Metabolic Potential of Epichloë Endophytes for Host Grass Fungal Disease Resistance.

Asexual species of the genus Epichloë (Clavicipitaceae, Ascomycota) form endosymbiotic associations with Pooidae grasses. This association is important both ecologically and to the pasture and turf industries, as the endophytic fungi confer a multitude of benefits to their host plant that improve competitive ability and performance such as growth promotion, abiotic stress tolerance, pest deterrence and increased host disease resistance. Biotic stress tolerance conferred by the production of bioprotective metabolites has a critical role in an industry context. While the known antimammalian and insecticidal toxins are well characterized due to their impact on livestock welfare, antimicrobial metabolites are less studied. Both pasture and turf grasses are challenged by many phytopathogenic diseases that result in significant economic losses and impact livestock health. Further investigations of Epichloë endophytes as natural biocontrol agents can be conducted on strains that are safe for animals. With the additional benefits of possessing host disease resistance, these strains would increase their commercial importance. Field reports have indicated that pasture grasses associated with Epichloë endophytes are superior in resisting fungal pathogens. However, only a few antifungal compounds have been identified and chemically characterized, and these from sexual (pathogenic) Epichloë species, rather than those utilized to enhance performance in turf and pasture industries. This review provides insight into the various strategies reported in identifying antifungal activity from Epichloë endophytes and, where described, the associated antifungal metabolites responsible for the activity.

1. Introduction

Pasture is one of the main food sources for livestock throughout the world. While pure pasture-based farming systems rely solely on pasture, mixed farming systems supplement with commodities such as cereals and grains. More consumers are shifting towards “grass-fed” livestock produce, as it caters for both increasing global food demands and social concerns for animal-welfare [1,2]. Novel solutions to sustainably manage food supply systems also address increasing demands for nutritious food [3,4]. There is a wide variety of pasture species used in agriculture including annual grasses, perennial grasses, legumes, and herbs. Forage grasses that are adaptable to a wide range of climatic conditions are extensively used in farming systems [5,6].

With changing climates and evolving pest and pathogen pressures, pasture and turf grasses are threatened by abiotic and biotic stresses with increasing severity [5,7,8,9]. It has long been known that Epichloë fungi of the family Clavicipitaceae, associated with Pooidae grasses such as Lolium spp. (e.g., perennial ryegrass, short-term ryegrass and tall fescue), improve host plant abiotic and biotic stress tolerance by producing bioactive metabolites [9,10,11,12]. Though Epichloë endophytes are historically well-characterised for their antimammalian and insecticidal alkaloid toxins, recent studies confirm that their bioactivities are not limited to these compounds alone [11,13,14,15]. Therefore, there is a requirement for an increase in applied research into Epichloë endophyte-derived bioactive metabolites to improve host grass performance and stress tolerance. This review focuses on prospects for Epichloë endophytes providing host plant disease resistance, and discusses modern experimental data analysis techniques for bioprospecting of antifungal metabolites in novel Epichloë strains.

2. Epichloë Endophytes

Epichloë fungi infecting grasses are categorized as sexual or asexual species based on their transmission mechanism. Sexually transmitting Epichloë spp. are known to cause choke disease in grasses during flowering, leading to reduced seed yield and aesthetic value of turf grasses [16,17]. In this process the fungus forms a stroma with spores and horizontal transmission occurs [16,17]. As they cause disease, sexual Epichloë spp. are considered pathogenic even though they live asymptomatically in the host plant during vegetative growth stages. Asexual Epichloë spp. are clonal, transmit vertically by infecting the seeds, and do not cause diseases to the host grass in any growth stage of the host plant [16]. Therefore, as they provide significant performance benefits, the pasture and turf grass industries utilize asexual (endophytic) strains of Epichloë species.

Asexual Epichloë spp. form endophytic associations with cool-season grasses by colonizing leaves, pseudostems, seeds and seedlings. In this mutualistic relationship, the endophyte relies on the plant for vertical dissemination while also gaining shelter and nutrients throughout its life [13]. The production of functional metabolites triggered by the association greatly benefits the host plant by conferring abiotic and biotic stress tolerances such as improved seedling vigour, persistence, and enhanced growth [13,18]. While most of these benefits are advantageous in an agricultural scenario, Epichloë endophytes are also well known for producing toxic antimammalian alkaloids (lolitrem B—ryegrass staggers, ergovaline—fescue toxicosis). These alkaloid toxins are extremely harmful to grazing animals and cause significant economic losses to the livestock industry [19,20,21]. Consequently, studies related to their biological and chemical properties, genetics, biosynthetic pathway and modes of action have been the focus from as early as 1980s. The biosynthetic intermediaries of these compounds have also been characterized for biological activity and many have been selected as candidates for insecticides [22]. Previous reports on Epichloë endophytes have demonstrated that the alkaloids lolitrem B, ergovaline, n-acetylloline, n-formylloline and peramine possess insecticidal and invertebrate pest deterrence properties [20,23,24]. However, the broad chemical diversity and metabolic capacity of Epichloë endophytes relating to antimicrobial activity remain largely unknown.

Field, glasshouse and lab-based studies have established that Epichloë endophytes improve host plant disease resistance, and a few studies have identified the production of metabolites with novel antimicrobial properties that may play a role in host plant protection from phytopathogens [11,12,25]. However, in most of these studies wild-type strains such as SE (Standard Endophyte; also referred to as wild type or common endophyte, CE) or Ky31 (Kentucky31) were investigated rather than the animal friendly strains utilized in pastoral agriculture. Thus, to better exploit endophyte-mediated disease resistance, improve pasture and turf quality, and reduce the impact of phytopathogen disease on animal welfare, Epichloë strains that are safe for animals, should be investigated.

3. Novel Endophytes

The search for endophytes that do not produce metabolites toxic to mammalian grazers led to the discovery of novel endophytes and their use commercially, as shown by the many novel strains for which Plant Breeders’ Rights has been granted worldwide, and those marketed with registered trademarks (Table 1) [26,27,28]. The significant economic benefits of animal-safe endophyte strains that also improve pasture persistence by reducing the impact of invertebrate pests underpinned the investigation and exploration of high-performing endophyte infected pasture grasses [29,30,31]. Novel endophytes are typically screened and selected based on alkaloid profiles in planta, in particular low/no lolitrem B and low/no ergovaline combined with bioprotective (insect deterring) lolines and/or peramine. However, more recently, screening methods have been extended to biosynthetic pathways and gene clusters associated with these known alkaloids [27,29,32,33].

Table 1

Novel Epichloë endophyte strains for which Plant Breeders’ Rights (PBR) have been granted worldwide 1.

Country	PBR Grant Date	Endophyte Strain(Market Name)	EpichloëSpecies	HostCommon Name	Known Alkaloid Profile 2	Applicant
New Zealand	23 April 1996 (expired)	AR1	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	P	Grasslanz Technology Ltd. (GTL; Palmerston North, New Zealand)
European Union	20 October 2003
Australia	26 October 2004
New Zealand	21 April 2015	AR1006	E. uncinata	Meadow fescue	L	GTL
New Zealand	5 October 2016	AR1017	E. uncinata	Meadow fescue	L	GTL
New Zealand	29 November 2018	AR127	E. festucae var. lolii (LpTG-1)	Perennial ryegrass		GTL
New Zealand	25 July 2008	AR37	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	J	GTL
Australia	30 March 2010
Uruguay	20 November 2011					Fischer Fleurquin Gustavo(FFG; Montevideo, Uruguay)
New Zealand	23 April 1996 (expired)	AR501	E. coenophiala (FaTG-1)	Tall fescue	LP	GTL
European Union	20 October 2003
New Zealand	1 February 1999 (expired)	AR542(MaxP® New Zealand, Australia;MaxQ® USA)	E. coenophiala (FaTG-1)	Tall fescue	LP	GTL
Australia	26 October 2004
Uruguay	6 December 2004					FFG
New Zealand	25 July 2008	AR584(MaxP® New Zealand, Australia;MaxQII™ USA)	E. coenophiala (FaTG-1)	Tall fescue	LP	GTL
Australia	29 September 2010
Uruguay	2 August 2013					FFG
Argentina	4 September 2014					Gentos SA(Buenos Aires, Argentina)
New Zealand	12 May 2010	AR601 (Avanex®)	E. coenophiala (FaTG-1)	Tall fescue	EL	GTL
Australia	19 August 2013
European Union	22 May 2017
New Zealand	12 May 2010	AR604	E. coenophiala (FaTG-1)	Tall fescue	EL	GTL
European Union	22 May 2017
New Zealand	28 August 2014	AR95 (Avanex®)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	E	GTL
European Union	22 May 2017
Australia	3 April 2018
Uruguay	21 November 2014					FFG
New Zealand	17 January 2019	CM142	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	J	Cropmark Seeds Ltd.(CSL; Rolleston, New Zealand)
Australia	17 August 2020					Cropmark Seeds Australia Pty Ltd. (CSA; South Melbourne, Australia)
New Zealand	27 August 2014	E815 (Edge)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	LtmEP	DLF Seeds A/S(DLF; Roskilde, Denmark)
Australia	23 October 2017					DLF
New Zealand	23 June 2010	Happe	E. siegelii		L	DLF
Australia	23 October 2017					DLF
New Zealand	25 July 2008	Nea 2 (NEA/NEA2/NEA4) 3	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	LtmEP	Barenbrug New Zealand Ltd. (BBNZ; Christchurch, New Zealand)
New Zealand	30 June 2009	Nea 3 (NEA4 3)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	BBNZ
New Zealand	25 July 2008	Nea 6 (NEA2 3)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	BBNZ
New Zealand	29 August 2014	Nea 10	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	BBNZ
New Zealand	13 August 2014	Nea 11	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	BBNZ
New Zealand	19 March 2021	Nea 12	Epichloë sp. (LpTG-3)	Perennial ryegrass	J	Agriculture Victoria Services Pty Ltd. (Bundoora, Australia)
New Zealand	29 August 2014	Nea 21	Epichloë sp. (FaTG-3)	Tall fescue	LP	BBNZ
New Zealand	29 August 2014	Nea 23	Epichloë sp. (FaTG-3)	Tall fescue	LP	BBNZ
New Zealand	10 July 2019	Nea 47 (NEA2 3)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	BBNZ
New Zealand	18 August 2014	PTK647 (Protek®)	E. coenophiala (FaTG-1)	Tall fescue	EL	DLF
Australia	23 October 2017					DLF
New Zealand	28 August 2014	U12	E. uncinata	Meadow fescue	L	CSL
Australia	12 August 2021					CSA
New Zealand	2 September 2016	U13	E. uncinata	Meadow fescue	L	CSL
New Zealand	25 July 2008	U2 (GrubOUT®)	E. uncinata	Meadow fescue	L	CSL
Australia	30 January 2014					CSA
Argentina	11 March 2015					Gentos SA
European Union	22 May 2017					CSL
New Zealand	14 October 2008	UNC1	E. uncinata	Meadow fescue	L	CSL
Australia	Not applicable 4	AR5 (Endo5)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	EP	GTL
USA	Not applicable 4	E34®	E. coenophiala (FaTG-1)	Tall fescue	LP	Barenbrug USA(Tangent, OR, USA)
USA	Not applicable 5	KY31 (Kentucky31)	E. coenophiala (FaTG-1)	Tall fescue	EPL	Not applicable
worldwide	Not applicable 5	SE (Standard endophyte)	E. festucae var. lolii (LpTG-1)	Perennial ryegrass	LtmEP	Not applicable

1 UPOV Pluto database search for Plant Breeders’ Rights granted to 26 November 2021 https://pluto.upov.int/search (Search: Botanical name -Epichloë). 2 Alkaloid profile, P = peramine, L = lolines, E = ergovaline, Ltm = lolitrem B, J = epoxy-janthitrems. 3 Strains are sometimes marketed as a combination e.g., NEA2 = Nea 2, Nea 6, Nea 47. 4 Commercial endophytes that were not listed in UPOV Pluto database search results. 5 Wildtype, toxic strains.

The search for animal-safe endophyte strains is important to the agricultural industry, with new endophytes being developed continuously (Table 1). Therefore, high-throughput methods and commercial standards are used regularly to detect the presence of alkaloids in pasture [34]. Those compounds that are not commercially available can also be isolated and purified using published methodologies [19,35]. The screening methods are, however, limited to the “known known” alkaloids—the well described Epichloë-derived compounds referred to in Table 1. Of the novel endophytes described on Table 1 only two, Nea 12 and Nea 23, have been investigated for production of antifungal metabolites in vitro and in planta [36,37]. Thus, further research is required to investigate the ‘known unknown’ metabolites, such as those responsible for disease resistance, that are beneficial to the pasture and turf grass related industries [38].

4. Disease Stress to Pasture Grasses

Pasture grasses are threatened by many pathogenic diseases, causing devastating losses to pasture yield and quality. These diseases include rusts, leaf spot diseases, blights, blotches, moulds, and wilts caused by pathogenic fungi, bacteria or viruses (Table 2).

Table 2

Common disease-causing phytopathogens in pasture and turf grasses.

Disease/Common Name	Causative Organism	Symptoms	Damage/Loss	Control Measures	References
Crown rust	Puccinia coronata	Reddish brown spores on leaf	Dry matter yield loss 30–40% and stock thrift	Fungicide; Judicious grazing management; Resistant varieties.	[41,45,76]
Grey leaf spot	Pyricularia grisea	Small water-soaked lesions on leaf blades gradually turning to dark necrotic spots and to grey spots	Up to 90% pasture loss	Fungicide; Controlled release of N fertiliser; Biocontrol, Resistant varieties.	[50,63,77,78]
Brown blight and net blotch	Drechslera sp.	Net lesions with small dark brown bars amphigenous lesions with dark brown margins, light brown center	Dry matter and herbage loss	Fungicide; Managed grazing before it spreads.	[49,60,79]
Stem end rust	Puccinia graminis	Reddish brown spores on sheath and stem	Seed yield loss, Dry matter loss	Fungicide; Judicious grazing management; Resistant varieties.	[49,80]
Blind seed disease	Gloeotinia temulenta	Fungal mycelia on seeds under microscopic observation	Seed yield loss, reduce seed germination 50–90%	Fungicide; Increased rate of N application.	[64,65,81]
Snow mould	Microdochium nivale	Dark brown lesions and pink sporodochia rows parallel to veins	Seedling damage leading to yield loss, seed loss and yield loss	Fungicide; Biological control; Compost application.	[51,60,82]
Yellow patch	Ceratobasidium cereale	Root pathogen	Yield loss	Fungicide	[83,84]
seedling pathogen and leaf spot	Fusarium solani	Wilting of seedlings and necrotic lesions in mature plants	Yield loss, dry matter loss	Fungicide	[85,86]
Bacterial wilt	Xanthomonas translucens	Water-soaked lesions and turning to bluish purple colour	Forage yield loss 20–40%	Biological control; Resistant varieties.	[52,87,88,89]
Ryegrass mosaic virus	RGMV	Light green-yellow streaky mosaic or brown necrosis on leaves	Dry matter yield loss 21–30%	Resistant varieties; Mixed pasture species.	[7,90]

In perennial ryegrass, crown rust (Puccinia coronata) is one of the most severe fungal disease-causing pathogens that affects foliage, causing substantial losses in pasture and turf grass industries. Severe rust infections can cause up to 37% loss in dry matter (DM) when infected plants are harvested and dried, and a 94% loss in fresh matter (FM) yield [39,40,41,42]. Water soluble carbohydrate content in rust-infected grasses is significantly lower compared to uninfected grasses, which in turn affects digestibility and palatability and can lead to low milk yields if used as feed [43,44,45]. Root growth is also reduced due to depleted carbohydrate reserves [46]. Further, the resultant increase in dead herbage tissue makes grass susceptible to Pithomyces chartarum, which causes serious disease of facial eczema in cattle and sheep [46,47].

Other foliage pathogens include Drechslera sissans, which causes significant production losses; disease incidence is significantly increased with high nitrogen (N) use conditions [44,48,49]. Though not prominent in Australia, snow mould (Microdochium nivale) and grey leaf spot (Pyricularia grisea) are another two serious diseases causing yield losses in perennial ryegrass and tall fescue in the northern hemisphere [50,51].

As well as the effect on foliage, several fungal pathogens also directly infect the inflorescence, thus altering seed yield. The most significant of these are stem rust (Puccinia graminis) and blind seed disease (Gloeotinia granigena), which both occur on seed heads on a wide range of pasture grasses. Stem rust infections can result in seed yield losses of up to 93% in turf-type perennial ryegrass cultivars, forcing seed producers, especially those growing late maturing cultivars, to use fungicides to prevent losses. Incidence of blind seed disease depends on environmental conditions. Infection usually results in seed death from heavily infected stands [44].

Pathogens are not limited to fungi. Bacterial pathogens include Xanthomonas translucens pv graminis, which causes bacterial wilt disease resulting in yield losses in ryegrass [52]. Pseudomonas syringae pv atropurpurea causes chlorosis in ryegrass [53]. Viral diseases can also infect perennial grasses, and occur in high incidences causing serious damage. These include barley yellow dwarf virus (BYDV), cereal yellow dwarf virus (CYDV), and ryegrass mosaic virus (RGMV). Upon infection, BYDV and CYDV cause leaf yellowing, stunting and tillering, and these symptoms have an effect on plant performance, productivity, quality and yield [7,54]. Mosaic streaking necrosis by RGMV leads to herbage yield loss [7]. Table 2 further describes common diseases in perennial ryegrass and tall fescue used for pasture and turf, detailing causative organism, symptoms, damage and current control measures.

Phytopathogenic fungi causing disease in grasses may also produce mycotoxins that pose a threat to animal health and wellbeing. Fusarium sp. derived toxins—T-2/HT-2, zearalenone, deoxynivalenol—affect food intake and animal performance [55,56]. Though actual toxins are not well characterized, Drechslera biseptata has been associated with acute bovine liver disease [57,58]. Rusts caused by Puccinia sp., pathogens reduce the palatability of infected grasses by altering total carbohydrate contents [41,59].

Disease outbreaks cause significant yield loss and degrade forage quality, and hence proper control measures are required. The most widely used method for disease control is chemical treatment, which includes spraying fungicides and insecticides for disease vectors [60,61,62]. Fungicides include dimethylation inhibitors, nickel salts and dithiocarbamates [63,64]. Another approach is to apply suitable fertiliser in desirable rates of application to enrich the growth medium with macro and micro nutrients necessary for plant growth and performance [63,65]. Cultural and mechanical control measures include grazing and irrigation management practices [49,60]. All these curative methods are costly and labour demanding. Continuous long term application of fungicides may have a negative impact on livestock health, as some fungicides are toxic when accumulated in large quantities [66,67]. Furthermore, fungicides have a negative impact on Epichloë endophytes in perennial ryegrass and tall fescue and may lead to loss of benefits from the symbiotic association [64,68]. Breeding of resistant varieties is time consuming and costly. Thus, introducing beneficial microorganisms as biocontrol agents may be a more cost and time-efficient response.

Bacteria have been reported to be effective biocontrol agents when applied to the plant or seeds. Pseudomonas aeruginosa reduces disease incidence and disease severity of grey leaf spot in perennial ryegrass when applied to seeds or the plant in controlled environment pot trials and field trials [69]. Paenibacillus elgii SD17 reduces disease severity of brown patch disease and pythium blight in turf grasses in both controlled chamber and field trials [70]. Inoculating novel Epichloë strains to grass populations would have relatively minimal ecological impacts as the symbiota are naturally occurring [71].

The asexual Epichloë sp. utilized in pastures and turf are known to produce antimicrobial metabolites and inhibit pathogen growth under both in vitro and in planta conditions [25,37,72]. With the availability of new high throughput technology, rigorous analytical methods are available to identify and quantify antimicrobial activity and detect the responsible molecules [73,74,75]. Early studies have shown the potential of Epichloë endophytes to inhibit pathogen infections, while more recent studies have focused on isolating and characterising responsible antimicrobial molecules [25,36,72].

5. Epichloë sp.-Derived Antifungal Activity and Host Disease Resistance Studies

There are limited reports of Epichloë endophyte-derived antifungal activity and host disease resistance. For this review we used the scientific search engine Scopus (www.scopus.com (accessed on 5 November 2021) to identify peer reviewed scientific reports with the terms “Epichloë” or “Neotyphodium” and “fungitoxic” or “antifungal” in the title, abstract, or keywords. Only publications in English and published in the last 25 years (1996–2021) were considered. Data integrity and collation were performed by either individually checking and filtering nonrelated articles or preserving and adding related articles cited in reviews and journal publications. Relevant secondary documents listed in Scopus but not indexed in the database were also included. Subsequently, only 46 journal articles, three letters, and one short survey (Supplementary Table S1) were found. During the same time period, the number of English journal manuscripts found using the terms “Epichloë” or “Neotyphodium” was 1313. Although only 3.5% of Epichloë fungi-related studies concerned antifungal activity and host disease resistance against fungal pathogens, this number has been increasing (Figure 1). This demonstrates that researchers are recognizing the importance of Epichloë endophyte derived antifungal activity as an important driver for better performing pasture and turf.

Figure 1

Number of documents (journal articles, letters and short surveys) published in last 25 years on Epichloë-derived antifungal activity and/or disease resistance.

6. Bioprospecting Antifungal Metabolites from Epichloë Endophyte Strains

Despite the paucity of publications on Epichloë-derived antifungal activity and host disease resistance, there have been papers describing various methodologies to demonstrate the presence of biological activity of Epichloë endophyte strains [12,25,91,92,93,94]. However, the methods utilized historically are subjective, qualitative and do not allow for accurate comparisons between studies. To address the current research gaps in Epichloë-mediated fungal disease resistance, it is necessary to establish rigorous analytical processes to characterize the bioactivity of Epichloë strains quantitatively and accurately to determine the effect of responsible compounds. Recent advances in analytical methods and software tools enable development of standardized protocols and processes to analyze antifungal activity. In this section we review current knowledge in the field, highlighting the methods and tools available, to create a schematic process for identification, characterization and application of antifungal compounds produced by Epichloë endophytes (Figure 2).

Figure 2

An experimental workflow proposed for bioprospecting antifungal metabolites from Epichloë endophytes using a stepwise process. (a) Epichloë strain identification, (b) Identification of bioactive strains using in vitro antifungal activity assays, (c) antifungal metabolite isolation and characterisation, (d) untargeted metabolite annotation for antifungal compound detection, (e) qualitative and quantitative confirmation of antifungal metabolites in planta.

6.1. Epichloë Endophyte Strain Identification

Epichloë strains are commonly tested in vitro using different plate-based assays to identify their antifungal activity [92,93,94,95]. Endophyte strains must be isolated, purified, and transferred to in vitro cultures prior to being tested (Figure 2a). As they are naturally found in symbiotic association with host plants, it is ideal to freshly isolate the Epichloë strain from an infected plant. [27]. Strain identity should be confirmed in advance of any testing, which generally involves DNA-based PCR analysis, such as an SNP diagnostic test, to enable identification of known Epichloë strains [28,96]. Ideally, novel endophyte strains would be defined by whole genome sequencing prior to testing bioactivity. Alkaloid profiling endophyte strains for currently known Epichloë-derived alkaloids will provide additional information on animal safety concerns [26,27].

6.2. Identification of Bioactive Strains Using In Vitro and in Planta Assays

Antifungal activities of endophyte strains are commonly detected using in vitro dual culture assays [37,92,93,94,95,97]. In dual culture assays, phytopathogens and endophytes are grown in close proximity to study their antagonistic reactions (Figure 2b). Growth parameters (growth area, mycelial density, growth direction) of phytopathogens are observed over a period of time to detect inhibitory activity. Earlier studies qualitatively analysed the results based purely on observation. However, with the availability and accessibility of imaging and image analysis software, growth parameter data such as pathogen growth area can easily be converted to quantitative data. Quantitative data can be analysed for precision and errors, and subjected to statistical analysis to detect the significance of antifungal activity [94]. Dual culture assays are suitable for detecting antifungal activity as they are quick, easily replicated and can be used to test against a range of pathogens. Importantly, the in vitro antifungal phenotypes observed are consistent and observed in independent isolates of the same strain and across duplicate assays [37,98,99].

In addition, dual culture assays provide information on direct interaction of the endophyte with phytopathogens, for example, Epichloë festucae strain E437 was shown to reduce hyphal tip growth of the pathogens Drechslera erythospila and Colletotrichum graminicola [95]. Fernando et al. (2020) observed differential bioactivity between three asexual Epichloë strains (Nea 12, Nea 21 and Nea 23) evaluated against three phytopathogens (Ceratobasidium sp., Drechslera sp. and Fusarium sp.), indicating that there is variation in the production of bioactive metabolites and their composition [37]. In vitro liquid culture extracts exhibited differential antifungal activity consistent with dual culture assays [37], establishing that the Epichloë strains produce and secrete antifungal metabolites.

However, it should be noted that virulence of a pathogen may be different when it is infecting a plant; therefore, dual culture assay results may not directly relate to pathogen inhibition in planta. Dual culture assays are also unsuitable for biotrophic pathogens (e.g., rusts such as Puccinia coronata) that are unable to be grown in pure cultures. Spore germination assays are less common but have also been used to study the antifungal activity of Epichloë strains and understand the physiological mechanisms [93,100,101,102]. It is noteworthy that Christensen and Latch’s study in 1991 is the only available in vitro study of Epichloë coenophiala (previously known as Acremonium/Neotyphodium coenophialum) demonstrating antifungal activity against spores of the rust pathogen Puccinia graminis by inhibition of urediniospore germination [101]. Spore germination assays are also a useful tool for testing antifungal compounds isolated from Epichloë sp. [102].

Detached leaf assays are another type of semi-in vitro assay used to overcome some of the limitations associated with dual culture and spore germination assays. In detached leaf assays, leaves from endophyte infected host plants are inoculated with phytopathogens and disease symptom development parameters (leaf spot number, area/size of leaf spot) are used to characterise the antifungal activity [25,95]. These assays are complex, do not account for direct interaction of the pathogen with the endophyte [11] and depend on tiller age, endophyte incidence, concentration of pathogen inoculum, and environmental conditions, which may lead to inconsistent results [92]. Nonetheless, detached leaf assays can be informative when performed in addition to dual culture assays.

There are a few instances where pot or field trials were conducted to detect endophyte mediated disease resistance in grasses [91,95,103,104,105,106,107,108,109]. Most studies investigated wild type endophyte strains, and others did not provide information on strain details. These in planta assays have confirmed that Epichloë endophytes improve host plant disease resistance. While some studies focused on measuring disease severity parameters (lesion numbers, lesion size) [91,110] other studies included leaf senescence characteristics (chlorophyll content, superoxide dismutase, peroxidase, catalase, and ascorbate peroxidase) to further understand the mechanism of disease resistance [95,103]. Conducting field and pot trials in quarantine is expensive and increases risk of outbreaks.

6.3. Antifungal Metabolite Isolation and Characterisation

While most studies have focused on detecting antifungal activity, a few characterised the mode of action and identify compounds responsible. Some studies provide evidence that Epichloë strains produce antifungal molecules in vitro liquid cultures and secrete to the culture media [37,97,111,112]. In these studies, the filtrate containing the fungal secretome can be extracted and bioassays performed to identify antifungal activity [37,111,112]. Three studies have extracted the secretome/culture filtrate using different solvent systems and tested their activity against selected grass pathogens [37,112,113]. These confirmed the ability of Epichloë strains to produce antifungal molecules in vitro even when not in contact with the pathogens. This characteristic is important to identify and isolate bioactive compounds (Figure 2c).

Historically, antifungal metabolites have been isolated from a few sexual Epichloë strains (Table 3), while antifungal metabolites produced by strains of asexual (endophytic) Epichloë species that are utilised by pasture and turf industries remain to be discovered. While most metabolites are isolated from in vitro cultures, others have been isolated from infected plants (Table 3 and Figure 3). Early studies identified a series of antifungal compounds from Epichloë typhina isolated from timothy grass (Phleum pratense) [114,115]. Yue et al. (2000) tested antifungal activity of three fractions of aqueous extracts from a range of Epichloë species and confirmed their antifungal activity against Cryphonectria parasitica, the causal phytopathogen of chestnut blight, using thin layer chromatography assay [112]. Subsequently, they used E. festucae (BM7, M. D. Richardson) from Festuca rubra to isolate six antifungal metabolites and confirmed their activity against C. parasitica, Lactisaria fusiformis, Magnaporthe poae, and Rhizoctonia solani using Potato Dextrose Agar (PDA) plate based assays [112]. Nuclear Magnetic Resonance (NMR) and Gas Chromatography-Mass Spectrometry (GC-MS) data were acquired to fully characterise the isolated compounds. Tian et al. (2017) isolated antifungal protein Efe-AfpA, active against Sclerotinia homoeocarpa [105]. They used E. festucae Rose City isolate (E. festucae RC) in association with Festuca rubra subsp. rubra (strong creeping red fescue) to isolate apoplastic proteins. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was used to isolate the proteins. The peptide sequence was determined by using Liquid Chromatography tandem Mass Spectrometry (LC/MS/MS2). Bioactivity of the isolated protein was confirmed using a PDA plate-based assay [105]. Purev et al. (2020) isolated antifungal ε-poly-L-lysines encoded by the VibA gene from E. festucae strain E437. They used NMR and MALDI-TOF MS to identify the molecule and determine the structure. Disk diffusion assays confirmed the antifungal activity against Drechslera erythrospila and Phytophthora capsica [116]. Structures of some of these antifungal metabolites are shown in Figure 3. A recent study by Fernando et al. (2021) confirmed currently known Epichloë-derived antimammalian and insecticidal alkaloids and their intermediates (peramine, n-formylloline, n-acetylloline, lolitrem B, epoxyjanthitrem I, paxilline, terpendole E, terpendole C, ergovaline) are not responsible for the antifungal activity observed by Epichloë endophytes [22].

Table 3

Summary of fungitoxic compounds isolated from sexual Epichloë sp. and their characteristics.

Fungitoxic Compound	ChemicalCharacteristics	ChemicalFormula	m/z or Molecular Weight	Epichloë sp.	Host Grass	Source Material	Tested Pathogens	Reference
Indole-3-acetic acid	IAA derivative	C10H9NO2	176.0667	E. festucae	Festuca rubra	purified mycelia	Lactisaria fusiformis,Magnaporthe poae,Rhizoctonia solani	[112]
Indole-3-ethanol	IAA derivative	C10H11NO	162.0874	E. festucae	Festuca rubra	purified mycelia	L. fusiformis, M. poae,R. solani	[112]
Methylindole-3-carboxylate	IAA derivative	C10H9NO2	175.0594	E. festucae	Festuca rubra	purified mycelia	L. fusiformis, M. poae,R. solani	[112]
Indole-3-carboxaldehyde	IAA derivative	C9H7NO	159.0684	E. festucae	Festuca rubra	purified mycelia	L. fusiformis, M. poae,R. solani	[112]
Cyclonerodiol	sesquiterpinoid	C15H28O2	165.0507	E. festucae	Festuca rubra	purified mycelia	L. fusiformis, M. poae,R. solani	[112]
Chokol AChokol BChokol CChokol DChokol EChokol FChokol G	sesquiterpinoid	C12H22O2C15H26O2C15H26O2C15H26O2C15H28O3C14H24O3C11H20O2	199.1693 239.2006 239.2006 239.2006 271.2228 241.1798 185.1536	E. typhina	Phleum pratense	Epichloë infected plant material (choke)	Cladosporium herbarum, Cladosporium phlei	[114,132]
Chokol K	sesquiterpinoid	C15H26O	222.19841 [M-H2O]− = 204	E. sylvatica E. clarkii	Brachypodium sylvaticum Holcus lanatus	Unfertilized stromata and unfertilized stromata head space	Stagonospora nodorum,Mycosphaerella graminicola	[102]
N,N-diacetamide	diactamide	C4H7NO2	102.0610	E. festucae	Festuca rubra	Epichloë infected plant material (leaves)	L. fusiformis, M. poae,R. solani	[112]
Gamahonolide AGamahonolide BGamahorin	gamahonolide	C12H2 3C18H28O6C12H14O4	213.1502 341.1952 222.0881	E. typhina	Phleum pratense	Epichloë infected plant material (choke)		[133]
5-hydroxy-4-phenyl-2(5H)-furanone		C12H10O3	177.0546	E. typhina	Phleum pratense	Epichloë infected plant material (choke)		[133]
Trans-p-coumaric acidCis-p-coumaric acidp-hydroxybenzoic acidp-hydroxyphenylacetic acidTyrosol	phenolic acid derivatives	C9H8O3C9H8O3C7H6O3C8H8O3C8H10O2	164.1580 164.1580 139.1220 153.0507168.0980	E. typhina	Phleum pratense	Epichloë infected plant material (choke)	C. phlei	[134]
Epichlicin	Cyclic peptide	C48H74N12O14	3 [M+Na+17]+ = 1082	E. typhina	Phleum pretense	purified mycelia	C. phlei	[135]
Fatty acids	C-18 and C-19 fatty acid	C18H32O3C19H34O3Cl9H36O3Cl9H36O3	297311312312	E. typhina	Phleum pratense	Epichloë infected plant material (choke)	C. herbarum, C. phlei	[115]
Efe-AfpA	protein	55 amino acids	6278 Da	E. festucae	Festuca rubra subsp. rubra	Epichloë infected plant material (tiller)	Sclerotinia homoeocarpa	[105,136]
Cyclosporin T	peptides	C61H109N11O12(11 amino acids)	1188.6 g/mol (2 MW)	E. bromicola	Elymus tangutorum	purified mycelia	Alternaria alternata,Bipolaris sorokiniana,Fusarium avenaceumCurvularia lunata	[136]
ε-poly-L-lysines	peptides	28–34 lysine sub-units		E. festucae	Festuca pulchella	purified mycelia	Drechslera erythrospila Phytophthora capsici	[116]

1m/z when ionised to [M-H2O]−, 2 Molecular weight indicated in g/mol, 3 m/z when ionised to [M+Na+17]+.

Figure 3

Structures of the fungitoxic compounds isolated from sexual Epichloë species that are listed and referenced in Table 3.

6.4. Untargeted Metabolite Annotation for Antifungal Compound Detection

Epichloë endophytes and their host plants are complex and sophisticated multicellular organisms, thus it is not unexpected that the metabolic profile of endophytes in planta is vastly different and more complex than endophytes in vitro [36,37]. In endophyte-infected plants, both the host plant and endophyte metabolome trigger secondary metabolite biosynthesis machinery that would not be otherwise observed individually [117]. It is important to study the metabolic profiles of these endophyte strains both in vitro, in planta and upon infection with diseases to understand the complex biological process involved in endophyte mediated disease resistance (Figure 2d).

With recent whole-genome sequencing strategies revealing that the number of genes encoding the biosynthetic enzymes in various fungi and bacteria are undoubtedly greater than the known secondary metabolites of these microorganisms, it is highly likely that most endophytes might actually express only a subset of their biosynthetic genes under standard in vitro laboratory conditions, such that only a minor portion of their actual biosynthetic potential is harnessed [118,119]. Thus, selection of the most appropriate method for metabolic profiling and isolation of metabolites, as well as consideration of the environmental conditions the endophytes are grown in, is important to identify the most agronomically important endophytes in the ecosystem.

Metabolomics is the study of metabolite profiles of a cell, tissue or an organism under given conditions [120]. Investigation of metabolic profiles of endophytes in response to a pathogen infection is the first step in understanding the mode of action and enabling their use efficiently against pathogens. Metabolome analysis may entail either a targeted analysis of a certain class of metabolite, or total (untargeted) metabolite profiling of a given sample. Preliminary screening of the total metabolome provides a view on overall performance, whereas targeted bioassay-guided isolations provide more specific details about metabolites and their potential uses [98]. It is important to realise that during the isolation procedure, bioactivity can disappear due to degradation of the bioactive compounds in the extract. Furthermore, the activity can be diluted due to inadequate chromatographic separation and poor fractionation, thus reducing the effective concentration [98].

The most widely used techniques for microbial extracts are High Performance Liquid Chromatography (HPLC), Gas Chromatography (GC), Mass Spectrometry (MS) and Nuclear Magnetic Resonance (NMR) [120]. The more advanced methods couple high resolution mass spectrometers to other analytical techniques, such as chromatography or collision-induced fragmentation, to obtain precision and structural information of compounds in a short period of time with less effort, for example LC-MS, MS2 and GC-MS [120,121]. A qualitative, as well as quantitative, understanding of microbial metabolites requires knowledge of both extracellular and intracellular metabolites [121,122,123]. Hence the microbes, as well as the media they are grown on, are analysed for metabolites. To identify these metabolites, it may be necessary to extract the microbial metabolites into a solvent that is suitable for the specific technique. To extract microbial metabolites, many techniques can be used and there is also a range of possible solvents available depending on the polarity of the metabolites of interest [123].

Pinu et al. describes metabolic footprint analysis as the global identification and quantification of the metabolites present in the spent culture medium of microbial cells using different analytical techniques. Both extracellular as well as intracellular metabolites are specific to a time point under a certain set of environmental conditions while the microbes are growing. Capturing these timepoints is important to conduct a detailed descriptive study; thus, quenching (stopping further biological activity) becomes an important step in microbial metabolite analysis. Depending on the aims, the study could be dynamic (gathering data on microbial growth at multiple time points) or time resolved (on a single time point) [124,125,126]. Selection of an appropriate extraction solvent is important because it affects the yield and polarity of metabolites extracted. Thorough metabolic profiling can unravel the potential of a microorganism (biocontrol agent, natural product uses) and complement genomic studies of the organism.

Untargeted metabolomic study approaches have been used to understand the metabolic potential of Epichloë strains relating to antifeeding (insect/invertebrate) activity [127,128,129]. Other studies have investigated the effect of endophytes on the root exudate metabolome [130] and in response to different field nutrient or environmental conditions [128,131]. Untargeted metabolite fingerprinting coupled to spectral data analysis software could be the solution to understanding the complex process of endophyte-host plant-phytopathogen relationships and the metabolic response or production of bioactive metabolites in the presence of a pathogen (Figure 2d). Green et al. (2020) studied the Lolium perenne apoplast to identify Epichloë festucae-derived novel metabolites [73] and recently Fernando et al. (2021) conducted bioassay guided extraction of antifungal metabolites and metabolite annotation of bioactive extracts from Epichloë strains [36].

The analytical tools used to detect metabolites in these studies include ultra-high performance liquid chromatography systems (UHPLC), photodiode array detector, Q Exactive Plus high-resolution mass spectrometer (QE-MS) and high-resolution orthogonal time-of-flight MS (e.g., HRqTOF-MS). There are many data mining software platforms available to obtain a list of metabolites or features including vendor specific software such as MARKERLYNX XS for MASSLYNX v.4.1. (Waters, MA, USA), or vendor neutral software such as Refiner MS and Analyst modules of Genedata Expressionist® (Genedata, Basel, Switzerland). Metabolites can be annotated based on accurate mass information and their identity defined by database searches. Green et al. (2020) used databases (KEGG, BioCyc and in-house databases) (http://www.kegg.jp) (http://biocyc.org) to annotate metabolites and identified a novel amino acid glycoside, a set of Epichloë cyclins, peramine, and putatively identified two peptides, a Kojibiose-related metabolite (322.0643 Da), N-(hydroxypentyl) acetamide (145.1101 Da), and a compound with an accurate mass or m/z of 471.1952 Da [73]. Fernando et al. annotated metabolites from two bioactive strains of asexual Epichloë species and used them as biomarkers for detection in planta. Bioassay guided fractionation, followed by metabolite analysis, identified 61 “known unknown” prospective antifungal metabolites (out of more than 20,000 in planta) that, either singly or in combination, are responsible for the observed bioactivity. The workflow developed in this study allows testing of endophyte bioactivity while ensuring that the metabolite is expressed in planta and so useful for field deployment (Figure 2).

These studies show how metabolite fingerprinting can be used for bioprospecting antifungal metabolites from Epichloë endophytes. While isolation and full chemical characterisation is still necessary to understand the role and mode of action of the bioactive metabolites, these robust novel methods help visualise the potential of novel Epichloë strains as biocontrol agents without conducting field or pot trials. Based on this, we propose a methodology of exhaustive testing and analysis to identify biologically active compounds (Figure 2).

6.5. Qualitative and Quantitative Confirmation of Antifungal Metabolites in Planta

It is a common practice to conduct routine alkaloid testing for the ‘known known’ Epichloë-derived toxic alkaloids before field deployment of novel associations [26,27,34]. Comprehensively characterized and purified compounds can be applied in routine diagnostics for the presence and abundance of Epichloë-derived antifungal metabolites in planta (Figure 2e) [36]. Availability of well tested antifungal compound standards, or availability of isolation methods through scientific publication, enables further investigation of abundance of antifungal compounds in response to environmental conditions, such as disease challenge by phytopathogens, and will confirm their role in improving host plant disease resistance.

7. Future Directions

In the past 25 years, the potential of asexual Epichloë sp. to provide bioprotection from fungal pathogens to the host grass has been noted frequently but not pursued. Wildtype strains have been shown to improve disease control in pot trials and field scenarios, often when outbreaks have occurred rather than by design. There is a strong market globally for novel, animal safe, endophyte strains that provide insect and disease control. Some of these strains exhibit strong antifungal activity against phytopathogens, as observed using in vitro bioassays. With the discovery of novel endophytes with favourable ‘known known’ alkaloid profiles, and significant advances in high-throughput analytical techniques and data analysis, the opportunity now arises to investigate endophyte-mediated disease resistance. This knowledge can then be applied to select superior strains for the pasture and turf industries, improving animal health and host grass performance through enhanced disease resistance. It is anticipated that the outcomes of these studies would expand the current screening methods for Epichloë sp. strains to include bioprotective compounds and their respective genes.