dilution rate used; Figure 130) demonstrating viability of the endophyte included in the Ca-alginate coating layer.

o

Method 3: Double-Coating of Artificial Seeds Generated from Endophyte Inoculated Isolated Embryos

Method 3, inoculation of isolated seed-derived embryos with endophyte mycelium and production of endophyte-infected artificial seeds in perennial ryegrass, is based 5 on double-coating of artificial seeds generated from endophyte-inoculated isolated embryos as follows:

Freshly isolated embryos of perennial ryegrass are coated with an endophyte suspension (1/16 dilution), mixed with alginate [6% Ca-alginate in modified MS 0 medium (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO4+1.95 g MES+ 1mg/l

BAP] to generate a first coating layer containing endophytes in individual wells of 96well plates.

Artificial seeds with a first endophyte-containing alginate layer coating freshly isolated 25 embryos of perennial ryegrass are blot-dried on filter paper in laminar air flow for 30 mins and then coated with a second alginate layer of 3% Ca-alginate without any nutrients.

Double-coated artificial seeds with endophyte-containing layer coated embryos of 30 perennial ryegrass are then germinated on MS medium

Second coating with nutrient deprived medium of endophyte-inoculated artificial

-1362017251752 25 Oct 2017 seeds aims to reduce endophyte out-growth during germination and restrict endophyte growth in close proximity to isolated perennial ryegrass embryo (Figure 131).

Artificial seeds with a first endophyte-containing alginate layer coating freshly isolated embryos of perennial ryegrass are blot-dried on filter paper in laminar air flow for 30 mins and then coated with a second alginate layer of 3% Ca-alginate without any nutrients.

Endophyte growth is mainly restricted to inner alginate coating layer for a period of up to 3 weeks (Figure 132).

Embryos of perennial ryegrass are freshly isolated directly in endophyte suspension (1/8 dilution), then partly air-died and coated with a first alginate layer [3% Ca5 alginate in modified MS medium (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO4+1.95 g MES+ 1mg/l BAP] in individual wells of 96-well plates.

Artificial seeds with directly endophyte-inoculated embryos of perennial ryegrass are stored at 4C overnight and then coated with a second alginate layer of 3% Ca0 alginate without any nutrients.

Double-coated artificial seeds with directly endophyte-inoculated embryos of perennial ryegrass are then germinated on MS medium.

Double-coated artificial seeds with directly endophyte-inoculated embryos of perennial ryegrass germinated on MS medium show germination rates comparable to the original seed batch used for embryo isolation (Figure 133).

Assessing Endophyte Presence in Seedlings Derived from Artificial Seeds with 30 Seed-Derived Embryos Inoculated with Novel Endophytes

In order to assess endophyte presence in seedlings derived from artificial seeds with seed-derived embryos inoculated with novel endophytes (e.g. NEA11) using Method

-1372017251752 25 Oct 2017

1, the following experiment was undertaken:

Endophyte presence in seedlings derived from artificial seeds produced with embryos from perennial ryegrass seed cv. Bronsyn E- (endophyte minus, 2668 5 seed batch) inoculated with novel endophyte NEA11

Following 6 week grow-out of seedlings derived from artificial seeds, endophyte presence was assessed based on endophyte-specific SSR test.

Twenty-three seedlings from germinated artificial seeds generated with o Bronsyn E minus (2668) seed-derived embryos inoculated with NEA11 using Method were transferred to soil. 6 of 23 seedlings (i.e. 26%) tested positive for NEA11 endophyte presence in the endophyte-specific SSR test demonstrating the establishment of symbiota (Table 51). Endophyte presence in symbiota established from germinated artificial seeds generated with perennial ryegrass seed-derived embryos inoculated with novel endophyte NEA11 using Method 1 was confirmed following 3 months after transfer to soil.

SSR Marker

NLESTA1QA09	NLESTA1NG03	NLESTA1CC05
	Seedling	Allele 1	Allele 2	Allele 1 Allele 2	Allele 1 Allele 2	Endophyte detected
0	2668_14	153	184	226	167	NEA11
2668_15	153	184	226	167	NEA11
	2668_1	153	184	226	167	NEA11
	22	153	184	226	167	NEA11
	2668 Bb1	153	184	226	167	NEA11
	2668_13	153	184	226	167	NEA11

Table 51 - Assessing Endophyte Presence in Seedlings Derived from Artificial 25 Seeds with Seed-Derived Embryos Inoculated with Novel Endophytes

Large-Scale Inoculation of Designer Endophytes in Perennial Ryegrass Artificial Seeds

Large-scale inoculation of designer endophtyes derived from induced mutagenesis 30 through colchicine-treatment (e.g. NEA12dh17) or derived from X-ray mutagenesis (e.g. IRM1-35) in perennial ryegrass artificial seeds is carried out using methods 1 to 3 described above.

-1382017251752 25 Oct 2017

Freshly isolated embryos of perennial ryegrass are incubated with designer endophyte (e.g. NEA12dh17, IRM1-35) suspension (1/16 dilution) for 30 mins at RT in individual wells of 96-well plates.

Inoculation suspension is removed from well and inoculated embryos are allowed to partly air-dry on filter paper disks.

Artificial seeds are produced with designer endophyte-inoculated embryos with 3% 0 sodium alginate-containing modified MS growth medium [MS (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO₄+1.95 g MES+ 1mg/l BAP],

Artificial seeds are allowed to germinate on MS medium for germination.

Freshly isolated embryos of perennial ryegrass are directly inoculated with designer endophyte (e.g. NEA12dh17, IRM1-35) suspension (1/8 dilution), partly air-dried and then coated with Ca-alginate in individual wells of 96-well plates.

Artificial seeds from perennial ryegrass directly inoculated with designer endophytes 0 (e.g. NEA12dh17, IRM1-35) and then coated with Ca-alginate layer are able to germinate on MS germination medium leading to the establishment of symbiota. Designer endophyte presence and identity in the symbiota generated following largescale inoculation of designer endophtyes derived from induced mutagenesis through colchicine-treatment (e.g. NEA12dh17) or derived from X-ray mutagenesis (e.g. 25 IRM1-35) in perennial ryegrass artificial seeds is demonstrated using an endophytespecific SSR test.

Large-Scale Inoculation of Transgenic Endophytes in Perennial Ryegrass Artificial Seeds

Large-scale inoculation of transgenic endophtyes derived from genetic transformation of NEA12 endophyte with plasmid containing a chimeric gene for expression of the DsRed fluorescent marker gene (e.g. NEA12-DsRed) in perennial ryegrass artificial

-1392017251752 25 Oct 2017 seeds is carried out using method 1 described above.

Freshly isolated embryos of perennial ryegrass are incubated with transgenic endophyte (e.g. NEA12-DsRed) suspension (1/16 dilution) for 30 mins at RT in 5 individual wells of 96-well plates.

Inoculation suspension is removed from well and inoculated embryos are allowed to partly air-dry on filter paper disks.

Artificial seeds are produced with transgenic endophyte-inoculated embryos with 3% sodium alginate-containing modified MS growth medium [MS (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO₄+1.95 g MES+ 1mg/l BAP],

Artificial seeds are allowed to germinate on MS medium for germination.

Freshly isolated embryos of perennial ryegrass are directly inoculated with transgenic endophyte (e.g. NEA12-DsRed) suspension (1/8 dilution), partly air-dried and then coated with Ca-alginate in individual wells of 96-well plates.

Artificial seeds from perennial ryegrass directly inoculated with transgenic endophyte (e.g. NEA12-DsRed) and then coated with Ca-alginate layer are able to germinate on MS germination medium leading to the establishment of symbiota with transgenic endophytes. Transgenic endophyte presence and identity in the symbiota generated following large-scale inoculation of transgenic endophtye (e.g. NEA12-DsRed) in 25 perennial ryegrass artificial seeds is demonstrated using an endophyte-specific SSR and transgene-specific PCR test.

Large-Scale Inoculation of Novel Endophytes in Tall Fescue Artificial Seeds

Large-scale inoculation of novel endophtyes from tall fescue (e.g. NEA17, NEA19, NEA20) in tall fescue artificial seeds is carried out using method 1 described above. Freshly isolated embryos of tall fescue are incubated with novel fescue endophytes

-140 2017251752 25 Oct 2017 (e.g. NEA17, NEA19, NEA20) suspension (1/16 dilution) for 30 mins at RT in individual wells of 96-well plates.

Inoculation suspension is removed from well and inoculated embryos are allowed to 5 partly air-dry on filter paper disks.

Artificial seeds are produced with novel endophyte-inoculated embryos with 3% sodium alginate-containing modified MS growth medium [MS (without CaCI₂) + 750 mg/L glutamine + 5μΜ CuSO4+1.95 g MES+ 1mg/l BAP],

Artificial seeds are allowed to germinate on MS medium for germination.

Freshly isolated embryos of tall fescue are directly inoculated with novel fescue endophytes (e.g. NEA17, NEA19, NEA20) suspension (1/8 dilution), partly air-dried 5 and then coated with Ca-alginate in individual wells of 96-well plates.

Artificial seeds from tall fescue directly inoculated with novel fescue endophytes (e.g. NEA17, NEA19, NEA20) and then coated with Ca-alginate layer are able to germinate on MS germination medium leading to the establishment of symbiota. 0 Novel endophyte presence and identity in the symbiota generated following largescale inoculation of novel fescue endophtyes (e.g. NEA17, NEA19, NEA20) in tall fescue artificial seeds are demonstrated using an endophyte-specific SSR test.

Example 17 - Method for Selection of Stable Symbiota

Generational Stability and Viability of Perennial Ryegrass Endophytes

Experience within the forage genetic supply industry (technically orientated pasture plant seed breeding commercial companies) has created a series of requirements for stability of endophyte presence in grass crops grown from certified seed batches. In terms of inter-generational stability (that is, the presence of endophyte in a seed 30 batch from one generation of varietal multiplication as compared to its offspring generation), the maximum level of loss that can be generated is 5%, but ideally not more than 2-3% loss per generation. Furthermore, due to requirements for

-141 2017251752 25 Oct 2017 warehouse storage of large-scale seed batches, the endophyte must be stable and viable (allowing recolonisation of the germinated seeding) in seed for at least 3 years after harvest.

Experimental strategies that can be applied to identify and predict intergenerational stability and viability include:

• The use of molecular genetic marker technology, such as endophyte-specific simple sequence repeat (SSR) assays, to identify the presence or absence of endophyte on the basis of a diagnostic PCR test (presence versus absence, and characteristic size of generated PCR product). This test is appropriate to track changes in stability in plant populations from successive populations. However, it does not confirm viability in the seed.

• Quantitative PCR (qPCR) is a method for detection of endophyte-specific gene expression, and is hence appropriate for retrospective confirmation of endophyte viability in seed, given germination of seedlings.

Accelerated ageing (AA) is an artificial method, based on application of specific temperature and relative humidity (RH) conditions (Gundel et al., 2009), that can mimic the effects of longer-term storage on endophyte viability over a short-period 0 of time (Happ et al., 1993). AA tests are hence suitable to differentiate seed quality among commercially acceptable perennial ryegrass seed lots (Wang etal., 2004) in order to provide a predictive assay within a time-frame suitable for commercial decisions.

The objective of the study in question was therefore to develop an AA method for seed of grass-endophyte symbiota, which can be applied to merit ranking of endophytes based on their predicted viability in stored seed. The method is also applicable for selection of stable symbiota with long-term viable endophytes for deployment as a tool in next-generation ab initio molecular breeding, selection and evaluation of grass-endophyte symbiota [rather than breeding and selection of grass host followed by endophyte inoculation and symbiotum evaluation only]. The method

-142 2017251752 25 Oct 2017 is applicable to symbiota of both perennial ryegrass and tall fescue [different host genetic backgrounds; different endophytes].

The AA method was developed using a range of humidity levels and treatment times 5 of accelerated ageing followed by different storage times and conditions (summarised in Table 52), as applied to a number of novel endophyte-grass associations (different endophytes i.e. ST, AR1, NEA3, NEA2, NEA6 and AR37; Table 52) in a single genetic background (i.e. cv. Bronsyn).

Experimental stage	Condition	Variation in condition
Accelerated Ageing	Temperature (°C)	42 °C
Humidity level (%)	100%	80%	Control
Time (days)	4 days	7 days	0 days
Storage	Temperature (°C)	15-25°C	40°C	3-5°C	-20°C
Time (months)	0 days (control)	8	14	ongoing
Table 52 - Conditions for accelerated	ageing

Figure 134 shows an outline of the AA process. In summary:

• Following treatment:

• 400 seed (100 per replication) per treatment were germinated without storage • 400 seed (100 per replication) per treatment were placed in storage • Following 14 days of germination, seedlings were counted and the germination rate was recorded • Seedlings were grown for 8 weeks prior to assessing endophyte presence

-1432017251752 25 Oct 2017

Endophyte strain	Type	Origin	Alkaloids
ST	Wild type		L+ E + P
NEA3	Novel	Agriseeds	E + P
NEA2	Novel	Agriseeds	E + P
NEA6	Novel	Agriseeds	E + P
AR1	Novel	Ag Research	P
AR37	Novel	Ag Research	Janthitrems

ETable 53 - Endophyte strains inoculated into sub-populations (non-isogenic) of perennial ryegrass cv. Bronsyn E = Ergovaline; L = Lolitrem Β; P = Peramine

Figure 135 shows the results of seed germination rate of perennial ryegrass cv. Bronsyn containing different endophytes after AA treatment of seed followed by seed storage.

These results indicated that:

• There was no effect of endophyte presence on symbiotum seed viability [see control treatment, no AA] • Accelerated ageing at high humidity [i.e. 80% HR] and storage at high temperature [i.e. 40C] has the highest impact on germination [i.e. loss of germination at 24 months in storage].

Figure 136 shows the corresponding data on endophyte viability in seed from perennial ryegrass cv. Bronsyn with different endophytes after AA treatment of seed followed by seed storage.

These results indicated that:

-1442017251752 25 Oct 2017 • AA [i.e. 80% RH for 7 days] is highly informative for prediction of endophyte viability in stored seed • Endophyte viability declines slightly during medium term storage [i.e. 14 months] at room temperature · Seed storage at high temperature [i.e. 40C] leads to a rapid decline in endophyte viability • Viability of the novel endophytes NEA3, NEA2 and NEA6 is lower than for ST, AR1 and AR37

The study was then extended to assess applicability of the AA method for prediction of endophyte viability in stored seed for different endophytes (i.e. ST, NEA10, NEA11, NEA12, E1 and AR1) in different host genetic backgrounds (i.e. cvs. Bronsyn, Alto, Trojan and Bealey) (Table 54, Table 55).

Host Cultivar	Endophyte Strain
	AR1	ST	NEA10 NEA11 NEA12	E1
Bronsyn Alto Trojan Bealey

Table 54 - Endophyte strains and host cultivars for accelerated ageing

Experimental stage	Condition	Variation in condition
Accelerated Ageing	Temperature (°C)	42 °C
Humidity level (%)	100% 80% Control
Time (days)	4 days 7 days 0 days

Table 55 - Variations in condition for accelerated ageing

Figure 137 shows the seed germination rates of symbiota representing different endophytes in different host genetic backgrounds after accelerated ageing treatment of seed.

-1452017251752 25 Oct 2017

Figure 138 shows endophyte viability values for different endophytes in different host genetic backgrounds after accelerated ageing treatment of seed.

These results indicated that:

• AA treatement has a high impact on endophyte viability.

· AA treatment at 100%RH for 4 -7 days is highly informative for prediction of endophyte viability in seed.

• AA allows to ranking of endophytes based on viability across a range of host distinct genetic backgrounds.

• The rank order of viability ranking among novel endophytes is E1 < NEA10 <

NEA11 and NEA12.

Figure 139 shows the endophyte viability values of different endophytes in different host genetic backgrounds after accelerated ageing treatment of seed.

AA treatment [i.e. 100%RH for 4d or 7d] was hence demonstrated to reduce endophyte viability thus allowing for its potential use as a selection tool for stable associations.

These results demonstrate that AA treatment can be used for selection of stable 0 symbiota generated by artificial inoculation.

• AA treatment offers the opportunity to be applied as a selection tool for stability of symbiota • AA treatment [i.e. 100%RH for 4 -7 days] reduces endophyte viability, and so can be used for highly effective counter-selection against unstable associations that would prove to be problematic in commercial practice.

• AA treatment permits symbiota to be ranked based on their stability • Identification or selection of stable symbiota was demonstrated for the following combinations: NEA10 in Alto, NEA11 in Bealey, NEA12 in Trojan and in Alto.

-1462017251752 25 Oct 2017

The AA protocol is extended to seed of tall fescue-endophyte associations, by extension of the protocol for prediction of host-endophyte association viability that was developed for perennial ryegrass seed; including testing of multiple fescue endophytes (AR542 [MaxP™ /MaxQ™ /ArkPlus™]; E34; KY31) in one or more tall 5 fescue genetic backgrounds. Conditions for AA are described in Table 56 and tall fescue-endophyte symbiota evaluated are described in Table 57.

Experimental stage	Condition	Variation in condition
Accelerated Ageing	Temperature (°C)	42 °C
Humidity level (%)	100%	80%	Control
Time (days)	4 days	7 days	0 days
Storage	Temperature (°C)	15-25°C.	40°C	3-5°C	-20°C
Time (months)	0 (control)	6	12	ongoing

Table 56 - Comparison of accelerated ageing and storage

Grass Host	Endophyte
Quantum II	Max-P
Bar Optima	E34
Jesup	Max-Q
Kentucky 31	KY31

Table 57 - Tall fescue-endophyte symbiota subjected to treatments for accelerated ageing and storage

-1472017251752 25 Oct 2017 ln conclusion:

• A method for AA treatment [i.e. 80% -100% RH for 4-7 days] was established.

• The method allowed prediction of endophyte viability in stored seed (based on a range of endophytes assessed in single and different host genetic backgrounds).

• The method permitted merit ranking of novel endophytes according to predicted viability in stored seed (based on a range of endophytes assessed in single host genetic background).

• The method permitted selection and ranking of symbiota according to stability, o

Example 18 - Method for Rapid Endophyte Viability assessment Objective:

The objective of the work was to develop a fast, reliable and low-cost method for determining endophyte viability in perennial ryegrass seed. The corresponding assay 5 requirements were defined as follows.

Assay Requirements:

1. Rapid determination - 3 to 5 day-old seedlings

2. Robust and reliable

3. Sensitive enough to use from single seed to seed batches

4. Specific to Neotyphodium endophytes (i.e. does not detect other fungi)

5. Detects live endophyte only

-1482017251752 25 Oct 2017

Early Endophyte Gene Expression-Based TaqMan Assay

The steps in the assay design were as follows:

a) Transcriptome sequence data available for Bronsyn +/- ST (6 week old tillers) plants were used to identify candidate endophyte-specific genes expressed in planta, through BLAST sequence analysis of data against database containing endophyte, other fungi, Brachypodium gene sequences;

b) Five endophyte-specific genes expressed in planta (6 week old tillers of perennial ryegrass plants from Bronsyn) were identified - LtmJ and 4 uncharacterised proteins (7490, 8263, 0005, 2232);

c) Primers were designed for the TaqMan assay as follows:

- DNA primers and probe - both dead and live endophyte samples will amplify

- RNA primers and probe - only live, viable endophyte samples will amplify

- Primers for Bronsyn GAPDH plant gene for seed viability assessment

-1492017251752 25 Oct 2017

- DNA primers and probe - both viable and unviable seed will amplify

- RNA primers and probe - only viable seeds will amplify

Figure 140 shows primers designed for TaqMan assay.

TaqMan Primer Functionality

TaqMan primer functionality was tested on samples from:

- 6 week-old tillers of Bronsyn + ST

- 6 week-old tillers of Bronsyn endophyte free as negative control

- ST mycelium from in vitro endophyte culture as positive control to enable addressing the following experimental hypotheses:

a) Can the endophyte-specific transcripts [i.e. LtmJ and 4 uncharacterised proteins (7490, 8263, 0005 and 2232)] be detected in separately extracted

DNA and RNA samples of ST endophyte mycelium?

b) Can the plant-specific gene [i.e. GAPDH] transcript be detected in samples of separately extracted RNA and DNA of perennial ryegrass leaf material?

c) Can endophyte-specific transcripts [i.e. LtmJ and 4 uncharacterised proteins (7490, 8263, 0005, 2232)] be detected in samples of separately extracted

RNA and DNA from endophyte-infected perennial ryegrass [Bronsyn + ST] leaf material?

Figure 141 shows shows TaqMan primer functionality test for the endophyte-specific gene LtmJ · There was No RNA in ST culture as expected • Only expressed in planta.

Figure 142 shows TaqMan primer functionality test for the endophyte-specific gene 7490

Figure 143 shows TaqMan primer functionality test for the endophyte-specific gene

8263

-150 2017251752 25 Oct 2017

Figure 144 shows TaqMan primer functionality test for the endophyte-specific gene 0005

Figure 145 shows TaqMan primer functionality test for the endophyte-specific gene 2232

Figure 146 shows TaqMan assay control - no template.

Figure 147 shows TaqMan assay control - Plant GAPDH.

The results obtained can be summarised as follows:

• LtmJ primers work well and LtmJ is not transcribed in endophyte (ST) in vitro culture as expected since Ltm gene is only expressed in planta · Primers designed for the other 4 endophyte-specific genes encoding uncharacterised proteins and expressed in tillers of established grassendophyte symbiota all function (endophyte viability assessment)

RNA/DNA Co-Extraction from Pooled Samples of Endophyte-Infected Perennial 0 Ryegrass Plants

In addition, experiments were undertaken using samples generated by co-extraction of both DNA and RNA in a single step from:

• Co-extraction of both DNA and RNA in single step

- 6 week-old tillers of perennial ryegrass cv. Bronsyn + ST endophyte

- 6 week-old tillers of perennial ryegrass cv. Bronsyn endophyte free

- 7 day-old isolated germinating seed-derived epicotyls (pooled samples) from perennial ryegrass cv. Bronsyn + ST endophyte.

· Test TaqMan primers on co-extracted DNA/RNA

Test primers for in p/anfa-expressed endophyte-specific genes

7490 and 0005.

-151 2017251752 25 Oct 2017 • Is the co-extracted DNA/RNA sample of sufficient quality for primers/probes to work?

• Do the endophyte-specific test genes expressed in 7 day-old epicotyls from grass-endophyte symbiota enable early in planta endophyte viability assessment?

Figure 148 shows Detection in Co-Extracted DNA/RNA Pooled Samples for the Endophyte-Specific Gene 7490.

Figure 149 shows Detection in Co-Extracted DNA/RNA Pooled Samples for the Endophyte-Specific Gene 0005.

Figure 150 shows a control (no template) for Detection in Co-Extracted DNA/RNA Pooled Samples.

RNA/DNA Co-Extraction from Individual 10 Day-Old Germinating Seed-Derived Epicotyls

In addition, experiments were undertaken using samples generated by co-extraction of both DNA and RNA in a single step from:

- 10 day-old isolated germinating seed-derived epicotyls (individual seeds)

- 7 day-old isolated germinating seed-derived epicotyls (pooled seeds)

These samples were used to test the designed TaqMan primers on co-extracted 25 DNA/RNA, specifically to test primers designed for in p/anta-expressed endophytespecific genes 7490, 0005 and 2232 to answer the following question:

• Can endophyte-specific test gene expression be detected in RNA/DNA coextracted samples from individual 10 day-old isolated germinating seedderived epicotyls?

Figure 151 shows detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls of the endophyte-specific gene 2232.

-152 2017251752 25 Oct 2017

Figure 152 shows detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls of the endophyte-specific gene 7490.

Figure 153 shows detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls of the endophyte-specific gene 0005.

Figure 154 shows a control (no template) for detection in co-extracted DNA/RNA samples from individual 10 day old epicotyls.

o

Is Endophyte-Specific Gene Expression Detectable at Earlier Developmental Stages?

In addition, to assess if endophyte-specific gene expression is detectable at earlier developmental stages, experiments were undertaken using samples generated by 5 co-extraction of both DNA and RNA in single step from:

- 7-day old isolated germinating seed-derived epicotyls (pooled)

- 5 day-old isolated germinating seed-derived epicotyls (pooled)

- 3 day-old isolated germinating seed-derived epicotyls (pooled)

These samples were used to test the designed TaqMan primers on co-extracted 0 DNA/RNA, specifically to test primers designed for in planta-expressed endophytespecific genes 7490, 0005 and 2232 to answer the following question:

• Are the endophyte-specific genes expressed earlier than 7 days?

Figure 155 shows detection in co-extracted DNA/RNA from 3, 5 and 7 day-old pooled 25 epicotyls of the endophyte specific gene 2232.

Figure 156 shows detection in co-extracted DNA/RNA from 3, 5 and 7 day-old pooled epicotyls of the endophyte specific gene 7490.

Figure 157 shows detection in co-extracted DNA/RNA from 3, 5 and 7 day-old pooled epicotyls of the endophyte specific gene 0005.

-1532017251752 25 Oct 2017 ls Endophyte-Specific Gene Expression Detectable at Earlier Developmental Stages in a Single Whole Germinating Seed?

In addition, to assess if endophyte-specific gene expression is detectable at earlier (<

days) developmental stages in a sample from a single whole germinating seed, 5 experiments were undertaken using samples generated by co-extraction of DNA and RNA from single-seed samples from:

- 3 day-old isolated germinating seed-derived epicotyls (control) day-old single whole germinating seed including epicotyl) day-old single whole germinating seed including epicotyl) day-old single whole germinating seeds seeds seeds (i.e. individual (i.e. individual (i.e. individual whole whole whole seed including epicotyl).

and to confirm that there sufficient signal from a single whole germinating seed and that the assay can be undertaken on 100 single whole germinating seeds individually to enable a quantitative assessment of endophyte viability at % level.

Assay Validation

Assay validation is undertaken by assessing:

a) endophyte-specific gene expression to be detected in symbiota with novel endophytes i.e. endophytes other than ST e.g. AR1, AR37, NEA2, NEA6, NEA10, NEA11, NEA12 and E1;

b) endophyte-specific gene expression to be detected in symbiota with different host genetic backgrounds i.e. perennial ryegrass from e.g. genotypes Bronsyn 25 and Alto;

Furthermore, for assay validation, results from the quantitative rapid assay (i.e. endophyte-specific gene expression-based assay at single whole germinating seed level; 3-10 days) are compared with assays detecting endophyte colonisation of 30 host and symbiota establishment (i.e. current endophyte DNA detection-based assay at 6 week-old plant level).

-154 2017251752 25 Oct 2017

In addition, the assay developed for perennial ryegrass-endophyte symbiota is assessed for its applicability with primers designed for detecting endophyte-specific gene expression in tall fescue-endophyte symbiota i.e. seeds from different endophytes (e.g. E34, E77, E79 and E80) in different tall fescue genetic backgrounds 5 (e.g. BarOptima, BarSelect and BarElite), as well as in grass-endophyte associations of Acremonium endophytes in Brachiaria and Urochloa grasses.

Example19 - Molecular Phenotyping as Tool in Ab Initio Breeding I

Significant genotype x genotype (G x G) effects have been observed in the process 0 of identifying stable host-endophyte associations for novel endophytes.

Designer associations are compatible, stable and may be selected for at a population level for additional phenotypes (performance, alkaloid production, etc.). For instance, particularly stable associations have been observed in the following combinations 5 (ryegrass variety - novel endophyte):

• Impact-NEA10 • Tolosa-NEA12 • Bealey - E1

The objectives of the study were therefore to develop molecular phenotyping as a tool in ab initio breeding and selection of symbiota, based on the following steps:

• Deployment of multiple endophytes across members of multiple host grass populations • Metabolic profiling of relevant alkaloid content.

· Selection on the basis of individual symbiota performance.

In order to identify individual plants within a population with optimal alkaloid production, 80 plants from a single population of perennial ryegrass variety Bealey, which contains a mixture of two distinct endophyte strains (NEA2 and NEA6) were subjected to semi-quantitative metabolic profiling. Variation of 30 alkaloid profile and content was observed across the population of plants (Fig.

158). The two endophyte strains have been observed to exhibit different characteristic alkaloid profiles: NEA2 has previously been described (van Zijll de

-1552017251752 25 Oct 2017

Jong et al. 2008) as producing lolitrem B, no ergovaline and moderate levels of peramine, while NEA6 produces low levels of lolitrem B, moderate ergovaline and moderate peramine. Identification of lolitrem B-producing plants within the population confirmed the presence of NEA2 (in a majority of the plants) rather 5 than NEA6.

Figure 158 shows metabolic profiling of a population of perennial ryegrass-endophyte symbiota

The study was then extended to identify individual plants within a population with favourable (high peramine, low ergovaline, no/low lolitrem B) toxin profiles. A genetic identity test based on SSR analysis was critical for identification of endophyte presence and identity and to enable accurate selection. Genetic variation within the plant host appeared to influence the favourable toxin profile within a sub-group characterised by a single endophyte genotype (e.g. NEA6), as shown by quantitative variation across a group of peramine-producing plants (Figure 159, Table 58). The data can be used for breeding selection processes.

Sample name	QA09	NG03	CC05	Endophyte detected	Quantitative score	peramine	ergovaline	lolitrem B
2B	171	328	165	NEA2	3	20024646	229761	2093005
4A	194	226	187	NEA6	3	4322114	1593878	0
11A	194	226	187	NEA6	3	3980110	1062058	0
15B		226	187	NEA6	2	0	0	0
27B	171	328	165	NEA2	3	20788755	206953	904985
30A	171	328	165	NEA2	3	11367920	161304	675423
44A	171	328	165	NEA2	3	9964578	233454	436925
46A	194	226	187	NEA6	3	9706442	1505812	0

Table 58 - SSR marker analysis of selected samples

-1562017251752 25 Oct 2017 ln summary, this study demonstrated:

1. Development of a molecular phenotyping method for alkaloid (P/E/L) production as a tool for ab initio molecular breeding of perennial ryegrass-endophyte symbiota.

2. Deployment of molecular phenotyping method in proof-of-concept ab initio selection of perennial ryegrass-endophyte symbiota consisting of two endophytes in one grass host population.

3. Effective selection of sub-population of perennial ryegrass-endophyte symbiota consisting of two endophyte and multiple grass host genotypes for optimal alkaloid (P/E/L) production.

Example 20- Molecular Phenotyping as Tool in Ab Initio Breeding II

Current forage grass breeding does not examine a wide range of compounds or molecules, largely due to cost restrictions. Breeding improved forages through the application of molecular phenotyping of the symbiota for improved herbage quality 5 can deliver forage cultivars with elite ruminant nutrient profile.

The objective of this work was to develop high throughput cost effective molecular phenotyping protocols that can be used as a tool in breeding symbiota. This was achieved through the following steps:

· Quantification of herbage quality, measured by water soluble carbohydrate (WSC): protein ratio

- Improved WSC : Protein ratio is expected to improve productivity as well as reducing the environmental impact of the farming system · Selection on the basis of individual symbiota performance

Identification of Individual Plants within a Population with Optimal WSC: Protein Ratio

Figure 160 shows the results of molecular phenotyping of a population of symbiota to 30 enable ab initio breeding and selection.

-1572017251752 25 Oct 2017

Water Soluble Carbohydrate Quantification

Several methods exist for the quantification of WSC in plants samples, however none have been widely adopted in forage breeding due to various limitations in throughput and cost.

The protocol developed is based upon an enzymatic assay, with the key improvement being the application of robotic automation to sample processing that delivers the following system:

• Fast and semi-automated • Provides individual concentrations for glucose, sucrose, fructose and fructans • Accurate, with demonstrated application in perennial ryegrass • Demonstrated correlation of the enzymatic method to HPLC results

Figure 161 shows the production of glucose-6-phosphate from glucose.

Figure 162 shows the presence of water soluble carbohydrates in a population of symbiota.

The protocol developed and has been exemplified by the following experiment:

· A collection of c. 1000 perennial ryegrass genotypes x 4 clonal replicates with technical replicates were sampled at the three leaf stage from an established field trial.

• Samples were heat-treated and then freeze dried before being processed through the semi-automated pipeline of carbohydrate extraction and enzymatic determination of the glucose, fructose, sucrose and fructan content.

• c. 400 plants a day processed using this protocol.

• Plants with high and low carbohydrate levels were identified and subsequently were taken forward for an experimental proof-of-concept breeding step

Figure 163 shows quantification of proteins.

-1582017251752 25 Oct 2017

Measurement of protein can be achieved using several methods currently, however none have been widely adopted in forage breeding due to various limitations in throughput and cost.

The protocol developed is based upon an established colorimetric assay, with the key improvement being the application of robotic automation to sample processing, that delivers the following system:

• Measuring true plant protein • Colorimetric quantification method · Extensive testing of different extraction buffers and colorimetric methods has identified the current method as most optimal - extraction with weak NaOH and quantification using Bradford assay, detection using plate reader.

Achievements

In summary, the following was achieved:

1. An automated molecular phenotyping method for water soluble carbohydrate and total protein was developed as a tool for ab initio molecular breeding of perennial ryegrass-endophyte symbiota;

2. These molecular phenotyping methods were deployed in proof-of-concept screens across large perennial ryegrass populations;

3. A sub-population of perennial ryegrass plants with higher water soluble carbohydrate levels was selected in a ‘proof-of-concept’ breeding step

-1592017251752 25 Oct 2017

Example 21 - Method for Endophyte Identification in Symbiota

Seed Bulk vs. Individual Plant Testing for Endophyte Presence, Identity and Incidence

Eight perennial ryegrass lines were tested for NEA2/NEA6 endophyte presence, identity and proportion using SSR markers.

plants per line

23x10 seed bulks per line

Cultivar/Line	Source
Bealey NEA2	Nucleus'11
Bealey NEA2	11235AT2
Bealey NEA2	12138A
Trojan NEA2/NEA6	951T1
Trojan NEA2/NEA6	13181A
Trojan NEA2/NEA6	13206
Bealey NEA2	Nucleus'02
Bealey NEA2	RG0339

Table 59 - Cultivars and sources

SSR Marker	Expected SSR profile (allele size in base pairs)
Standard Toxic (ST)	NEA2	NEA3	NEA6	NEA10	NEA11	NEA12	RBG- E1
NLESTA1QA09	149	171	198	194	194	153,184	188	160
NLESTA1NG03	226	328	226	226	226	226	217	217
NLESTA1CC05	164	167	203	187	187	167	138	135
B11	176	176	208	199	183	356,360	131	164

Table 60 - Expected SSR profile

-160 2017251752 25 Oct 2017

Line/ Cultivar	Source	Endophyte Identification	Proportion of NEA2 & NEA6 in endophyte positive samples
NEA2	NEA6	E	Major endophyte detected	Proportion of NEA2 (%)	Proportion of NEA6 (%)
Bealey NEA2	Nucleus'11	65	11	9	NEA2	85.5	14.5
Bealey NEA2	11235AT2	57	12	16	NEA2	82.6	17.4
Bealey NEA2	12138A	39	3	42	NEA2	92.9	7.1
Trojan NEA2/NEA6	951T1	28	19	39	NEA2	59.6	40.4
Trojan NEA2/NEA6	13181A	26	34	25	NEA6	43.3	56.7
Trojan NEA2/NEA6	13206	18	46	21	NEA6	28.1	71.9
Bealey NEA2	Nucleus'02	63	11	11	NEA2	85.1	14.9
Bealey NEA2	RG0339	63	8	14	NEA2	88.7	11.3
Total		359	144	176

Table 61 - Individual plants

-161 2017251752 25 Oct 2017

Line/ Cultivar	Source	Endophyte Identification (in no. of seed bulks)	Major endophyte detected in seed lot
NEA2	NEA2/ NEA6	NEA6	E
Bealey NEA2	Nucleus'11	6	17	0	0	NEA2
Bealey NEA2	11235AT2	7	16	0	0	NEA2
Bealey NEA2	12138A	13	10	0	0	NEA2
Trojan NEA2/NEA6	951T1	4	19	0	0	NEA2
Trojan NEA2/NEA6	13181A	0	23	0	0	Both NEA2 & NEA6
Trojan NEA2/NEA6	13206	0	23	0	0	Both NEA2 & NEA6
Bealey NEA2	Nucleus'02	10	13	0	0	NEA2
Bealey NEA2	RG0339	12	11	0	0	NEA2

Table 62 - Seed bulks

Summary • NEA2 was identified as the predominant endophyte in Bealey NEA2/NEA6 · Proportions of NEA2/NEA6 in Bealey have not altered over the 2002-2011 period.

• Roughly equal proportions of NEA2/NEA6 were observed in Trojan.

• Tillers testing is a robust method for unequivocally assessing endophyte presence, identity and incidence.

· Determines the exact percentage of specific endophyte strain incidence in each line.

• Identifies endophyte free plants.

• Bulked seed testing is an indicative method for assessing endophyte incidence.

-162 2017251752 25 Oct 2017 • Determines the relative proportion of each endophyte in a particular line.

• Does not allow to identify endophyte free individuals.

Genetic Purity of Endophyte-Inoculated Perennial Ryegrass Lines · Seed samples from five lines

- Coded as Lp513, Lp539, Lp613, Lp660, Lp667 • Multiple seed bulks (23 per line) analysed with 3 markers

- Diagnostic capacity to discriminate ST, NEA2, NEA6 • Lp513, Lp613 - contain NEA6 > NEA2 · Lp660 - contains NEA2, ST, trace NEA6 • Lp667 - contains ST, NEA6 • Lp539 - contains ST only

Relationship to previous assessments of NEA2, NEA6 incidence? Origin of ST endophyte incursions?

Achievements

1. Established methods for determining endophyte presence, identity and incidence in individual tiller samples

2. Established methods for determining endophyte presence, identity and incidence in bulked seed samples.

:0

Example 22 - Method for allelic content characterisation in symbiota genotypes

Method for Allelic Content Characterisation in Host Genotypes

SNP based marker tools have been developed in perennial ryegrass, transferred to 25 Italian ryegrass and developed in tall fescue.

Comprehensive cultivar catalogues have been developed of genetic diversity and relatedness between available germplasm.

Clonal breeding nurseries have been established and genotyped to assist with breeding decisions.

-1632017251752 25 Oct 2017

SNP based marker tools have been developed to co-genotype host plant and endophyte.

SNP tools have been developed to assess seed batch purity and endophyte content 5 on c. 1,000 seed in a sequence based assay system.

Co-genotyping of the symbiota has been developed and exemplified.

• Protocol:

- Grind 2.5g seed

- Extract DNA

- Perform 384 PCRs on DNA

- Pool PCR products

- Ligate on illumina adaptors

- Amplify sample

- Sequence on the illumina HiSeq2000 - 1-2% spike on a single lane

- Process data using a two step pipeline created in LINUX

- Analyse data using a pair of custom R scripts

Figures 164 and 165 show genetic diversity and relatedness between germplasm.

o

Genotyping By Sequencing

Protocols under development - application of reduced costs of sequencing to:

• Saturate genome with molecular markers • Low cost per sample · High throughput possible • Genome sequence data will assist development • Initial study in maize and barley • 24,186 markers mapped onto the barley genetic map • c. 200,000 markers developed in maize · Published protocol - Elshire et al May 2011 PlosOne • 96 initial custom barcodes developed inhouse.

-164 2017251752 25 Oct 2017

Figure 166 shows genotyping by sequencing.

Figure 167 shows procedure for genotyping by sequencing.

Breeding Selection Program

With expanding data sets available to breeding programs selection of plants will require sophisticated computational tools for maximal gain.

Program Method:

Selection of a sub-set of phenotypically elite individuals that conserve all allelic variants.

Program design:

User defines the size of a sub-set to select.

Program selects a sub-set of individuals that preserve all allelic variation seen in the population.

Program selects to the size of the user defined sub-set elite individuals.

:0

Outputs plant identities, mean phenotype value and number of missing alleles.

Phenotype value is a single figure, but composite values can be generated.

Example: Selection of 150 plants from 1,000 plants - processing time c. 1 hour.

Potential Schema For Cultivar Sub-Selection • From cultivars represented in the clonal nursery PG one50 and LP443 were chosen for sub-selection.

· Select for within cultivar based on phenomic performance and or associated markers, while maintaining overall population allelic frequencies.

-1652017251752 25 Oct 2017 • Expected that c. 6% elite plants would be identified, while c. 20-30 individuals would be needed to retain population diversity.

• Target trait in the initial attempt was quality - digestibility and WSC.

• Identified cohorts of plants from the nursery that would be polycrossed.

· Crossing performed - seeds harvested and counted - germination to follow.

• Trial of PG one50 vs PG one50 subselect as well as LP443 vs LP443 subselect to follow.

• Application amenable to many traits greater success likely with higher heritabilities.

Achievements

1. Established methods for genotyping symbiota from leaf tissue samples.

2. Established methods for genotyping symbiota from seed bulk samples.

3. Established computational breeding tools for selection of elite individuals whilst preserving genetic diversity in a population.

Example 23 - Genomic Selection in Forage Plant Species Abstract

Genomic selection (GS) is a powerful new method for exploitation of DNA sequence polymorphisms in domestic animal and crop plant improvement, through prediction of breeding values based on all markers distributed genome-wide. Forage grasses and legumes, which support a broad range of global pastoral industries, provide important targets for GS implementation, as many key target traits for improvement are difficult 25 or expensive to measure, and are measured late in the life-cycle of selection candidates. A number of biological factors that vary across the range of candidate species, such as reproductive mode and polyploid genome architecture, present both challenges and opportunities for GS. The generic attributes of forage breeding programs are described, along with the status of genomic resources for a 30 representative species group (ryegrasses). A breeding scheme for ryegrass incorporating GS is described, allowing two rounds of selection for key agronomic traits within a time-period previously required for a single round, potentially leading to

-1662017251752 25 Oct 2017 doubling of genetic gain rate for such characters. An extremely limited extent of linkage disequilibrium, which is the major challenge for GS implementation in ryegrass breeding, is addressed. The strategy also incorporates recent advances in DNA sequencing technology to minimise costs.

An alternative approach is to simultaneously use all genome-wide distributed markers to predict breeding values, in an approach known as genomic selection (GS) (Meuwissen et al. 2001). GS uses a panel of markers which are sufficiently dense such that all QTLs are expected to be in LD with at least one marker locus. There 0 are two advantages of GS over MAS. Firstly, all of the genetic variance may potentially be captured by the markers, as marker effects do not need to exceed the significance threshold in order to be used to predict breeding value. Secondly, marker effects are not biased upwards by multiple testing. This is in contrast to marker effects obtained from GWAS: due to the large number tested, it is likely to be 5 those markers with favourable error terms that exceed the significance threshold (Beavis 1994).

In order to implement GS, a reference population of individuals with both known genotypes and phenotypes is used to derive a prediction equation which predicts 0 genomic estimated breeding values (GEBVs) from marker genotypes. This prediction equation can then be used to estimate breeding values for selection candidates for which genotypes, but potentially no phenotypes, have been obtained. In simulation studies the accuracy of GEBV (the correlation of the GEBV with the true breeding value, which determines the rate of genetic gain) for selection candidates 25 may be as high as 0.85 (Meuwissen et al. 2001). In practice, accuracy values as large as this have not yet been reported, although accuracies of GEBV up to 0.71 have been reported in Holstein-Friesian dairy cattle (Van Raden et al. 2009).

The implication of the availability of GEBVs for improvement programs can be 30 profound. GS enables accurate selection decisions to be made early in the life-cycle of the target organism, at the price only of genotyping with the marker panel. It

-1672017251752 25 Oct 2017 follows that the impact of GS on rate of genetic gain will be highest for traits which are either:

• Expensive or difficult to measure, such as herbage quality in forages or grain quality in cereals.

· Measurable only by significant damage to or destruction of the individual, such as nematode infection resistance in cereals, hence preventing subsequent breeding from that individual.

• Expressed late in life, such as crop grain yield, or after the individual has been selected for breeding, such as survival or vegetative persistence.

o

The potential value of GS for crop plant breeding has been recognised (Heffner et al. 2009; Janninck et al. 2010). In forages, yield and quality of herbage are the key traits for forage improvement, and support feed conversion into unit quantities of meat or milk. These characters provide good candidates for GS, as they require expensive 5 and/or destructive late-life measurement. Traits such as in-field persistence are also of obvious interest. This article provides a review of current knowledge relevant to the prospects for GS in forages, and describes possible strategies for implementation.

o Biology of Target Species

In considering the prospects for applying principles of GS to forage species, relevant aspects of biology such as reproductive mode, polyploid status and symbiotic interactions must be considered. These factors may differ markedly from the 25 properties of those outcrossing, diploid animal species for which the approach was pioneered. The major pasture species for current and future use in Australian pastoral agriculture are classified with respect to these multiple biological factors in Table 63, and commentary is made on implications of each factor for GS application in subsequent sections.

-1682017251752 25 Oct 2017

Species	Perennial ryegrass	Italian ryegrass	Warmseason grasses	Tall fescue	White clover	Alfalfa
Attribute
Taxonomic group	Grass	Grass	Grass	Grass	Legume	Legume
Reproductiv e mode	SI	SI	Apomictic	SI	SI	SI
Annualityperenniality	Perennial	Biennial	Perennial	Perennial	Perennial	Perennial
Ploidy level	Diploid	Diploid	Diploid Autotetrapl oid	Allohexaploi d	Allotetraploid	Autotetrapl oid
Symbiotic plantmicrobe association	Fungal (Neotyphodi um Ioiii, LpTG2)	Fungal (Neotyphodi um occultans)	Fungal (Acremoniu m implicatum - Brachiaria spp.)	Fungal (Neotyphodi um coenophialu m, FaTG2, FaTG3)	Bacterial (Rhizobium leguminosaru m)	Bacterial (Rhizobium meliloti)
Key Issues for implementati on of GS	Non-extensive LD, likely high Ne, development of efficient GBS methods	Capacity to select at sexual level and transition elite genotypes to apomictic level	Limited knowledge of LD and Ne, technical constraints on SNP discovery and validation, development of efficient GBS methods	Limited knowledge of LD and Ne, developme nt of efficient GBS methods capable of determining allele dosage

Table 63: Summary information for biological factors influencing feasibility of GS implementation in forage species

-1692017251752 25 Oct 2017

1.1 Taxonomic classification

A number of plant species are important for global pastoral agriculture, the majority belonging to either the grass (Poaceae) or legume (Fabaceae) families. Grasses and 5 legumes are either cultivated separately, or in combination as companion species in a mixed sward. The most important temperate pasture grass species are members of the Lolium-Festuca species complex (ryegrasses and fescues), two closely related genera residing within the Poeae tribe of the cool-season grass sub-family Pooideae, In contrast, various species of warm-season pasture grass (members of the genera 0 Brachiaria, Paspalum, Eragrostis and Pennisetum) belong to several tribes within the sub-family Panicoideae.

The key cool-season species predominantly belong to the genera Medicago (medics, including alfalfa [Medicago sativa L.]) and Trifolium (clovers, including white clover 5 [Trifolium repens L.] and red clover [Trifolium pratense L.]) within the Trifolieae tribe of the Fabaceae clade Hologalegina. There are also a number of agronomically important warm-season pasture legumes, including members of the genera Chamaecrista, Macroptilium, Stylosanthes, Centrosema and Desmodium.

1.2 Reproductive mode

The majority of temperate forage species (with the exception of subterranean clover, T. subterraneum L.) exhibit obligate outbreeding (allogamy), dependent on crosspollination. Examples include perennial ryegrass, Italian ryegrass, tall fescue, meadow fescue (Festuca pratensis Huds. syn. L. pratense), white clover and alfalfa.

Inhibition of inbreeding is due to genetically-controlled self-incompatibility (SI) systems, in which incompatible pollen-stigma interactions arise from specific matching of SI locus alleles in male and female reproductive tissues.

Gametophytic SI (GSI) systems, in which the male gamete phenotype is determined 30 solely by its own haploid genotype, are characteristic of forage grasses and legumes.

For a pasture grass species such as perennial ryegrass, SI contributes to a high level of heterozygosity and heterogeneity, such that intra-population diversity typically

-170 2017251752 25 Oct 2017 exceeds inter-population variability (Guthridge et al. 2001: Wang et al. 2009). This is important for prospects of successful GS application in these species, as discussed below. Limited knowledge of SI allele content for specific genotypes is a potential obstacle to targeted varietal development, especially for restricted-base cultivars 5 (Guthridge et al. 2001), which are based on small numbers of foundation clones.

Effects on fertility and seed yield and constraints on permutation of beneficial genetic variants are most acute under self-fertilisation, or crosses between close-bred genotypes such as full-sibs. In the context of GS implementation, obligate outbreeding leads to high levels of genotypic heterozygosity and rates of 0 polymorphism that are likely to be higher than in livestock species (e.g. The Bovine HapMap Consortium 2009) to which GS has been successfully applied.

Apomixis represents a different reproductive mode, in which embryonic development proceeds in the absence of a paternal contribution from an unreduced egg cell. The 5 products of apomixis are therefore clones of the maternal genotype. A parthenogenic propagation system is highly attractive to plant breeders, due to the ability to preserve and multiply elite genotypes (Spielman et al. 2003). Apomixis is usually associated with hybridity and/or polyploidy, and expression of the trait is generally incomplete, such that a mixture of sexual and asexual fertilisation products 0 is obtained from a single plant. Variation is also observed within species, with fully sexual diploid genotypes and facultative apomictic polyploid genotypes within a single population. Apomictic warm-season grasses which are targets for breeding improvement include several species of Brachiaria (signal grass), Paspalum (dallis grass) and Pennisetum (relatives of cultivated pearl millet), as well as Eragrostis 25 curvula (Schrad.) Nees (weeping lovegrass) (Pessino et al. 1999). For GS, selection at the allogamous sexual diploid level for optimal genotypes could be very effectively linked with large-scale propagation based on transition to a polyploid, parthenogenic level.

Target species also vary in terms of annual or perennial growth habit, the majority being short- or long-lived perennials. This property is potentially advantageous for GS, as clonal replicates from a single fully genotyped individual may be evaluated for

-171 2017251752 25 Oct 2017 multiple traits across a number of distinct environments, providing increased confidence in phenotypic evaluation for derivation of GEBVs. Parthenogenic propagation of single genotypes through apomixis may permit a similar quality of information to be obtained.

1.3 Polyploidy

For allopolyploids, which derive from hybridisation events between distinct taxa, the most important effects on feasibility of GS are two-fold; the significant challenge presented to the process of SNP discovery and validation; and the requirement to 0 reliably assay SNPs associated with causal polymorphisms if such mutations only segregate from a single sub-genome (Kaur et al. 2012). The former arises due to a requirement to distinguish between genuine SNPs (homologous variants between alleles within a sub-genome), homoeologous sequence variants (HSVs) (arising among corresponding loci in each sub-genome) and paralogous sequence variants 5 (PSVs) (arising among duplicated genes both within and between sub-genomes) (Hand et al. 2008; Lawless et al. 2009). Partial solutions for this problem have been established for allotetraploid (2n = 4x = 32) white clover (Hand et al. 2008) and allohexaploid tall fescue (2n = 6x = 42) (Hand et al. 2010), based on depth of sequencing (Margulies et al. 2005), and comparisons to diploid (2x) progenitor taxa 0 for prediction of sub-genome structure.

Autopolyploid forage species, typically arising from chromosome doubling events within single taxa, include alfalfa and artificially generated varieties of ryegrasses, as well as species such as cocksfoot (Dactylis glomerata L.), bird’s foot trefoil (Lotus 25 corniculatus L.) and smooth brome grass (Bromus inermis L.). Although SNP discovery in autopolyploids is likely to be less complex than for allopolyploids, additional challenges for GS studies may be expected to arise due to requirements for quantitative genotyping (determination of allelic dosage), accurate haplotype reconstruction and verification of marker-trait gene linkage.

-172 2017251752 25 Oct 2017

1.4 Symbiotic plant-microbe associations

Both pasture grasses and legumes form mutually beneficial symbiotic associations with asexual microbial species. Schemes for GS implementation could hence be designed to target co-selection of adapted symbiota, in order to optimise the compatibility and stability of associations. The best-characterised of grass endophytes are fungal species of the order Ascomycota and family Claviceptaceae. In the temperate grasses, anamorph-teleomorph relationships, due to presence and absence of a sexual state, are observed between members of the genera Neotyphodium and Epichloe. Neotyphodium endophytes confer beneficial effects on the host grass through deterrence of invertebrate herbivory and amelioration of abiotic stress tolerance, in return for physical protection and nutrition.

Legume species form symbiotic interactions with a range of proteobacterial species, most belonging to the order Rhizobiales. The rhizobia are sequestered within specialised nodules on the roots of legume plants, and perform biological fixation of atmospheric nitrogen (N₂) to ammonium (NH₄ ⁺) in an oxygen-limited environment (Sessitsch et al. 2001).

1.5 Target traits

Herbage yield and quality are the primary production traits for forage species. The latter is primarily controlled by variation for digestibility (associated with cellulose and lignin polymer content in cell walls) and availability of oligosaccharide carbohydrates for energy provision to the grazing animal (Cogan et al. 2005). Other aspects of quality include reduction of bloating in ruminants, associated with protein-rich 25 leguminous diets and ameliorated by the presence of condensed tannins. Biotic stress resistance characters include responses to viral, bacterial and fungal pathogens, and to invertebrate pests. Examples include alfalfa mosaic virus (AMV) of both lucerne and white clover and the crown pathogen of ryegrasses (Puccinia coronata f.sp. lolii) (Dracatos et al. 2010). Abiotic stress-related traits include 30 tolerance to waterlogging (Pearson et al. 2010) and drought in forage grasses, and to aluminium toxicity, phosphorus deficiency and saline stress in clovers. Capacity to produce sufficient quantities of viable seed is important for both grass and pasture

-1732017251752 25 Oct 2017 legume varieties. All of these traits are good targets for GS, due to requirements for laborious, expensive, destructive or late life-cycle measurement.

Structure and Objectives of Pasture Plant Breeding Programs

Many different forms of breeding program have been implemented for pasture plant species. A generic scheme, which describes most of the relevant features of such breeding programs, is depicted in Fig. 168. The majority of commercial activities have been based on a strategy that begins with establishment of a base population of up to 10,000 individuals, followed by seed multiplication within families, generating up o 100,000 individuals for mass selection. Evaluation for persistence under grazing pressure or visual assessment (as spaced plants in the field) is used for sub-selection with a reduction by a factor of 10 in the number of genotypes evaluated. A surviving group of up to 1,000 potential parental clones undergo further assessment for key performance characteristics such as yield and persistence. Determination of parental breeding value may be obtained through the use of a number of experimental methods (Vogel and Pedersen 1993). In half-sib progeny testing (HSPT), maternallyderived progeny from a selected genotype are evaluated in replicated phenotypic trials to minimise confounding environmental variation, obtaining information on the general combining ability (GCA) of the parent. Full-sib progeny testing (FPST) o measures the specific combining ability (SCA) of a parental genotype through identification of superior pair-cross derived families. Within- and between-family selection (WBFS) involves establishment of multiple polycrosses (random intermatings between selected individuals), harvest and bulking of equal seed numbers from each mother plant, and establishment and evaluation of replicated spaced-plant half-sib progeny nurseries for key production characters.

Following selection of foundation (SynO) genotypes, a synthetic 1 (Syn1) population is generated, usually by polycrossing, less commonly by combination of Fi seed from crosses between each parent in a diallel structure. The number of foundation 30 individuals may vary from as low as 4 for perennial ryegrass (although higher for Italian ryegrass), to 50-100 for polyploid species such as tall wheat grass and alfalfa (Bray and Irwin 1999). Synthetic 2 (Syn2) populations are then obtained by Syn1

-174 2017251752 25 Oct 2017 multiplication, through seed harvest from maternal parents fertilised within a single common pollen cloud. The Syn2 populations are then assessed in multiple environments for key traits, leading to the selection of one population for commercial release as a variety.

Commercial breeding programs that implement specific versions of this generic scheme will typically involve two cycles of selective genetic recombination and subsequent selection/evaluation within a 6-9 year period (Fig. 168). Given the extended time-frame for these stages, there are major opportunities for acceleration 0 of genetic gain based on GS strategies. However, implementation of GS in commercial forage breeding programs is likely to require major structural alterations to fully exploit the technology, at least at the existing stage of development of genomic technologies. In current systems, foundation clones have rarely been retained, and details of lineage structure have not generally been recorded, 5 precluding pedigree-based calculation of estimated breeding values. The value of individual genotypes in this paradigm has been relatively low, limiting the opportunity to apply strategies that depend on expensive genotyping analysis and/or phenotypic characterisation. A transition to pedigree-based breeding (or at least the use of marker information to capture pedigree structure) through clonal nursery evaluation 0 of specific genotypes is hence an important step towards implementation of GS in pasture plant breeding.

Requirements for GS application to breeding of ryegrasses

Of the major forage species, the ryegrasses and alfalfa currently possess the best25 developed suites of genetic and genomic resources. However, as the genetic systems of ryegrasses are arguably more similar (as outbreeding, diploid taxa) to those of domestic animals, these species provide a potential test-case for development of GS strategies in forages. In order for this objective to be achieved, a number of key limitations must be overcome.

-1752017251752 25 Oct 2017

1.6 Availability of sequence polymorphisms

GS requires a panel of sequence polymorphism distributed genome-wide capable of genotypic assay across a large number of individuals at reasonable cost. SNPs have 5 become established as the marker of choice for the GS application in most species, given high abundance in the genome, and the availability of accurate and inexpensive detection technologies. However, other types of molecular markers could theoretically also be used for GS subject to meeting the criteria described above.

Development of genomic resources for perennial ryegrass has until recently been relatively slow compared to major cereal species such as rice, maize, wheat and barley. Initial large-scale SNP discovery for perennial ryegrass was performed through functional selection of gene sequences, amplicon generation from parents of full-sib mapping families, followed by cloning, sequencing, alignment and validation through the use of a Mendelian transmission test (Cogan et al. 2006). This method was applied to genes from multiple functional classes, detecting high levels of nucleotide diversity from limited numbers of sampled haplotypes. The consensus from multiple SNP discovery studies (Cogan et al. 2006; Xing et al. 2007; Dracatos et al. 2008; Dracatos et al. 2009; Brazauskas et al. 2010; Fiil et al. 2011) is that SNP 0 frequency typically varies between 1 per 20-150 bp, depending on gene identity and number of genotypes. DNA sequencing of pooled gene-specific amplicons from multiple (c. 500 genotypes) further suggests a high ‘global’ average value of between 1 per 20-25 bp (Cogan et al. 2010a).

A number of approaches have been implemented to generate larger numbers of SNPs with genome-wide distribution. In one early approach, groups of candidate genes were selected and pooled amplicons from these regions were sequenced using the second-generation Roche GS FLX platform. Derived SNPs were formatted for genotyping using the Illumina GoldenGate™ oligonucleotide ligation-amplification

384-plex assay (Cogan et al. 2010c). An approach that allows discovery of a larger number of SNPs, with broader genome coverage, is based on comparative genomics to select exon regions that are regularly distributed across the ryegrass genome for

-1762017251752 25 Oct 2017 pooled sequencing. ‘Proof-of-concept’ for this activity has come from fine-scale genetic and physical mapping of the SI loci (Shinozuka et al. 2010). A more general strategy entailed the exploitation of physical mapping data from bread wheat, for which conserved synteny with perennial ryegrass has been previously defined (Jones 5 et al. 2002a), to align with the draft genome of B. distachyon, and hence use regularly spaced Brachypodium gene templates to selected orthologous ryegrass ESTs for amplicon design and pooled sequencing (Cogan et al. 201 Od).

Complexity reduction methods are also useful to survey the whole gene space, using 0 methods such as Cot filtration and hypomethylation (Forster et al. 2010). Collectively, the discovery activities described above have delivered c. 20,000 predicted SNP loci, sufficient to support an initial design for an integrated high-density oligonucleotidebased genotyping chip, analogous to those used in livestock GWAS and GS studies. However, the dramatic increase in power of current sequence technologies suggest 5 that whole genome sequencing per se, coupled with comparison between contrasted genotypes, is now a more cost effective route to genome-wide SNP discovery. Complete assembly of a complex Poaceae genome is still highly challenging for second-generation platforms (Gupta 2008). However, assembly of the gene-space as unigene contigs is now feasible based on sequencing with Illumina Hi-Seq, which has 0 delivered c. 60 X coverage of the perennial ryegrass genome (Cogan et al. 2011).

In the near future, SNP discovery and subsequent genotyping in species such as ryegrasses are likely to converge through a transition from genotyping per se to ‘genotyping-by-sequencing’ (GBS) methods (Huang et al. 2009,, Elshire et al. 2011).

For instance, the restriction site-associated DNA (RAD) method (Baird et al., 2008) provides access to an essentially unlimited set of sequence polymorphisms through second-generation sequencing of reduced complexity representations from specific genotypes (Wang et al. 2011). This approach is likely to be essential for ryegrass and other forage species, in order to obtain the large numbers of markers and low-cost of 30 genotyping that are necessary given the population structures and current breeding practices for these crops.

-1772017251752 25 Oct 2017

1.7 The challenge of limited LD in ryegrasses

Two key questions for design of any GS experiment are the number of required SNPs, and the number of individuals that must be genotyped and phenotyped in the reference population in order to predict GEBVs for selection candidates with useful 5 accuracy. The answer in both instances depends critically on extent of LD in the target species. LD is commonly measured by the r² statistic, which may also be interpreted as the proportion of variance explained by a SNP which has an association with a QTL. The extent of genome wide LD is largely determined by past effective population size, particularly in outbreeding species with a large N_e. The ? 1 expectation of r is-------- where N_e is effective population size and c is the map

4N_ec +1 distance between loci in Morgans (Sved 1971). Meuwissen (2009) demonstrated by simulation that to achieve very accurate genomic estimated breeding values (GEBV), 10*N_e*L markers are required, where L is the genome length in Morgans. If estimates of N_e are not available, and r² between markers has been estimated, Calus 5 et al. (2008) demonstrated that GEBV with an accuracy greater than 0.7 could be predicted provided r² between adjacent markers was on average above 0.27. Given this threshold as a guideline, what are the prospects for GS in perennial ryegrass?

Results from multiple linkage disequilibrium (LD) studies (Ponting et al. 2007;

Auzanneau et al. 2007; Xing et al. 2007, Fiil et al. 2011) have revealed a typically rapid decay pattern, over less than 1kb. For example, Ponting et al. (2007) demonstrated that LD, as measured by r², decayed to less than 0.27 within 1000 bp in most population types (including ecotypes and cultivars). Fiil et al. (2011) also found that LD decayed rapidly within the representative genes, although some genes 25 exhibited much slower rates of decay.

Both the extremely high frequency of polymorphism, and rapid decay of LD, suggest a very large past N_e value for perennial ryegrass. For comparison, in humans, LD decays to 0.27 at approximately 25 kb, so N_e for ryegrass is likely to be larger than 30 the tens of thousands estimated for the human population (e.g Tenesa et al. 2007).

In cattle, LD decays to 0.27 at approximately 50 kb, and ancestral population sizes

-1782017251752 25 Oct 2017 are in the order of 2000-3000 (Bovine Hap Map Consortium 2009). These comparisons suggest that very dense markers will be necessary to implement GS in ryegrasses. Strategies to deal with this problem are discussed below.

It should be noted that the pattern of LD in ryegrass is very different to that observed in crops like maize, for which LD within heterotic groups, and even across highly diverse germplasm, is extensive (Van Inghelandt et al. 2011; Yan et al. 2009). This property allows accurate estimation of GEBVs; for instance, Riedelshiemer et al. (2012) reported accuracies of GEBV of 0.72 for biomass in maize using a set of 0 56,110 SNPs, while Albrecht et al. (2012) reported accuracies of GEBV of 0.72-0.74 for grain yield in test cross progeny, using only 1,152 SNPs.

The second element for design of a GS experiment is to determine the number of fully phenotyped and genotyped individuals that are required in the reference 5 population. The accuracy of GEBVs in individuals with no phenotype of their own can be derived deterministically and depends on number of individuals genotyped and phenotyped in the reference population, heritability of the trait, and number of loci affecting the trait (Goddard 2008; Daetwyler et al. 2008). Given that little knowledge is available regarding the number of loci affecting most agronomically 0 important traits, equivalence between number of loci and number of independent chromosome segments in the population is a conservative assumption, and this can be calculated from N_e and L. This deterministic prediction suggests that large reference populations are required to predict accurate GEBVs, particularly for low heritability traits (Fig. 169), and agrees well with accuracies that have been achieved 25 in dairy cattle breeding experiments (Hayes et al. 2009). Once again, given that N_e is likely to be very large in ryegrass populations, the deterministic prediction would indicate that extremely (and impractically) large reference populations would be required to implement GS for ryegrass, in the absence of any strategy to reduce LD. Such strategies hence become a key component of GS implementation for 30 ryegrasses. Furthermore, it is important to recognise that GS may deliver gains even if marker density is not as high as suggested values, and the size of the reference population is lower than desired, provided some structure is present within the

-1792017251752 25 Oct 2017 population (such as, for example, if full-sib or half-sib matings are common). SNP markers may also be used to trace the pedigree of the population and the inheritance of large chromosomal blocks from parents and other recent ancestors. Habier et al (2008) demonstrated that a considerable proportion of GEBV accuracy was actually 5 derived from these sources, particularly when N_e has recently contracted, as must have occurred in most forage breeding programs. In the strategy outlined below, an artificial reduction in N_e and therefore increased extent of LD is specifically exploited.

Strategies for Application of GS to Forage Breeding Programs

As described above, experimental evidence suggests that LD in perennial ryegrass is limited, and strategies are required to mitigate this effect in order to effectively implement GS. One strategy is to consider the LD which exists within families (within which N_e is very small), which is equivalent to the use of linkage information to predict GEBV. Algorithms which use LD and linkage simultaneously to predict GEBV 5 are already available (Meuwissen and Goddard 2004). Furthermore, although requiring definitive demonstration, recent decline in N_e values is anticipated for forage species, due to a focus on use of specific germplasm pools within breeding programs. This is an advantage for GS, as evaluation of fewer chromosome segments is required in the reference population (although inbreeding effects must 0 be carefully managed). Whether the decline in N_e has been sufficiently large to permit use of reference populations of practical size may only be answered through interpretation of experimental data, and this is one of the first questions that will be addressed when genome-wide SNP data becomes available. Irrespective of the outcome, the scheme presented here begins with a further reduction of N_e.

A potential GS scheme for ryegrass which largely exploits within-family information for genomic predictions is now described (Fig. 170). A key feature of our scheme is a reduction in generation interval, which in existing breeding programs is usually very long, because only completed varieties or newly-accessed germplasm pools are 30 generally used for the generation of first-phase crosses (see Fig. 168). An opportunity to accelerate genetic gain with GS through shortening of the generation interval could be achieved through introduction of elite genotypes from the first round

-180 2017251752 25 Oct 2017 of selection directly into the second and subsequent rounds of selection. The identity of elite genotypes to include at this stage would be based on GEBVs for traits such as herbage yield, quality and persistence.

Another feature of the scheme is the use of GBS. Although chip-based formatting of currently known SNP loci for highly multiplexed genotyping is logistically feasible, setup costs of design (due to the relatively small-size of the international research and development community) and current processing costs (corresponding to several hundreds of US$ per sample) are unlikely to be acceptable for forage species. In 0 contrast, bar-coded multiplexed samples on the Illumina Hi-Seq platform could potentially deliver high-density SNP data at price points in the vicinity of an order of magnitude lower (e.g. c. US$20 per sample), with costs likely to decline further in the near future (Elshire et al. 2011). Given provision of data from such an Open’ genotyping system, the large number of polymorphisms necessary to enable GS would become readily accessible.

Establishment of the base population

1. The scheme begins with evaluation of available elite germplasm. This is achieved by establishment of a spaced plant in-field nursery incorporating individual genotypes (c. 1,000) from multiple elite germplasm sources, with a moderate level of clonal replication (e.g. four-fold), to obtain c. 4,000 ramets. Direct phenotypic evaluation of genotypes in the nursery is performed for yield, herbage quality and other traits which can be measured at this stage (Fig. 170, box A). As many current elite varieties of perennial ryegrass have a restricted25 base, being derived from a small number of parents (4-6), the inclusion of commercial germplasm will pre-dispose towards a reduction of N_e from the outset.

2. Selection of the 150 best-performing plants is performed based on nursery 30 data, augmented by historical data on elite genotype performance and maximised diversity indices for paircrossing, in order to reduce N_e. A series of paired ‘top-crosses’ between these selected superior individuals is made. In

-181 2017251752 25 Oct 2017 order to allow more accurate prediction of breeding values between progeny sets, and to further reduce N_e, 50 genotypes are designated as ‘bigamous’ parents, as each is mated to two separately selected ‘monogamous’ parents. This procedure obtains 50 pairs of Ft families, each pair being half-sibs based 5 on the common bigamous parent (Fig. 170, box B)

3. GBS of the 150 parents is performed based on a complexity reduction sequencing method (e.g. Baird et al. 2008; Elshire et al. 2011) to identify genome-wide SNP variation within the parental group. (Fig.3, box B) o

Mini-sward evaluation and selection

4. Seed from each full-sib family is harvested for establishment of mini-swards for evaluation¹ of persistence, disease resistance and performance in a sward environment (in order to allow estimation of competition effects). Spaced-plant mini-swards are established containing 100 Ft genotypes from each family, with 10 cm spacing in a 1 m² area, corresponding to 10,000 plant genotypes in total². Depending on logistical constraints, each Fi family-specific mini-sward may be replicated for additional evaluation. For instance, three-fold replication would require establishment of 30,000 distinct plant genotypes in mini-swards. 0 Phenotypic evaluation is performed for the Fi families in the mini-sward setting for target traits (e.g. yield, quality, disease resistance and persistence) (Fig.3, box C). The large number of plants in the mini-sward may only require genotypic analysis using low-density SNP assays (e.g. 384-plex Illumina GoldenGate™ assays), as imputation can be used to infer their genotypes at 25 the SNPs identified in step 3. Habier et al. (2009) demonstrated that this process can be achieved very accurately if parents are genotyped for the ¹The capacity to obtain sufficient seed for establishment of mini-swards is based on the following estimate: number of seed produced by a single reproductive tiller = 50; number of reproductive tillers per plant may be as high as 50 (large pot grown plants) or as low as 4, but c. 10 is taken as a reasonable minimum expectation when clones are prepared for crossing; so number of seed produced per plant = c. 500; so number of seed produced by typical pair-cross = 1,000; minimum germination rate = 75%; so 500-750 potential germinants available.

²Due to the likelihood of pronounced edge-effects, a guard of plants from a generic variety would be established around each 10x10 array.

-182 2017251752 25 Oct 2017 densely spaced markers, as is the case in this example (for which parents have been genotyped for a very large number of loci by GBS).

5. Using the imputed data and individual plant performance data from the mini5 swards, SNP prediction equations for persistency, competition effects and other traits are updated (Fig. 170, boxC).

6. A new parental population with 150 genotypes is selected (using GEBVs for all traits) from the first-round (from 30,000, at a selection frequency of 0.5%), which are maintained with minimal replication in order to mitigate plant loss (Fig. 170, box D). These individuals are subjected to GBS as described in step 3.

Nursery evaluation step

7. Crossing of these selected individuals is performed in the next round of the selection cycle, in order to generate a full replacement nursery population of 4,000 ramets (Fig. 170, box B). These individuals are genotyped using the low-density marker panel, genotypes are imputed and GEBVs are calculated for persistence and disease resistance. These data are combined with

GEBVs for yield and herbage quality (which include information from the individual’s own record), and 150 individuals are selected to generate progeny for the mini-sward step.

Steps 4-7 as described above may be repeated with periodic evaluation of genetic 25 gain. When sufficient genetic gain has been made to clearly distinguish a new ‘prevariety’ then diversion of a sub-set of the 150 selected genotypes into synthetic population development can occur. A total of 16 genotypes (c. 10% selection intensity) may be chosen to derive 4 polycross groups of four parental genotypes each, on the basis of predicted performance and genetic distance, to produce 30 restricted-base pre-varieties. Multi-site evaluation will be used to measure phenotypic attributes such as yield, quality and persistence of these synthetics (Fig.3, box E).

-1832017251752 25 Oct 2017

This scheme is intended to deliver maximum genetic gain through rapid cycling of crossing activity based on GS predictions, particularly for persistence and disease resistance. In contrast to the scheme presented in Fig.1, recombination and selection takes place as part of each cycle, rather than intermittently within a more protracted 5 period. The major logistical requirements, which are for mini-sward establishment and evaluation, may be reduced by a more radical strategy, in which seed from all initial pair-crosses is pooled and sowed in a single large-scale closed-spaced trial. Elite individuals would then be identified by phenotypic evaluation and assigned to cross-of-origin through low-density SNP-based genotypic analysis.

o

Although derivation of GEBVs would reduce the requirements for full-scale phenotyping, apart from during the first round of selection, it is anticipated that retention of the parental genotypes and maintenance of the spaced plant nursery would allow refinement of predictions through evaluation of additional traits, or long5 term monitoring of previously measured traits. The value of a spaced plant nursery system would also be enhanced through evaluation of genotype x environment (G x E) effects. This objective would ideally be achieved through establishment of clonal replicates in different target environments, and GEBVs could then be generated for each environment.

:0

One of the key objectives of this proposed scheme is to enhance the significance of individual genotypes during breeding practice, which have traditionally been negligible compared to the population as a whole. Retention, intensive phenotypic characterisation and use in a fully recorded (based on SNP genotype) pedigree 25 structure will deliver at least part of this worth.

Conclusions

Although forage species have been relatively undeveloped in terms of molecular breeding compared to other major crop plants, GS implementation has the potential 30 to deliver major advances, mainly through the capacity to complete multiple selection rounds within the time conventionally used for single rounds. This outcome is possible if accurate GEBVs can be predicted for traits such as persistence and

-184 2017251752 25 Oct 2017 disease resistance. Information on these traits is currently obtained only after potential selection candidates are up to five years of age. However, if GS is to be implemented in forage species, a number of challenges must be overcome. The relative deficiencies of DNA marker resources, and influence of polyploid genome 5 structures for some species, constitute the first challenge. Barriers to marker availability will rapidly disappear as GBS becomes less expensive, while enhanced methods for genotypic analysis of polyploid genomes have also been developed (Gidskehaug et al. 2011).

The two major remaining challenges are the very limited extent of LD in forage species such as ryegrasses, and restricted opportunities to implement GS in current breeding programs. In this review, the first factor is addressed through use both of linkage information within families to increase the accuracy of GEBV prediction, and, in the longer-term, reduction of N_e in populations by breeding from elite varieties.

Undesirable correlated effects of inbreeding depression under such schemes could be managed through incorporation of measures of genomic diversity into the selection criteria (e.g. Pryce et al. 2012). To address the second factor, a breeding scheme has been proposed which permits GS to accelerate genetic gain through compression of selection round duration. However, implementation of this scheme 0 would require significant restructuring of current breeding schemes.

Although field- and glasshouse-based evaluation systems using largely visual assessment criteria have been sufficient to enable varietal improvement to date, effective use of molecular breeding systems is also dependent on access to more 25 accurate and detailed phenotypic values (Furbank et al. 2009; Houle et al. 2010), for instance, those delivered by ‘whole-of-lifetime’ measurements of plant growth and performance from automated glasshouse and image analysis phenomics platforms. The accurate quantification of phenotypic variation using next-generation technologies is a major challenge and opportunity for forage breeding (Walter et al. 30 2012).

-1852017251752 25 Oct 2017

Example 24 - Genomic selection of plant-symbiome associations

Artificial selection methods are dependent on a two-step process: firstly, recombination to generate new arrangements of genetic elements, and secondly, differential sampling of newly generated genotypes on the basis of superior 5 phenotypic selection. Asexual symbionts (e.g. endphytes, bacteria) do not participate in the recombination events during sexual reproduction of the host plant. Generation of variability in host-symbiont pairs is hence combinatorial in nature, and the full power of genomic selection for favourable expression of the symbiotic association will depend on capability to generate as broad a range of pair-wise combinations 0 between host genotype and symbiont genotype as possible. The potential diversity space for co-selection with the symbiont may be estimated as: [Number Target germplasm pools] x [Number Distinct symbiont genotypes]. As this number is very large, selective sampling for maximum symbiont diversity using a genetic algorithm is desirable.

Symbionts which reproduce sexually (e.g. some fungi) provide an additional opportunity to select symbionts within populations for most beneficial effect on the host plant. Here, genomic characterisation and subsequent genomic selection of symbionts would be needed. The combinatorial nature would remain at least as 0 complex, but it could potentially increase to: [Number target germplasm pools] x [Number distinct symbiont genotypes] x [Number variants within symbiont genotype group].

Such methods are also applicable in the case of hosts which differ in potential to 25 attract symbiont populations (both quantity and diversity). Hence, genomic selection may be used to determine host affinity to symbiont spectra, diversity and quantity.

Such methods are also applicable to co-select of legume-rhizobia symbioses, symbioses with mycorrhizae, favourable host-symbiont genome interactions and 30 host-symbiont-environmental interactions.

-1862017251752 25 Oct 2017

Example 25 - Leaf and root microbiomes in temperate grass: Perennial ryegrass (Lolium perenne)

Clonal Lolium perenne plants, both with and without the fungal endophyte Standard 5 Toxic Neotyphodium lolii were grown in 6 different hydroponic solutions representing different nutritional conditions. No carbon source was present, so that the endophyte must grow on carbon from the plant.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion 0 concentrations, plants were washed to remove residual media and roots and leaves were snap frozen in liquid N₂. RNA was extracted from leaves and roots of 5 replicate plants per treatment (120 samples).

This RNA was used to construct cDNA RNA seq libraries which were analysed by

RNAseq analysis (llummina HiSeq sequencing). The libraries were analysed using blastn and a single database hit recorded per sequence read. The database queried contained plant nuclear genes, a representative plant chloroplast, and mitochondrial genome, the fungal endophyte nuclear and mitochondria genome, and sequences representing other fungi, bacteria, algae, insects and rhizaria. A second database 0 contained approx. 120,000 named 16S sequences. Hits were recorded as being above a cut off of 80% identity and 40 bp overlap and grouped as to their origin.

The vast majority of reads mapped to plant database sequences. However significant reads mapped to the other classes of organisms. Figure 171 shows the percent of 25 total reads mapped mapping to algae, bacteria, fungi (other than endophyte), insects and rhizaria in the leaves and roots of endophyte free and endophyte containing plants in the full hydroponics media. This reveals a higher level of microbes associated with roots compared to leaves.

-1872017251752 25 Oct 2017

Example 26 - 16S analysis of microbiome in perennial ryegrass (Lolium perenne)

Clonal Lolium perenne plants, both with and without Standard Toxic Neotyphodium 5 lolii were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per 0 treatment. This RNA was used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S 5 rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. Database sequences were extracted if there were hits recorded in at least 10 different libraries.

:0

These database sequences were classified taxonomically using the Ribiosomal Database Project classifier http://rdp.cme.msu.edu/ and grouped to show the taxonimic distribution of the microbes represented in the Lolium perenne microbiome. As the plant chloroplast is of microbial origin, plant chloroplast sequences are present 25 in the 16S sequence database. Reads mapping to these sequences were not included in the taxonomic summary.

Figure 172 shows matches counted (16S blastn hit >98% identity over an overlap of >40bp; mapping >10 samples). 11,000 - 164,000 reads were assigned to 16S per 30 sample. Distribution of 2774 bacterial phyla (non-plant chloroplast).

-1882017251752 25 Oct 2017

Example 27 - Leaf and root microbiomes in perennial ryegrass (Lolium perenne)

Clonal Lolium perenne plants, both with and without Standard Toxic Neotyphodium 5 lolii were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per 0 treatment. This RNA was used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S 5 rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. The matrix of 16S counts for 16S sequences that had hits in 10 or more libraries was analysed as to the sum of the 0 counts for each database sequence by plant organ. These sums were themselves summed by taxonomic class of bacteria.

Figure 173 plots the sum by bacterial class of read counts from leaves and roots respectively. This reveals that some classes of bacteria are predominantly found in 25 leaves (eg Cyanobacteria) or roots (eg Sphingobacteria) whereas other classes are found in both leaves and roots (eg Gammaproteobacteria). 470 16S sequences had hits in at least 10 leaf and 10 root libraries.

-1892017251752 25 Oct 2017

Example 28 - Leaf and root microbiomes in perennial ryegrass (Lolium perenne). Metagenomics reveals differences in bacterial species predominance in shoot and root microbiomes

Clonal Lolium perenne plants, both with and without Standard Toxic Neotyphodium lolii were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion 0 concentrations, RNA was extracted from leaves and roots of 5 replicate plants per treatment. This RNA was used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit 5 recorded per sequence read. In this case the database consisted of 117101 16S rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. The counts values in this matrix (present 0 in at least 10 samples) were log transformed to base 2 and this matrix was loaded into the the MultiExperimentviewer software (part of the TM4 software suite www.tm4.org). This log transformed matrix was used for hierarchical clustering of the sample sets using average linkage with Pearson correlation, and the order of the samples optimised.

This produced the sample tree shown in Figure 174. The clustering demonstrates that the bacterial counts matrix from the roots contains a signature that can classify the nutritional status of the plant-microbiome associations (i.e. symbiota). In addition the leaf bacterial microbiome clearly differentiates roots from leaves (ie. different 30 bacteria are predominant in different organs eg. leaves versus roots). However the leaf bacterial microbiome does not classify the nutritional status (hierarchical clustering of bacterial counts classifies root treatments but not leaf treatments).

-190 2017251752 25 Oct 2017

Example 29 - Leaf and root microbiomes in perennial ryegrass (Lolium perenne). Differences in bacterial species predominance depending on symbiotum nutritional status

Clonal Lolium perenne plants, both with and without Standard Toxic Neotyphodium lolii were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per treatment. This RNA was used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. The counts values in this matrix were log transformed to base 2 and this matrix was loaded into the the MultiExperimentviewer software (part of the TM4 software suite www.tm4.org). The 16S sequences were then used for Cluster Affinity Search Technique analysis using Pearson correlation and a threshold of 0.75. The resulting clusters were then individually hierachically 25 clustered using average linkage with Pearson correlation and the 16S sequence order optimised.

This produced 1036 clusters. Of these clusters 11 contained more than 10 genes and contained 47% of the genes in the matrix. Examples of 2 of these 11 clusters are 30 shown in Figures 175 and 176. These display (clusters) bacterial sequences whose abundance is increased in response to a low ammonia hydroponic treatments, and (cluster2) bacterial sequences whose abundance is repressed in response to low

-191 2017251752 25 Oct 2017 ammonia and low nitrate treatment.

Example 30 - Leaf and root microbiomes in perennial ryegrass (Lolium perenne). Metagenome analysis of bacterial microbiome in symbiota reveals 5 range of bacterial species known to be N fixers and phytostimulators of grasses

Clonal Lolium perenne plants, both with and without Standard Toxic Neotyphodium lolii were grown in 6 different hydroponic solutions representing different nutritional 0 conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per treatment. This RNA was used to construct cDNA RNA seq libraries which were 5 analysed by llummina HiSeq sequencing. The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library 0 and a counts matrix of hits per database sequence per library analysed was created.

This matrix was filtered to show reads mapping to sequences annotated as Azospirillum species. The counts mapping to these 12 sequences per library are shown in Figure 177. Azospirillum species are known to associate with 25 monocotyledenous plants, and also to fix atmospheric nitrogen and produce phytohormones (phytostimulators of grasses). At least one of the 12 species can be seen to have counts that increase in low all low ammonia samples, where fixation of atmospheric nitrogen would be of utility to plants associated with the bacteria fixing the nitrogen. Thus, Azospirillum species were induced in number under low N.

-192 2017251752 25 Oct 2017

Example 31 - Leaf and root microbiomes in Antarctic hairgrass (Deschamsia Antarctica). Metagenomics reveals diverse plant bacterial microbiome

Deschampsia antartica plants, from two different accessions (Da2 and Da17) were grown in 6 different hydroponic solutions representing different nutritional conditions. These were used to examine abundant bacteria in leaf and root microbiomes through metagenomics.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion 0 concentrations, RNA was extracted from leaves and roots of 5 replicate plants per treatment. This RNA was pooled and used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria in plants grown under different nutritional regimes. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created.

Database sequences were extracted if hit count per library analysed was greater than or equal to 100 counts.

These database sequences were classified taxonomically using the Ribiosomal Database Project classifier http://rdp.cme.msu.edu/ and grouped to show the 25 taxonimic distribution of the microbes represented in the Deschampsia antartica leaf and root microbiome across bacterial phyla (Figure 178). There were 203 16S sequences out of 7109 sequences identified.

-1932017251752 25 Oct 2017

Example 32 - Leaf and root microbiomes in Antarctic hairgrass (Deschamsia antarctica). Metagenomics reveals defferences in bacterial species predominance in shoot and root microbiomes

Deschampsia antartica plants, from two different accessions (Da2 and Da17) were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per 0 treatment. This RNA was pooled and used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S 5 rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. Database sequences were extracted if hit count per library analysed was greater than or equal to 100 counts.

:0

This counts matrix was modified, by each database sequence by library count value being divided by the average counts value for that sequence across all the libraries. This modified matrix was loaded into the the MultiExperimentviewer software (part of the TM4 software suite www.tm4.org) and used for hierarchical clustering of the 16S 25 sequences using average linkage with Pearson correlation.

The heat map shown in Figure 179 displays differences in bacterial species predominance in shoots and roots.

-194 2017251752 25 Oct 2017

Example 33 - Leaf and root microbiomes in Antarctic hairgrass (Deschamsia antarctica). Differences in bacterial species predominace in shoot and root microbiomes depending on symbiotum nutritional status

Deschampsia antartica plants, from two different accessions (Da2 and Da17) were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per 0 treatment. This RNA was pooled and used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S 5 rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. Database sequences were extracted if hit count per library analysed was greater than or equal to 100 counts.

:0

This counts matrix was modified, by each database sequence by library count value being divided by the average counts value for that sequence across all the libraries. This modified matrix was loaded into the the MultiExperimentviewer software (part of the TM4 software suite www.tm4.org) and used for hierarchical clustering of the 16S 25 sequences using average linkage with Pearson correlation.

Figure 180 shows a zoomed region of the heat map, displaying differences in bacterial species predominance in response to differing nutritional status in root samples.

-1952017251752 25 Oct 2017

Example 34 - Leaf and root microbiomes in Antarctic hairgrass (Deschamsia antarctica). Differences in bacterial species predominace in shoot and root microbiomes depending on symbiotum nutritional status

Deschampsia antartica plants, from two different accessions (Da2 and Da17) were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion concentrations, RNA was extracted from leaves and roots of 5 replicate plants per 0 treatment. This RNA was pooled and used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S 5 rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. Database sequences were extracted if hit count per library analysed was greater than or equal to 100 counts.

:0

This counts matrix was modified, by each database sequence by library count value being divided by the average counts value for that sequence across all the libraries. This modified matrix was loaded into the the MultiExperimentviewer software (part of the TM4 software suite www.tm4.org) and used for hierarchical clustering of the 16S 25 sequences using average linkage with Pearson correlation.

Figure 181 shows a zoomed region of the heat map, displaying differences in bacterial species predominance in response to differing nutritional status in leaf samples. The predominant bacterial species in leaf microbiome are predominantly 30 Cyanobacteria. The bacterial species differ in response to N in leaves (ie. low N).

-1962017251752 25 Oct 2017

Example 35 - Leaf and root microbiomes in Antarctic hairgrass (Deschamsia antarctica). Metagenome analysis of bacterial microbiome in symbiota reveals range of bacterial species known to be N fixers and phytostimulators of grasses

Deschampsia antartica plants, from two different accessions (Da2 and Da17) were grown in 6 different hydroponic solutions representing different nutritional conditions.

After 3 weeks of treatment with biweekly replenishment of media to maintain ion 0 concentrations, RNA was extracted from leaves and roots of 5 replicate plants per treatment. This RNA was pooled and used to construct cDNA RNA seq libraries which were analysed by llummina HiSeq sequencing.

The reads from these libraries were analysed using blastn and a single database hit recorded per sequence read. In this case the database consisted of 117101 16S rRNA sequences. The rRNA sequences are used to taxonomically identify bacteria. A hit was counted above a threshold of 98% identity and 40 bp. The hits were counted per database sequence per library and a counts matrix of hits per database sequence per library analysed was created. Database sequences were extracted if hit count per library analysed was greater than or equal to 100 counts.

This counts matrix was modified, by each database sequence by library count value being divided by the average counts value for that sequence across all the libraries. This modified matrix was loaded into the the MultiExperimentviewer software (part of 25 the TM4 software suite www.tm4.org) and used for hierarchical clustering of the 16S sequences using average linkage with Pearson correlation.

Figure 182 shows a zoomed region of the heat map displaying Azospirillum species clustered together. These bacteria are known to associate with monocotyledenous 30 plants, and also to fix atmospheric nitrogen and produce phytohormones (phytostimulator of grasses). These bacteria are shown to be more predominant in low nitrogen nutrient solutions in root samples. Thus, Azospirillum species were

-1972017251752 25 Oct 2017 induced in number under low N.

Example 36 - Summary

These examples demonstrate that metagenomics of grass symbiomes reveals compex endo/epiphytic microbiomes and species diversity in grass microbiomes. There is also variation in microbiome between grass species, between roots and shoots, and across environmental conditions. The microbiome profile reveals species diversity for symbiome performance enhancement, microbiome profiling 0 across genotypes and environments, microbiome monitoring and microbiome inoculation.

Example 37 - Maximising performance of ruminants through genomic selection for optimum interaction of individual ruminant genomes and rumen microflora 5 profiles

Ruminants are unique in their ability to turn low quality forage into meat, milk and fibre. This unique ability is as a result of the rumen, an organ of the digestive tract which is host to vast multitudes of bacteria and other micro-organisms. The profile of 0 rumen micro-organisms can be obtained by taking a sample of rumen fluid, or alternatively profiling the faeces, and counting the abundance of species, for example through meta-genome sequencing. This profile differs between ruminant species and between individuals. Differences in rumen profile may be associated with differences in feed conversion efficiency, milk composition and health outcomes. Thus, selecting 25 to maximise phenotypic performance of ruminants may be achieved through genomic selection for optimum individual ruminant genome by rumen profile. Given the number of micro-organisms in the rumen is very large, and the number of genes in the genomes of ruminants is approximately 20,000, a genetic algorithm is required to prioritise sample combination based on diversity.

-1982017251752 25 Oct 2017

Example 38 - Endophyte inoculation method in perennial ryegrass

This example describes enhancement of endophyte inoculation frequency following 5 puncturing isolated embryos of perennial ryegrass with an hypodermic needle prior to inoculation using method 1 (direct inoculation) or method 2 (coating with endophyte containing Ca-alginate layer).

Embryos isolated from perennial ryegrass seeds were inoculated with endophyte ο NEA11 using either methods 1 or 2, with endophyte suspensions at different dilution rates (1/4, 1/8, 1/16; see Figure 183) subjected, with and without wounding of embryos with an hypodermic needle. Puncturing of embryos prior to inoculation greatly enhanced inoculation efficiency, demonstrated by SSR-based endophyte detection in 6 week old symbiota recovered from artificial seeds derived from 5 inoculated embryos (see Table 64).

Endophyte Detected Treatment Method RDB cone Wounding NEA11 E- lnoculation%

A	1	1/16	No	0	42	0
B	1	1/8	No	0	42	0
c	1	V8	Puncture	11	29	27.5
D	2	1/16	No	0	42	0
E	2	1/8	No	0	42	0
F	2	1/8	Puncture	9	9	50

Method 1:Direct Inoculation of Isolated Embryos with Endophyte Suspension Prior to Ca-Alginate Coating

Method 2: Direct Coating of Isolated Embryos with Endophyte-Containing Ca-Alginate Layer

Table 64: Number and frequency of endophyte-inoculated perennial ryegrass plants recovered following different endophyte inoculation treatment methods

Example 39 - Endophyte inoculation method in perennial ryegrass and tall fescue

This example describes enhancement of endophyte inoculation frequency following 25 puncturing isolated embryos of perennial ryegrass (L. perenne) and tall fescue (F.

arundinacea) with an hypodermic needle prior to inoculation using method 1 (direct inoculation) or method 2 (coating with endophyte containing Ca-alginate layer).

-1992017251752 25 Oct 2017

Method 1: Direct Inoculation of Isolated Embryos with Endophyte Suspension Prior to Ca-Alginate Coating

Method 2: Direct Coating of Isolated Embryos with Endophyte-Containing Ca5 Alginate Layer

Embryos isolated from seeds from different varieties were inoculated with different endophytes (NEA11 and NEA17) using either methods 1 or 2, with and without wounding of embryos with hypodermic needle. Puncturing of embryos prior to 0 inoculation greatly enhanced inoculation efficiency, demonstrated by SSR-based endophyte detection in 6 week-old symbiota recovered from artificial seeds derived from inoculated embryos (see Table 65).

Species Variety Experiment Endophyte No. of artificial seeds No. of artificial seeds Method 1 Method 1 plus wounding

				Total	Negative	Positive	Total	Negative	Positive
L. perenne	Alto	1	NEA11 (LpTG-2)	42	42	0	20	16	4
		2	NEA11 (LpTG-2)	21	21	0	21	20	1
		3	NEA11 (LpTG-2)	84	84	0	40	29	11
F. arundinacea	Dovey	1	NEA11 (LpTG-2)	42	42	0	40	39	1
L. perenne	Alto	1	NEA17 (FaTG-2)	42	42	0	42	42	0
F. arundinacea	Dovey	1	NEA17 (FaTG-2)	70	70	0	35	35	0
	Finesse	2	NEA17 (FaTG-2)	42	42	0	70	70	0

Species Variety Experiment Endophyte No. of artificial seeds No. of artificial seeds ________________________________________________Method 2_______________Method 2 plus wounding Total Negative Positive Total Negative Positive

L. perenne Alto 1 NEA11 (LpTG-2) 84 84 0 18 9 9

Table 64: Number and frequency of endophyte-inoculated perennial ryegrass and tall fescue plants recovered following different endophyte inoculation treatment methods

Example 40 - A method to discover host genotype by microbiome profile interaction in dairy cattle

Here, we describe a method to map regions of the host genome that affect the rumen microbial profile, and illustrate this with an example.

-200 2017251752 25 Oct 2017

Methods

The method consists of efficiently testing large numbers of DNA polymorphisms of the host for their effect on rumen microbiome profiles.

A rumen microbiome profile consists of, for each cow, the counts of sequence reads derived from a rumen sample that map to either the species, or contigs of a rumen metagenome assembly, or 16S sequences (for example see Brule et al. 2009, Hess etal. 2011).

The matrix of counts per cow mapping to each of n contigs or species, or partial assemblies of species, can then be treated as phenotypes in a genome wide association with DNA polymorphisms of the hosts own genome, so a model can be fitted to the data:

y = mu + Xb + e

Where for each contig/species/partial species assembly, y is the vector of counts for m cows, mu is the mean, X is m x 1 design vector assigning the genotype of the host to the record, and e is a vector of random residual effects.

:0

As n, the number of contigs/species or partial species assemblies is likely to be large, the power to identify DNA polymorphisms in the host genome may be increased by considering associations across the contigs/species/partial assemblies. This may be done efficiently by conducting the genome wide association study for each contig, 25 then accumulating the information across contigs, by taking the most significant region for each contig, and then assessing the proportion of times across all contigs that the region is most significant. This may identify master host regulators of rumen microbiome profiles. This may be done in sliding windows of 100kb, to take into account the effect of highly variable linkage disequilibrium in cattle populations.

Example

The data were from 15 Holstein -Fresian dairy cows, grazing at DPI Ellinbank

-201 2017251752 25 Oct 2017 research station. Each cow had genotypes for 624,930 SNP, from the Illumina Bovine HD array.

Rumen samples were taken from each of the cows during a methane measurement 5 experiment. DNA was extracted for each sample, and libraries were prepared of

300bp random fragments. Then paired end sequencing was performed on Illumina HiSeq. The sequence reads were mapped to the rumen metagenome assembly (8092 contigs) described by Hess et al. 2011.

Five example genome wide association studies (for five different contigs) are shown in Figure 184.

A region on chromosome 21 was identified that may be a master regulator of rumen microbiome profiles, as it associated with sequence read counts in 3 out of the five 5 contigs.

-202 2017251752 25 Oct 2017

References

Albrecht, T, V. Wimmer, H.J. Auinger, M. Erbe, C. Knaak, M. Ouzunova, H. Simianer, and C.C. Schon, 2011: Genome-based prediction of testcross values in maize.

Theor. Appl. Genet. 123, 339-350.

Auzanneau, J., C. Huyghe, B. Julier, and P. Barre, 2007: Linkage disequilibrium in synthetic varieties of perennial ryegrass. Theor. Appl. Genet. 92, 837-847.

Baird, N.A., P.D. Etter, T.S. Atwood, M.C. Currey, A.L. Shiver, Z.A. Lewis, E.U.

Selker, W.A. Cresko, and E.A. Johnson, 2008: Rapid SNP discovery and genetic 0 mapping using sequenced RAD markers. PLoS One 3, e3376.

Beavis W.D., 1994: QTL analysis: power, precision and accuracy. In: A. Paterson (ed.) Molecular Dissection of Complex Traits, 145-162. CRC, Boca Raton, Louisiana, USA.

Braazauskas, G., I. Pasakinskiene, T. Asp, and T. LObberstedt, 2010: Nucleotide 5 diversity and linkage disequilibrium in five Lolium perenne genes with putative role in shoot morphology. Plant Sci. 179, 194-201.

Bray, R.A., and J.A.G Irwin, 1999: Medicago sativa L. (lucerne) cv. Hallmark. Aus. J. Exp. Agric. 39, 643-644.

Brule, J.M., et al, (2009) Gene-centric metagenomics of the fiber-adherent bovine 0 rumen microbiome reveals forage specific glycoside hydrolases. PNAS. 106, 19481953.

Cogan, N.O.I., K.F. Smith, T. Yamada, M.G. Francki, A.C. Vecchies, E.S. Jones, G.C. Spangenberg, and J.W. Forster, 2005: QTL analysis and comparative genomics of herbage quality traits in perennial ryegrass (Lolium perenne L.). Theor. Appl.

Genet. 110, 364-380.

Cogan, N.O.I., R.C. Ponting, A.C. Vecchies M.C. Drayton, J. George, M.P. Dobrowolski, T.I. Sawbridge, G.C. Spangenberg, K.F. Smith, and J.W. Forster, 2006: Gene-associated single nucleotide polymorphism (SNP) discovery in perennial ryegrass (Lolium perenne L.). Mol. Genet. Genom. 276, 101-112.

Cogan, N.O.I., R.C. Baillie, M.C. Drayton, K. Savin, C. Anderson, G.C. Spangenberg, and J.W. Forster, J.W., 2010a: Massively-parallel sequencing technology permits direct detection of novel SNP haplotypes in perennial ryegrass (Lolium perenne L.).

-2032017251752 25 Oct 2017

Molecular Breeding of Forage and Turf 2010, Abstract P-6, 51-52. Buenos Aires, Argentina.

Cogan, N.O.I., J. Wang, G.Ye, C. Bandaranayake, M.L. Hand, R.C. Baillie, M,C. Drayton K. Lawless, M.P. Dobrowolski S. Erb, K.F. Smith, and J.W. Forster, J.W., 5 2010b: Association mapping of forage quality traits in perennial ryegrass (Lolium perenne L.). Molecular Breeding of Forage and Turf 2010, Abstract P-148, 211. Buenos Aires, Argentina.

Cogan, N.O.I., R.C. Baillie, M.L. Hand, M.C. Drayton P. Tian, T. Webster R. Chapman. G.C. Spangenberg, and J.W. Forster, 2010c:Accelerated SNP discovery 0 in perennial ryegrass based on pooled amplicon resequencing. Molecular Breeding of Forage and Turf 2010, Abstract P-4, 47-48. Buenos Aires, Argentina.

Cogan, N.O.I, M.L. Hand, T.l. Sawbridge, R.C. Baillie, and J.W. Forster, 2010d: Comparative genomics in the grass family Poaceae for structured genome-wide SNP discovery in perennial ryegrass (Lolium perenne L.). Molecular Breeding of Forage 5 and Turf 2010, Abstract P-5, 49-50. Buenos Aires, Argentina.

Cogan, N.O.I., J. Wang, C. Inch, D. Pickett, R. Elton, R.C. Baillie, M.C. Drayton, H. Shinozuka, and J.W. Forster, 2011: Genomic tools for pasture plant breeding. Plant and Animal Genome XIX, Abstract W335. San Diego, California, USA.

Daetwyler, H.D., B. Villanueva, B., and J.A. Woolliams, 2008: Accuracy of predicting 0 the genetic risk of disease using a genome-wide approach. PLoS One 3, e3395.

Dekkers, J.C.M., 2004: Commercial application of marker- and gene assisted selection in livestock: strategies and lessons. J Anim. Sci. 82, E-Suppl: E313-E328.

Dracatos, P.M., N.O.I. Cogan, M.P. Dobrowolski, T.l. Sawbridge, G.C. Spangenberg,

K.F. Smith, and J.W. Forster, 2008: Discovery and genetic mapping of single 25 nucleotide polymorphisms in candidate genes for pathogen defence response in perennial ryegrass (Lolium perenne L.). Theor. Appl. Genet. 117, 203-219.

Dracatos, P.M., N.O.I. Cogan, T.l. Sawbridge, A.R. Gendall, K.F. Smith, G.C. Spangenberg, and J.W. Forster, 2009: Molecular characterisation and genetic mapping of candidate genes for qualitative disease resistance in perennial ryegrass 30 (Lolium perenne L.). BMC Plant Biol. 9, 62.

Dracatos, P.M., N.O.I. Cogan, P.J. Keane, K.F. Smith, and J.W. Forster, 2010: Biology and genetics of crown rust disease in ryegrasses. Crop Sci. 50, 1605-1624.

-204 2017251752 25 Oct 2017

Dumsday, J.L., K.F. Smith, J.W. Forster, E.S. Jones, 2003: SSR-based genetic linkage analysis of resistance to crown rust (Puccinia coronata Corda f. sp. lolii) in perennial ryegrass (Lolium perenne L.). Plant Pathol. 52, 628-637.

Elshire, R.J., J.C. Glaubitz, Q. Sun, J.A. Poland, K. Kawamoto, E.S. Buckler, and

S.E. Mitchell, 2011: A robust, simple genotyping-by-sequencing (GBS) approach for high diversity species. PLoS One 6, e19379.

Faville, M., A.C. Vecchies, M. Schreiber, M.C. Drayton, L.J. Hughes, E.S. Jones,

K. M. Guthridge, K.F. Smith, T. Sawbridge G.C. Spangenberg, G.T. Bryan, and J.W. Forster, 2004: Functionally-associated molecular genetic marker map construction in perennial ryegrass (Lolium perenne L.). Theor. Appl. Genet. 110, 12-32.

Fiil, A., I. Lenk, K. Petersen, C.S. Jensen, K.K. Nielsen, B. Schejbel, J.R. Andersen, and T. LObberstedt, 2011: Nucleotide diversity and linkage disequilibrium of nine genes with putative effects on flowering time in perennial ryegrass (Lolium perenne

L. ). Plant Sci. 180, 228-37.

Forster, J.W., N.O.I. Cogan, M.P. Dobrowolski, E. van Zijll de Jong, G.C. Spangenberg, and K.F. Smith, 2008a: Molecular breeding technologies for forage and turf plants. In: C. Kole and A. Abbott (eds.) Principles and Practices of Plant Genomics Volume 2: Molecular Breeding, 395-430. Science Publishers, Inc., New Hampshire, USA.

Forster, J.W., N.O.I Cogan, M.P. Dobrowolski, M.G. Francki, G.C Spangenberg, and K.F. Smith, 2008b): Functionally-associated molecular genetic markers for temperate pasture plant improvement. In: R.J. Henry (ed.) Plant Genotyping II: SNP Technology, 154-187. CABI Press, Wallingford, Oxford, UK..

Forster, J.W., N.O.I.Cogan, J. Wang, G. Ye, C. Bandaranayake, M.L. Hand, R.C.

Baillie, M.C., Drayton, K. Lawless, M.P. Dobrowolski, S. Erb, and K.F. Smith, 2010: Development and implementation of molecular genetic tools for enhancement of herbage quality and other agronomic traits in perennial ryegrass (Lolium perenne L.). Molecular Breeding of Forage and Turf 2010, Abstract, 18-19. Buenos Aires, Argentina.

Furbank, R.T., 2009: Plant phenomics: from gene to form and function. Funct. Plant Biol. 36, v-vi.

Ganal, M.W., G. Durstewitz, A. Polley, A. Berard, E.S. Buckler, A. Charcosset, J.D.

-2052017251752 25 Oct 2017

Clarke, E.M. Graner, M. Hansen, J. Joets, M.C. Le Paslier, M.D. McMullen, P. Montalent, M. Rose, C.C. Schon, Q. Sun, H. Walter, O.C. Martin, and M. Falque, 2011: A large maize (Zea mays L.) SNP genotyping array: development and germplasm genotyping, and genetic mapping to compare with the B73 reference 5 genome. PLoS One. 6, e28334.

Gidskehaug, L, M. Kent, B.J. Hayes, and S. Lien, 2011: Genotype calling and mapping of multisite variants using an Atlantic salmon iSelect SNP array. Bioinformatics 27, 303-310.

Goddard, M.E., 2008: Genomic selection: Prediction of accuracy and maximisation of 0 long term response. Genetica 136, 245-257..

Gout, M., and S. Jones, 2006: Estimates of the annual sales of proprietary cultivars in Australia, and the value of pastures to the livestock and cropping industries. Presentation made to Pastures Australia. Adelaide, Australia, February 2006.

Gupta, P.K., 2008: Single-molecule sequencing technologies for future genomics 5 research. Trends Biotechnol. 26, 602-611.

Guthridge, K.M., M.P. Dupal, R. Kolliker, E.S. Jones, K.F Smith,., and J.W. Forster, 2001: AFLP analysis of genetic diversity within and between populations of perennial ryegrass (Lolium perenne L.). Euphytica 122, 191-201.

Habier, D., R.L. Fernando, and J.C.M. Dekkers, 2009: Genomic selection using lowfi density marker panels. Genetics 182, 343-353.

Hand, M.L., R.C. Ponting, M.C. Drayton, K.A. Lawless, N.O.I. Cogan, E.C. Brummer, T.l. Sawbridge, G.C. Spangenberg,., K.F. Smith, and J.W. Forster, 2008: Identification of homologous, homoeologous and paralogous sequence variants in an outbreeding allopolyploid species based on comparison with progenitor taxa. Mol.

Genet. Genom. 280, 293-304.

Hand, M.L., N.O.I. Cogan, A.V. Stewart, and J.W. Forster, 2010: Evolutionary history of tall fescue morphotypes related to molecular phylogenetics of the Lolium-Festuca species complex. BMC Evol. Biol. 10, 303.

Hayes, B.J., H.D. Daetwyler, P. Bowman, G. Moser, B. Tier, R. Crump, M. Khatkar, 30 H. Raadsma, and M.E. Goddard, 2009: Accuracy of genomic selection: comparing theory and results Proc. 14th Conf. Assoc. Advanc. Anim. Br. Genet. 14, 34.

Hayward, M.D., J.W. Forster, J.G. Jones O. Dolstra, C. Evans, , N.J. McAdam, K.G.

-2062017251752 25 Oct 2017

Hossain, M. Stammers, J.A.K., Will M.O. Humphreys, and G.M. Evans, (1998) Genetic analysis of Lolium. I. Identification of linkage groups and the establishment of a genetic map. Plant Breed. 117, 451-455.

Heffner, E.L., M.E. Sorrells, and J.- L. Jannink, (2009) Genomic selection for crop 5 improvement. Crop Sci. 49, 1-12.

Henderson, C.R., 1984: Applications of linear models in animal breeding. University of Guelph, Ontario, Canada.

Hess, M. et al, (2011) Metagenomic Discovery of Biomass-Degrading Genes and Genomes from Cow Rumen. Science 331, 463-467.

Houle, D., D.R. Govindaraju, and S. Omholt, 2010: Phenomics; the next challenge. Nature Rev. Genet. 11, 855-866.

Huang, X., Q. Feng, Q. Qian, Q. Zhao, L. Wang, A. Wang, J. Guan, D. Fan, Q. Weng, T. Huang, G. Dong, T. Sang, and B. Han, 2009: High-throughput genotyping by whole-genome sequencing. Genome Res. 19, 1068-1076.

Huang, X., X. Wei. T. Sang, Q. Zhao, Q. Feng, Y. Zhao, C. Li, C. Zhu, T. Lu, Z. Zhang, M. Li, D. Fan, Y. Guo, A. Wang, L. Wang, L. Deng, W. Li, Y. Lu, Q. Weng, K. Liu, T. Huang, T. Zhou, Y. Jing, W. Li, Z. Lin, E.S. Buckler, Q. Qian, Q.F. Zhang, J. Li, and B. Han, 2010: Genome-wide association studies of 14 agronomic traits in rice landraces. Nat. Genet. 42, 961-967.

Janninck, J.-L., A.J. Lorenz, and H. Iwata, 2010: Genomic selection in plant breeding: from theory to practice. Briefings Funct. Genom. 9, 166-177.

Jones, E.S., M.P. Dupal, R. Kolliker, M.C. Drayton, and J.W. Forster 2001: Development and characterisation of simple sequence repeat (SSR) markers for perennial ryegrass (Lolium perenne L.). Theor. Appl. Genet. 102, 405-415.

Jones, E.S., N.L. Mahoney, M.D. Hayward, LP. Armstead, J.G. Jones, M.O. Humphreys, LP. King, T. Kishida, T. Yamada, F. Balfourier, G. Charmet, and J.W. Forster, 2002a: An enhanced molecular marker-based map of perennial ryegrass (Lolium perenne L.) reveals comparative relationships with other Poaceae species. Genome 45, 282-295.

Jones, E.S., M.P. Dupal, J.L. Dumsday, L.J. Hughes, and J.W. Forster, 2002b: An SSR-based genetic linkage map for perennial ryegrass (Lolium perenne L.). Theor. Appl. Genet. 105, 577-584.

-2072017251752 25 Oct 2017

Kaur, S., M.G. Francki, and J.W. Forster, 2012: Identification, characterisation and interpretation of single nucleotide sequence variation in allopolyploid crop plant species. Plant Biotech. J. 10, 125-138.

Lawless, K.A., M.C. Drayton, M.L. Hand, R.C. Ponting, N.O.I Cogan, T.I. Sawbridge,

K.F. Smith, G.C. Spangenberg, and J.W. Forster, J.W., 2009): Interpretation of SNP haplotype complexity in white clover (Trifolium repens L.), an outbreeding allotetraploid species. In: T. Yamada, and G. Spangenberg (eds.) Molecular Breeding of Forage and Turf: The Proceedings of the 5^th International Symposium on the Molecular Breeding of Forage and Turf, 211-221. Springer, New York, USA.

Lango Allen, Η, K. Estrada, G. Lettre, S.l. Berndt, M. Weedon, et al., 2010: Hundreds of variants clustered in genomic loci and biological pathways affect human height. Nature 467, 832-838.

Mackay, T.F.C., 2001: The genetic architecture of quantitative traits. Ann. Rev. Genet. 35, 303-309.

Margulies, Μ., M. Egholm, W.E. Altman, S. Attiya, J.S. Bader, L.A. Bemben, J. Berka,

M.S. Braverman, Y.-J. Chen, Z. Chen, S.B. Dewell, L. Du, J.M. Fierro, X.V. Gomes, B.C. Godwin, W. He, S. Helgesen, C.H. Ho, G.P. Irzyk, S.C. Jando, M.L.I. Alenquer, T.P. Jarvie, K.B. Jirage, J.-B. Kim, J.R. Knight, J.R. Lanza, J.H. Leamon, S.M. Lefkowitz, M. Lei, J. Li, K.L. Lohman, H. Lu, V.B. Makhijani, K.E. McDade, M.P.

McKenna, E.W. Myers, E. Nickerson, J.R. Nobile, R. Plant, B.P. Puc, M.T. Ronan, G.T. Roth, G.J. Sarkis, J.F. Simons, J.W. Simpson, M. Srinivasan, K.R. Tartaro, A. Tomasz, K.A. Vogt, G.A. Volkmer, S.H. Wang, Y. Wang, M.P. Weiner, P. Yu, R.F. Begley, and J.M. Rothberg, 2005: Genome sequencing in microfabricated highdensity picolitre reactors. Nature 437, 376-380.

Meuwissen, T. Η. E., B. J. Hayes, and Μ. E. Goddard, 2001: Prediction of total genetic value using genome-wide dense marker maps. Genetics 157, 1819-1829. Meuwissen, T.H.E., and M.E. Goddard, 1996: The use of marker haplotypes in animal breeding schemes. Gen. Sei. Evol. 28, 161-176.

Meuwissen, T.H., and M.E. Goddard, 2004: Mapping multiple QTL using linkage disequilibrium and linkage analysis information and multitrait data. Genet. Sei. Evol. 36, 261-79.

Meuwissen, T.H.E., 2009: Accuracy of breeding values of 'unrelated' individuals

-2082017251752 25 Oct 2017 predicted by dense SNP genotyping. Genet Sei Evol. 41, 35.

Odegard, J., M.H. Yazdi, A.K. Sonesson, and T.H.E. Meuwissen, 2008: Incorporating desirable genetic characteristics from an inferior into a superior population using genomic selection. Genetics 181, 737-745.

Pearson, A. N.O.I. Cogan, R.C. Baillie, M.L. Hand, H. Shinozuka S. Erb, T. Wilkinson, G.A. Kearney, A.R. Gendall, K.F. Smith, and J.W. Forster, 2010: Identification of QTLs for morphological traits influencing waterlogging tolerance in perennial ryegrass (Lolium perenne L.). Theor. Appl. Genet. 122, 609-622.

Pessino, S.C., J.P.A. Ortiz, M.D. Hayward, and C.L. Quarin, 1999: The molecular 0 genetics of gametophytic apomixis. Hereditas 130, 1-11.

Ponting, R.C., M.C., Drayton, N.O.I. Cogan, M.P. Dobrowolski, K.F. Smith, G.C. Spangenberg, and J.W. Forster, 2007: SNP discovery, validation, haplotype structure and linkage disequilibrium in full-length herbage nutritive quality genes of perennial ryegrass (Lolium perenne L.). Mol. Genet. Genom. 278, 589-597.

Pryce, J.E., B.J. Hayes, and M.E. Goddard, 2012: Novel strategies to minimize progeny inbreeding while maximizing genetic gain using genomic information. J. Dairy Sci. 95, 377-88.

Riedelsheimer, C, A. Czedik-Eysenberg, C. Grieder, J. Lisec, F. Technow, R. Sulpice, T. Altmann, M. Stitt, L. Willmitzer, and A.E. Melchinger, 2012: Genomic and 0 metabolic prediction of complex heterotic traits in hybrid maize. Nat Genet. 15, 21720.

Sanna, S., A.U. Jackson, R. Nagaraja, C.J Wilier, et al., 2008: Common variants in the GDF5-UQCC region are associated with variation in human height. Nat. Genet. 40, 198-203.

Sessitsch, A., J.G. Howieson, X. Perret, H. Antoun, and E. Martinez-Romero, 2002: Advances in Rhizobium research. Crit. Rev. Plant Sci. 21, 323-378.

Shinozuka, H., N.O.I. Cogan, K.F. Smith, G.C. Spangenberg, and J.W. Forster, 2010: Fine-scale comparative genetic and physical mapping supports map-based cloning strategies for the self-incompatibility loci of perennial ryegrass (Lolium perenne L.). 30 Plant Mol. Biol. 72, 343-355.

Skot, L., J. Humphreys, M.O. Humphreys, D. Thorogood, J. Gallagher, R. Sanderson, I.P. Armstead, and I.D. Thomas, 2007: Association of candidate genes

-2092017251752 25 Oct 2017 with flowering time and water-soluble carbohydrate content in Lolium perenne (L.). Genetics 177, 543-547.

Skot, L., R. Sanderson, A. Thomas, K. Skot, D. Thorogood, G. Latypovea, T. Asp, and I. Armstead, 2010: Allelic variation in the perennial ryegrass FLOWERING 5 LOCUS T gene is associated with changes in flowering time across a range of populations. Plant Physiol. 155, 1013-1022.

Spelman, R. J., D. J. Garrick, and J. A. M. van Arendonk, 1999: Utilisation of genetic variation by marker assisted selection in commercial dairy cattle populations. Livest. Prod. Sci. 59: 51-60.

Spielman, M., R. Vinkenoog, and R.J. Scott, 2003: Genetic mechanisms of apomixis. Phil. Trans. Roy. Soc. Lond. B. 358, 1095-1103.

Sved, J.A., 1971: Linkage disequilibrium and homozygosity of chromosome segments in finite populations. Theor. Popul. Biol. 2, 25-141

Tian, F., P.J. Bradbury, P.J. Brown, H. Hung, Q. Sun, S. Flint-Garcia, T.R.

Rocheford, M.D. McMullen, J.B. Holland, and E.S. Buckler, 2011: Genome-wide association study of leaf architecture in the maize nested association mapping population. Nature Genet. 43, 159-162.

The Bovine Hapmap Consortium, 2009: The genetic history of cattle. Science 324, 528-532.

VanRaden, P. M., C.P. Van Tassell, G.R. Wiggans, T.S. Sonstegard, R.D. Schnabel, J.F. Taylor, and F. Schenkel, 2009: Invited review: Reliability of genomic predictions for North American Holstein bulls. J. Dairy Sci. 92, 16-24.

Turner L.B., M. Farrell, M.O. Humphreys, and O. Dolstra, 2010: Testing water-soluble carbohydrate QTL effects in perennial ryegrass (Lolium perenne L.) by marker 25 selection. Theor. Appl. Genet. 121, 1405-1417.

Vogel, K.P., and J.F. Pedersen, 1993: Breeding systems for cross-pollinated forage grasses. Plant Breed. Revs. 11, 251-274.

Walter, A., B. Studer, and R. Kolliker, 2012: Advanced phenotyping offer opportunities for improved breeding of forage and turf species. Ann. Bot., available 30 on-line: doi:10.1093/aob/mcs026.

Wang, J., M.P. Dobrowolski, N.O.I. Cogan, J.W. Forster, and K.F. Smith, 2009: Assignment of individual genotypes to specific forage cultivars of perennial ryegrass

-210 2017251752 25 Oct 2017 (Lolium perenne L.) based on SSR markers. Crop Sci. 49, 49-58.

Wang, L., A. Wang, X. Huang, Q. Zhao, G. Dong, Q. Qian, T. Aang, B. Han, 2011: Mapping 49 quantitative trait loci at high resolution through sequencing-based genotyping of rice recombinant inbred lines. Theor. Appl. Genet. 122, 327-340.

Wilkins, P.W., and M.O. Humphreys, 2003: Progress in breeding perennial forage grasses for temperate agriculture. J. Agric. Sci. 140, 129-150.

Van Inghelandt, D., J.C. Reif, B.S. Dhillon, P. Flament, and A.E. Melchinger, 2011: Extent and genome-wide distribution of linkage disequilibrium in commercial maize germplasm. Theor Appl Genet. 123, 11-20.

Visscher, P. M., 2008: Sizing up human height variation. Nat. Genet. 40, 489 - 490. Xing, Y., U. Frie, B. Schejbel, T. Asp, and T. LObberstedt, 2007: Nucleotide diversity and linkage disequilibrium in 11 expressed resistance candidate genes in Lolium perenne. BMC Plant Biol. 7: 43.

Yan, J., T. Shah, M.L. Warburton, E.S. Buckler, M.D. McMullen, and J. Crouch, 2009:

Genetic characterization and linkage disequilibrium estimation of a global maize collection using SNP markers. PLoS One 4, e8451.

Claims (15)

1. THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

   1. A method for producing an improved plant, said method including:

   (v) providing a library of germplasm of an organism; and

   5 a library of symbionts;

   (vi) inoculating the plant germplasm library with one or more symbionts selected from the symbiont library to generate symbiota;

   (vii) selecting, screening and/or evaluating the inoculated plant germplasm library or the symbiota for desired symbiota characteristic(s); and

   0 (viii) subsequently identifying, culturing or otherwise using the symbiota exhibiting desired characteristic(s) to produce the improved plant, wherein:

   said symbiont(s) are endophytes and/or epiphytes and/or plant-associated microbiomes;

   5 said library of germplasm includes 100s to 1000s of genotypes;

   said library of symbionts includes 10s to 100s of genotypes; and said step (iii) includes a step of genetic analysis and/or metabolic analysis; and said plant germplasm is present as an embryo; and said inoculation step

   0 includes coating said embryo with a symbiont-containing coating layer to form an artificial seed.
2. 2. A method according to claim 1, wherein the symbiont is selected from one or more of fungi, viruses, bacteria and other microbes.
3. 3. A method according to claim 2, wherein the symbiont(s) includes a genetic variation selected to enhance symbiont trait introgression.
4. 4. A method according to claim 3, wherein the genetic variation is introduced via

   30 one or more of random mutagenesis, di-poly-ploidisation, targeted mutagenesis, cisgenesis, transgenesis, and intragenesis.

   -212 2017251752 07 Oct 2019
5. 5. A method according to claim 1, wherein the plant has symbiotic compatability with an endophyte.
6. 6. A method according to claim 5, wherein the embryo is treated to create one or 5 more points of entry for the symbiont.
7. 7. A method according to claim 1, wherein the screening step (iii) includes screening artificial seeds for compatibility and/or stability by accelerated ageing and selecting symbiota exhibiting desired characteristic(s).

   o
8. 8. A method according to claim 7, further including subjecting the selected symbiota to a rapid endophyte viability assay.
9. 9. A method according to any one of claims 1 to 8, wherein the method further 5 includes:

   growing the artificial seeds to form plantlets or seedlings; and screening the plantlets or seedlings for symbiont presence.
10. 10. A method according to claim 9, wherein the plant is a grass or a legume.

    :0
11. 11. An improved plant with a symbiont, produced using a method according to any one of claims 1 to 10.
12. 12. A plant, plant seed or other plant part derived from a plant according to claim

    25 11 and stably infected with a symbiont.
13. 13. A method of genomic selection of plant-symbiont associations, said method including:

    (i) providing a library of nucleic acid samples from said plant-symbiont 30 association obtained by the method according to claim 1;

    (ii) analysing said samples using metagenomics to obtain a genetic profile for each sample; and

    -2132017251752 07 Oct 2019 (iii) selecting plant-symbiont associations with a desired genetic and metabolic profile.
14. 14. A method according to claim 13, including the preliminary steps of:

    5 (i) providing a plant-symbiont association;

    (ii) subjecting said plant-symbiont association to one or more environmental conditions; and (iii) preparing a library of nucleic acid samples from each of the environmentally treated plant-symbiont associations.

    o
15. 15. A method according to claim 14, wherein said symbiont is a bacterial microbiome.