Comparative Genomics of Nectriaceae, including Freshwater Fungi Show Environment Adaptations and Conservation Strate-Gies for Fungi

1. Introduction

The Nectriaceae family was introduced in 1865 to accommodate the hypocrealean species having ascomata that are generally yellow, orange-red to purple and usually change colour in potassium hydroxide and lactic acid [1,2]. Nectrioid fungi were mainly characterized by slightly bright-colored and generally superficial perithecia containing ascospores with articulated cells, and their various phialidic anamorph. Some genera affiliated with Nectriaceae have straight asci and conidia, while those of other genera are strongly curved [1].

The reexamination of nectrioid fungi has been carried out continuously since 1950 by several taxonomists [3,4,5,6,7,8,9,10,11,12].

The genus Neonectria is characterized by well-developed or minute stromata, subglobose to broadly pyriform and gregarious perithecia that are laterally or not collapsing when dry, 2- or 3-layered perithecial wall, smooth to warted perithecial surface, cylindrical to clavate asci with or without an apical ring, and uni- or multiseptate ascospores with spinulose or striate surface [13].

Nectriaceae is one of the largest families of Hypocreales (Sordariomycetes, Ascomycota) [14], with 167 genera described in Mycobank (09.02.2024, Mycobank). They are frequently found on both living and decaying woody materials, soil, fruiting bodies of other fungi, and insects and occur on various substrates in tropical and subtropical regions around the world [8]. Several of them are documented as endophytes, or opportunistic plant-pathogens, pathogens of crops and humans [15]. They have potential to degrade resistant plant material [16], and are used in industrial and commercial applications [17].

The genera Fusarium and Neonectria are the most studied taxa in Nectriaceae due to their relevant economic importance as plant-pathogen. The genus Fusarium is one of the largest genera in the family Nectriaceae, and more than 1,000 species have been reported in 2022 [14]. Fusarium was included in the top 10 globally most important genera of plant pathogenic fungi based on scientific and economic importance [18]. The genus Neonectria has 57 species reported in Mycobank (09.02.2024, Mycobank). Neonectria species are known for infection of the branch and trunk which causes lesions known as cankers, they are normally associated with beech and fruit trees bark canker [19]. The most known Neonectria plant-pathogens are N. coccinea (European beech bark disease); N. faginata (American beech bark disease); N. ditissima (hardwood canker disease) [20,21].

Neonectria is also a common genus of endophytic fungi [19,22]. Fungal endophytes of plants are widespread and important for host plant health [23]. They spend all or part of their life, residing in healthy plant tissue, often in non-pathogenic mutualistic relationships, often protecting the plant against biotic and abiotic stress and promoting plant growth [22,24,25]. Their presence and abundance are often associated with the development stage of the plant, season, and environment [25,26]. When in association with the roots, fungal endophytes play important roles in ecosystem processes and nutrient cycling, with beneficial symbiotic relationships with many plants [27,28]. Endophytic fungi can be latent pathogens [29,30], mutualists, for example mycorrhizal fungi [31], and/or saprobes [32,33], but should be detected within the tissues of healthy host plants [34,35].

The Nectriaceae family also includes fungi with saprobe capabilities, in soil and freshwater [17,36]. The Freshwater fungi are believed to be originated from soil fungi approximately 1.5 billion years ago due to a similar microbial machinery found in both groups and evolutionary studies [37,38]. In soil, Nectriaceae saprobes have a cosmopolitan distribution around the world and colonize newly dead, organic plant material [39]. Freshwater fungi are characterized as having partial or complete lifecycle in freshwater environments [40], where they act primarily in the decomposition of leaf litter, particularly aquatic hyphomycetes [41]. The aquatic hyphomycetes are characterized by tetraradiate, branched, and sigmoid conidial shapes that facilitate the attachment on the surface of aquatic leaf litter [42,43]. However, in aquatic Nectriaceae the typical conidial shape can be clove--shaped with or without minute outgrows [43,44]. Freshwater ascomycetes possess distinctive adaptations that enable their survival in freshwater environments for example their unique lignocellulose enzymes that work by softening the leaves and wood, which is particularly fundamental for breaking down wood underwater [37].

According to many studies, in order to adapt to colder environments, certain freshwater fungi have developed unique mechanisms, such as cold-active enzymes and antifreeze proteins [45,46,47,48,49,50], these characteristics help the freshwater fungi to survive in extreme environmental conditions like arctic and subarctic streams and as well in freshwater streams. However, the escalating temperatures caused by global warming have emphasized the need to comprehend and forecast the reaction of microbial life and ecosystems to severe drought and warm weather occurrences. These events are one of the major environmental stresses experienced by microorganisms such as fungi [51] and can affect the activity of their enzymes and specialized metabolites. For example, the production of specialized metabolites in lichens is affected by light, UV radiation, altitude, temperature fluctuations, and seasonality [52,53,54]. In addition, global warming may have significant consequences in human health, crop production, and the well-being of forests and other habitats [53]. Therefore, research mechanisms to identify temperature adaptations and the possible impacts of climate change on fungal species have been implemented not only to predict ecosystem impact but also to target conservation goals [47,55,56].

In this manner, we present the first genome of the saprobe freshwater fungus Neonectria lugdunensis and an up-to-date phylogenomics tree of the Nectriaceae family observing the divergences in genome traits according to fungal lifestyle. We emphasized their machinery for the breakdown of plant-litter and their enzyme capability to resist and be active in colder and warmer temperatures, which is an important mechanism to identify not only the environment adaptation for temperature fluctuations but also may indicate possible target species for conservation goals in times of climate change-related impacts.

2. Materials and Methods

The Neonectria lugdunensis strain was obtained from the internal collection of DSMZ (Leibniz Institute DSMZ, German Collection of Microorganisms and Cell Cultures) with the assigned identifier DSM 113088. Originally this strain was derived from ascospores of the perithecial morph on a twig in Slovakia (GenBank accession number of ITS- sequence DQ247777) by Ludmila Marvanová and deposited in the Czech Collection of Microorganisms under CCM F-13783. The fungus was cultivated on agar plates containing 2% malt extract (Feelwell, Germany) using Oxoid brand agar (Germany) at a temperature of 16°C within a designated cooling facility for DNA extraction. To confirm the taxonomic assignation of the strain cultures, we employed Sanger sequencing of the Beta Tubulin marker gene. This was accomplished using the pair primers T1 (5′-AAC-ATG-CGT-GAG-ATT-GTA-AGT-3′) and T2 (5′-TAG-TGA-CCC-TTG-GCC-CAG-TTG-3′). The obtained sequences were manually curated using the Sequencher tool v5.4.6 (http://www.genecodes.com). Subsequently, a BLAST analysis was conducted against the National Center for Biotechnology Information (NCBI) nucleotide database (https://www.ncbi.nlm.nih.gov/).

After taxonomy confirmation of the strain, the biomass was transferred to 1 litter flask each with potato-glucose liquid medium (Carl Roth, Germany), and transferred to a 16°C cooling room over a shaker (Bottmingen, Switzerland) at 120 RPM. The samples were kept in a shaker until enough biomass (14 grams) for genome extraction (see Supplementary File). The genomic DNA with fragment sizes greater than fifteen kilobases, verified with an Agilent Femto Pulse (Agilent, United States), were sent to Hifi genome sequencing at Macrogen, Netherlands. The raw sequence files were obtained from Macrogen, using Flye v2.9 [57], and the quality was accessed with Busco v5.2.2 [58] using the ascomycota_odb10 database.

For the comparative genomics of the Nectriaceae family, we have obtained 17 genomes of Nectriaceae from the NCBI database (see Table 1). We aim to select representative genomes of each genus at random but taking into consideration the lifestyle of each fungal species, and we added all genomes available for the Neonectria genus. We also added 2 outgroup genomes for phylogenomics analysis belonging to the Ophiocordycipitaceae family, phylogenetically closely related to Nectriaceae [37], and the newly assembled genome of Neonectria lugdunensis. The genomes were verified according to their genome completeness to assure values higher than 95%, according to the results obtained with BUSCO and the ascomycota_odb10 database.

An all-vs-all Genome-to-genome alignment and comparison analysis was performed using the DNADIFF program from MUMMER3 [59]. All genomes were aligned and compared against each other to obtain the average nucleotide identity (ANI) and genomes with similarity higher than 95% were considered of the same species [60]

The integrative method of RepeatModeler2 [61] was employed to generate a comprehensive denovo species-specific repeat library for all genomes. The identification of new repeats by RepeatModeler v2.05 resulted in a library of consensus sequences for each species combined with repeat annotation carried out using RepeatMasker v4.1.5 [62], via sequence comparison against the Dfam library, generating transposable elements results, while soft masking the genome. The soft-masked genome was used to predict the tRNA sequences determined by tRNAscan-SE 2.0.12 [63] with default parameters for eukaryotic organisms.

The soft-masked genomes from NCBI (17 Nectriaceae and 2 outgroup taxa) together with the newly assembled genome of Neonectria lugdunensis, were then submitted to the Braker3 v3.0.3 [64] pipeline to perform ab initio gene prediction with parameters “--esmode” and “--fungus”, using therefore, the software GeneMarker-ES v4.71_lic [65], to produce denovo hints to train Augustus v3.5.0 [66] which after that, predicted the aminoacid sequences of each genome.

The produced aminoacid sequences were used in Orthofinder v3.0.0 [67]with the “-og” option to predict only the Single-Orthologs common to all species of this study, generating then sequence clusters. Each cluster was aligned using MAFFT v7.520 [68] with the option “--auto” and trimmed using Trimal v1.4.rev15 [69] with the option “-automated1” to remove poor alignment regions. The phylogenetic clusters were than merged into a unique phylogenetic tree using IQTREE2 v2.2.2.7 [70] with the options –bb 1000 -m TEST. The -m TEST parameter stands for ModelFinder [71], which searches for the best evolutive model for each gene cluster, respecting the evolutionary specificities of each gene.

To investigate the genetic potential of the Nectriaceae fungi for secondary metabolite production, the AntiSMASH fungal cluster predictor (v7.0.1) [72] was implemented. The presence of genes and clusters associated with the biosynthesis of secondary metabolites was assessed in every genome. Fasta files containing genome sequences were utilized as inputs, with the default search parameter set to "relaxed." Every additional feature was enabled, including Cluster-border prediction by utilizing transcription factor binding sites (CASSIS).

The predicted aminoacid sequences from the Braker3 pipeline were used for secretome prediction using a combination of tools. To predict where the provided protein sequence possesses a signal peptide, at its N-terminus, that directs the protein towards secretion, we use the consensus sequences obtained from the tools Signalp 6.0 [73] and Targetp2 [74]. In addition, the sequences were also filtered out using DeepTMHMM [75]to remove transmembrane domain signals.

Secreted, not secreted and effector proteins were then annotated for Carbohydrate-active enzyme (CAZymes) with run_dbCAN4 (https://github.com/linnabrown/run_dbcan), a standard alone tool of dbCAN3 web server (https://bcb.unl.edu/dbCAN2/), using the tools HMMER v3.3.2 [76], DIAMOND v2.1.8 [77], and dbCAN_sub with the dbCAN3 [78] database v12. The analysis was performed against all 6 available classes of carbohydrate-active enzymes (CAZys): carbohydrate-binding module (CBM), glycoside hydrolases (GHs), polysaccharide lyases (PLs), auxiliary activities (AAs), carbohydrate esterases (CEs), and glycosyl transferases (GTs). Only sequences that were identified as CAZy enzymes with all 2 or more tools (HMMER, DIAMOND, and dbCAN_sub) were considered for the results, as recommended by dbcan3.

Putative secreted effector proteins were identified from the secreted proteins pool using EffectorP v3.0 [79] where proteins were predicted to have or not interact with plants at the apoplast and cytoplasmic level, through the tool, we can observe the secreted effector proteins associated with plant infection by Fungi including the main local of infection of the protein.

To better understand the adaption of the freshwater hyphomycetes to the environment, proteins with psychrophilic and thermophilic characteristics were predicted using the machine learning tool ThermoProt [80], for the secretome, and effector proteins.

3. Results

The newly assembled genome of Neonectria lugdunensis, has 44.78 Mbp and 97.6% genome completeness with N50 of 44.7 Mbp, showing similar characteristics from many genomes available in NCBI from Nectriaceae as shown in this study (Table 1). We detected 4.38% of repetitive elements in N. lugdunensis, 2.04% were retroelements, and 0.33% were DNA transposons. Only 207 sequences of tRNA were identified in N. lugdunensis, which is the smallest amount in comparison with the freshwater saprobe fungi Aquanectria penicillioides (222) and the soil saprobe Thelonectria discophora (225).

The N. lugdunensis has fewer predicted genes (11480) than the freshwater saprobe Aquanectria penicillioides with 12575 although higher than the soil saprobe Thelonectria discophora with 10364 genes. This result is also reflected in the number of secreted proteins of N. lugdunensis being smaller (797) than freshwater A. penicillioides (820) and higher than the soil saprobe Thelonectria discophora (655) and also reflected the number of effector genes (229 in N. lugdunensis, 254 in A. penicillioides, and 191 in soil saprobe T. discophora) (Table S1). Regarding BGCs, N. lugdunensis, had higher amounts among saprobe fungi with 48 predicted clusters, followed by the other freshwater A. penicillioides with 40, and the soil saprobe T. discophora with 33. (Table S2). N. lugdunensis was the only freshwater fungi in this study to have a BGC for NRP-Metallophore.

The species Ilyonectria robusta (current name according to Mycobank: Ramularia robusta) is known to be a plant pathogen and endophyte. However, the genome strain was isolated as an endophyte (https://mycocosm.jgi.doe.gov/Ilyrob1/Ilyrob1.home.html) and therefore presented in this study as such (Table S1)

We obtained 2684 Single-copy ortholog genes yielded a robust, well-resolved, and comprehensive phylogeny of the Nectriaceae family (Figure 1). The internodes in the tree received a strong bootstrap value higher than 95%, characterizing a strongly supported tree and not particularly associated with the lifestyle of the fungi.

The species Neonectria galligena is nowadays a known synonym of Neonectria ditissima and has been described as being the same species since 1995 [8] although, in the NCBI database, both names are represented with different genomes. The genome-to-genome analysis through ANI confirmed that they are the same species (ANI value of 98.91%), but the enzymatic profile suggests being from different strains, presenting different phenotypes, and therefore the synonymy was kept in the study. All other genome comparisons using ANI show values smaller than 95% between species genomes.

Our results show that the genome length of endophyte organisms was on average bigger (49.24 Mbp) than plant-pathogens (48.04 Mbp) and saprobes (46.71 Mbp), despite the largest genome in the study being Dactylonectria alcacerensis which is a plant pathogen with 61.76 Mbp (Figure S1). This value was also reflected in the average number of aminoacid sequences predicted where endophytes (12971) were the highest followers by plant-pathogens (12798) and saprobes (11473) (Figure S2). However, we observed a change in the GC content where the highest average in the plant pathogens than saprobe fungi and endophytes (51.54, 51.42, 51.07, respectively) (Figure S3).

The plant pathogenic fungi in this study presented more tRNA on average than saprobe and endophyte (218.42, 218.00, 212.67 respectively). This is contradictory to what was found in other studies [81,82] (Figure S4).

We observed a large variation in repetitive elements according to the fungi lifestyle (Figure S5), where on average the saprobe (7.35) has more repetitive elements than plant pathogens (5.28) and endophytes (3.13). When observing the repetitive elements sequences with origins in retroelements, we found statistically significant differences between endophytes (0.69) and plant pathogens (2.60) (Figure 2). The difference in retroelements between saprobe, the largest on average with 3.67, and plant pathogens, is not significant, as well as saprobe and endophytes. Regarding the repetitive elements classified as DNA transposons (Figure S6), we observed a shift where saprobes have more repetitive DNA transposons on average (0.75) than endophytes (0.68) and plant pathogens (0.53) (Figure S6).

The number of genes present in endophytes was also higher than in plant-pathogens and saprobes (Table S1), The number of secreted CAZy enzymes was higher in plant pathogens than in endophytes and saprobe fungi, (953.33, 907, and 757.33, respectively) (Figure S7). The secreted CAZy enzymes with cold properties are higher in endophytes (382.67) than in plant pathogens (359.92) and saprobes (313.33) (Figure S8). On the other hand, the secreted CAZy enzymes with thermo resistance were higher in plant pathogens (217.17) than endophytes (196.33) and statistically significant among plant pathogens and saprobes (146.00) (Figure 3).

The average number of effector proteins secreted by endophytes (278.33) is higher than in plant pathogens (268.17) and saprobes (224.67) (Figure S9). The temperature adaptation for effector genes shows statistical significance between endophytes and saprobes for cold and thermos adaptations, where the effector genes adapted to cold was higher in endophytes (166.33), followed by plant pathogens (145.42) and saprobes (126.67) (Figure 4). The endophytes also show higher potentiation in adapting to warmer temperatures, than plant pathogens and saprobes (65.00, 59.08, and 44.67, respectively) (Figure 5).

The endophytes have shown to have more effector genes on average with apoplastic characteristics (237.00) than plant pathogens (223.92) and saprobes (190.67) (Figure S10). On another hand, the effector genes with cytoplasmatic characteristics are shown to be statistically significant between endophytes (99.93 on average) and saprobes (71.00 on average), despite that, plant pathogens (101.58) had higher amounts than endophytes and saprobes (Figure S11).

Regarding Biosynthetic Gene Clusters, endophytes were shown to produce on average more biosynthetic metabolites based on cluster results than plant pathogens and saprobes (54, 53.25, 40.33, respectively) (Figure S12).

We observed also the formation of clusters based on fungal lifestyle when considering their genome products associated with plants (enzymes and metabolites) (Figure 6). In the central region of the figure, we have saprobe organisms surrounded by plant pathogens above and below them, while two endophytes are located on the right side and Mariannaeae in the extreme north of the graph.

4. Discussion

The endophytes exist within plant tissues without causing plant apparent harm to the host, whereas plant pathogens infect and cause illness in plants. Both of these groups work closely with living plant cells, overcoming the plant's immunological defenses and adapting to its metabolic environment [83,84]. These interactions need a complex set of genes that create enzymes, toxins, and other compounds that aid the invasion of plant cells, suppressing the plant immune system responses, and manipulating plant metabolism. As a result, their genomes frequently contain genes for complicated secondary metabolite production, a variety of transporters, and genes that confer resistance to plant defense mechanisms [85,86]. On the other hand, saprobe fungi, as decomposers of organic material, require a different set of enzymatic tools focused primarily on breaking down cellulose, lignin, and other plant structural components. While this still requires a diverse set of enzymes, the interaction with non-living material is less complex than manipulating and responding to living cells, and therefore, is expected that saprobes have fewer genes than endophytes and plant pathogens [87,88,89].

The repetitive elements are microsatellite regions, or more often transposable elements (transposons) which increase their number in the genome mainly due to cellular stress, which is not a characteristic of endophyte environments [90,91,92]. This corroborates our findings for fewer repetitive elements in endophyte genomes likely influenced by their lifestyle since being endophytic is a very cost-effective strategy. When establishing themselves within plants, they are shielded from abiotic and biotic challenges being less subject to stress, which is a factor for the multiplication of transposable elements [93].

Our study confirms the results of Queiroz, 2020 [94] where the number of repetitive elements was a key element in distinguishing between pathogenic and endophytic fungi, despite non-statistical significance. Several studies found Transposable elements abundant in plant-pathogenic fungi [95,96,97]. In our study, this case can be even better observed with retrotransposons, where the different occurrence between endophytes and plant pathogens was statistically significant. The amount of retroelements identified in endophytes is fewer than in saprobe and statistically significantly less than in plant pathogens.

The transposable elements contribute significantly to genomic plasticity, which is crucial for adapting to host defenses. For instance, in the case of the plant pathogen Verticillium dahliae, TEs exert an influence on gene expression variations essential for pathogenicity enabling the fungus to adapt rapidly to host-induced stresses or resistances [98]. Meanwhile, saprobe fungi are not known for presenting large amounts of TEs [96,99,100], since they are less subject to adaptation to host living cells and may reside in less stressful environments [101,102,103,104], although, the higher number of TEs for saprobes in our study could indicate potential adaptation to more dynamic or challenging environmental conditions [96].

The secretome of fungi, which includes CAZy enzymes, plays a pivotal role in the interaction between fungi and plants. These enzymes are crucial not only for breaking down the complex polysaccharides found in plant cell walls, which aids in nutrient acquisition and infection processes but also for helping fungi evade plant immune responses, facilitating their survival and propagation in various environments [104,105,106]. The CAZymes of classes CE, GH, and PL are often referred to as plant-litter degrading enzymes because they play crucial roles in the degradation of plant biomass by fungi and bacteria [107,108].

In our study, saprobe fungi produced less secreted CAZy enzymes than endophytes and plant pathogens. According to Zhao (2013) [109], saprobe fungi are expected to only degrade plant-litter complex polysaccharides from dead material without having to interact with the plant-living cell defense, and for that, typically require a narrower range of enzymes. In contrast, endophytes and pathogens interact directly with living plant tissues, requiring a broader array of enzymes not only to breakdown living plant tissues but also evade or suppress plant immune responses, and successfully infect their hosts. These interactions demand more specialized enzymatic functions, such as those that degrade pectin, hemicelluloses, and other plant cell wall components under varying physiological conditions as to successfully establish their mutualistic relationships with their hosts [89,110–1142.

The CAZy enzymes when secreted have several functions other than the degradation of plant cell-wall. They are also involved in promoting attachment, invasion, colonization, and nutrient acquisition from the hosts and that’s why numerous studies suggest that endophytes and plant pathogens can manufacture the same leaf-degrading enzymes as closely related saprobic fungi [89,108,110,113,114,115,116,117,118].

Pathogens and endophytes must overcome various layers of plant protection to successfully infect the plant [119,120]. The first immunological response of the plant is through receptor-like kinases that can detect pathogen-associated molecular patterns called microbe-associated molecular patterns (MAMPs) which activate the MAMP-triggered immunity or pathogen-triggered immunity, which is effective against a wide range of microorganisms [121,122,123,124]. However, microorganisms can bypass this mechanism by producing effector proteins that alter cellular processes in the host, leading to effector-triggered susceptibility [122,125,126,127,128]. In addition, the plant has a second layer of receptors called resistant proteins or R proteins which also recognize patterns in effector proteins in the host generating a resistance response that promotes the death of infected cells containing the infection [122,129,130,131,132,133]. Hence, effector proteins have distinct properties to act in the host extracellular space (apoplast) and intracellular space (cytoplasm) during infection. The cytoplasmic effectors have a higher proportion of positively charged amino acids, whereas apoplastic effectors are enriched for cysteine residues [79]. Therefore, it is expected that the more interactions needed between the fungi and plants to overcome their immune system, the more equipped the plant pathogen with effector genes should be the fungi.

However, the cell defense system is not the only obstacle to the survival and adaptation of fungi. The environmental conditions play a major role in protein synthesis, sporulation, and diversity of their communities. Currently, the world is undergoing a significant and fast-paced climate change caused by human activities, specifically global warming. This has resulted in a rise in temperatures, which are presently approximately 1.5 °C higher than they were during the pre-industrial period [134]. This even raises concerns about the adaptation of fungal species with pathogenic capabilities, such as those that infect plants, since is expected an increase in plant diseases in crops under projected climate change scenarios [134,135,136]

In our study, the pathogenic fungi from the Nectriaceae family have more secreted enzymes adapted to higher temperatures than endophytes and saprobes which indicates a higher potential to adapt to global warming situations. The potential for growth and infection of Nectriaceae fungi, are observed in many of the most important economical plant-pathogens from Nectriaceae family belonging to the genera Neonectria and Fusarium [137,138,139,140,141,142,143]. In our study plant-pathogens presented less cold-adapted and more thermo-adapted proteins indicating a potential to perform better in higher temperatures, which reinforces the findings of [139] showing that phytopathogens from Nectriaceae order are organisms that may prevail in global warming increasing their distribution and impact.

On the other hand, in some cases, the temperature can be a comprehensive factor that impacts cellular metabolism and functions. This encompasses essential biochemical parameters, such as the rates at which reactions occur, molecule binding, and the flexibility of cell membranes [134,144]. As example, the freshwater saprobe fungal communities are susceptible to being affected by climate change in species composition and abundance and therefore, impact freshwater ecosystem function [145,146,147,148]. They are richer in cold regions or environments and have genome machinery more adapted to colder temperatures [47,149]. Therefore, the increase in the temperature from global warming can cause a reduction in their activity and could negatively impact energy, carbon, and nutrient cycling, threatening the delivery of ecosystem services to higher organisms [150,151,152,153].

The newly assembled genome of Neonectria lugdunensis presented in the study, was first isolated from a submerged decaying twig in a stream bed (personal communication L. Marvanová) and is commonly reported in aquatic ecology studies [154] but has also been identified in soil [155]. The N. lugdunensis has been shown to be resistant to long drought at 25°C and with the potential to endure even higher temperatures for a limited time, as shown in [153]. Other freshwater fungi have better adaptation to colder temperatures showing growth peaks at temperatures between 15°C and 25°C [146,153]

The low amount of thermo-adapted protein observed in the saprobe fungi from this study is a concerning situation when the increase in world temperature can also represent a loss in saprobe biodiversity, growth, and survival. The temperature is one of the most important factors influencing the structure of saprobe communities, which may lead to severe consequences in the ecological process of decomposition [156]. Therefore, monitoring their biodiversity and the implementation of conservation efforts are vital to preserve the biodiversity of saprobe fungi.

The protein synthesis is a central cellular process that is partly regulated by the availability of tRNA molecules. The tRNAs can pre-present in the genome and spread in multiple families in multiple copies or single genes [81]. They are vital for the breakdown of organic matter and the synthesis of proteins and metabolites that can interact with living organisms. In endophytes and plant pathogens, they are essential for the organism virulence and evasion of host defense and are evolved with pathogenicity [82,157,158]. Therefore, the number of copies of tRNA and their variability may be utilized to predict the efficiency with which genes may be translated. This allows for the estimation of protein synthesis rates, cell responsiveness to external influences, and eventually, evolutionary adaptations to novel environments and has been considered a determinant for lifestyle transitions among basidiomycetes fungi [81,159]. In our data, tRNA did show to be significant in differentiating fungal lifestyle.

The fungi have developed distinct mechanisms for producing CAZyme and secondary metabolites to suit their lifestyles [160,161,162,163]. The secondary metabolites are primarily encoded by biosynthetic gene clusters (BGCs), that are jointly controlled and located very close together in a certain genome region, and as a result, the BGCs in charge of a given specialized metabolite are either "silent" or upregulated [52,164,165,166].

These metabolites play a crucial role in several key adaptive processes associated with ecological interactions and stress responses within their environment and are not directly associated with the growth or reproduction of the fungi [164,167,168,169] being often linked to communication and the defense and/or attack against their surrounding organisms or hosts [52], therefore expected to be present in higher quantities in endophytes and phytopathogens, as presented in this study.

Many secondary metabolites act as pathogenicity factors and have detrimental impacts on host health [163] largely due to the production of mycotoxins – toxic secondary metabolites encoded by BGCs. Mycotoxins can act as virulence factors weakening or killing host plants, and aiding colonization [170,171,172]. The Nectriaceae fungi have been noted for their biological activities, for parasitism on plants, fungi, and insects, and as producers of antibiotics and/or mycotoxins [173], although not much research was implemented in the study of the mycotoxins of the genus Neonectria during plant infection.

The non-ribosomal peptide synthetases (NRPS) and polyketide synthases (PKS) are two types of large, modular enzyme complexes (megasynthases) involved in the biosynthesis of non-ribosomal peptides and polyketides, respectively. The NRPS creates molecules by joining amino acids. In contrast, PKS creates molecules by joining acyl groups to form large polypeptide chains responsible for many different catalytic domains, each performing a different chemical reaction. This allows them to create highly complex and diverse molecules [54]. In plant-associated fungi, NRPS and PKS have distinct functions in the synthesis of phytotoxins, mycotoxins, and antibiotics [174]. According to Yoder (2001) [175] numerous virulence factors were identified in NRPS clusters, demonstrating them as necessary mechanisms in fungal pathogenesis. They were found in higher concentrations in plant pathogenic and endophytes in this study and lower amounts in freshwater saprobes.

Terpene cyclases are enzymes found in plants and microorganisms that help to form monoterpenes, sesquiterpenes, and diterpenes by converting prenyl diphosphate chains. Terpenoids come from isopentenyl diphosphate (IPP), which prenyl transferases modify to create geranyl diphosphate (GPP), farnesyl diphosphate (FPP), and geranyl diphosphate (GGPP). These compounds serve as starting points for making monoterpenes (from GPP), sesquiterpenes (from FPP), and diterpenes (from GGPP) with the help of terpene cyclases enzymes [176,177,178]. Currently, over 80,000 terpenoids are known [179]. Our study found the highest amount of terpene clusters in plant pathogenic fungi. The terpene is known in pathogenic fungi for producing many mycotoxins of the class sesquiterpenoids which play a vital role in fungal virulence [180]. The terpenes can also have effects on fungal growth and protective effect against oxidative stress and UV radiation by the production of carotenoids [178,181,182]. Different environmental factors including light and temperature were shown to change the production level and composition of carotenoids [183].

The ribosomally synthesized and post-translationally modified peptides (RiPPs) were found without many variations in this study. are an increasingly important group of natural products known for their powerful biological activities [184]. They are produced through a straightforward process. The initial step involves the synthesis of a precursor peptide by the ribosome, which consists of leader, core, and follower amino acid sequences. Subsequently, the core sequence undergoes specific post-translational modifications, guided by the leader and follower sequences. The final bioactive RiPP is released after these sequences are removed [185].

Metal ions play a vital role in numerous enzymatic processes, but their excess can be detrimental to the growth and development of various organisms [186]. Research has shown that siderophores, originally known for their iron-binding capacity, can bind to a variety of metals. This multi-faceted functionality has prompted the broader classification of these compounds as "metallophores," referring to secondary metabolites capable of binding a range of metal(loid) cations [187,188]. Metallophores are low-molecular-weight organic ligands that facilitate the delivery of essential metal ions to an organism, while the organism regulates the production and release of these ligands based on its metal ion requirements [187]. In our study, we detected the Siderophore bind with transporter nickel (NI-Siderophore) in only three species (C. pteridis, I. robusta, and F. nematophilium), while Metallophore (NRP- Metallophore) was found in 11 species in this study with one cluster per organism.

The BGC indole is a volatile compound that has been associated with effective fungicides to face continuous fungal infections [189,190] and was not identified on the saprobes genomes in this study showing that it might be a crucial cluster for the fungi infection and virulence in plants.

The phosphate compounds produced in BGCs are chemicals with phosphorus-carbon bonds with the general chemical formula C−PO(OH)2 or C−PO(OR)2 where R is an alkyl or aryl functional group [191]. They have a wide range of bioactivities, such as antibiotics, antivirals, pesticides, and antiparasitics, where approximately 15% of all phosphonate natural products are commercialized [192]. In our study, it was found in 5 plant pathogens and 1 endophyte showing to be an additional feature in fungal infection.

Isocyanides, also known as isonitriles, are a group of microbial secondary metabolites that have been extensively studied due to their wide range of pharmacological applications such as antifungal, antibacterial, antitumor, and antiprotozoal bioactivity [193,194,195,196]. In addition, they play an important role in the pathogenesis of insect, plant, and human diseases [196,197,198] showing to be an additional accessory for fungal infection found in 10 plant pathogen fungi and 2 endophytes in this study. These compounds are distinguished by the highly reactive isocyanide functional group (R≡N+-C−), which originates from the conversion of specific amino acids within the compound [199].

For the endophytes, mycotoxins can also cause herbivory limitation: Ergot alkaloids produced by endophytes limit herbivory, protecting the plant and increasing the fitness of both the plant and the fungus [200]. In addition, the endophyte mycotoxins can also prevent their predation by insects, aiding in the competition against these pests [201], as a consequence protecting the plant and the fungi itself from predation. Biosynthetic gene clusters such as betalactone are known to be an antiviral heterocyclic compound contributing to biocontrol activity and in addition, acting in insect immune suppression, [202,203,204]. This might be important for endophytes and plant pathogens to avoid predation of the leaf and competition by other organisms.

5. Conclusions

We found that endophytes and plant pathogen fungi carry more effector genes and genes dedicated to secondary metabolism than saprobes, likely due to their higher interaction with the defense mechanisms of living cells of plants.

Despite not finding many variables with statistical significance, except the number of retroelements and thermo-effector genes, we can still differentiate fungal lifestyle based on the combination of characteristics associated with plant interaction of enzymes and metabolites shown in this study. This suggests that we can determine the lifestyle of the fungi based on this data and therefore, machine learning algorithms could be applied efficiently for lifestyle prediction in larger datasets. In addition, the low abundance of effector genes adapted to higher temperatures found in the freshwater saprobes in this study may suggest that they would be important organisms for targeting preservation goals due to global warming events. In contrast, the higher amounts of thermo-adapted effector genes in plant pathogens suggest a greater capacity to enhance their bioactivity as temperatures rise, and therefore a potential to escalate the economic impact of these fungi when proliferating in agricultural systems.

A lot still needs to be explored for Nectriaceae: Within the genomes, the laboratory, and the field, such as the secondary metabolites during fungal infection of plants and their adaptations to different temperatures due to global warming. In the next years, many plant pathologists may study effector proteins and metabolites that contribute to effector functions in plant cell-wall to better understand the mechanisms of infection of fungi and reduce their economic impact. In the same direction, freshwater fungi show a lot of potential to be explored in enzymes adapted to cold temperatures and secondary metabolites that can have biotechnological importance in different areas.