Skip to main content
Download PDF

The impacts of ecological disturbances on the diversity of biosynthetic gene clusters in kauri (Agathis australis) soil

Environmental Microbiome volume 19, Article number: 69 (2024) Cite this article

Abstract

Background

The ancient kauri (Agathis australis) dominated forests of Aotearoa New Zealand are under threat from a multitude of ecological disturbances such as forest fragmentation, biodiversity loss, climate change, and the spread of the virulent soil pathogen Phytophthora agathidicida. Taking a wider ecosystem-level approach, our research aimed to explore the impacts of forest disturbance and disease outbreaks on the biosynthetic potential and taxonomic diversity of the kauri soil microbiome. We explored the diversity of secondary metabolite biosynthetic gene clusters (BGCs) in soils from a range of kauri forests that varied according to historical disturbance and dieback expression. To characterise the diversity of microbial BGCs, we targeted the non-ribosomal peptide synthetase (NRPS) and polyketide synthetase (PKS) gene regions for sequencing using long-read PacBio® HiFi sequencing. Furthermore, the soil bacterial and fungal communities of each forest were characterized using 16 S rRNA and ITS gene region sequencing.

Results

We identified a diverse array of naturally occurring microbial BGCs in the kauri forest soils, which may offer promising targets for the exploration of secondary metabolites with anti-microbial activity against P. agathidicida. We detected differences in the number and diversity of microbial BGCs according to forest disturbance history. Notably, soils associated with the most undisturbed kauri forest had a higher number and diversity of microbial NRPS-type BGCs, which may serve as a potential indicator of natural levels of microbiome resistance to pathogen invasion.

Conclusions

By linking patterns in microbial biosynthetic diversity to forest disturbance history, this research highlights the need for us to consider the influence of ecological disturbances in potentially predisposing forests to disease by impacting the wider health of forest soil ecosystems. Furthermore, by identifying the range of microbial BGCs present at a naturally high abundance in kauri soils, this research contributes to the future discovery of natural microbial compounds that may potentially enhance the disease resilience of kauri forests. The methodological approaches used in this study highlight the value of moving beyond a taxonomic lens when examining the response of microbial communities to ecosystem disturbance and the need to develop more functional measures of microbial community resilience to invasive plant pathogens.

Background

New Zealand kauri (Agathis australis) is one of the world’s largest and longest-living tree species, reaching heights up to 60 m, diameters up to 5 m, and ages up to 1700 years [1]. Endemic to Aotearoa New Zealand, kauri function as an ecological foundation species, significantly influencing the surrounding plant and soil environment and supporting one of the most floristically diverse forest types in Aotearoa New Zealand [2]. Additionally, kauri have an immense cultural value and are regarded as a taonga (treasured) species by indigenous Māori communities [3]. Despite their ecological and cultural significance, the long-term health of kauri forests is under critical threat. A long history of extensive deforestation means that less than 1% of primary old-growth kauri forest remains [1]. These highly fragmented old-growth kauri forests are now under greater pressure from the spread of kauri dieback [4]. Kauri dieback is caused by the highly destructive, soil-borne pathogen Phytophthora agathidicida and symptoms include lower trunk gummosis, chlorosis, leaf loss, and tree death [5, 6]. Dieback disease is distributed across most of the kauri tree species’ natural remaining range [7]. Although short-term control options have been identified [8, 9], no long-term control options currently exist for the treatment of the disease.

The soil microbiome provides important functional roles in supporting plant health and providing defense against plant pathogens [10]. Plant-associated soil microorganisms can defend plants against pathogen attack by inducing plant systemic resistance, preventing pathogen establishment, and through the release of anti-microbial compounds that directly antagonize the pathogen [11, 12]. Our previous research identified differences in the soil microbiome associated with non-symptomatic and symptomatic kauri, indicating that the soil microbiome may be an important environmental factor influencing dieback expression [13]. Later research isolated microbial strains from kauri soils that inhibited the growth of P. agathidicida [14] and subsequent whole genome analysis identified numerous gene regions that encoded the production of secondary metabolites, which may have been responsible for the anti-microbial activity observed against P. agathidicida [15].

Although valuable, using culture-dependent methods to assess the bioactivity of soil microorganisms is resource-intensive and limits the degree of replication required for most systems-level microbial ecology research [16]. Furthermore, culture-dependent methods often examine the bioactivity of a few single strains against a target pathogen, whereas the suppression of soil pathogens is often mediated by the collective activities of the microbial consortium [17]. Thus, exploring the biosynthetic potential of the collective soil microbiome may be more effective for systems-level research that aims to explore the impacts of ecosystem disturbance on forest resilience. Culture-independent, genomic sequencing of environmental DNA to study the diversity and distribution of naturally occurring biosynthetic gene clusters (BGCs) can be applied as an effective tool to explore the biosynthetic potential of the soil microbiome over large spatial scales [18, 19]. Microbial BGCs are groups of genes physically clustered within the microbial genome that encode the operation of major biosynthetic pathways [20, 21]. Non-ribosomal peptide synthetases (NRPSs) and polyketide synthases (PKSs) are major mega-enzyme complexes encoded by BGCs that are responsible for the biosynthesis of many anti-microbial secondary metabolites [16, 18]. Indeed, NRPS and PKS gene regions were the most common types of BGCs detected in the genomes of the microbial strains antagonistic to P. agathidicida [15].

Using culture-independent sequencing methods, this research aimed to explore how successive ecological disturbances have impacted the diversity and abundance of microbial NRPS and PKS BGCs in kauri forest soils. A range of kauri forests across the North Island of Aotearoa New Zealand were sampled that varied according to the degree of disturbance and the distribution of dieback disease. To survey the biosynthetic diversity of the kauri soil microbiome, NRPS and PKS BGCs were targeted using PacBio long-read amplicon sequencing [22]. To survey the taxonomic diversity of the kauri soil microbiome, bacterial and fungal communities were targeted using Illumina 16 S rRNA and ITS gene region sequencing [13]. This research will enable us to begin to explore how successive ecological disturbances have impacted the biosynthetic diversity of the kauri soil microbiome and the potential implications that this may have for kauri forest health and resilience.

Methods

Study sites

Six kauri-dominated forests were targeted for soil sampling which varied according to the degree of historical disturbance and distribution of dieback disease. These sites were named Oratia (West Auckland), Laingholm (West Auckland), Silverdale (North Auckland), Waipoua (West Northland Region), and Puketi (East Northland Region) (Fig. S1, Supplementary File 1). The most fragmented and disturbed forest sites were Oratia, Laingholm, and Silverdale – these sites were all small pockets of remnant kauri forest located on private landowner properties with a widespread distribution of kauri dieback. The sites sampled in Waipoua Forest had originally been investigated in a previous study by Byers et al. [13] and sat in the middle of the high-to-low disturbance spectrum. Declared a forest sanctuary in 1952, Waipoua Forest is one of the largest remaining contiguous tracts of kauri forest [1, 23]. However, before protection, Waipoua Forest was subject to a long history of fragmentation and land use change [13, 24, 25] and there is now an extensive distribution of dieback disease across the area [5, 7]. Puketi Forest represented the least disturbed forest site in our study. Like Waipoua, Puketi Forest is one of the largest remaining tracts of kauri forest remaining [26]. Puketi Forest is one of two kauri forests to contain areas of unlogged kauri [27] and has an exceptionally high plant biodiversity, with over 360 indigenous vascular plants recorded across the area [26, 28]. The high biodiversity of Puketi is actively protected via national legislation, as well as local community efforts (i.e., mammalian pest control) [26, 29, 30]. Although isolated cases of kauri dieback have been reported within Puketi, the extent of dieback is restricted and no kauri across the forest area sampled in our study displayed any symptoms of dieback [24, 27, 31].

Soil sampling and DNA extraction

The soil DNA extracts originally collected by Byers et al. [13] were retrieved from long-term storage at -80°C and a subset was randomly selected for genomic analysis. Soil samples were collected from Puketi, Oratia, Silverdale, and Laingholm Forest in the summer of 2022–2023. The number of kauri trees sampled per forest varied according to the forest area and sampling permissions (Table S1, Supplementary File 1). Due to the challenging physical terrain, a random sampling technique was used to select kauri for sampling. Each kauri sampled was assessed for dieback symptomology and classified as dieback ‘symptomatic’ or ‘non-symptomatic’ [5]. The methods for soil sampling followed those described by Byers et al. [13]. Briefly, the litter layer was removed, and four 10 cm-depth soil samples were collected from around the base of each kauri and combined to form a single composite sample. One composite soil sample was collected for each kauri targeted for sampling. Once in the laboratory, soil samples were sieved to less than 2 mm and stored at -20oC until required for DNA extraction. Soil DNA was extracted from 250 milligrams (mg) of each soil sample using the QIAGEN DNeasy PowerSoil Pro Kit (Hilden, Germany) following the manufacturer’s instructions. Each soil sample was tested for the presence of P. agathidicida using the TaqMan real-time qPCR assay [32] or loop-mediated isothermal amplification (LAMP) assay [33].

Long-read amplicon sequencing of NRPS and PKS BGC regions

Long-read amplicon sequencing was used to survey the diversity of NRPS and PKS BGC regions in kauri soil using the PacBio Single Molecule Real Time (SMRT®) sequencing platform [34]. This enabled the complete capture of the large gene amplicons produced by the NRPS and PKS primer sets [22], which were the same as those originally developed by Charlop-Powers et al. [35]. A two-stage PCR protocol was performed to attach the adaptors and barcodes before SMRT® sequencing. The methodology used for the first-stage PCR is detailed in Supplementary File 1. First-stage PCR libraries were sent to the Australian Genome Research Facility (Melbourne, Australia) for SMRTbell® library preparation and sequencing on the PacBio Sequel II platform (Pacific Biosciences, California, USA). Quality-controlled, demultiplexed PacBio sequencing reads were processed into circular consensus sequence (CCS) reads using in-house software. CCS sequencing reads were processed into amplicon sequence variants (ASVs) using a DADA2 workflow amended for handling PacBio CCS long-read data [36]. The NRPS and PKS ASVs were clustered based on 97% sequence similarity using the DECIPHER R package [37]. Protein annotations of the NRPS and PKS BGCs were performed using BLASTX sequence similarity searches against the complete NCBI protein database [22, 38] and the Minimum Information about a Biosynthetic Gene cluster (MIBiG) v3.1 database [18, 39,40,41]. BGCs were taxonomically annotated at the phylum level based on the taxonomy of the closest match hit on the NCBI database. BGCs sharing less than 70% similarity to genes in the NBCI database was classified as phylum “Unidentified”. Furthermore, BGCs sharing ≥ 70% similarity to genes that were annotated as “unclassified” (or otherwise not reported) on the NCBI database were also annotated as phylum “Unidentified”.

Short-read amplicon sequencing of 16 S rRNA and ITS gene regions

Illumina amplicon sequencing was used to survey the diversity of bacterial and fungal communities in the kauri soil DNA [13]. The bacterial 16 S rRNA V3-V4 gene region was targeted for sequencing using the 341 F/806R primer pair [42]. The fungal ITS2 gene region was targeted for sequencing using the ITS3/ITS4 primer pair [43]. Genomic DNA was sent to NovogeneAIT Genomics (Singapore) for Illumina amplicon library preparation and sequencing using the Illumina NovaSeq PE250 platform. Quality-controlled, demultiplexed raw FASTQ sequencing reads were processed into ASVs using the standard DADA2 v1.8 workflows designed for 16 S rRNA and ITS gene sequencing reads [36]. Taxonomies were assigned to the 16 S rRNA and ITS ASVs using the Ribosomal Database Project Classifier (RDP trainset 18, release 11.5) [44] and the UNITE v10.05.2021 reference database [45].

Statistical analysis

Handling and analysis of the sequencing datasets were primarily performed using phyloseq R [46]. BGC and ASV count tables were preprocessed to remove singletons and non-target taxa (e.g., Chloroplasts). To adjust for differences in sequencing depth between samples, count tables were rarefied prior to diversity analysis (NRPS BGCs = 4100, PKS BGCs = 2200, fungal ASVs = 19000, bacterial ASVs = 15000). In addition, sample-based rarefaction curves were used to confirm that the sequencing depth sufficiently captured the diversity of each target gene region (Fig. S2 to S5, Supplementary File 1). The number of Observed ASVs and Shannon Diversity were calculated as metrics of alpha diversity using vegan R [47]. Generalized linear mixed-effects models (GLMMs) were fitted using a Gamma (log = link) distribution to examine the effects of forest and disease expression on alpha diversity [48]. The statistical significance of each GLMM was tested using Type II Wald Chi-square tests and post-hoc p-values were adjusted using the Holm method [49]. Beta diversity was calculated using the Bray-Curtis Dissimilarity Index and visualized using nonmetric multidimensional scaling (NMDS) [47]. Significant differences in beta diversity were determined using PERMANOVA and pairwiseAdonis, with p values adjusted using the Holm method [50]. ANCOM-BC2 analysis was performed to identify BGCs and ASVs that were differentially abundant between symptomatic and non-symptomatic kauri [51]. Before ANCOM-BC2 analysis, fungal and bacterial taxa were glomerated at the genus level and Puketi Forest was removed from the dataset as no symptomatic kauri were present in this forest. The ANCOM-BC2 models were run using default parameters and p values were adjusted using the Holm method. To identify BGCs that were significantly enriched in each forest, microbial biomarker analysis was performed using the microbiomeMarker R package via the run_ancom function using default parameters [52].

Results

Pathogen DNA bioassays

All soil samples collected from non-symptomatic kauri in Puketi Forest tested negative for the presence of P. agathidicida (Table 1). In Waipoua Forest, all soil samples collected from non-symptomatic kauri tested negative for P. agathidicida. Only 55% of the soils collected from symptomatic kauri in Waipoua Forest tested positive for P. agathidicida. The results of the pathogen DNA bioassays for Silverdale and Oratia Forest suggest a very patchy and complex distribution of P. agathidicida relative to the expression of the dieback (Table 1). In Silverdale Forest, over 71% of the symptomatic kauri tested positive for P. agathidicida. However, 75% of the non-symptomatic kauri soil samples from Silverdale also tested positive, which suggests that the pathogen was widely distributed in soils across the entire forest. The P. agathidicida detection rate in Laingholm Forest was very low – less than 15% of symptomatic kauri tested positive for the presence of P. agathidicida. In Oratia Forest, 75% of soil samples from symptomatic kauri tested positive for P. agathidicida. The patchy distribution and inconsistent detection of P. agathidicida in the forest soils investigated in this study align with the findings of previous diagnostic surveys which identified a high spatial heterogeneity of P. agathidicida relative to kauri dieback expression [7, 53, 54]. For example, diagnostic surveys by Bellgard et al. [53] detected P. agathidicida in only 32.5% of soil samples associated with symptomatic kauri. There is a strong body of evidence that has consistently identified P. agathidicida as the causal agent of kauri dieback through various methods [4, 5, 53,54,55]. However, the confidence, reproducibility, and sensitivity of P. agathidicida soil testing remain a key issue in kauri dieback research more widely [7]. Considering these issues, the classification of symptomatic and non-symptomatic kauri in this study remained based on kauri dieback disease expression (see Methods for more details). This well-matched our study’s aims, as we were primarily interested in studying functional microbial indicators associated with kauri tree health, not just the presence or distribution of P. agathidicida in kauri forest soils.

Table 1 The results of pathogen DNA bioassays showing the percentage of symptomatic or non-symptomatic kauri that tested either positive or negative for the presence of P. agathidicida DNA in each kauri forest
Full size table

The biosynthetic gene diversity of the kauri soil microbiome

We detected significant differences in the alpha diversity of microbial NRPS BGCs between the kauri forests (Observed: Chisq = 246.44, DF = 4, p < 0.001, Shannon: Chisq = 94.73, DF = 4, p < 0.001). Puketi Forest had a significantly higher number and Shannon Diversity of microbial NRPS BGCs than all other kauri forests (Fig. 1). In addition, the alpha diversity of microbial PKS BGCs exhibited significant differences between kauri forests (Observed: Chisq = 18.22, DF = 4, p < 0.001, Shannon: Chisq = 21.54, DF = 4, p < 0.001). The number of observed microbial PKS BGCs was significantly lower in Puketi Forest than in Waipoua, Laingholm, and Silverdale. Furthermore, Puketi Forest had a lower Shannon Diversity of microbial PKS BGCs than Waipoua and Silverdale (Fig. 1).

Fig. 1
figure 1

Log values for the alpha diversity of NRPS BGCs, PKS BGCs, fungal ASVs, and bacterial ASVs in each kauri forest. The compact letter display denotes significant differences between forests, with forests not sharing any common letter determined as significantly different (p-adjusted < 0.05)

Full size image

The beta-diversity of microbial NRPS BGCs (R2 = 0.22, DF = 4, p < 0.001) and PKS BGCs (R2 = 0.24, DF = 4, p < 0.001) exhibited significant differences between kauri forests, with all forests sharing significant pairwise differences (Fig. S6, Supplementary File 1). Notably, there were large differences in the beta diversity of biosynthetic genes between Puketi versus Oratia, Silverdale, and Laingholm Forest.

Kauri dieback expression did not impact the biosynthetic diversity of the kauri soil microbiome, as there were no significant differences in the alpha diversity of microbial NRPS (Observed: Chisq = 0.09, DF = 1, p = 0.76; Shannon: Chisq = 0.42, DF = 1, p = 0.52) or PKS BGCs (Observed: Chisq = 0.05, DF = 1, p = 0.82; Shannon: Chisq = 0.004, DF = 1, p = 0.95) between non-symptomatic and symptomatic kauri (Fig. 2a, b). Furthermore, there were no significant differences in the beta-diversity of microbial NRPS BGCs (R2 = 0.02, DF = 1, p = 0.38) and PKS BGCs (R2 = 0.03, DF = 1, p = 0.12) between symptomatic and non-symptomatic kauri (Fig. S7, Supplementary File 1).

Fig. 2
figure 2

Log values for the alpha diversity of NRPS BGCs, PKS BGCs, fungal ASVs, and bacterial ASVs in symptomatic and non-symptomatic kauri. Significant differences between non-symptomatic and symptomatic kauri are denoted by *, whereby * p < 0.05, ** p < 0.01, *** p < 0.001

Full size image

The biosynthetic domain composition of the kauri soil microbiome

The mean sequence similarity shared between microbial NRPS and PKS BGCs versus genes deposited in the NCBI protein database was 71.0% and 78.8%, respectively (Fig. S8, Supplementary File 1). We observed 49.3% of NRPS BGCs and 81.4% of PKS BGCs to share ≥ 70% similarity to BGCs in the NCBI protein database. When screened against the MIBiG database, the mean sequence similarity of NRPS BGCs (58.8%) and PKS BGCs (66.9%) was much lower (Fig. S9, Supplementary File 1). Similar to findings by Dror et al. [56], we observed only 12.1% of NRPS BGCs and 41.3% of PKS BGCs to share ≥ 70% similarity to BGCs in the MIBiG database.

When the taxonomic affiliation of the microbial BGCs was examined, NRPS BGCs affiliated with the phylum Actinomycetota had the highest mean relative abundance across all forests (38.3%), followed by Pseudomonadota (10.2%), and Acidobacteriota (5.1%; Fig. 3a). The most abundant phyla affiliated with PKS BGCs were Actinomycetota (57.2%) and Pseudomonadota (16.9%; Fig. 3b). In addition, the phylum-level affiliations for a high proportion of the NRPS and PKS BGCs were Unidentified (43.5% and 19.9%, respectively).

Fig. 3
figure 3

The relative abundance (%) of the phylum-level taxonomic affiliations of microbial NRPS BGCs (plot A) and PKS BGCs (plot B) in each kauri forest

Full size image

The NRPS and PKS BGCs with the highest relative abundance in kauri forest soils

Prior to rarefaction, a total of 9418 different NRPS BGCs were detected across the five forests. The number of NRPS BGCs varied across the forests, being highest in Puketi (5570), followed by Waipoua (2212), Silverdale (2082), Laingholm (2008) and Oratia (1400). The top 10 most abundant microbial NRPS BGCs in each kauri forest are presented in Fig. 4a and the protein annotations for these BGCs are presented in Table 2Footnote 1. Overall, the NRPS BGC with the highest average relative abundance across all the kauri soils was BGC_1238, which had a particularly high abundance in Laingholm and Oratia Forest (Fig. 4a). As displayed in Fig. 4b, many more microbial NRPS BGCs were significantly enriched in Puketi Forest (e.g., BGC_1089, BGC_1645, BGC_460) and Silverdale Forest (e.g., BGC_1086, BGC_1095, BGC_636) compared to the other forest sites. Almost all the most highly abundant microbial NRPS BGCs in the kauri forest soils shared a low sequence similarity (< 70%) to genes in the MIBiG database (Table 2), which suggests that the kauri soils were dominated by NRPS BGCs with a high biosynthetic novelty.

Fig. 4
figure 4

Plot A displays the mean relative abundance (%) of microbial NRPS BGCs in each kauri forest, each colour represents a different BGC. Only the top 10 most abundant microbial NRPS BGCs in each forest are displayed. Plot B displays the microbial NRPS BGCs that were identified to be significantly enriched in each forest, with BGCs coloured by forest site. Waipoua Forest is not presented in plot B as no NRPS BGCs were identified to be significantly enriched in this site

Full size image
Table 2 The protein annotations for the most abundant microbial NRPS BGCs in the kauri soils when screened against genes in the NCBI and MIBiG references database using BLASTX. Only the protein annotations for microbial NRPS BGCs that shared at least a 70% amino acid sequence similarity to genes in the NCBI database are presented
Full size table
Table 3 The protein annotations for the most abundant microbial PKS BGCs in the kauri soils when screened against genes in the NCBI and MIBiG references database using BLASTX. Only the protein annotations for microbial PKS BGCs that shared at least a 70% amino acid sequence similarity to genes in the NCBI database are presented
Full size table

In total, 9458 different PKS BGCs were detected across the five forests prior to rarefaction. The number of PKS BGCs varied across the forests, being highest in Waipoua (5295), followed by Silverdale (3048), Laingholm (2807), Puketi (2486) and Oratia (1871). As displayed in Fig. 5a, the microbial PKS BGC with the highest average relative abundance across all the kauri forests was BGC_1647 – this BGC had a particularly high abundance in Laingholm, Oratia, and Puketi. When screened against the MIBiG database, BGC_1647 shared an 85% sequence similarity to BGC0001646, which encodes for the biosynthesis of lagriamide (Table 3). Compared to the other forest sites, Puketi Forest had an enriched abundance of many PKS BGCs such as BGC_659, BGC_1419, BGC_1124, BGC_1056, BGC_1124, and BGC_1419 (Fig. 5b). This suggests that the abundance profile of PKS BGCs in Puketi Forest was more distinctive compared to the other forest sites. However, all these PKS BGCs shared a low sequence similarity to genes in the MIBiG database, making it difficult to assess their biosynthetic origin (Table 3).

Fig. 5
figure 5

Plot A displays the mean relative abundance (%) of microbial PKS BGCs in each kauri forest, each colour represents a different BGC. Only the top 10 most abundant microbial PKS BGCs in each forest are displayed. Plot B displays the microbial PKS BGCs that were identified to be significantly enriched in each forest, with BGCs coloured by forest site

Full size image

Differential abundance of microbial NRPS and PKS BGCs between non-symptomatic and symptomatic kauri soils

In total, 60 microbial NRPS BGCs exhibited a significant differential abundance between non-symptomatic and symptomatic kauri (Fig. 6); 11 were higher in symptomatic kauri soils and 45 were higher in non-symptomatic kauri soils. The microbial NRPS BGCs with the highest differential abundance in symptomatic kauri soils included BGC_1645, BGC_1518, and BGC_1410 (Fig. 6). The microbial NRPS BGCs with the greatest differential abundance in non-symptomatic kauri soils included BGC_3214, BGC_665, and BGC_2004 (Fig. 6a). Consistent with other findings of this study, most of the microbial NRPS BGCs that were differentially abundant between symptomatic and non-symptomatic kauri soils shared a low sequence similarity to genes in the MIBiG database (49 out of 60), making it not possible to accurately predict the compounds encoded by these gene regions. Nonetheless, the protein annotations for the differently abundant NRPS gene regions are provided in Table S3, Supplementary File 2.

A further 15 microbial PKS BGCs exhibited significant differential abundances between non-symptomatic and symptomatic kauri soils (Fig. 6b); nine were higher in symptomatic kauri soils and six were higher in non-symptomatic kauri soils. The microbial PKS BGCs with a high differential abundance in symptomatic kauri soils included BGC_1459, BGC_1396, BGC_1230, and BGC_43 (Fig. 6b). When screened against the MIBiG database, BGC_1459 and BGC_1396 shared a 70% sequence similarity to the mycobactin-encoding gene BGC0001021 (Table S4, Supplementary File 2). Nearly all the microbial PKS BGCs that were significantly higher in non-symptomatic kauri soils shared a low sequence similarity to genes in the MIBiG database, except for BGC_1026 which shared a 70% similarity to BGC0001330 (Table S4, Supplementary File 2).

Fig. 6
figure 6

The differential abundance of NRPS BGCs (plot A) and PKS BGCs (plot B) between symptomatic and non-symptomatic kauri soils. The red horizontal line indicates the -log10 p-value threshold (p-adjusted < 0.05) and microbial BGCs above the line had a significant differential abundance. Positive W-values indicate microbial BGCs with a greater abundance in symptomatic kauri and negative W-values indicate microbial BGCs with a greater abundance in non-symptomatic kauri

Full size image

The fungal and bacterial diversity of the kauri soil microbiome

There were significant differences in the alpha diversity of fungal (Observed: Chisq = 109.38, DF = 4, p < 0.001; Shannon: Chisq = 21.80, DF = 4, p < 0.001) and bacterial (Observed: Chisq = 607.32, DF = 4, p < 0.001; Shannon: Chisq = 132.83, DF = 4, p < 0.001) communities between the kauri forests (Fig. 1). However, these differences were primarily observed in Waipoua Forest, which had significantly lower richness and diversity of soil fungal and bacterial communities than the other forest sites (Fig. 1). There were significant differences in the beta-diversity of fungal (R2 = 0.23, DF = 4, p < 0.001) and bacterial communities (R2 = 0.31, DF = 4, p < 0.001) between forest sites, with all forests sharing significant pairwise differences in community dissimilarity (Fig. S6, Supplementary File 1).

Kauri dieback expression significantly impacted fungal alpha diversity, as there were significant differences in the number of observed ASVs (Chisq = 15.83, DF = 1, p < 0.001) and Shannon Diversity (Chisq = 8.15, DF = 1, p = 0.004) of fungal communities between symptomatic and non-symptomatic kauri (Fig. 2a, b). As identified in a previous study by Byers et al. [13], symptomatic kauri in Waipoua Forest had a higher alpha diversity of soil fungal communities than non-symptomatic kauri. Furthermore, the Shannon Diversity of soil fungal communities in Silverdale Forest was significantly higher in symptomatic kauri (Fig. 2b). Significant differences were identified in fungal beta diversity between symptomatic and non-symptomatic kauri (Fig. S7, Supplementary File 1); however, the magnitude of these differences was relatively small (R2 = 0.03, DF = 1, p = 0.03). When examining these differences individually for each forest site, disease expression had a significant impact on fungal beta diversity in Waipoua (R2 = 0.14, DF = 1, p < 0.001) and Oratia Forest (R2 = 0.31, DF = 1, p = 0.03) – no significant differences were observed in Silverdale or Laingholm.

Soil bacterial communities exhibited no significant differences in the number of observed ASVs between symptomatic and non-symptomatic kauri (Chisq = 1.41, DF = 1, p = 0.24) and only marginal significant differences in Shannon Diversity (Chisq = 3.84, DF = 1, p = 0.05) (Fig. 2b). There were no significant differences in the bacterial beta diversity (R2 = 0.02, DF = 1, p = 0.13) between symptomatic and non-symptomatic kauri (Fig. S7, Supplementary File 1).

The taxonomic composition of the kauri soil microbiome

In total, 1055 different fungal genera were detected in the five different forests prior to rarefaction. The highest number of fungal genera were detected in Puketi (663), followed by Laingholm (656), Silverdale (622), Oratia (545), and Waipoua (489). Fungal phyla with the highest average relative abundance across all the kauri forests were the Basidiomycota (45.2%), Ascomycota (35.6%), and Mortierellomycota (7.8%). The Basidiomycota had the highest relative abundance in Laingholm (49.1%), Oratia (43.7%), Silverdale (48.6%), and Waipoua (64.5%), whilst the Ascomycota had the highest relative abundance in Puketi (52.6%). Compared to the soil bacterial communities, there was greater variability in the makeup of dominant fungal genera across the different kauri forests (Fig. 7a). However, the fungal genera Mortierella (7.8%), Penicillium (7.7%), and Trechispora (5.3%) were consistently amongst the most abundant fungal taxa identified in each forest.

Fig. 7
figure 7

The mean relative abundance (%) of fungal genera (plot A) and bacterial genera (plot B) in each kauri forest, annotated based on the most refined taxonomic ranking that could be assigned. Only the top 10 most abundant genera in each kauri forest are displayed. Within each plot, colours represent different microbial genera

Full size image

In total, 1585 different bacterial genera were detected across the five forests prior to rarefaction. The highest number of bacterial genera were detected in Silverdale (1299), followed by Laingholm (1137), Oratia (989), Waipoua (700), and Puketi (657). Bacterial phyla with the highest average relative abundance were the Pseudomonadota (47.2%), Actinomycetota (20.4%), and Acidobacteriota (20.2%). Bacterial genera with a high average relative abundance across all the kauri forests included Bradyrhizobium (4.7%) and Rosiearcus (4.8%), as well as unidentified genera belonging to Bradyrhizobiaceae (5.7%), Rhizobiales (5.1%), Acidobacteria Gp2 (4.1%), Acidobacteria Gp3 (3.4%), and Acidobacteria Gp1 (3.2%) (Fig. 7b).

The results of ANCOM-BC2 analysis identified 21 fungal genera to have a significant differential abundance between symptomatic and non-symptomatic kauri; three fungal genera had a significantly higher abundance in symptomatic kauri and 18 fungal genera had significantly higher abundance non-symptomatic kauri (Fig. 8a). In addition, 10 bacterial genera had a significantly higher abundance in symptomatic kauri soils whilst 16 had a significantly higher abundance in non-symptomatic kauri soils (Fig. 8b). Most of the bacterial genera positively associated with symptomatic kauri soils were members of the order Clostridiales, belonging to the families Lachnospiraceae (Anaerostipes, Anaerobutyricum, Mediterraneibacter, and Blautia) and Ruminococcaceae (Sporobacter, Gemmiger, Faecalibacterium, and Butyricicoccus).

Fig. 8
figure 8

The differential abundance of fungal (plot A) and bacterial genera (plot B) between symptomatic and non-symptomatic kauri soils. The red horizontal line indicates the -log10 p-value threshold (p-adjusted < 0.05) and genera above the line had a significant differential abundance. Positive W-values indicate genera with a greater abundance in symptomatic kauri and negative W-values indicate genera with a greater abundance in non-symptomatic kauri

Full size image

Discussion

A high diversity of microbial NRPS BGCs was observed in the least disturbed kauri forest soils

A significant finding of our research is that soils from the least disturbed kauri forest (Puketi Forest), had a markedly higher number and diversity of microbial NRPS BGCs compared to the other more highly disturbed kauri forests that had a widespread distribution of dieback disease. These results have important implications for our understanding of how ecological disturbance may impact the resilience of the soil microbiome, as NRPS BGCs encode the production of non-ribosomal peptides (NRPs) which have important roles in supporting plant health and suppressing plant pathogens [57]. Previous research has identified a significantly greater diversity of microbial NRPS BGCs in soils suppressive to Fusarium wilt disease; the suppressive activity of these soils was proposed to be due to the anti-microbial products encoded by NRPS BGC regions [58]. Despite our finding that less disturbed kauri forest soils had a greater diversity of microbial NRPS BGCs, we did not observe any differences in the diversity of NRPS BGCs between non-symptomatic versus symptomatic kauri. At a finer scale within forest sites, this complicates our interpretation of how the diversity of microbial NRPS BGCs may influence the resilience of individual kauri to environmental stressors such as dieback. More research is required to untangle these dynamics, potentially building on our findings by examining differences in the active expression of microbial BGCs (e.g., targeting RNA) between non-symptomatic and symptomatic kauri.

Interestingly, Puketi Forest had a lower number of observed PKS BGCs than Waipoua, Laingholm and Silverdale, and a lower diversity of PKS BGCs than Waipoua and Silverdale. The greater alpha diversity of microbial PKS BGCs in more disturbed forests with greater expression of kauri dieback could potentially indicate that PKS-type BGCs are implicated in the “soil immune response” to tree disease [11]. Previous research identified that the genomes of fungal strains antagonistic towards P. agathidicida be enriched in PKS gene regions, which may suggest that PKS BGCs form an important part of the microbial defense against kauri dieback [15]. However, the differences observed in PKS BGC alpha diversity between disturbed vs. less disturbed kauri forests were relatively small and inconsistent, which makes it difficult to interpret the biological importance of this finding.

Further research is needed to understand what environmental factors are influencing the diversity of microbial BGCs in the kauri soils. The high diversity of microbial NRPS BGCs in Puketi Forest soils was unrelated to microbial taxonomic diversity, as there were no differences observed in the diversity of soil fungi and bacteria between Puketi Forest versus the more disturbed kauri forests in Auckland (e.g., Laingholm, Silverdale, and Oratia). Previous research has identified microbial biosynthetic diversity to be influenced by various environmental factors, such as soil abiotic properties (e.g., pH, moisture, carbon content), latitude, climate, and plant cover [59, 60]. The high plant diversity previously recorded in Puketi Forest [26] may be an important factor influencing the high diversity of microbial NRPS BGCs, however, more formal investigations are required to test the influence of plant diversity on the diversity of microbial NRPS BGCs in kauri forest soils. Our results indicate that the diversity of microbial NRPS BGCs may serve as a potential indicator of microbial resilience to pathogen invasion in kauri soils. To validate our research findings, quantitative research in controlled laboratory environments (e.g., plant-soil-pathogen bioassays) is required to investigate the relationships shared between the diversity of NRPS BGCs in kauri soil, the suppression of plant pathogens, and plant disease resistance.

Identification of microbial BGCs with a high relative abundance in kauri forest soils

As discussed, the diversity of microbial BGCs shared no clear relationship with microbial taxonomic diversity, which suggests that the genomes of certain taxa may be biosynthetically enriched and harbour multiple BGCs. Similar findings have been identified in previous research, which identified large differences in the distribution of BGCs across bacterial taxa [61, 62]. This is important, as the presence and abundance of biosynthetically enriched microbial taxa could function as bioindicators of microbial resilience in soils. Most of the microbial BGCs identified in this study were taxonomically affiliated with the bacterial phyla Actinomycetota and Pseudomonadota. Furthermore, many of the most abundant microbial BGCs identified in each kauri forest were affiliated with the Actinomycetota. Considering that microbial taxonomic analysis identified the Pseudomonadota and Actinomycetota to be the most abundant bacterial phyla in the kauri forest soils, the high contribution of these phyla to the biosynthetic ‘imprint’ of the microbiome may be simply due to their high community abundance in general. Taking a conservative approach, we only annotated the microbial NRPS and PKS BGCs identified in this study to phylum-level, as most of the microbial BGCs exhibited a high degree of novelty and shared a low sequence similarity to genes deposited in the reference databases. However, previous research has identified bacterial isolates belonging to the Pseudomonadota (genus Burkholderia) from kauri soils that displayed significant antagonism towards P. agathidicida [14]. Follow-up whole genome sequencing of these Burkholderia strains identified NRPS BGCs regions that shared 100% sequence similarity to BGCs that encode the biosynthesis of ornibactin and pyochelin [15]. Thus, based on our research, we propose that future research may streamline the discovery of natural products in the kauri soil microbiome to bacterial taxa belonging to the phyla Pseudomonadota and Actinomycetota.

We identified several NRPS and PKS BGCs that had a high abundance in kauri soil, such as the NRPS BGC ‘BGC_1238’ and the PKS BGC ‘BGC_1647’. Owing to their high natural abundance in kauri soils, which may indicate functional importance within the resident soil microbiome, these microbial BGCs warrant further examination to investigate their biosynthetic activity in more detail, including the determination of the natural products for which they encode. Interestingly, PKS BGC_1647 shared a relatively high (85%) sequence similarity with BGC0001646, an antifungal polyketide gene cluster detected in Burkholderia gladioli which encodes for the biosynthesis of lagriamide [63]. In contrast, NRPS BGC_1238 shared less than 50% sequence similarity with NRPS BGCs in the MIBiG database. This was part of a wider trend, as many of the microbial BGCs shared a low sequence similarity to genes deposited in the protein reference databases, which is consistent with previous research studies focused on the discovery of microbial BGCs in natural environments [40, 56, 62]. For example, Zhang et al. [62] identified less than 60% of microbial BGCs to match BGCs deposited in the MIBiG database. In addition, Dror et al. [56] identified less than 20% of NRPS BGCs and less than 40% of PKS BGCs to share a 70% similarity with genes in the MIBiG database, which is very similar to the values observed in our study (12.1% of NRPS BGCs and 41.3% of PKS BGCs). Such findings indicate a high degree of biosynthetic novelty in the kauri soil microbiome, which is unsurprising considering the high rates of endemism present in kauri forests [2, 7, 64], as well as the underexplored research environment surrounding kauri soil ecology in general.

Dieback expression was not associated with large shifts in the biosynthetic or taxonomic diversity of the kauri soil microbiome

We did not identify any major differences in the biosynthetic diversity of the kauri soil microbiome between kauri that were symptomatic and non-symptomatic for dieback disease. Similar patterns were observed in microbial taxonomic diversity, which displayed inconsistent and mostly minor responses to tree disease expression. The results presented in this study suggest that within an infected forest, disease expression in individual kauri trees did not drive large shifts in the biosynthetic or taxonomic diversity of the soil microbiome, or conversely, microbial biosynthetic and taxonomic diversity did not significantly influence the susceptibility of individual kauri to dieback expression. The microbial taxonomic analysis carried out in this research builds on findings by Byers et al. [13], which identified pronounced differences in the diversity and composition of soil microbial communities between non-symptomatic and symptomatic kauri in Waipoua Forest. Interestingly, such pronounced differences were not observed to the same magnitude when multiple forests were investigated in this study, highlighting the complex challenge and variable nature of examining soil microbes to understand the environmental drivers of tree dieback. By examining the differential abundance of individual microbial BGCs between non-symptomatic and symptomatic kauri, this research enabled us to identify microbial BGCs that may be part of the soil immune response to pathogen invasion and disease outbreak [17, 65]. Our research identified numerous microbial BGCs whose abundances were either higher in the soils associated with diseased kauri (e.g., NRPS BGC_1645 and PKS BGC_1459) or higher in the soils associated with healthy kauri (e.g., NRPS BGC_3214 and PKS BGC_1026). Targeting these BGC regions for further study may be of interest to streamline the discovery of microbial BGCs that are on the frontline in responding to plant disease expression [66] or those that are involved in suppressing the establishment of plant pathogens [58]. As discussed, many of these differentially abundant microbial BGCs exhibited a high degree of biosynthetic novelty. More research is needed to characterize the natural products encoded by these microbial BGCs so that their effects on the growth, development, and virulence of P. agathidicida can be investigated.

In this study, kauri were classified as symptomatic or non-symptomatic based on disease symptomology and tree health, as opposed to using the results of the pathogen DNA testing of the soil samples associated with each tree. This was decided to be the most appropriate and consistent method of classifying each tree, considering the issues surrounding the high spatial heterogeneity of P. agathidicida in kauri forests and the low sensitivity of the pathogen bioassay testing methods [7, 53, 54]. Furthermore, this study was primarily focused on identifying microbial indicators of kauri tree health. Thus, understanding the biosynthetic diversity of soil microbes associated with non-symptomatic “healthy” kauri was more appropriate for the aims of this study. However, we need to acknowledge that the pathogen DNA testing identified a very patchy and inconsistent distribution of P. agathidicida in the soil, with some soils collected from non-symptomatic kauri also testing positive for the presence of P. agathidicida. The inconsistent patterns we observed between kauri tree health and the detection of P. agathidicida may explain the inconsistent patterns we observed in soil microbial diversity between non-symptomatic and symptomatic kauri. To build upon our research findings – which focused on studying the biosynthetic diversity of the soil microbiome in relation to tree health – further diagnostic surveys are required at a larger scale to study the relationship between microbial BGC diversity and P. agathidicida presence in soils.

Conclusion

A primary aim of this research was to develop new approaches for studying the kauri soil microbiome, moving beyond a purely taxonomic focus by examining specific functional attributes of the microbiome related to forest disease resilience. This justified our decision to explore the diversity of microbial NRPS and PKS BGCs in kauri soil, as these gene regions are responsible for the production of anti-microbial secondary metabolites that have been linked to pathogen suppression and plant defense [16, 56, 66]. Furthermore, the application of culture-independent, long-read sequencing enabled us to screen the biosynthetic potential of the kauri soil microbiome across a large spatial scale and in a more resource-efficient manner than culture-dependent methods [16]. This approach enabled us to link patterns in microbial biosynthetic diversity to ecological disturbances (for example, microbial NRPS BGC diversity was positively associated with low forest disturbance), as well as identify a vast catalogue of NRPS and PKS BGCs present in naturally high abundances in kauri soils that may be of interest for future research. The findings of this research highlight the need for us to consider the influence of ecological disturbances in predisposing forests to disease, as well as support the discovery of natural products produced by the indigenous soil microbiome that may enhance the disease resilience of kauri forests.

Data availability

The datasets generated during and analysed during the current study are available in the Aotearoa Genomic Data Repository (https://data.agdr.org.nz) under the project code TAONGA-AGDR00055, https://doi.org/10.57748/Y6AE-6X10 [67].

Notes

  1. Only the protein annotations for microbial NRPS and PKS BGCs that shared a match hit of at least 70% against the NCBI protein database are presented in Tables 2 and 3. However, the protein annotations for all the microbial BGCs displayed in Figs. 4 and 5 are provided in Tables S1 and S2, Supplementary File 2.

References

  1. Steward GA, Beveridge AE. A review of New Zealand kauri (Agathis australis (D. Don) Lindl.): its ecology, history, growth and potential for management for timber. N Z J Forest Sci. 2010;40.

  2. Wyse SV, Burns BR, Wright SD. Distinctive vegetation communities are associated with the long-lived conifer Agathis australis (New Zealand kauri, Araucariaceae) in New Zealand rainforests. Austral Ecol. 2014;39:388–400. https://doi.org/10.1111/aec.12089.

    Article  Google Scholar 

  3. Lambert S, Waipara N, Black A, Mark-Shadbolt M, Wood W. Indigenous biosecurity: Māori responses to kauri dieback and myrtle rust in Aotearoa New Zealand. Hum Dimensions for Tree Health: Global Perspect. 2018;109–37. https://doi.org/10.1007/978-3-319-76956-1_5.

  4. Beever RE, Waipara NW, Ramsfield TD, Dick MA, Horner IJ. Kauri (Agathis australis) under threat from Phytophthora. Phytophthoras Forests Nat Ecosyst. 2009;74:74–85.

    Google Scholar 

  5. Waipara N, Hill S, Hill L, Hough E, Horner I. Surveillance methods to determine tree health distribution of kauri dieback disease and associated pathogens. New Z Plant Prot. 2013;66:235–41. https://doi.org/10.30843/nzpp.2013.66.5671.

    Article  Google Scholar 

  6. Weir BS, Paderes EP, Anand N, Uchida JY, Pennycook SR, Bellgard SE, Beever RE. A taxonomic revision of Phytophthora Clade 5 including two new species, Phytophthora Agathidicida and P. cocois. Phytotaxa. 2015;205:21–38. https://doi.org/10.11646/phytotaxa.205.1.2.

    Article  Google Scholar 

  7. Bradshaw R, Bellgard S, Black A, Burns B, Gerth M, McDougal R, Scott P, Waipara N, Weir B, Williams N. Phytophthora Agathidicida: research progress, cultural perspectives and knowledge gaps in the control and management of kauri dieback in New Zealand. Plant Pathol. 2020;69:3–16. https://doi.org/10.1111/ppa.13104.

    Article  Google Scholar 

  8. Horner I, Hough E, Horner M. Forest efficacy trials on phosphite for control of kauri dieback. New Z Plant Prot. 2015;68:7–12. https://doi.org/10.30843/nzpp.2015.68.5791.

    Article  Google Scholar 

  9. Horner I, Hough E. Phosphorous acid for controlling Phytophthora taxon Agathis in Kauri glasshouse trials. New Z Plant Prot. 2013;66:242–8. https://doi.org/10.30843/nzpp.2013.66.5673.

    Article  Google Scholar 

  10. Mendes R, Garbeva P, Raaijmakers JM. The rhizosphere microbiome: significance of plant beneficial, plant pathogenic, and human pathogenic microorganisms. FEMS Microbiol Rev. 2013;37:634–63. https://doi.org/10.1111/1574-6976.12028.

    Article  CAS  PubMed  Google Scholar 

  11. Raaijmakers JM, Mazzola M. Soil immune responses. Science. 2016;352:1392–3. https://doi.org/10.1126/science.aaf3252.

    Article  CAS  PubMed  Google Scholar 

  12. Schlatter D, Kinkel L, Thomashow L, Weller D, Paulitz T. Disease suppressive soils: new insights from the soil microbiome. Phytopathology. 2017;107:1284–97. https://doi.org/10.1094/PHYTO-03-17-0111-RVW.

    Article  PubMed  Google Scholar 

  13. Byers AK, Condron L, O’Callaghan M, Waipara N, Black A. Soil microbial community restructuring and functional changes in ancient kauri (Agathis australis) forests impacted by the invasive pathogen Phytophthora Agathidicida. Soil Biol Biochem. 2020;150:108016. https://doi.org/10.1016/j.soilbio.2020.108016.

    Article  CAS  Google Scholar 

  14. Byers AK, Condron L, O’Callaghan M, Waipara N, Black A. Identification of Burkholderia and Penicillium isolates from kauri (Agathis australis) soils that inhibit the mycelial growth of Phytophthora agathidicida. 2021;74:42–54. https://doi.org/10.30843/nzpp.2021.74.11736

  15. Byers AK, Condron L, O’Callaghan M, Waipara N, Black A. Whole genome sequencing of Penicillium and Burkholderia strains antagonistic to the causal agent of kauri dieback disease (Phytophthora Agathidicida) reveals biosynthetic gene clusters related to antimicrobial secondary metabolites. Mol Ecol Resour. 2023;00:1–17. https://doi.org/10.1111/1755-0998.13810.

    Article  CAS  Google Scholar 

  16. Dror B, Jurkevitch E, Cytryn E. State-of-the-art methodologies to identify antimicrobial secondary metabolites in soil bacterial communities-A review. Soil Biol Biochem. 2020;147:107838. https://doi.org/10.1016/j.soilbio.2020.107838.

    Article  CAS  Google Scholar 

  17. Berendsen RL, Vismans G, Yu K, Song Y, de Jonge R, Burgman WP, Burmølle M, Herschend J, Bakker PA, Pieterse CM. Disease-induced assemblage of a plant-beneficial bacterial consortium. ISME J. 2018;12:1496–507. https://doi.org/10.1038/s41396-018-0093-1.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Borsetto C, Amos GCA, da Rocha UN, Mitchell AL, Finn RD, Laidi RF, Vallin C, Pearce DA, Newsham KK, Wellington EMH. Microbial community drivers of PK/NRP gene diversity in selected global soils. Microbiome. 2019;7:78. https://doi.org/10.1186/s40168-019-0692-8.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Katz M, Hover BM, Brady SF. Culture-independent discovery of natural products from soil metagenomes. J Ind Microbiol Biotechnol. 2016;43:129–41. https://doi.org/10.1007/s10295-015-1706-6.

    Article  CAS  PubMed  Google Scholar 

  20. Cimermancic P, Medema Marnix H, Claesen J, Kurita K, Wieland Brown Laura C, Mavrommatis K, Pati A, Godfrey Paul A, Koehrsen M, Clardy J, et al. Insights into secondary metabolism from a global analysis of Prokaryotic Biosynthetic Gene clusters. Cell. 2014;158:412–21. https://doi.org/10.1016/j.cell.2014.06.034.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Liu Z, Zhao Y, Huang C, Luo Y. Recent advances in silent gene cluster activation in Streptomyces. Front Bioeng Biotechnol. 2021;9:632230. https://doi.org/10.3389/fbioe.2021.632230.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Benaud N, Zhang E, van Dorst J, Brown MV, Kalaitzis JA, Neilan BA, Ferrari BC. Harnessing long-read amplicon sequencing to uncover NRPS and type I PKS gene sequence diversity in polar desert soils. FEMS Microbiol Ecol. 2019;95:fiz031. https://doi.org/10.1093/femsec/fiz031.

    Article  CAS  PubMed  Google Scholar 

  23. Bassett C, Miers KH. Scientific reserves in state forest. J Royal Soc New Z. 1984;14:29–35. https://doi.org/10.1080/03036758.1984.10421725.

    Article  Google Scholar 

  24. Beachman J. The introduction and spread of kauri dieback disease in New Zealand–did historic forestry operations play a role? Forestry MPI Technical Paper. 2017. https://www.kauriprotection.co.nz/assets/Research-reports/Surveillance-Detection-Diagnostics-and-Pathways/the-introduction-and-spread-of-kauri-dieback-disease-in-new-zealand_final.pdf

  25. Lewis KS, Black A, Condron LM, Waipara NW, Scott P, Williams N, O’Callaghan M. Land-use changes influence the sporulation and survival of Phytophthora Agathidicida, a lethal pathogen of New Zealand kauri (Agathis australis). Forest Pathol. 2019;49:e12502. https://doi.org/10.1111/efp.12502.

    Article  Google Scholar 

  26. Conning L, Moors F. Natural areas of Puketi Ecological District: Reconnaissance survey report for the Protected Natural Areas Programme. Department of Conservation (Northland Conservancy). 1998. https://www.doc.govt.nz/documents/conservation/land-and-freshwater/land/puketi-ecological-district.pdf

  27. Forest and Bird. Pigs spread Kauri dieback in Northland. https://www.forestandbird.org.nz/resources/pigs-spread-kauri-dieback-northland (2020). Accessed 10 July 2024.

  28. Jager MM. Plant traits and their role in determining forest community structure along a soil fertility gradient. Master of Science (MSc). University of Waikato (Hamilton, New Zealand). 2014. https://hdl.handle.net/10289/8774

  29. Northland Regional Council. 15.8 Case study 2: Puketi Forest. https://www.nrc.govt.nz/resource-library-archive/environmental-monitoring-archive2/state-of-the-environment-report-archive/2007/state-of-the-environment-monitoring/indigenous-biodiversity/158-case-study-2puketi-forest/ (2023). Accessed 10 July 2024.

  30. Puketi Forest Trust. Puketi Forest Restoration Project. https://puketi.org.nz/forest-restoration/ (2023). Accessed 10 July 2024.

  31. Department of Conservation. DOC confirm Kauri dieback in Puketi Forest, Bay of Islands. https://www.doc.govt.nz/news/media-releases/2020-media-releases/doc-confirm-kauri-dieback-in-puketi-forest-bay-of-islands/ (2020). Accessed 10 July 2024.

  32. Than D, Hughes K, Boonhan N, Tomlinson J, Woodhall J, Bellgard S. A taq Man real-time PCR assay for the detection of Phytophthora ‘taxon Agathis’ in soil, pathogen of Kauri in New Zealand. Forest Pathol. 2013;43:324–30. https://doi.org/10.1111/efp.12034.

    Article  Google Scholar 

  33. Winkworth RC, Nelson BC, Bellgard SE, Probst CM, McLenachan PA, Lockhart PJ. A LAMP at the end of the tunnel: a rapid, field deployable assay for the kauri dieback pathogen, Phytophthora Agathidicida. PLoS ONE. 2020;15:e0224007. https://doi.org/10.1371/journal.pone.0224007.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Charnaud S, Munro JE, Semenec L, Mazhari R, Brewster J, Bourke C, Ruybal-Pesántez S, James R, Lautu-Gumal D, Karunajeewa H, et al. PacBio long-read amplicon sequencing enables scalable high-resolution population allele typing of the complex CYP2D6 locus. Commun Biology. 2022;5:168. https://doi.org/10.1038/s42003-022-03102-8.

    Article  CAS  Google Scholar 

  35. Charlop-Powers Z, Pregitzer CC, Lemetre C, Ternei MA, Maniko J, Hover BM, Calle PY, McGuire KL, Garbarino J, Forgione HM. Urban park soil microbiomes are a rich reservoir of natural product biosynthetic diversity. Proceedings of the National Academy of Sciences. 2016; 113:14811–14816. https://doi.org/10.1073/pnas.1615581113

  36. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: high-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3. https://doi.org/10.1093/nar/gkz569.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Wright ES. Using DECIPHER v2.0 to analyze big biological sequence data. in R R J. 2016;8. https://journal.r-project.org/archive/2016/RJ-2016-025/RJ-2016-025.pdf

  38. Sayers EW, Bolton EE, Brister JR, Canese K, Chan J, Comeau DC, Connor R, Funk K, Kelly C, Kim S, et al. Database resources of the national center for biotechnology information. Nucleic Acids Res. 2022;50:D20–6. https://doi.org/10.1093/nar/gkab1112.

    Article  CAS  PubMed  Google Scholar 

  39. Medema MH, Kottmann R, Yilmaz P, Cummings M, Biggins JB, Blin K, de Bruijn I, Chooi YH, Claesen J, Coates RC, et al. Minimum information about a Biosynthetic Gene cluster. Nat Chem Biol. 2015;11:625–31. https://doi.org/10.1038/nchembio.1890.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Geers AU, Strube ML, Bentzon-Tilia M. Small Spatial Scale Drivers of Secondary Metabolite Biosynthetic Diversity in Environmental Microbiomes. mSystems. 2023;8:e00724–00722. https://doi.org/10.1128/msystems.00724-22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Terlouw BR, Blin K, Navarro-Muñoz JC, Avalon NE, Chevrette MG, Egbert S, Lee S, Meijer D, Recchia Michael JJ, Reitz Zachary L, et al. MIBiG 3.0: a community-driven effort to annotate experimentally validated biosynthetic gene clusters. Nucleic Acids Res. 2022;51:D603–10. https://doi.org/10.1093/nar/gkac1049.

    Article  CAS  PubMed Central  Google Scholar 

  42. Fadrosh DW, Ma B, Gajer P, Sengamalay N, Ott S, Brotman RM, Ravel J. An improved dual-indexing approach for multiplexed 16S rRNA gene sequencing on the Illumina MiSeq platform. Microbiome. 2014;2:1–7. https://doi.org/10.1186/2049-2618-2-6.

    Article  Google Scholar 

  43. Yang RH, Su JH, Shang JJ, Wu YY, Li Y, Bao DP, Yao YJ. Evaluation of the ribosomal DNA internal transcribed spacer (ITS), specifically ITS1 and ITS2, for the analysis of fungal diversity by deep sequencing. PLoS ONE. 2018;13:e0206428. https://doi.org/10.1371/journal.pone.0206428.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  44. Wang Q, Garrity GM, Tiedje JM, Cole JR. Naive bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol. 2007;73:5261–7. https://doi.org/10.1128/aem.00062-07.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Abarenkov K, Zirk A, Piirmann T, Pöhönen R, Ivanov F, Nilsson RH, Kõljalg U. UNITE general FASTA release for Fungi. 2021. https://doi.org/10.15156/BIO/1280049

  46. McMurdie PJ, Holmes S. Phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013;8:e61217. https://doi.org/10.1371/journal.pone.0061217.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O’hara R, Simpson GL, Solymos P. Community Ecology Package. R package version 2.5.4. 2019. https://cran.r-project.org/web/packages/vegan/index.html. Accessed 10 July 2024.

  48. Bates D, Mächler M, Bolker B, Walker S. Fitting Linear mixed-effects models using lme4. J Stat Softw. 2015;67:1–48. https://doi.org/10.18637/jss.v067.i01.

    Article  Google Scholar 

  49. Hothorn T, Bretz F, Westfall P, Heiberger RM, Schuetzenmeister A, Scheibe S, Hothorn MT. Package ‘multcomp’. Simultaneous inference in general parametric models. Project for Statistical Computing. 2019. https://ftp5.gwdg.de/pub/misc/cran/web/packages/multcomp/multcomp.pdf

  50. Martinez Arbizu P, pairwiseAdonis. Pairwise multilevel comparison using adonis. R package version 0.4. https://github.com/pmartinezarbizu/pairwiseAdonis (2020). Accessed 10 July 2024.

  51. Lin H, Peddada SD. Analysis of compositions of microbiomes with bias correction. Nat Commun. 2020;11:3514. https://doi.org/10.1038/s41467-020-17041-7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Cao Y, Dong Q, Wang D, Zhang P, Liu Y, Niu C. microbiomeMarker: an R/Bioconductor package for microbiome marker identification and visualization. Bioinformatics. 2022;38:4027–9. https://doi.org/10.1093/bioinformatics/btac438.

    Article  CAS  PubMed  Google Scholar 

  53. Bellgard SE, Weir BS, Pennycook SR, Paderes EP, Winks C, Beever RE, Williams S. Specialist Phytophthora Research: Biology. Pathology Ecology and Detection of PTA. Final report for the Kauri Dieback Joint Agency Response MPI Contract. 2013; 11927:104. https://www.kauriprotection.co.nz/assets/Research-reports/Understanding-the-disease/Specialist-Phytophthora-Research-Biology-Pathology-Ecology-and-Detection-of-PTA.pdf

  54. Froud KJ, Chew YC, Kean JM, Meiforth J, Killick S, Ashby E, Taua-Gordon R, Jamieson A, Tolich L. 2021 Waitakere Ranges kauri population health monitoring survey. 2022. https://www.knowledgeauckland.org.nz/media/2392/tr2022-08-2021waitakere-ranges-kauri-population-health-monitoring-survey.pdf

  55. Horner I, Hough E. Pathogenicity of four Phytophthora species on kauri in vitro and glasshouse trials. New Z Plant Prot. 2014;67:54–9. https://doi.org/10.30843/nzpp.2014.67.5722.

    Article  Google Scholar 

  56. Dror B, Wang Z, Brady SF, Jurkevitch E, Cytryn E, Medema M. Elucidating the diversity and potential function of nonribosomal peptide and polyketide biosynthetic gene clusters in the Root Microbiome. mSystems. 2020;5:e00866–00820. https://doi.org/10.1128/mSystems.00866-20.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Ranjan A, Rajput VD, Prazdnova EV, Gurnani M, Bhardwaj P, Sharma S, Sushkova S, Mandzhieva SS, Minkina T, Sudan J, et al. Nature’s Antimicrobial Arsenal: non-ribosomal peptides from PGPB for Plant Pathogen Biocontrol. Fermentation. 2023;9:597. https://doi.org/10.3390/fermentation9070597.

    Article  CAS  Google Scholar 

  58. Zhao M, Yuan J, Zhang R, Dong M, Deng X, Zhu C, Li R, Shen Q. Microflora that harbor the NRPS gene are responsible for Fusarium wilt disease-suppressive soil. Appl Soil Ecol. 2018;132:83–90. https://doi.org/10.1016/j.apsoil.2018.08.022.

    Article  Google Scholar 

  59. Sharrar Allison M, Crits-Christoph A, Méheust R, Diamond S, Starr Evan P, Banfield Jillian F. Bacterial secondary Metabolite Biosynthetic potential in soil varies with phylum, depth, and vegetation type. mBio. 2020;11. https://doi.org/10.1128/mbio.00416-00420.

  60. Lemetre C, Maniko J, Charlop-Powers Z, Sparrow B, Lowe AJ, Brady SF. Bacterial natural product biosynthetic domain composition in soil correlates with changes in latitude on a continent-wide scale. Proceedings of the National Academy of Sciences. 2017; 114:11615–11620. https://doi.org/10.1073/pnas.1710262114

  61. Waschulin V, Borsetto C, James R, Newsham KK, Donadio S, Corre C, Wellington E. Biosynthetic potential of uncultured Antarctic soil bacteria revealed through long-read metagenomic sequencing. ISME J. 2022;16:101–11. https://doi.org/10.1038/s41396-021-01052-3.

    Article  CAS  PubMed  Google Scholar 

  62. Zhang Z, Zhang L, Zhang L, Chu H, Zhou J, Ju F. Diversity and distribution of biosynthetic gene clusters in agricultural soil microbiomes. mSystems. 2024;9:e01263–01223. https://doi.org/10.1128/msystems.01263-23.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Flórez LV, Scherlach K, Miller IJ, Rodrigues A, Kwan JC, Hertweck C, Kaltenpoth M. An antifungal polyketide associated with horizontally acquired genes supports symbiont-mediated defense in Lagria villosa beetles. Nat Commun. 2018;9:2478. https://doi.org/10.1038/s41467-018-04955-6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  64. McKenzie EHC, Buchanan PK, Johnston PR. Checklist of fungi on kauri (Agathis australis) in New Zealand. N Z J Bot. 2002;40:269–96. https://doi.org/10.1080/0028825X.2002.9512788.

    Article  Google Scholar 

  65. Bakker PAHM, Pieterse CMJ, de Jonge R, Berendsen RL. The soil-borne legacy. Cell. 2018;172:1178–80. https://doi.org/10.1016/j.cell.2018.02.024.

    Article  CAS  PubMed  Google Scholar 

  66. Carrión VJ, Perez-Jaramillo J, Cordovez V, Tracanna V, de Hollander M, Ruiz-Buck D, Mendes LW, van Ijcken WFJ, Gomez-Exposito R, Elsayed SS, et al. Pathogen-induced activation of disease-suppressive functions in the endophytic root microbiome. Science. 2019;366:606–12. https://doi.org/10.1126/science.aaw9285.

    Article  CAS  PubMed  Google Scholar 

  67. Byers AK. Biosynthetic gene diversity of kauri soil microbiome. Aotearoa Genomic Data Repository. 2023. https://doi.org/10.57748/Y6AE-6X10.

    Article  Google Scholar 

Download references

Acknowledgements

Thank you to Mels Barton, Fredrick Helm, Tania Penie (Te Rūnanga-Ā-Iwi O Ngāpuhi), Taoho Patuawa (Te Iwi o Te Roroa), and the private landowners across Auckland for your invaluable contribution to this project – without you this research would not have been possible. The authors would like to thank the two anonymous reviewers for their constructive feedback and insightful suggestions, which have greatly contributed to improving the quality of this manuscript.

Funding

This research was funded by the Royal Society of New Zealand Rutherford Discovery Fellowship, grant number LIU2101.

Author information

Authors and Affiliations

  1. Bioprotection Aotearoa, Lincoln University, P.O. Box 85084, Lincoln, 7647, New Zealand

    Alexa K. Byers & Amanda Black

  2. Plant and Food Research, Sandringham, Auckland, 1025, New Zealand

    Nick Waipara

  3. Faculty of Agriculture and Life Sciences, Lincoln University, P.O. Box 85084, Lincoln, 7647, New Zealand

    Leo Condron

Authors
  1. Alexa K. Byers
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Nick Waipara
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Leo Condron
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Amanda Black
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

A.K.B designed and led the research methodology, bioinformatics analysis, statistical analysis, original writing of the manuscript, and preparation of the manuscript for publication. A.B. contributed to project conceptualisation, supervision, administration, and funding acquisition. N.W. & L.C. contributed to project conceptualisation and research supervision. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Alexa K. Byers.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Byers, A.K., Waipara, N., Condron, L. et al. The impacts of ecological disturbances on the diversity of biosynthetic gene clusters in kauri (Agathis australis) soil. Environmental Microbiome 19, 69 (2024). https://doi.org/10.1186/s40793-024-00613-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40793-024-00613-1

Keywords

Environmental Microbiome

ISSN: 2524-6372