Seasonal and Spatial Dynamics of Fungal Leaf Endophytes in Eucalyptus crebra (Narrow-Leaved Ironbark) Authors: Neysa Mai, Md Javed Foysal, Verlaine J Timms, Leanne A Pearson, Caitlin S Romanis, Toby J T Mills, Jeff R Powell, Brett A Neilan Published: 2024 ABSTRACT AbstractFungal endophytes play an important role in improving the health and productivity of native and cultivated plant species. Despite their ecological and industrial importance, few eucalypt species have been studied in terms of their endophyte communities. We examined the seasonal and spatial dynamics of fungal leaf endophytes in the model species, Eucalyptus crebra (narrow-leaved ironbark), using ITS-based amplicon sequencing. Alpha and beta diversity analyses revealed significantly higher species richness in summer compared to autumn and spring. Similarly, two-way ANOVA analysis showed significantly higher species diversity in summer compared to autumn (observed p < 0.001, Chao1 p < 0.005) and spring (observed p < 0.005, Chao1 p < 0.005). No difference in Shannon index was observed among different canopy levels across the season. Beta-diversity showed differences in fungal composition across the seasons and at various canopy levels based on unweighted UniFrac distance metric (PERMANOVA season p < 0.001, canopy p < 0.05), signifying distinct separation of fungi based on presence-absence. Ascomycota was the most abundant and diverse phylum and was present throughout the year. In contrast, Basidiomycota was only observed during cooler and drier seasons. Neofusicoccum was the most abundant genus, but distribution fluctuated significantly across the seasons. Pestalotiopsis and Neopestalotiopsis were most abundant in the low leaf canopy, whereas Pseudosydowia was most abundant in the high canopy. This study indicates that the diversity and abundance of endophytic fungi in the leaves of healthy E. crebra trees fluctuate seasonally and across canopy levels. The data generated can be used as a baseline for assessing and potentially modulating the health of E. crebra and other important Eucalyptus spp.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00248-024-02455-3.Keywords: Eucalyptus crebra, Endophytes, Fungal communities, Season, Canopy level, Potential pathogens Introduction Endophytes are a class of endosymbiotic microbes that colonise the intercellular and intracellular spaces of plants at a particular time of their life cycle, without harming their host [1–3]. They have been identified as a potentially rich source of biologically active metabolites [3] and can improve a plant’s capacity to withstand harsh climatic conditions [4], unfavourable soil properties [5], phytopathogens [6], and herbivores [7]. By improving plant growth and resilience, endophytes can have ecosystem-wide effects and are important drivers of biodiversity [8]. While endophytes have been isolated from all plant tissue types, fungal leaf endophytes are particularly important for plant fitness and productivity. They can compete with or inhibit phytopathogens [9] and enhance photosynthesis by enriching the internal CO2 concentration [10]. Previous studies of the distribution, diversity, and abundance of plant microbes have demonstrated that environmental conditions, geography, and host intrinsic factors play a major role in shaping endophyte communities [11, 12]. Seasonal dynamics and weather patterns have a particularly high impact on the growth and diversity of endophytes [13], with high humidity and warm temperatures favouring fungal taxa [14, 15]. Geographic location also influences endophyte communities, where the probability of recovery of different endophytic taxa increases with distance between sites [16]. Leaf age is another important factor, with old leaves being more susceptible to colonisation by fungi than younger ones, due to their reduced immune defences [17]. Endophyte communities associated with agriculturally relevant plants have been intensively studied and manipulated to improve crop quality and yields [18–21]. In contrast, those from native species are poorly understood, despite their potential applications in aiding reforestation and restoring degraded ecosystems [22]. Eucalypts (Eucalyptus spp.) are among the most ecologically and economically important plants, particularly in Australia. They provide food and shelter to wildlife and are used widely in timber, pulp, paper, cellulose, and charcoal production [23–25]. The few species that have been characterised, with respect to their microbiome, are rich in fungal endophytes covering a broad range of taxa (around 1400 species in their leaves alone) [26], suggesting that the microbiome plays an important role in the ecological success of these hardy trees. Previous studies on fungal endophytes from Eucalyptus spp. exploited both traditional and high throughput sequencing technologies to understand their ecology [27, 28] and industrial [29, 30] significance. These studies identified Capnodiales as a core fungal order in Eucalyptus leaves, as well as other diverse taxa from the orders Diaporthales, Eurotiales, and Xylariales [31, 32]. Amplicon-based high-throughput sequencing (HTS) of the internal transcribed spacer (ITS) identified 13 unique endophytic fungal isolates from E. exserta. Among these, Lophiostoma sp., Pestalotiopsis sp., and Chaetomium sp. exhibited strong antimicrobial activity [30], suggesting a possible role in protection against phytopathogens. Similarly, culture-based ITS sequencing was used to classify endophytes from E. globulus [33], and culture-independent HTS was used to characterise fungal endophytes from E. grandis [34] and examine their roles in insect [35] and disease-resistance [36]. However, none of these studies examined fluctuations in endophyte communities in eucalyptus native species over different seasons or at different canopy positions. This study aimed to characterise the fungal endophytic microbiome of E. crebra leaves over space and time. Also known as the “narrow leaved iron bark”, this Australian Eucalyptus species thrives across New South Wales, Queensland, and Victoria. It is a primary food tree for koalas, an important source of nectar in the honey industry, and its strong, hard timber is used widely in construction. Furthermore, it grows well in low nutrient conditions and is competitive among weeds [37, 38], making it an excellent candidate for revegetating degraded lands. Despite its economic and environmental significance, little is known about the fungal endophyte community of E. crebra, the forces shaping it, or the role it plays in the ecological success of this important Eucalyptus species. By studying E. crebra fungal endophytes, researchers can uncover novel bioactive compounds that could have therapeutic or agricultural uses and gain a deeper understanding of symbiotic relationships between plants and endophytes. To address these knowledge gaps and provide a spatial–temporal baseline for subsequent studies, we examined seasonal fluctuations in fungal endophytes in the low, mid, and high leaf canopy of an E. crebra plantation within its native range using HTS-based amplicon sequencing. Materials and Methods Site Description and Sampling Fresh Eucalyptus crebra leaf samples were collected at the Hawkesbury Forest experiment site (33°36′40″S, 150°44′26.5″E) in Richmond, NSW, Australia, from November 2015 to July 2016. The experimental site consists of a series of plots, each planted with 160 trees of various Eucalyptus species. E. crebra leaves were harvested from a mixed culture plot. Trees were planted at a density of 1000 trees ha−1 (10 m2 per tree) giving a space of 3.85 m between rows and 2.6 m within rows. The mean annual temperature at this site is 17 °C, and the mean annual rainfall is 801 mm. The ratio of precipitation to evapotranspiration at the site is 0.6; therefore, the site is classified as a dry sub-humid environment under United Nations Environmental Programme (UNEP) classification (Millennium Ecosystem Assessment, 2005 cited previously) [39]. Leaves were harvested from four healthy E. crebra trees, selected as shown in Fig. S1, at three different canopy levels—low, mid, and high canopy. Leaf samples were packed in zip lock bags, labelled and transported to the University of Newcastle for processing. In total, 88 samples were processed for low (n = 29), mid (n = 29), and high (n = 30) canopy leaves. Samples were collected during four seasonal intervals: spring (n = 21, November 2015), summer (n = 22, January 2016), autumn (n = 21, April 2016), and winter (n = 24, June 2016). The temperature and humidity from different canopy levels, followed by average rainfall and wind speed, were calculated using a portable Kestrel 5500 Weather Meter during sampling (Table 1). Table 1Environmental conditions during E. crebra leaf samplingSeasonTemperature (°C)Humidity (%)Rainfall (mm)Wind speed (km/h)Summer35 ± 0.563 ± 1.286.43.5 ± 0.2Autumn20 ± 0.587 ± 1.511.81.8 ± 0.2Winter9 ± 0.251 ± 0.85.65.4 ± 0.3Spring28 ± 0.572 ± 1.524.91.8 ± 0.2*Day averages (mean ± SE) for the respective sampling duration. Temperature and humidity measurements were taken from the canopy after sample collection Environmental conditions during E. crebra leaf sampling *Day averages (mean ± SE) for the respective sampling duration. Temperature and humidity measurements were taken from the canopy after sample collection Leaves were surface sterilised according to the method described by Arnold et al. (2007) [40]. Leaf samples (0.1 g) were cut up, immersed in a 95% ethanol solution for 5 s, followed by 2 min in a NaOCl solution (0.5% free Cl2) and finally 2 min in 70% ethanol. Samples were then dried by gently blotting with a sterile paper towel. Dried leaf samples were cut into small pieces using sterile scissors, pulverised using liquid nitrogen, and stored at − 20 °C before DNA extraction. In total, 88 E. crebra leaf samples were processed for fungal endophyte analysis. DNA Extraction, Amplification, and Sequencing DNA was extracted from each leaf sample using the FastDNA Spin Kit for Soil (MP Biomedicals, Australia). Cell lysis was performed by vigorous shaking in a FastPrep® 24 bead beating instrument (MP Biomedicals) at an intensity of 6 ms−1 for 40 s. DNA was purified via solid phase extraction, following the manufacturer’s instructions and eluted in 100 μL of elution buffer. DNA quantity and purity were determined with the NanoDrop spectrophotometer. The DNA samples were stored at − 20 °C before polymerase chain reaction (PCR) amplification. Amplification of the fungal ITS2 region was performed using the fungal specific barcoding primers f/gITS7 (5′-GTGAATCATCGAATCTTTG-3′) [41] and ITS4 (5′-TCCTCCGCTTATTGATATGC-3′) [42]. Each 50 µL PCR reaction contained 25 µL of Phusion 2 × High-Fidelity (HF) PCR Master Mix with HF Buffer (New England BioLabs Inc., USA), 1 µL forward primer (10 µM), 1 µL reverse primer (10 µM), 21 µL of DEPC-treated water, and 2 µL of 20 ng template DNA. Thermocycling conditions were as follows: 98 °C for 2 min, 30 cycles of 98 °C for 30 s, 56.5 °C for 30 s, and 72 °C for 15 s, followed by 72 °C for 10 min. The PCR products were normalised using the SequalPrep™ Normalization Plate (96-well) Kit (Thermofisher Scientific, USA) following the manufacturer’s instructions. Library preparation of 88 samples using the Nextera XT Index Kit (Illumina, USA) and paired end (2 × 250 bases) sequencing of 95 samples was performed at the Ramaciotti Center for Genomics (Australia). ITS Data Processing The ITS sequence data were analysed using QIIME 2 (v2021.11) [43]. The adapter and primer sequences were trimmed using Cutadapt (v4.4) and paired. Quality filtering, trimming demultiplexed reads, identification of chimeras, and merging of non-chimeric reads were performed using the DADA2 plugin in QIIME 2 to obtain the feature frequency amplicon sequence variants (ASVs) [44]. The parameters were set to trim the first 10 bases from the forward and reverse reads and truncate the forward and reverse reads at 240 and 200 bases, respectively. The feature table was filtered with –p-min-frequency 10 to remove singletons. Subsequently, ASVs were annotated into fungal taxa against the UNITE (v10.0) database [45]. Each sample was normalised to an even rarefied depth of 8226 (Fig. 1A–B) followed by collapsing of the feature ASV table and taxonomy table (levels 1–7) using QIIME 2 taxa collapse to avoid any mismatch. Alpha–beta diversity of fungal endophytes in E. crebra leaves. Rarefaction curve showing species richness, depth, and saturation level for different seasons (A) and canopy levels (B). Distribution of shared and unique ASVs across the season (C) and at different canopy level (D). Alpha-diversity two-way ANOVA measurements in terms of observed species (E), Chao1 (F), and Shannon Diversity Index (G) for different seasons and canopy levels. Fungal community dispersion for different seasons and canopy levels as beta-diversity PCoA plot based on unweighted (H) and weighted (I) UniFrac distance metric. Axes 1 and 2 represent the percentage of variation in terms of community composition explained by PCoA Bioinformatic and Statistical Analyses Observed ASV richness, predicted ASV richness (Chao1 index), and Shannon Diversity Index were considered for the measurement of alpha metrics using the phyloseq R package [46]. Two-way ANOVA with Tukey’s multiple comparison tests was performed to identify significant differences between alpha-diversity measurements. The composition relatedness of leaf samples was analysed using unweighted and weighted UniFrac distance metrics to observe similarity and presence-absence. Principal coordinate analysis (PCoA) was used to plot and visualise beta-diversity [46]. The statistical significance was measured using a permutational multivariate analysis of variance (PERMANOVA) as adonis2 function in the vegan package to determine if the separation of groups was significant (R version 4.2.2) [47]. Statistically significant phyla and genera in different groups were identified using Kruskal–Wallis followed by Dunn’s post hoc test with “Bonferroni” correction to avoid the false-discovery rate [48]. Correlations between microbial communities and environmental conditions including temperature, rainfall, humidity, and wind were analysed as “Pearson” correlation co-efficient and plotted as a matrix using the microbiomeSeq R package [49]. Results Sequence Statistics, Alpha–Beta Diversity After filtering, 3.8 million quality reads were retained, ranging from 7703 to 144,796 reads per sample. After removal of host sequences (14,336), 389 ASVs were generated and further classified into four phyla, 47 orders, and 204 genera. A stationary rarefied curve was observed for most of the samples demonstrating depth and saturation level in terms of diversity (Fig. 1A–B). A greater number of ASVs representing unshared taxa were identified in summer, compared to other seasons, and in low canopy leaves compared to mid and high canopy leaves (Fig. 1C–D). According to Tukey’s multiple comparison test, summer samples showed the highest ASV richness compared to autumn (p < 0.001) and spring samples (p < 0.005). Summer samples also showed higher Chao1 diversity (p < 0.005) compared to autumn and spring samples. There was no difference in Shannon diversity based on season or canopy level, indicating similar evenness of the major fungal taxa across all samples. Observed and Chao1 diversity were significantly greater in low canopy leaves than in high (Fig. 1E–G). PCoA plots generated using the UniFrac distance metric showed the distinct separation of samples based on season and canopy level. Fungal communities were significantly different across the seasons and in different canopy levels, both in terms of presence-absence of taxa (unweighted UniFrac) and their relative abundance (weighted). The presence-absence (unweighted) however showed more pronounced effects (PERMANOVA R2 = 0.0774, p < 0.001 for season and R2 = 0.0374, p < 0.05 for canopy) than relative abundance (PERMANOVA R2 = 0.0704, p < 0.05 for season and R2 = 0.0341, p > 0.05 for canopy). The PCoA plot also showed that the weighted UniFrac can explain more changes in fungal communities in both axes than in the unweighted counterpart. Unweighted and weighted UniFrac distance metrics indicated that the fungal communities in summer and winter were similar (p > 0.05), in terms of diversity and abundance, while the fungal communities in summer and autumn were significantly dissimilar (p < 0.001). Weighted UniFrac, however, indicated that the fungal communities in summer and spring were similar in terms of relative abundance (p > 0.05) (Fig. 1H–I). The interaction between season and canopy level was insignificant for all measurements of alpha and beta-diversity (Fig. 1E–I). Fungal Composition More than 75% of reads from all samples (regardless of season or canopy level) were classified into the phylum Ascomycota. Twenty percent of reads from the summer and winter samples, regardless of canopy level, were classified as Basidiomycota. However, the abundance of Basidiomycota reads dropped below 10% in autumn and spring samples. The Ascomycota reads were classified into nine fungal classes. Of these, ASVs classified as Dothideomycetes were the most abundant (72.8% average across all samples), followed by Sordariomycetes (24.5%), and Eurotiomycetes (2%). The Basidiomycota comprised only two classes: Agaricomycetes (89.5%) and Exobasidiomycetes (9.8%) (Fig. 2A). Thirty-eight fungal genera displayed > 2% abundance in at least one sample. Neofusicoccum abundance (19.5% average across all samples) varied greatly from one season to another, reaching 28% during autumn and spring, but dropping to 11% in summer and winter. Other major genera were Cladosporium (11.7% average across all samples), Crustomyces (6.5%), Fusarium (1.9%), Pseudosydowia (4.9%), Mycosphaerella (4.8%), Schizophyllum (4.9%), Pestalotiopsis (5.1%), Teratosphaeria (3.4%), and Neopestalotiopsis (1.5%). The remaining fungal genera represented less than 35% of the total population (Fig. 2B). Cladosporium showed homogenous relative abundance throughout the summer and winter (regardless of canopy level) and was identified as a major core genus for these two seasons (Fig. S2). Neofusicoccum showed identical relative abundance throughout autumn and spring, as did Cladosporium in summer and winter. Regardless of their relative abundance over the seasons and canopy levels, these genera were present (> 1% relative abundance) in more than 90% of samples and therefore can be categorised as core genera in E. crebra tree leaves (Fig. S3). Fungal endophyte composition in E. crebra leaves. A Relative abundance of fungal phyla across seasons. B Relative abundance of fungal genera across seasons and canopy levels. Genera with ≥ 2% abundance in a season are shown in the figure. Genera with less than 2% abundance and unclassified reads are grouped as “ < 2% & other” in the plot Indicator Fungal Genera The abundance of Neofusicoccum was significantly higher in autumn and spring compared to summer and winter, indicating a preference for milder climatic conditions. The abundance of Pestalotiopsis was significantly reduced in the high-canopy levels compared to the low and mid-canopy levels. High canopy also significantly reduced Neopestalotiopsis abundance and positively influenced Pseudosydowia richness, compared to low canopy leaves (Fig. 3). Differential abundance of major fungal genera in E. crebra leaves over different seasons and canopy levels. Error bars denote standard deviation and jitters represent relative abundance in each sample. **Significant at an alpha level of 0.005. *Significant at an alpha level of 0.05 Correlations Between Taxon Abundance and Environmental Factors Pearson correlation revealed that the overall fungal abundance was correlated to wind and rainfall, particularly during autumn and spring. Low rainfall in winter positively correlated to Crustomyces abundance and moderate rainfall in spring negatively correlated to Alternaria abundance. Low wind speed was positively correlated with the abundance of Pseudosydowia (p < 0.005), Fusarium (p < 0.05), and Schizophyllum (p < 0.05). Low canopy and low wind speed were positively correlated to Pseudosydowia abundance (Fig. 4). Pearson correlation between major fungal genera and environmental factors. The legend on the right side indicates the correlation coefficient between the top 30 genera and environmental factors. Red represents a positive correlation, and blue represents a negative correlation. *p < 0.05, **p < 0.005 Discussion This is the first study to examine the endophytic fungal microbiome of Eucalyptus crebra leaves. Leaves are the primary site of photosynthesis and gas exchange; therefore, microorganisms that impact leaf health and performance can have a major effect on plant growth and fitness [50]. Our findings showed that this community is complex and dynamic, varying seasonally, in response to rainfall and wind speed, and in some cases, spatially across the canopy. Neofusicoccum dominated the fungal communities in autumn and spring. Neofusicoccum species are normal microbial flora of Eucalyptus leaves [51, 52], frequently associated with diseases of woody plants [53], and are the causal agent of Botryosphaeria blight and dieback in a range of horticultural crops [54, 55]. However, Neofusicoccum consists of more than 50 species, and many are non-pathogenic [56]. Despite collecting samples from fresh and visually disease-free leaves, several other potentially pathogenic genera were identified in high abundance, including Crustomyces [57], Cladosporium [58], Pestalotiopsis [59, 60], Teratosphaeria [61], and Schzophyllum [62]. Whether the endophytes colonising E. crebra are latent pathogens or non-pathogenic remains to be determined. In this study, the diversity of fungal communities in E. crebra was higher in low canopy leaves during summer. Autumn and spring showed high and uniform abundance of Neofusicoccum and this is in line with a previous study that reported this genus as part of the core fungal consortia of Eucalyptus microcorys [32]. A reduction in the abundance of Neofusicoccum was apparent in summer and winter but no seasonal-specific trends were observed for the other fungal genera. Interestingly, Cladosporium was found consistently in all summer and winter samples based on both detection threshold and abundance. Given its consistent presence during warm and cool weather, its role could be ecologically important for E. crebra. To our best knowledge, this is the first report of Cladosporium endophytes from Eucalyptus leaves. However, the Cladosporium-like fungi, Cladoriella eucalypti and Toxicocladosporium, have been identified in Eucalyptus leaves previously and may serve a similar function in the fungal microbiome [58, 63]. Torres et al. (2017) found that two Cladosporium species had the potential to act as biological control agents for chrysanthemum white rust, Puccinia horiana [64]. In all other plants, Cladosporium has been associated with pathologies such as leafspot, blight, and mould diseases [65–67]. The role of Cladosporium in E. crebra leaves is therefore unclear and further research is required to determine its functional properties, route of transmission, and potential for pathogenicity. In this study, canopy levels differentiated fungal communities only based on the presence of certain taxa and only in winter. No study yet differentiates endophytic Eucalyptus fungi based on canopy level; however, higher endophytic fungi diversity was observed in the high canopy of European maple (Acer pseudoplatanus L.), oak (Quercus robur L.), and linden (Tilia cordata L.) [68]. These trends likely differ from those observed in our study due to different geoclimatic factors between Europe and Australia. The growth of some abundant fungi in the low canopy (Fusarium, Hormonema, Schizophyllum, and Pseudosydowia) was positively correlated to low wind speed in autumn and spring for E. crebra. In contrast to the high canopy, which is comprised of mostly new vegetation and is exposed to high wind speeds, the low canopy is comprised of older leaves and is exposed to lower wind speeds, which could explain the increased fungal growth. Not all the fungal taxa identified in E. crebra are associated with pathogenicity and some of the endophytic genera reported here are known to be involved in host protection and development. Neofusicoccum is highly specific to Eucalyptus spp., widely distributed in the eastern parts of Australia, and is probably endemic to Australia [55]. Most Neofusicoccum species, specifically N. parvum, are categorised as aggressive plant pathogens linked to dieback and cancer in many plant species [69]. However, other species of Neofusicoccum, such as N. austral, N. luteum, and N. vitifusiforme, are potentially beneficial for plants as they produce diverse secondary metabolites with antibacterial, cytotoxic, and phytotoxic activities [70]. Due to the limited resolution of Illumina amplicon reads, we could not differentiate Neofusicoccum at the species level and therefore failed to predict its functional role in E. crebra. Apart from Neofusicoccum, we identified several common beneficial plant endophyte genera, including Alternaria, Cladosporium, Epicoccum, and Penicillium [71, 72]. Epicoccum is ubiquitous in air and soil, commonly displays an endophytic lifestyle in plants, and produces secondary metabolites that promote plant development and help to protect plants from phytopathogens [73, 74]. Penicillium species play a similar role to Epicoccum with additional protective roles in salt, excessive heat, and heavy metal stress [75–77]. Though the relative abundance of these beneficiary endophytes was quite low compared to that of the potential pathogens (Neofusicoccum), even low abundant microbes can play important roles in plant health [78]. The leaf samples in our study were collected fresh and were free from visual signs of disease. It is therefore possible that the Epicoccum, Penicillium, and some of the Neofusicoccum species reported here help to protect E. crebra leaves from pathogenic microbes. In the present study, we used very stringent statistical parameters to exclude outliers and minimised the false discovery rate to extract the most significant abundance scores. The differential abundance analysis for indicator taxa identified four highly abundant (> 5%) genera influenced by seasons and canopy levels. This clearly indicates that seasonality and canopy level have a major effect on the most abundant fungal taxa present in the leaves. Neofusicoccum and most of the Botryosphaeriaceae genera do not favour very high or very low temperatures and therefore grow slowly in the height of summer and winter [79, 80]. Pestalotiopsis and Neopestalotiopsis were present all year round but were significantly more prevalent at the low canopy level compared to the high canopy. Both genera are associated with leaf fall diseases in plants [81, 82] and their high abundance in E. crebra leaves suggests disease susceptibility in older leaves that fall from the lower canopy. Pseudosydowia displayed opposite drift of relative abundance compared to Pestalotiopsis and Neopestalotiopsis, i.e. it was more prevalent in the high canopy. Pseudosydowia can potentially be beneficial or harmful to the host plant. One study reported P. eucalypti as a pathogen of Eucalyptus globolus, which was identified from leaf and stem [83]. In another study, Pseudosydowia was the most dominant and beneficial endophytic fungus in the leaves of Eucalyptus spp., protecting its hosts from infection [84]. However, very little is known about this genus and its specific role in plant health and immunity. Interestingly, in the present study, statistically significant changes in relative abundance occurred only in the dominant fungal genera (> 5% abundance) known to play major roles (positive or negative) in disease predisposition and protection in other plants. Our results also showed that environmental factors such as precipitation and wind influence the diversity and abundance of the endophytic fungal community in the leaves of E. crebra. Crustomyces positively correlated to low rainfall in winter and Alternaria was negatively correlated to high rainfall in spring. Pseudosydowia abundance was correlated to wind and canopy level (low to high). For E. crebra and other Eucalyptus species, no information is available on taxa-environmental factor correlations. The observation from this study could be used to differentiate E. crebra from other Eucalyptus species leaves based on season and canopy levels. Specifically, the correlation of Pseudosydowia with low wind speed is significant and is supported by its differential abundance at different canopy levels, which are more or less exposed to wind. Conclusion This study suggests that the endophytic fungal community of E. crebra leaves fluctuates throughout the year in response to changes in temperature, rainfall, humidity, and wind. However, it appears stable across different leaf locations and does not seem to be affected by sun exposure or shading. While some of the classified endophytes belong to possibly beneficial taxa, others are potentially pathogenic, and their specific roles remain unclear. For a better understanding of their ecological roles in E. crebra leaves, isolation, identification, and pathogenicity assays of Alternaria, Cladosporium, Neofusicoccum, Penicillium, and Pseudosydowia from various seasons and at different canopy levels are required. The functional and pathogenic potential of these endophytes in E. crebra could be resolved via metagenomic studies. Subsequently, the challenge ahead is to improve technical limitations to suppress or eliminate plant DNA in metagenomic workflows, enhancing the recovery of the fungal reads and allowing more accurate tracking of the endophyte communities in response to environmental stimuli. To our knowledge, this is the first study to examine fluctuations in Eucalyptus endophyte communities over different seasons and canopy levels. The data generated can be used as a spatial–temporal baseline for assessing and potentially modulating the health of E. crebra and other ecologically and economically important Eucalyptus species. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 1957 KB) Author Contributions N.M. collected samples, curated data, and performed the laboratory assays. M.J.F. performed data analysis, prepared tables and figures, and wrote the main manuscript. V.J.T. validated data and reviewed the manuscript. L.A.P. assisted with writing and reviewing. C.S.R. and T.J.T.M. were involved in sample collection and method optimisation. J.P. and B.A.N. were involved in conceptualization, fund acquisition, resources, and supervision. All authors read and approved the final manuscript. Funding Funding for this study was gained through an Australian Research Council Federation Fellowship (FF0883440) and the Australian Research Council Centre of Excellence in Synthetic Biology (CE200100029). NM was supported by an Australia Awards Scholarship. Data Availability Raw 16S rRNA paired-end amplicon data files have been deposited to the National Centre for Biotechnology Information (NCBI) and are currently available under the Bioproject accession number PRJNA1098976. Declarations Competing Interests The authors declare no competing interests. Footnotes Neysa Mai and Md Javed Foysal contributed equally to this study. References Associated Data Supplementary Materials Supplementary file1 (DOCX 1957 KB) Data Availability Statement Raw 16S rRNA paired-end amplicon data files have been deposited to the National Centre for Biotechnology Information (NCBI) and are currently available under the Bioproject accession number PRJNA1098976.