Table of contents


Introduction


Importance of endophytes


Fungal sources as natural products


Plant–endophyte interaction


Specificity in endophytes


Evidence that endophytes are ancestral

a. Fossil evidence

b. Evidence from co-evolution with plants

c. Evidence from evolution of pathogens

d. Evidence from the ability to colonize plants without being rejected

e. Evidence from endophytes producing enzymes that are not needed for an endophytic lifestyle

f. Evidence from appressoria

g. Evidence from high-throughput analyses

h. Evidence from endophytes having many lifestyles

i. Evidence from the large numbers of fungi

j. Evidence from chemical diversity


Conclusion


Acknowledgments


References

Introduction

Fungi are a diverse and heterogeneous group of organisms that can occur as endophytes, epiphytes, pathogens or saprobes (Hyde et al. 2020b). The term ‘endophyte’ was coined by de Bary (1886) which simply means an organism inhabiting living tissue. Many definitions have since been proposed for endophytes (Siegel et al. 1984; Carroll 1986; Petrini 1986); the most commonly used is all organisms, including bacteria inhabiting plant organs that at some time in their life can colonize internal plant tissues without causing apparent harm to the hosts (Petrini 1991). However, it can be difficult to distinguish latent pathogens from endophytes (Hyde and Soytong 2008). Therefore, to verify the endophytic lifestyle, it must be demonstrated that the microorganism can successfully be reintroduced into endophyte-free seedlings, thereby fulfilling Koch’s postulates (Hyde and Soytong 2008; Bhunjun et al. 2021a).

Endophytes are ubiquitous in nature as they have been recovered from plants adapted to deserts, Arctic tundra, temperate and tropical forests, grasslands, and croplands (Arnold 2007; Arnold and Lutzoni 2007). Endophytes are also found in aquatic plants, including algae (Deutsch et al. 2021). Endophytes have been associated with all parts of plants, including stems, roots, petioles, leaf segments, fruit, buds, and seeds (Gouda et al. 2016). Endophytes can be divided into two groups, clavicipitaceous and non-clavicipitaceous, based on phylogeny and life history (Purahong and Hyde 2011). Clavicipitaceous endophytes are mostly systemic, transmitted vertically through seeds, and have a narrow host range; they are especially associated with grasses in mutualistic symbiotic relationships (Purahong and Hyde 2011). Non-clavicipitaceous endophytes have evolved to adapt to broad host ranges found in nonvascular and vascular plants (Purahong and Hyde 2011). Endophytes may exhibit a range of symbiotic relationships with their hosts which provide resilience and a better opportunity for plants to adapt to changes in the environment (Willis 2018; Hatfield et al. 2020). As mutualists, they can provide long-term benefits to hosts, but some endophytes can exhibit a mutualistic interaction for one plant species, but not for another (Hardoim et al. 2015).

Fungal endophytes have been well-studied due to their potential to produce novel compounds with biotechnological, industrial and medicinal properties (Strobel and Daisy 2003; Aly et al. 2011; Hyde et al. 2019; Mapook et al. 2022). This paper highlights the importance of endophytes and provides examples of their secondary metabolites. We discuss factors that influence the specificity in endophytes and critically evaluate the influence of co-evolution. The ability of endophytes to switch lifestyles to become pathogens or saprobes is also discussed. We argue that the endophytic lifestyle is ancestral to fungi. The implication of specificity is discussed in terms of the numbers of endophytic fungi and we argue that studying endophytes for novel compounds has limitations.

Importance of endophytes

Fungi play vital roles in several aspects of human life as food, medicine, decomposers, pathogens and as future environmental-friendly textiles (Hyde et al. 2019, 2020b; Mapook et al. 2022). Endophytes can provide several advantages to host plants and, in return, plants provide spatial structure, nutrients and, in the case of vertical transmission, dissemination to the next generation of hosts (Schulz 2006; Aly et al. 2011). Endophytes are thought to boost host plant growth, fitness and nutrient gain (Redman et al. 1999; Rodriguez and Redman 2008). Endophytes may improve the plant’s ability to tolerate various types of abiotic and biotic stresses (Redman et al. 2001; Waller et al. 2005; Rodriguez et al. 2009). Abiotic stresses include nutrient limitation, drought, salination and extreme pH values and temperatures (Bacon and Hill 1996). Endophytes may also improve the host’s ability to tolerate biotic stresses by altering interactions with insects, pests and pathogens (Oono et al. 2015; Clay et al. 2016). Therefore, endophytes may improve the sustainability of agriculture by reducing the need for fertilisers and pesticides through enhanced nutrient uptake and increased resilience to stresses (De Silva et al. 2019). Decreased water availability and increased soil salinization pose a challenge to crop growth in many parts of the world. The endophytes, Curvularia protuberata and Fusarium culmorum, have been used to confer salt and drought tolerance to rice varieties (Redman et al. 2011). Endophytes also possess metal sequestration or chelation systems that increase host tolerance to the presence of heavy metal, thereby enabling hosts to survive in contaminated soil (Weyens et al. 2009). Therefore, the incorporation of fungal symbionts is an important strategy to mitigate the impacts of climate change on crops and to expand agricultural production to marginal lands (Redman et al. 2011).

Plant–endophyte interaction

Several factors shape plant–endophyte interactions including the mode of transmission, the pattern of infection, plant age, environmental conditions and genetic background (Saikkonen et al. 1998). The propagation and transmission of several endophytes remain to be determined, however, most endophytes propagate horizontally or vertically (Lugtenberg et al. 2016). Most endophytes are transmitted horizontally through airborne spores (Aly et al. 2011). Vertical dissemination is growth within the seed or embryo which is transmitted to the next generation of hosts (Johnson et al. 2013). Endophytes can also be present in a dormant form within the seed or embryo. Some endophytes can propagate both horizontally and vertically, for example, Epichloё species (Aly et al. 2011). Epichloё species form systemic associations with the above ground tissues of grasses and are one of the most economically important examples of plant–endophyte interactions (Johnson et al. 2013). In vertical propagation, asexual Epichloё species grow in the embryo of a developing seed and as the seed germinates, hyphae colonize the seedlings (Philipson and Christey 1986). Vertical propagation can also occur via seed coat which has been reported in Curvularia protuberata, an endophyte that provides high soil temperature tolerance to tropical grass hosts in geothermal habitats (Redman et al. 2002). In horizontal transmission, propagation is dependent on the dispersal of the reproductive structures of the endophyte, such as spores by wind, rain, or a vector, from plant to plant (Lugtenberg et al. 2016). Some non-clavicipitaceous endophytes can colonize plants via appressoria or penetrate plant tissues directly with the hyphae (Chethana et al. 2021a). For example, several Colletotrichum species produce appressoria as endophytes (Ma et al. 2018; Jayawardena et al. 2021). These fungi produce fewer appressoria in healthy plants and sporulate rapidly during host senescence (Rodriguez et al. 2009).

Specificity in endophytes

Host-preference in antagonistic and mutualistic symbioses is likely based on several factors such as ecology and evolution (Arnold 2007; Saikkonen 2007). Host-specificity is based on a close adaptation between the host and fungal species as a result of ancient co-habitation and co-evolution (Aly et al. 2011). This adaptation results in complementary genetic systems that become permanently imprinted in the host and the fungal species (Moricca and Ragazzi 2008). Host-specificity is based on factors, such as host microorganism community, biochemical nature of host, ecology and evolution of the host (Arnold and Lutzoni 2007; Hyde et al. 2007). Endophytes can also exhibit tissue-specificity to adapt to different physiological conditions in hosts (Rodrigues and Samuels 1999). Selective pressures occurring in different plant tissues influence the endophytic community and therefore, the diversity of fungi within a plant can vary in terms of tissue and function (Lugtenberg et al. 2016). For example, the leaves, stems and roots of alfalfa plants are colonized by distinct fungi that produce different ranges of secondary metabolites (Weber and Anke 2006; Porras-Alfaro and Bayman 2011). This is also supported by the comparison of foliar and root endophyte communities in several studies which has little overlap (Suryanarayanan and Vijaykrishna 2001; Kumar and Hyde 2004; Ghimire et al. 2011). Many endophytes in leaves and woody tissues are considered to be host-specific, genus-specific or family-specific (Arnold 2007; Hyde and Soytong 2008). For example, the endophyte Diaporthe apiculata has been isolated only from healthy leaves of Camellia sinensis (Gao et al. 2016). Host-specificity in leaves and woody tissues may depend on several factors, such as initial endophyte colonization as well as substances within these tissues (Paulus et al. 2006; Arnold 2007; Hyde et al. 2007; Hyde and Soytong 2008).

The composition of microorganisms in plants is affected by several factors, such as the genotype of the host, host developmental stage, physiological status of the host, type of host tissue, and environmental conditions (Peršoh 2013; Golinska et al. 2015). Different environmental factors due to occurrence in particular geographic sites can influence the diversity of endophytes in the host (Vega et al. 2010). For example, the incidence, diversity and host breadth of endophytes increases with higher latitudes (Arnold 2007; Arnold and Lutzoni 2007). Studies have shown variability in the composition of endophytes from host plants of a taxon, due to the influence of host physiological conditions, even within homogenous sites (Ghimire et al. 2011; González and Tello 2011). The ability of a fungus to inhabit a host depends on its ability to suppress the immune system of the host and compete with the diverse community of microorganisms present in the plant (Hardoim et al. 2015). Fungi have developed several mechanisms to elude recognition by plant receptors or disturb the signalling cascades of plants’ innate defences (Ryder and Talbot 2015; Anjago et al. 2018). When fungal spores land on the cuticle or the surface wax layer, they may sense the molecules that inhibit their entrance into the hosts. Fungi then release adhesives such as glycoproteins, hydrophobins and chitosan depending on the taxon, plant surface and environmental signals (Newey et al. 2007; Geoghegan and Gurr 2016; Anjago et al. 2018). Adhesives such as hydrophobins also act as a shield, preventing the plant immune system from recognising the spores (Anjago et al. 2018). The spores and various fungal structures are also often surrounded by an extracellular matrix which consists of glycoproteins and lytic enzymes (Chethana et al. 2021a). These enzymes facilitate the breakdown of the cuticle, allowing the absorption of nutrients. During appressorium maturation, fungi synthesise and use melanin to form a thick layer on the inner appressorial wall to combat the plant innate defences (Chethana et al. 2021a). Fungi can also deploy a repertoire of effectors to suppress the host immune system through penetration pegs (Irieda et al. 2014). The length of the endophytic phase may also be involved in host-specificity as it should match the morphological and traits of the host to be able to persist and evolve (Saikkonen et al. 2004).

Endophytes can also confer habitat-specific stress tolerance to hosts, which can affect the composition of the microorganisms as a result of habitat-specific stresses (Rodriguez et al. 2008). For example, endophytes present in some wild plants are often lacking in their domesticated relatives (Lugtenberg et al. 2016). It is likely due to selective pressure, under variable environmental and physiological conditions, to maintain endophytes with beneficial activity for the host (Lugtenberg et al. 2016). Therefore, plants may lose some of the endophytic community when they no longer benefit from their presence. In cases with no selective pressure, endophytes could be less effective due to the loss of specific secondary metabolite production (Lugtenberg et al. 2016). This is supported by studies that have shown that domestication has reduced chemical resistance in hosts against herbivorous insects as compared to their wild relatives (Chen et al. 2015). In another study, rice plants sensitive to salt and drought were inoculated with selected endophytes and grown under continual stress (Redman et al. 2011). These plants maintained 100% colonization levels whereas inoculated plants grown in the absence of stresses showed a decrease in the selected endophyte colonization to 65% (Redman et al. 2011). These studies show that the composition of endophytes can vary due to environmental conditions.

Endophytes enable hosts to better adapt to their environments and become successful competitors (Clay et al. 1993). For example, the endophytes Epichloё coenophiala (= Neotyphodium coenophialum) and E. festucae (= N. lolii) increase allelopathic and the competitive ability of tall fescue (Lolium arundinaceum) and ryegrass (Lolium perenne), respectively (Clay and Holah 1999; Sutherland et al. 1999). This reduces the diversity of the surrounding plant community and these interactions affect ecosystem processes such as soil respiration and decomposition (Hector et al. 2000; Madritch and Hunter 2003; Gartner and Cardon 2004). Therefore, endophytes can affect decomposition and/or alter the diversity of living plants, affecting the soil detritivore and microbial assemblages (Coûteaux et al. 1995; Lemons et al. 2005). Endophytes can alter the metabolism of hosts, causing changes in litter components, for example by the production of alkaloids (Lyons et al. 1990). These changes influence the decomposition rates and can affect invertebrate detritivores, microbial decomposers and the decomposition microenvironment, which can subsequently result in some level of specificity of the microorganisms in these habitats (Omacini et al. 2004; Lemons et al. 2005).

Evidence that endophytes are ancestral

In this review, we provide evidence that fungal endophytes are ancestral to all other life modes and we provide arguments based on the following:

Fossil evidence

The earliest fungi may have evolved about 1000 MYA and lived at the mouth of a river (Loron et al. 2019). Soil surface microbial communities containing fungi and algae were possibly the first terrestrial association between fungi and photosynthetic organisms (Gehrig et al. 1996; Evans and Johansen 1999). Fungi were likely associated with multiple origins of green algae prior to the origin of land plants (Lutzoni et al. 2018). Algae first evolved in the sea and primaeval seas were probably freshwater or only slightly salty (Schilling et al. 1978; Swenson 1983). It is hypothesised that having a symbiotic association with fungi allowed some algae to colonize terrestrial habitats over 500 MYA (Du et al. 2019). It is hypothesised that members of Glomeromycota played a pivotal role in the colonization process, as Glomeromycota members which lived in symbiotic relationships with cyanobacteria or algae eventually were symbionts of early land plants (Schüßler 2002; Lutzoni et al. 2018). In the late Paleozoic (541–251 MYA), the supercontinent drift began, which was followed by the Pangea plate disintegration in the Middle Jurassic (176 to 161 MYA) (Peace et al. 2020). These events led to the formation of continental amalgamation in the early Cretaceous and plants became widespread during this period. Therefore, it is likely that the intimate association between plants and fungi may have facilitated the colonization of land plants (Heckman et al. 2001; Hyde and Soytong 2008) as well as their ability to adapt to different conditions throughout the continents. The earliest record of mutualistic symbiosis was found in the roots of the fossil tree Amyelon radicians from the Paleozoic era (Bacon and Hill 1996). There is limited fossil evidence of Paleozoic fungi, especially endophytic fungi as the identification of fungal endophytes in fossil material is hampered by the inherent difficulty of determining the condition of the host at the time of colonization, i.e. whether it was alive and functional or in the process of senescence or decay (Taylor et al. 2005; Krings et al. 2009). Therefore, Krings et al. (2009) recommended that with fossils, the term fungal endophyte should be understood as strictly descriptive, and should be used for fungi that occur within intact plant cells or tissues without any visible disease symptoms.

Kidston and Lang (1921) described various species of Palaeomyces from early land plants from the Rhynie Chert, that are now regarded as specialized endophytes. Hass et al. (1994) indicated that all fungi described from this locality consist of different chytridiomycete forms (Taylor et al. 1992). Harvey et al. (1969) found a fossil from the Early Devonian Rhynie Chert which appears to be like that of modern Apodachlya pyrifera Zopf. Molecular evidence suggests that the major groups of fungi were already well diversified by the time the first land plants with conducting elements appeared on Earth during the Silurian Period (Heckman et al. 2001; Bhattacharya et al. 2009; Blair 2009; Taylor et al. 2009). Although land plants with true leaves and roots became widespread during the Devonian (Beerling et al. 2001; Raven and Edwards 2001; Gensel 2008), fungal endophytes can be documented from the oldest land plants that lacked a differentiation of the plant body into roots, shoot axes and leaves. Taylor et al. (2004) stated that Early Devonian Rhynie chert contains well-preserved direct evidence for fossil fungi and their interactions with other elements. While there are numerous examples of fungal endophytes associated with the Rhynie chert land plants, most of these are represented by isolated parts or stages of the life cycle. Fungal endophytism in early land plants is the endomycorrhizal symbiosis occurring in Aglaophyton major (Taylor et al. 1995; Remy and Hass 1996; Taylor and Krings 2005). The stomata on the prostrate axes provide entrances for fungal endophytes, including the endomycorrhizal fungus Glomites rhyniensis (Taylor et al. 1995).

Glomeromycota can be traced back to the Early Devonian (Kidston and Lang 1921; Remy et al. 1994; Taylor et al. 1995, 2005) and also in slightly younger Devonian (Stubblefield and Banks 1983) and Carboniferous (Wagner and Taylor 1981, 1982; Krings et al. 2011). The latest Cretaceous (Maastrichtian) records of Glomeromycota include specimens found in the fossilized dung of herbivorous dinosaurs from the Lameta Formation in Central India (Kar et al. 2004; Sharma et al. 2005). Three types of glomeromycotan spores have been described from the Rhynie chert, e.g., i. resembling Glomus Tul. & C. Tul. (Taylor et al. 1995); ii. similar to the extant genus Scutellospora C. Walker & F.E. Sanders (Dotzler et al. 2006); and iii. with germination shield usually tongue-shaped with infolded margins as in the modern genus Acaulospora Gerd. & Trappe (Dotzler et al. 2009). Krings et al. (2012) opined that Glomeromycota were relatively diverse by the Rhynie chert time, and became well-established as a group even before true roots evolved since all of the Rhynie chert plants and many other early land plants at that time lacked roots. The description of Glomites sporocarpoides producing spores in sporocarps from the Rhynie chert, adds further support to the early diversification of Glomeromycota (Karatygin et al. 2006).

Combresomyces cornifer, an important example of an intracellular endophyte, occurring in the Visean lycophyte periderm, is interpreted as Peronosporomycetes based on the presence of specimens displaying oogonia with attached paragynous antheridia (Dotzler et al. 2008). Peronosporomycetes (Oomycota) are believed to be among the oldest eukaryotes on Earth (Pirozynski 1976), but the fossil record of this group remains inconclusive (Johnson et al. 2002). The Rhynie Chert (Early Devonian) has been critical in documenting early land plant–fungal interactions. Krings et al. (2012) studied petrographic thin sections of the Rhynie chert plant Nothia aphylla. There are several records of fossil fungal endophytes that have been documented throughout the Phanerozoic from all over the world. Selected taxa of fossil fungal endophytes are illustrated in Fig. 1 and a summarized account of important fossil endophytic fungi is given in Table 1.

Fig. 1
figure 1

Selected taxa of fossil fungal endophytes 1. Glomites cycestris Carlie J. Phipps & T.N. Taylor 1996; 2. Glomites rhyniensis T.N. Taylor et al. 1995; 3. Gigasporites myriamyces Carlie J. Phipps & T.N. Taylor 1996; 4. Endochaetophora antarctica J.F. White & T.N. Taylor 1988; 5. Milleromyces rhyniensis T.N. Taylor et al. 1992; 6. Paleoblastocladia milleri W. Remy et al. 1994, 6a. Complete mycelium, 6b. Details of fruiting body; 7. Palaeodikaryomyces baueri Dörfelt in Dörfelt & Schäfer 1998; 8. Palaeomycites acinus (S.K. Srivast.) Kalgutkar & Janson. 2000; 9. Palaeomycites bharwainensis (H.P. Singh & R.K. Saxena) Kalgutkar & Janson. 2000; 10. Palaeomycites robustus (R.K. Kar) Kalgutkar & Janson. 2000; 11. Rhizophagus fasciculatus (Thaxt.) C. Walker & A. Schüßler in A. Schüßler & C. Walker 2010; 12. Chlamydospora dichotoma R. Kar et al. 2010, 2011; 13. Palaeogigaspora excellens R. Kar et al. 2020; 14. Palaeomyces horneae Kidst. & W.H. Lang 1921; 15. Palaeomycites minnesotensis (F. Rosend.) Kalgutkar & Janson. 2000. Scale bars: 1–3, 6b, 7, 9 = 20 μm; 4 = 200 μm; 5, 6a = 40 μm; 8 = 30 μm; 10–11, 14–15 = 50 μm; 12–13 = 10 μm

Full size image
Table 1 A summarized account of important fossil endophytes
Full size table

Evidence from co-evolution with plants

Coevolution is the reciprocal adaptation among interacting organisms whereby one species evolves in response to evolutionary changes in another (Thompson and Cunningham 2002). Fungi and plants are intricately linked throughout evolutionary history and it has been hypothesized that plants may have never colonized land without fungi (Pirozynski and Malloch 1975). The lack of plant-defence reactions against the presence of endophytes could occur as a result of co-evolution (Aly et al. 2011). Co-evolution is also demonstrated by the synchronization of fungal and host reproductive systems, as in the case of vertical transmission of fungal spores with the seeds whereby successful host reproduction is beneficial to both partners (Schardl et al. 1991; Moricca and Ragazzi 2008).

Co-evolution events have led to changes in the genome of the host and fungus over several millions of years of generations (Naranjo-Ortiz and Gabaldón 2020). Co-speciation is likely to have congruent phylogenies in the host and fungal species with similar divergence times (De Vienne et al. 2013; Phukhamsakda et al. 2022). For example, Didymellaceae species have been associated with endophytic, pathogenic and saprobic lifestyles (Hongsanan et al. 2020). Several Didymellaceae species are associated with Clematis (Ranunculaceae) and divergence time studies have estimated that Didymellaceae and Clematis species diverged in the Miocene (Xie et al. 2011; Phukhamsakda et al. 2020). The largest monocotyledon family Arecaceae (palms) has over 2600 species belonging to over 200 genera (Dransfield et al. 2008). Apiosporaceae species have been associated with endophytic, pathogenic and saprobic lifestyles (Hyde et al. 2020a; Phukhamsakda et al. 2022). The ancestor lineage of Apiosporaceae evolved around 94 MYA, while Arecaceae diversified around 66–100 MYA. Apiosporaceae species may have co-evolved with Arecaceae which could explain the occurrence of most Apiosporaceae species on Arecaceae. Therefore, these studies suggest that the study of hosts that co-evolved with fungal groups will likely result in a large diversity of novel fungi (Hyde et al. 2020b; Phukhamsakda et al. 2022).

Evidence from evolution of pathogens

Endophytes can switch to pathogens upon host ageing or senescence, thus becoming more widespread (Saikkonen et al. 1998; Hyde and Soytong 2008; Álvarez-Loayza et al. 2011). The transition to a pathogenic lifestyle could be because endophytes become more widespread upon host ageing, thus enabling them to cause visible external infections (Mattoo and Nonzom 2021). A pathogenic lifestyle can be facultative or obligate (Delaye et al. 2013; Bhunjun et al. 2021a). Obligate pathogens require a host to fulfil their life cycle as they rely on living plant cells for nutrients whereas facultative pathogens usually live in the soil and can complete their life cycle without a host (Barna et al. 2012; Kemen and Jones 2012). Fungi with obligate pathogenic lifestyles can rarely return to other lifestyles as they have lost important biosynthetic pathways responsible for the synthesis of important secondary metabolites whereas facultative pathogens can switch lifestyles (Barna et al. 2012; Kemen and Jones 2012). The majority of fungal species have some part of their life cycle either in, or directly associated with the soil environment (Bridge and Spooner 2001). Fungi play a complex role in the soil ecosystem and soil ecosystem harbour a large diversity of fungi (Tedersoo et al. 2021), including taxa present as resting spores or as mycelia (Bridge and Spooner 2001). As a result, fungi isolated from soil samples have been associated with endophytic lifestyles, for example, Apiospora species (Pintos and Alvarado 2021). This suggests that some endophytic fungi can originate from the soil ecosystem, which can result in some level of specificity due to habitat-specific soil ecosystems.

Evidence from the ability to colonize plants without being rejected

Plants can detect signals associated with invaders and respond by activating their immune systems (Zipfel and Oldroyd 2017). Plants can recognise these signals through receptor-like kinases that can detect non-self-molecules called microbe-associated molecular patterns (MAMPs) and respond by activating MAMP-triggered immunity (Boller and Felix 2009; Mattoo and Nonzom 2021). Plant receptors can also recognize effectors produced by invaders and respond by activating effector-triggered immunity (Mendoza-Mendoza et al. 2018). Endophytes can colonize internal plant tissues as they have developed unique colonization strategies which are based on various factors including the host genotype (Hardoim et al. 2015). Endophytes produce several molecules to colonize hosts for immunoevasion and immunosuppression as well as for scavenging nutrients within the host (Mattoo and Nonzom 2021). Endophytes avoid detection by producing evolving variant MAMPs or by disposing burst of reactive oxygen species (Lopez-Gomez et al. 2012). Piriformospora indica produces lectins and small secreted protein (SSP) effectors which may be involved in the evasion and suppression of host defences during the early stages of colonization (Lahrmann and Zuccaro 2012). Endophytes can silence pathways targeted by microRNAs (miRNAs) which play a vital role in plant defence (Plett and Martin 2018). This is remarkable considering that plant genome can encode several hundreds of miRNAs genes (Budak and Akpinar 2015). Endophytes can also secrete low levels of deleterious hydrolytic enzymes, thus avoiding being detected by the hosts’ defence (Mattoo and Nonzom 2021). Therefore, it can be hypothesised that endophyte represents an ancestral state since endophytes can survive within hosts without being rejected by the host defences and without causing any symptoms.

Evidence from endophytes producing enzymes that are not needed for an endophytic lifestyle

Endophytes can produce a wide range of enzymes (Rashmi et al. 2019; El-Gendi et al. 2022). Several studies have shown that endophytes can produce the same leaf-degrading enzymes as their saprobic counterparts (Promputtha et al. 2010). For example, endophytic species of Eurotium, Geniculosporium and Xylaria have the potential to degrade lignin (Bucher et al. 2004; Koide et al. 2005; Purahong and Hyde 2011). Xylaria species also have the potential to degrade cellulose (Pointing et al. 2003, 2005). The ability of endophytes to degrade cellulose and lignin could be vital to allow endophytes to break down host tissue for nutrients and persist as saprobes (Lumyong et al. 2002; Promputtha et al. 2007, 2010; Oses et al. 2008). Several endophytes such as Aspergillus japonicus, Cladosporium cladosporioides and Pestalotiopsis guepinii can produce xylanase which is involved in the degradation of plant matter into usable nutrients (Bezerra et al. 2012). The ability to produce these degrading enzymes that are not needed for an endophytic lifestyle likely suggests endophytic lifestyle to be an equilibrium stage which is activated upon host senescence or host ageing (Kogel et al. 2006; Moricca and Ragazzi 2008). Alternatively, the ability to produce these enzymes suggest that they had a saprobic ancestral lifestyle (Knapp et al. 2018).

Evidence from appressoria

Appressoria are the best-characterized infection structures found in plant pathogens, however, they can also be produced by endophytes (Hyde et al. 2007; Demoor et al. 2019). In endophytes, appressoria grow through plant tissues and are predominantly inter-cellular with little to no impact on host cells (Rodríguez-Gálvez and Mendgen 1995; Talbot 2003). Appressoria formation is important for endophytes to colonise and obtain nutrients from living hosts and appressoria produced by endophytes likely results in a certain level of host-specificity (Zhou and Hyde 2001; Chethana et al. 2021a, b). A list of genera that produce appressoria is given in Table 2 and the majority of these appressoria-producing genera are endophytes. The presence of appressoria in genera with different lifestyles provide strong evidence for the phenomenon of lifestyle switching. For example, Colletotrichum is one of the most important plant pathogenic genera worldwide (Bhunjun et al. 2019; Hyde et al. 2020b; Jayawardena et al. 2021) and it produced appressoria during endophytic, pathogenic and saprobic lifestyles. This could suggest that as endophytes, these Colletotrichum species may be inactive in the healthy plant until triggered to change to saprobe or pathogen.

Table 2 List of appressoria-producing genera with endophytic, saprobic or pathogenic lifestyles. * denotes endophytic interaction capability based on FungalTraits (Põlme et al. 2020)
Full size table

Evidence from high-throughput analyses

High throughput sequencing (HTS) can provide in-depth analysis of fungal communities as it can detect slow-growing species, species with low competitive ability, unculturable communities, recently evolved species and species complexes (Arnold 2007; Nilsson et al. 2019a; Tedersoo et al. 2021). These approaches have revealed an enormous and unprecedented magnitude of fungal diversity, thus providing a more comprehensive idea of fungal communities compared to traditional techniques (Nilsson et al. 2019b; Tedersoo et al. 2020). HTS can yield about 15-fold increase in species richness compared to culture-dependent methods (U’Ren et al. 2014).

The endophytic community recovered from HTS is dominated by Ascomycota (Lucero et al. 2011; Kemler et al. 2013; Peršoh 2013; Dissanayake et al. 2018; Parmar et al. 2018). Botryosphaeriaceae, Mycosphaerellaceae, Nectriaceae, Pleosporaceae and Teratosphaeriaceae were particularly dominant in trees of Eucalyptus grandis (Kemler et al. 2013). However, Teratosphaeriaceae was not recovered from trees of Eucalyptus using culture-dependent methods (Fisher et al. 1993; Smith et al. 1996). Members of Dothideomycetes including taxa of Alternaria, Aureobasidium and Cladosporium were particularly dominant in Atriplex canescens and Atriplex torreyi (Lucero et al. 2011). Dothideomycetes and Leotiomycetes were commonly recovered from asymptomatic foliage of loblolly pine (Pinus taeda) (Arnold et al. 2007). Ascomycota was the most abundant, followed by Basidiomycota, while Chytridiomycota and Zygomycota were found to be rare in HTS of the endophytic community of Dysphania ambrosioides (Parmar et al. 2018). Cladosporium was the most abundant genus, but species of Alternaria, Aureobasidium, Filobasidium, Fusarium and Purpureocillium were also abundant (Parmar et al. 2018). The endophytic community of Quercus macrocarpa was dominated by ascomycetes with Alternaria, Epicoccum, Erysiphe and Microsphaeropsis as the most abundant genera (Jumpponen and Jones 2009). Dissanayake et al. (2018) compared the diversity of endophytes in stems of grapevine using culture-dependent and HTS. All the culturable fungi were ascomycetes, however, HTS recovered Ascomycota, Basidiomycota and Zygomycota (Dissanayake et al. 2018). Dothideomycetes and Eurotiomycetes were dominant among ascomycetous taxa, followed by Sordariomycetes, Leotiomycetes and Pezizomycetes. Cadophora and Cladosporium were frequently detected in Dissanayake et al. (2018). There was an overlap of 53% in fungal genera with the detection of Acremonium, Aspergillus, Botryosphaeria, Botrytis, Cladosporium, Lasiodiplodia and Phoma in both approaches (Dissanayake et al. 2018).

Culture-independent methods have revealed different species diversity compared to culture-dependent methods, yielding several distinct clades not recovered from culture-dependent methods (Arnold et al. 2007; Kemler et al. 2013; U’Ren et al. 2014; Dissanayake et al. 2018). The lower diversity of endophytic fungi recovered from culture-dependent methods could be because HTS can detect fungal species based on a minute amount of DNA (Lindahl et al. 2013). However, in culture-dependent method, these endophytes are likely detected only when they switch lifestyles, forming fruiting bodies or spores when causing diseases. In some studies, several isolated strains were not detected in culture-independent methods (Arnold 2007; U’Ren et al. 2014). Sordariomycetes was commonly isolated from Pinus taeda, but was not detected in HTS (Arnold et al. 2007). Leotiomycetes were not found in HTS data from Pinus leiophylla needles despite being common in cultures and the overlap between culture-based and HTS data was very low (U’Ren et al. 2014). As a result, these two approaches are often considered complementary and together they are more likely to give a more reliable estimate of the true diversity (U’Ren et al. 2014; Dissanayake et al. 2018). A large number of unclassifiable OTUs are commonly recovered from HTS, which cannot be assigned even to phylum level (Jumpponen and Jones 2009; Lucero et al. 2011; Kemler et al. 2013; Peršoh 2013; U’Ren et al. 2014) and could therefore represent novel lineages (Nilsson et al. 2019a).

The application of FUNGuild (Nguyen et al. 2016) showed that several endophytes recovered using HTS were classified as potential saprobes or pathogens (Dissanayake et al. 2018). The endophytes Botryosphaeria, Curvularia, Lasiodiplodia and Penicillium were classified as potential pathogens while Acremonium, Aspergillus, Eupenicillium, Kernia, Lophiostoma, Penicillium, Phialosimplex and Pyrenochaeta were classified as potential saprobes (Dissanayake et al. 2018). This possibly suggests that these species remain quiescent as endophytes, but they become activated on receiving a specific signal, which allows them to complete their life cycle (Peršoh 2013). Guild assignment of the foliar communities of invasive pines showed that several species were classified as potential saprobes or pathogens (Steel et al. 2022). Species of Aspergillus, Biscogniauxia, Parastagonospora, Penicillium, Phaeotrichum, Preussia, Sydowia and Xylaria were classified as potential saprobes while Austroafricana, Didymella, Lophodermium, Nothophaeocryptopus, Phaeomoniella, Pyrenophora and Ramularia were classified as potential pathogens (Steel et al. 2022). Alternaria, Coniochaeta, Phaeomoniellales and Xylariaceae species were classified as endophytes, saprobes and pathogens (Steel et al. 2022). This contributes to the hypothesis that the endophytic community are involved in the decomposition process and as pathogens. However, guild assignment is based on accurate identification of OTUs, which is problematic due to the presence of misidentified species on public databases such as GenBank and the lack of reference sequences (Nilsson et al. 2006; Bhunjun et al. 2020). Using a similarity threshold for species identification is problematic as accurate species identification in most genera relies on multi-gene phylogenetic analyses. Another limitation is that these ecological conclusions are based on data of 10% of the expected diversity of fungi based on the relatively conserved estimate of 1.5 million (Hyde et al. 2020b; Bhunjun et al. 2022).

Evidence from endophytes having many lifestyles

Many fungi can switch lifestyles which can represent evolutionary transitions (Rodriguez and Redman 2008). Alternatively, this can indicate that these fungi have achieved significant ecological plasticity, thus ensuring optimal growth and reproduction in a variety of hosts, which ultimately results in the expansion of their bio-geographic distribution (Rodriguez and Redman 2008). Some endophytes can switch to pathogens due to a host shift. For example, Colletotrichum tropicale which is a common endophyte in Panamanian trees, has been isolated as a pathogen from the leaves of Persea americana and fruits of Annona muricata (Rojas et al. 2010). Endophytes can switch to pathogens due to an imbalance in nutrient exchange or environmental variations (Kogel et al. 2006; Moricca and Ragazzi 2008). For example, the endophyte Botryosphaeria stevensii (Diplodia mutila) is associated with mature plants of Iriartea deltoidea, which usually occur under shady conditions (Álvarez-Loayza et al. 2011). Álvarez-Loayza et al. (2011) found that low light supports the endophytic state whereas high light triggers the pathogenic state, causing cell death and tissue necrosis. Endophytes can also switch lifestyles due to mutations (Redman et al. 1999). For example, disruption of a single genetic locus in pathogenic isolates of Colletotrichum magnum can result in the isolate switching to an endophytic lifestyle (Redman et al. 1999; Rodriguez and Redman 2008).

Morphology-based studies have shown that the proportion of endophytes ranges from 0 to 92% frequency in decomposer communities (Osono 2006; Peršoh 2013, 2015; Guerreiro et al. 2018; Jayawardena et al. 2018). Most colonizers in the early stages of decomposition, and many saprobes on leaf litter, may have originated from endophytes (Osono et al. 2004; Koide et al. 2005; Osono 2006; Promputtha et al. 2007; Peršoh 2013). By switching to saprobes, endophytes become part of the decomposer community which can increase the diversity of saprobes and thus accelerate decomposition rates (Purahong and Hyde 2011). Phoma medicaginis, which is a dominant endophyte of Medicago lupulina and Medicago sativa, accelerates growth and sporulation after host death (Weber et al. 2004). Phoma medicaginis also produces brefeldin A, an antifungal metabolite, which facilitates its establishment as well as the transition from an endophytic to saprotrophic lifestyle (Weber et al. 2004). Several studies have reported the ability of endophytes to switch to saprobes after the hosts senescence in both grass and non-grass species (Osono 2006; Hyde et al. 2007; Promputtha et al. 2007; Purahong and Hyde 2011). This could be due to modification of host tissue during senescence, allowing mycelium to penetrate the epidermis and colonize the host surface (Dickinson 1976). This could also be because once inside the host tissue, endophytes assume a quiescent state either for the whole lifetime of the host or until environmental conditions are favourable for the fungus or the ontogenetic state of the host changes to the advantage of the fungus (Sieber 2007). This suggests that the evolved ability to withstand host defences is responsible for the different taxa that occur as saprobes on different plants.

The most frequently reported endophytes usually belong to groups composed of saprobes or pathogens. Several genera have been associated with all lifestyles based on FungalTraits (Table 3) and several of these genera produce appressoria (Table 2). However, there is a lack of evidence for different lifestyles in all species which could be because HTS data are not currently accommodated in existing fungal classification systems. For example, Colletotrichum artocarpicola was isolated as a pathogen using culture-dependent method (Bhunjun et al. 2019), but it also has close relatives to endophytes (> 99% alignment score: UDB0754027, UDB0750279) and saprobes from soil samples (≥ 90% alignment score: UDB0413257, UDB0579972) based on a BLASTn search in UNITE. Neocatenulostroma castaneae was isolated as a saprobe and it produced appressorial pegs at the tips or in between the hypha cells in the axenic culture, which suggests an endophytic or pathogenic life stage (Phukhamsakda et al. 2022). This is supported by data in UNITE which shows that Neocatenulostroma castaneae has close relatives to endophytes (> 90% alignment score: UDB0747801) based on a BLASTn search. The most abundant endophytic OTUs recovered from Peršoh (2013) were close relatives (≥ 90% alignment score) to saprobes from soil and/or dead plant materials (Nilsson et al. 2019b), therefore possibly representing latent decomposers. Several endophytic taxa have close relatives to saprobes based on HTS data from environmental samples. For example, Acremonium sclerotigenum (KM215633), Aspergillus iizukae (AB859956), Cladosporium colombiae (KM215637) and Fusarium fujikuroi (KM215642) were isolated as endophyte from Silybum marianum (Raja et al. 2015) and these taxa have close relatives to saprobes from soil samples (≥ 90% alignment score) based on BLASTn searches in UNITE (Nilsson et al. 2019b). Several endophytes can degrade cellulose and lignin (Promputtha et al. 2010), which contributes to the hypothesis that endophytes can switch to saprobes. Several studies have shown that many endophytes are from important pathogenic genera such as Alternaria, Colletotrichum, Fusarium, Penicillium and Phoma (Ganley et al. 2004; Promputtha et al. 2007; Dissanayake et al. 2018; Rashmi et al. 2019). Appressoria formation has also been reported in several of these genera such as Cladosporium, Colletotrichum and Fusarium species as pathogens and saprobes (Table 1). Alternaria alternata, Bipolaris sorokiniana and Cladosporium herbarum are common endophytes identified from wheat leaves based on morphology, but these species are also common wheat pathogens (Larran et al. 2002). Several endophytes isolated from western white pine (Pinus monticola) were from Rhytismataceae and Mycosphaerellaceae based on the ITS region, which are important pathogens (Ganley et al. 2004). However, the isolated endophytes were not closely related to the three pathogenic species associated with Pinus monticola (Ganley et al. 2004), which could be due to the lack of sampling. Several Rhytismataceae (for example AY465451, AY465440, AY465473) and Mycosphaerellaceae endophytes (for example AY465456, AY465457) from Ganley et al. (2004) have close relatives to saprobes from soil based on BLASTn searches in UNITE (> 90% alignment score). This shows that the majority of endophytes have close relatives to pathogens and saprobes.

Table 3 List of 40 genera with all lifestyles from FungalTraits (Põlme et al. 2020)
Full size table

Evidence from the large numbers of fungi

Fungal diversity ranges from the conservative estimate of 1.5 million species up to 12 million based on metabarcoding approaches (Hawksworth 1991; Wu et al. 2019; Baldrian et al. 2021), but only about 150,000 species have been named and classified to date (Hyde et al. 2020b; Phukhamsakda et al. 2022). The estimation of species number is vital for systematics, resources and classification (Hawksworth 1991). The correct estimation and description of new species are important as undescribed fungi may produce novel metabolites with biotechnological, industrial and pharmaceutical applications (Hyde et al. 2019). To conserve species it is first necessary to discover and recognise them (Bhunjun et al. 2022). Vascular plants can host ten to 100 different endophytic species, with two to five being host-specific (Dreyfuss and Chapela 1994). Therefore, based on the ratio of host to species of 1:4, about 1 million endophytic species were estimated based on the presence of about 250,000 vascular plants (Dreyfuss and Chapela 1994; Ganley et al. 2004). The estimate of one million endophytic species still accounts for 1 in 14 species based on the estimated 14 million species on Earth (Purvis and Hector 2000). This magnitude of diversity signifies the importance of endophytes as crucial components of fungal biodiversity (Lugtenberg et al. 2016). Such a large diversity is also important, considering that many plants are unable to endure abiotic and biotic stresses in the absence of endophytes (Redman et al. 2002).

The colonization rate of endophytes varies among ecosystems, ranging from 1 to 44% in arctic and boreal ecosystems up to more than 90% in tropical ecosystems (Higgins et al. 2007). A single tropical leaf can harbour over 90 endophytic species and 50 distinct genera in grassland species (Bayman 2006; Porras-Alfaro et al. 2008). The high diversity of endophytes in tropical forests is attributed to the large diversity of hosts (Hyde and Soytong 2008). Most endophyte studies have been on angiosperms, followed by gymnosperms (Rashmi et al. 2019). The diversity of fungi on many hosts remains to be studied as there are about 2,100 plant species described every year (Bhunjun et al. 2022). Most endophyte studies have been reported from Asia, which is also the continent with the highest fungal species discovery in the last few years (Bhunjun et al. 2022). Therefore, extensive studies of other continents are likely to yield a large number of novel endophytes. Some endophytes show some level of specificity whereas some are ubiquitous. The ubiquitous nature of endophytes could be due to adaptation to stresses and environmental conditions, thus leading to the occurrence of generalist endophytes (Govinda Rajulu et al. 2013; Suryanarayanan 2013). Most studies have focused on leaf tissues as they harbour a rich diversity of fungi and they are relatively easy to handle compared to other tissues (Rashmi et al. 2019). However, there is little overlap in endophyte species between different host tissues, therefore a large diversity of distinct host and tissue-specific endophytes is likely to be found (Rashmi et al. 2019). We hypothesise that the total diversity of endophytes is currently underestimated at 1 million if the endophytic lifestyle is common in all fungi. There are about 450,000 plant species (Pimm and Joppa 2015), which indicates about 1.6 million fungal species based on the ratio of host to species (1:4). Insect fungi are estimated at 1.5 million (Hywel-Jones 1993), marine fungi about 12,500 species (Jones 2011), freshwater fungi about 4,100 species (Jones et al. 2014), lichens and lichenicolous fungi about 20,000 species (Feuerer and Hawksworth 2007). Therefore, the diversity of endophytes is estimated at 3.1 million species.

Evidence from chemical diversity

Endophytes show great potential as a major source of biologically active compounds with biotechnological, medicinal or agricultural applications (Aly et al. 2011; Hyde et al. 2019). Natural products are vital in the discovery and development of drugs due to their structural ability to regulate the body’s defence and compete with pathogens (Atanasov et al. 2021). The study of endophytes has led to the discovery of thousands of metabolites, which can be attributed to the large diversity of endophytes and their ability to quickly grow on artificial media (Gouda et al. 2016; Rashmi et al. 2019). Endophytes produce several novel natural compounds with unique chemical structures that may have been optimized by evolution for biological and ecological relevance (Gunatilaka 2006; Rashmi et al. 2019). It is hypothesised that these natural products play a vital role in communication or adaptation as a response to habitat and environmental changes (Gunatilaka 2006; Kuhnert et al. 2021). Secondary metabolite production can also be triggered by the limitation of food sources, competition with other organisms and the presence of other fungi and pathogens (Cueto et al. 2001; Ho et al. 2003; Glauser et al. 2009). Several of these secondary metabolites have antimicrobial, antiparasitic, cytotoxic, neuroprotective, antioxidant or immunosuppressant properties (Gunatilaka 2006; Zhang et al. 2006; Aly et al. 2011; Hyde et al. 2019). For example, enfumafungin, the first pharmaceutical drug from an endophyte (Hormonema carpetanum) shows antifungal activity against Candida and Aspergillus (Pelaez et al. 2000). Plants use the chemical camptothecin to defend against insects and pathogens (Kusari et al. 2013). The camptothecin-producing endophyte Fusarium solani was isolated from the inner bark tissues of the plant Camptotheca acuminata which produces camptothecin. The presence of specific amino acid residue alterations in Fusarium solani ensured protection against its own and the plant camptothecin (Kusari et al. 2011). This suggests that the ability to colonize plants without being rejected is due to these metabolites which allow endophytes to withstand host defences.

Conclusion

Fungi-plant relations can be regarded as a flexible interaction, whose directionality is determined by slight differences in fungal gene expression in response to the host, or by host recognition and response to the fungi (Aly et al. 2011). Therefore, genetic differences in the genome of fungi and plants control the outcome (positive or negative) of the symbiosis (Moricca and Ragazzi 2008). For example, Moniliophthora perniciosa is responsible for witches’-broom disease of Theobroma cacao, but it was also isolated as an endophyte from the same host (Lana et al. 2011). There was no genetic difference between the endophytic and pathogenic isolates of M. perniciosa based on RAPD analysis and enzyme production assay (Lana et al. 2011). It is hypothesised that the pathogenic state is a result of miscommunication between fungi and host (Osono 2006; Hyde et al. 2007; Promputtha et al. 2007; Rodriguez and Redman 2008). This is supported by studies whereby different tomato cultivars were inoculated with Colletotrichum magnum, which expressed a mutualistic lifestyle in one cultivar and a pathogenic lifestyle in another (Rodriguez and Redman 2008). Mutualistic interactions between fungi and host can therefore be considered as a balanced state, which is controlled by environmental, physiological and genetic factors (Kogel et al. 2006). The endophytic lifestyle is likely ancestral as endophytes can colonise plant tissues without causing any symptoms and they can easily switch to saprobes or pathogens by activating additional genes. When endophytes switch to saprobes, they activate genes such as carbohydrate-hydrolyzing enzymes, proteases and transporters to acquire nutrients (Promputtha et al. 2007, 2010). When endophytes switch to pathogens, they activate additional genes such as effectors, carbohydrate-hydrolyzing enzymes, transposons, proteases, transporters, siderophores and toxins (Aylward et al. 2017). Appressorium formation also supports endophytic lifestyle to be ancestral. For example, appressorium formation has been observed in several genera as saprobes (Table 2) despite not requiring living hosts for their nutritional requirements (Brun et al. 2009; Phukhamsakda et al. 2022). This raises the question “why do saprobes need appressoria unless they are endophytes?”. There is further evidence based on the ability of endophytes to produce enzymes that are not needed for an endophytic lifestyle and based on ancestral character state reconstruction. Xylariomycetidae species have variable stromatic characters and they are endophytes, pathogens and saprobes (Rashmi et al. 2019; Hyde et al. 2020a). The ancestral character of Xylariomycetidae ascomata was determined as being immersed (Samarakoon et al. 2022), which suggests endophytes as being ancestral. Therefore, there is strong evidence that suggests that endophytes are ancestral.

Culture-dependent methods usually result in a low diversity of endophytes as it can result in bias towards fast-growing and ubiquitous species such as Colletotrichum, Diaporthe, Pestalotiopsis and Phyllosticta (Doilom et al. 2017; Dissanayake et al. 2018; Tibpromma et al. 2020; De Silva et al. 2021). The low diversity could also be due to limitations of using the surface sterilization method, which can influence the diversity of endophytes (Promputtha et al. 2007). Culture-dependent methods rely on the ability of the fungi to grow on artificial media and usually result in taxa from about 16 genera, with Diaporthe species being extremely common (Ganley et al. 2004; Rashmi et al. 2019; Tibpromma et al. 2020; De Silva et al. 2021). Detection and identification of microorganisms can also be affected by the isolation method, selection of media, size of host tissue fragments and the time since sample collection (Arnold 2007; Sun et al. 2011; Bhunjun et al. 2021b). This is supported by HTS which can recover a higher diversity of fungi compared to culture-dependent methods from the same substrates (Arnold et al. 2007; Kemler et al. 2013; U’Ren et al. 2014; Dissanayake et al. 2018). The diversity of pathogenic and saprobic taxa from culture-dependent methods is usually also higher compared to the diversity of endophytes (Larran et al. 2002; Jayawardena et al. 2018; Rashmi et al. 2019; Phukhamsakda et al. 2020). As a result, the chemical diversity recovered from endophytes is lower compared to that of saprobes as the genera are limited. The low chemical diversity could also be because the majority of studies have focused on endophytes compared to saprobes (Schulz et al. 2002; Rashmi et al. 2019). Therefore, if saprobes were instead chosen, the species diversity and chemical diversity would be much higher. However, improvement in the methods of isolating and culturing endophytes will result in a wider diversity, thus avoiding isolating the same genera/species.

Fungal ancestors likely had an endophytic lifestyle for easy access to hosts and switched to saprobes or pathogens. This is supported by the hypothesis that pathogenesis likely evolved in the later stages of the evolutionary process (Burdon and Laine 2019). However, character state reconstruction of Dothideomycetes estimated saprotrophic or the rock-inhabiting lifestyle to be ancestral based on genome data (Ametrano et al. 2019). Several genera have only been associated with an endophytic lifestyle, such as Bifusisporella, Coccogloeum, Creodiplodina, Dictyoarthrinium and Oblongocollomyces (Põlme et al. 2020) which suggests that these fungi die with their hosts. However, several genera that have been identified and classified as saprobes or pathogens have not been isolated as an endophyte, for example, Delphinella, Kabatina and Phaeocryptopus (Põlme et al. 2020). This could be due to having a dominant phase among the lifestyles which is better suited to adapt to the host and environmental factors. This could also be because several species have probably lost their ability to be endophytes, for example, insect associated fungi (Triolet et al. 2022).