Core hyphosphere microbiota of Fusarium oxysporum f. sp. niveum

doi:10.21203/rs.3.rs-3693966/v1

Download PDF

Research Article

Core hyphosphere microbiota of Fusarium oxysporum f. sp. niveum

https://doi.org/10.21203/rs.3.rs-3693966/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 09 Mar, 2024

Read the published version in Environmental Microbiome →

You are reading this latest preprint version

Bacteria and fungi are dynamically interconnected, leading to either beneficial or antagonistic relationships with the plants. Within this interkingdom interaction, the microbial community that are directly associated with the pathogen makes up the pathobiome. While the overall soil bacterial community associated with the Fusarium wilt diseases is widely examined, the specific bacterial populations that directly interact with the Fusarium wilt pathogens are yet to be discovered. In the study presented here, we define the bacterial community associated with the hypha of Fusarium oxysporum f. sp. niveum race 2 (FON2). Using the 16S rRNA gene metabarcoding, we describe the hyphosphere pathobiome of three isolates of FON2.

Interkingdom interactions

Bacterial-fungal interactions

pathobiome

Fusarium wilt

The hyphosphere, the zone of influence that is immediately around a fungal hypha, represents a major hotspot of interkingdom interactions between fungi and bacteria. Within the soil ecosystem, the hyphosphere functions as a major microniche, where the exometabolic chemistry is influenced by both the host fungi and its hyphal-associated bacteria [1–3]. These molecular exchanges include genetic information, metabolic exchange and chemotaxis [3] Whereas the bacterial physical association within the hyphosphere can be variable, including: (i) freely motile and moving along the hyphae attached as single cells, (ii) within the hyphae as endohyphal bacteria, or (iii) forming biofilms along the hyphae [3, 4]. The planktonic bacterial movement on the hyphae, termed the “hyphal highway” or “fungal highway” [5, 6], has been observed in several soil fungi. For example, the hyphosphere of soil fungi Lyophyllum sp. allows several individual bacteria to move along the hyphae and form biofilms [7]. The most well-studied bacterial-fungal interactions (BFI) system is between ectomycorrhizal fungi and its “helper bacteria” [8–11]. The triangular interactions of bacteria, fungi and plants have been demonstrated in such systems. For instance, fungal-associated bacteria can fix nitrogen and mobilize phosphates to enhance plant growth [10, 12]. While it is well established that beneficial fungal-associated bacteria can assist in plant nutrition, it is unclear whether such effects of bacteria are associated with pathogenic fungi. As a first step towards this, we aim to define the microbial community structure of a soil born pathogen system.

Previous research within the hyphosphere of pathogenic fungi has been to elucidate pathogen antagonism towards designing biocontrol measures [13]. However, the hyphosphere bacteria associated with pathogen success has been understudied. While there are some evidence that bacteria can influence the virulence of pathogenic fungi [14], it is reasonable to hypothesize that such effects are likely more pronounced in hyphosphere interactions, where the bacterial partner can help the host-fungi to express or enhance their virulence. Evidence of such interactions is known from endohyphal bacteria. For instance, Burkhholderia endosymbiont of rice seedling blight (Rhizopus microspores) produces phytotoxins that allow the fungal pathogen to infect the plant host [15]. Further studies within this field include Enterobacter sp. with soil-borne phytopathogen Rhizoctonia solani [16]. While research like this point to fungal-helper bacteria within the realm of pathogenesis, these studies are restriced to singular bacteria associated with one pathogen, and not consortia of microorganisms as expected in a pathobiome.

The microbiome associated with the pathogen has been discussed as part of the biotic component of a relatively newer concept of pathobiome [17]. Although the functions of the microbial community within the pathobiome are unknown we suggest the microbiota recruited in the pathobiome assist in infecting the hosts. It has been proposed to change the disease triangle to the disease tetrahedral to include a susceptible host biotic factor/microbiota (or symbiome) [18, 19]. Recently reviewed [17], the concept of pathogens being lone invaders is limited and the field is shifting towards the pathobiome as a holistic approach of studying pathogen success [18, 20].

In the study presented here, we set-out to investigate the hyphosphere bacterial diversity of Fusarium wilt pathogen of watermelon. Fusarium wilt affects a wide range of crop plants and are caused by fungi that belong to the species-clade Fusarium oxysporum species complex (FOSC). The FOSC strains are one of the most devastating plant pathogens and is ranked 5th among the top 10 plant pathogenic fungi [21]. FOSC are further grouped into formae speciales (f. sp.) based on host specificity [22, 23]. FOSC strains that are specific to watermelon are called F. oxysporum f. sp. niveum (FON). FON is found in states with significant watermelon cultivation, such as Florida, Georgia, and Texas, where farmers can experience 30–80% or at times total yield loss to this soil pathogen [24, 25]. As a soil-borne pathogen, it is reasonable to assume that FON2 will interact with a diverse bacterial community, with some involved in assisting in its pathogenicity. As an initial step towards evaluating the hyphosphere bacteria acting as a FON’s pathobiome, we investigated the hyphosphere bacterial community diversity of the fungal inoculum from the infected tissues. As infected tissues as inoculum within plant residues, we envisaged that the bacterial associates in this fungal inoculum would best represent the pathobiome. We will specifically answer the following questions: 1) does FON sustain a hyphosphere bacterial community? 2) if so, can the fungi maintain a consistent bacterial diversity through subsequent subcultured generations? 3) are various FON isolates have a central core bacterial community? Finally, 4) is there evidence of strain-specific interactions between fungi and bacteria? In the study presented here, we define the hyphosphere bacterial assemblage of watermelon Fusarium wilt in three different FON isolates from different counties in Texas from 2002–2020 (COM 2020, W2 2018, and HAR 2002). This study sets the stage to study hyphosphere bacteria of soil-borne fungi as potential pathobiome of the fungal pathogen.

Seed and Soil Sterilization:

Watermelon seeds (Super Gold) received from Dr. Thomas Isakeit (Texas A&M AgriLife Extension) were washed with Tween 20® (0.002%), rinsed with 10% bleach, surface sterilized with 70% ethanol and rinsed with sterile water: with each done separately for 2 minutes. After sterilization, watermelon seeds were placed on sterile 2% Gelrite (RPI) in Petri dishes. Aliquots of 100 ul of ¼ strength Hoagland’s solution were added to each seed [26]. Seeds were incubated in a dark box for up to 7 days at room temperature. The soil mixture with 1-part commercial sand, and 1-part commercial soil (Jolly Gardener Pro-Line C/25 Growing Mix) in square pots (9.8cm x 9.8cm x 8cm) was autoclaved (gravity cycle 60 min) once per day for three consecutive days. The autoclaved soils in square pots were moistened with 5ml of sterile ¼ strength Hoagland’s solution into which the germinated seeds were planted. Plants were kept in a growth chamber with light intervals of 11 hours of dark and 13-hour light.

Inoculation of watermelon plants

FON2 inoculum preparation:

We used 3 FON2 isolates (COM, HAR, and W2) from our earlier study [27] where we demonstrated genetic differences between these strains. Briefly, COM was isolated in 2020, W2 in 2018, and HAR in 2002. All Isolates were initially isolated from symptomatic watermelon plants. In this study, the FON2 inoculum was prepared by culturing the fungal strains in tryptic soy broth (TSB) on an orbital shaker for 7 days at room temperature.

Inoculation into the sand-oatmeal matrix

A growth matrix consisting of soil and oatmeal was prepared. This matrix included an equal mass of oatmeal (The Quaker Oat Bran™) and play sand. This mixture was dispensed as 10 grams per glass Petri plates and sterilized by autoclave. Following this, 200 ul of FON2 inoculum and 5 ml of 1x phosphate buffer solution (PBS) were dispensed into the sand-oatmeal matrix and incubated at 27°C for up to 7 days. To inoculate the soil, a part of the soil near the rhizosphere was removed without disturbing or touching the roots. This soil was mixed with the sand-oatmeal fungal mix, and reintroduced to the initial site from where the soil was removed.

Tissue sampling, and generation culturing from the hyphal tip

After the 10-day post-inoculation, stem slices (> 3mm thick) from the infected (Fig. 1A) and non-infected plants were removed from above the soil line. The infected plants symptoms included leaf wilt and discoloration consistent with Fusarium wilt (Fig. 1A). The tissue samples were rinsed with 10% bleach, surface sterilized with 70% ethanol and washed with sterile water, each step done separately for 2 minutes before aseptically transferring to the center of a half-strength potato dextrose agar (½ PDA). The plates were incubated upright in an incubator at 27°C for seven days. These plates were labeled “Generation 1” cultures. On day seven, when hyphae had grown from diseased tissue, sterile toothpicks were used to carefully collect the hyphal tips at the edge of the colony from the surface layer of the agar media at 4 quadrants to represent 4 replicates per plate. Hyphal tip microbiota collected from sterilized toothpicks were suspended in molecular-grade water (1 mL). Then, 20ul of the FON2 water suspension was dispensed to the middle of ½ PDA plate labeled as Generation 2. Finally, 500ul of the water and FON2 solution was mixed with 500ul of 60% glycerol and stored at -80°C for preservation, and the remaining amount was stored as our DNA material for amplicon-sequencing. Of the four Generation 2 plates 3 were resampled at 4 quadrants as earlier and labeled as Generation 3 plates. This resulted in 13 microbiota suspensions that were later used for microbial community analyses.

Microscopy

Throughout the incubation of FON2 hyphosphere, culture plates of the three fusaria were regularly inspected for hyphal growth. Hyphal tips were examined under an upright Olympus BX60 compound microscope using 40×/0.65 Ph2 objective. Images were acquired using an Olympus camera model LC30 (Olympus Soft Imaging Solutions GMBH, Munster, Germany). Sample images of all hyphosphere of FON2 isolates were grown on 0.5X PDA for a minimum of three days. Under the microscope images of the hyphal tip were taken to view bacterial hyphal interactions.

DNA extraction and SSU rRNA library preparation

Aliquots of 200ul of 13 suspensions were used as DNA samples which were placed at -18°C for freeze-thaw lysis-based approach. DNA was initially quantified in a SpectraMax ® QuickDrop™ Micro-volume spectrophotometer (Molecular Devices, USA) and then in a Qubit® 2.0 Fluorometer (Invitrogen, Life technologies, USA) with AccuGreen™ High Sensitivity dsDNA Quantitation Kit (Biotium, USA). The presence of fungal and bacteria DNA within the samples were assessed by PCR using established universal primers: ITS1 (5’-TCC GTA GGT GAA CCT TGC GG-3’) and ITS4 (5’-TCC TCC GCT TAT TGA TAT GC-3’) for fungal DNA, and 16S rRNA gene amplification using 27F (5’-AGA GTT TGA TCM TGG CTC AG-3’) and 1492R (5’-GGT TAC CTT GTT ACG ACT T-3’) primers for bacteria [28, 29]. A nested amplification approach was used to prepare amplicon sequencing libraries for 16S rRNA. The first step of nested PCR utilized primers 27F and 1492R, samples with 16S rRNA gene amplification and were visually confirmed after electrophoresis on 1% agarose gel containing GelGreen (Biotium, USA). The second PCR reaction consisted of 16S rRNA gene primers without overhang 341F (5’-CCT ACG GGN GGC WGC AG-3’) and 785R (5’-GAC TAC HVG GGT ATC TAA TCC-3’) [30]. PCR conditions followed the KAPA HIFI Hotstart Readymix PCR kit protocol (KAPA Biosystems). Final nested PCR products were sent to Novogene (Novogene Corporation Inc., Sacramento, CA, USA) for 16S rRNA gene sequencing with Illumina NovaSeq PE250 platform (Illumina, San Diego, CA).

Sequence analysis

Metabarcoding sequence analysis were processed in Mothur v.1.48.0 [31] following the MiSeq SOP pipeline with modifications [32]. The modifications introduced were based on our initial assessment with sequences obtained from the 13 hyphosphere of one of the tests FON2 strains, COM. Based on the initial rarefaction curves (Figure S1) we determined that sequence depth of 50,000 reads per sample would suffice for hyphosphere metabarcoding. Additionally, in the initial test the align.seqs step resulted in loss of several reads which was rectified when the “flip” option was set to “flip = T” (to reverse complement the reads that did not align in the forward direction). Based on these observations, a modified pipeline incorporating these two changes was applied to the final analyses of the 39 samples (13 samples per FON2 hyphosphere). The OTUs thus obtained were classified with SILVA v132 bacterial reference trained with our v3-v4 primers.

Statistical Analysis

Mothur OTU results were analyzed and visualized using R version 4.2.1 (RStudio, Boston, MA, USA). Alpha-diversity indices were calculated within Mothur to obtain the richness and evenness measures within our samples, including Shannon, Chao, and Simpson evenness. Tabulated alpha-diversity output from Mothur was compared among isolates with ANOVA and Tukey HSD test and visualized using ggplot2, stats, and multicompview packages [33, 34]. Relative abundance stats obtained from Mothur were utilized in co-occurrence network analysis. Three different networks were generated for each of the individual hyphosphere populations. The networks were generated as Spearman correlations (coefficients > 0.7) within the Microeco package and visualized with Gephi 0.10.1 [35, 36]. FASTA output from “get.oturep” command to get list of OTU and representative sequences were utilized for Maximum-likelihood phylogenetic trees and calculated within molecular evolutionary genetics analysis program (MEGA11) were calculated with default bootstrap value of 500 and utilized within the interactive tree of life program for tree visualization of the core microbiome among isolates (iTOL v6) [37, 38]. Weighted and unweighted UniFrac were visualized with r packages phyloseq, tidyverse and ggplot packages with maximum-likelihood phylogenetic trees made from MEGA11 program. Linear Discriminant Analysis (LDA) with logarithmic score significant value of 0.05 and LDA cut off of 3 was performed at OTU level to identify taxa that are significantly unique to our isolates hyphosphere; results were visualized with microbiomeMarker package [39].

Core microbiota associated with the hyphal tip was determined by OTUs that occurred in at least 3 samples of the 39 iterations across all three isolates samples. Once the core microbiome was identified, the selected OTUs underwent arcsine transformation and presented as a ternary plot to visualize the distribution of the microbial communities among our three isolates through the ggplot2 extension ggtern package [40]. Heatmaps were generated from the core microbiota to visualize the relative abundances of the OTUs among the test fungi using the phyloseq package [41]. An OTU was deemed to be core if it was observed in three or samples within the 13 samples. These were visualized as correlation plots to distinguish positive and negative interactions within each FON2 isolates core microbiota; this was generated with corrplot package [42]. Since the hyphosphere samples consist of very low biomass the DNA yield was too low to perform shotbun metagenomics analyses. Hence, the functional prediction of bacterial communities within individual hyphospheres was performed using PICRUST2 based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database[43, 44]. The t-test was utilized to determine significant differences in functional potential of the hyphosphere microbiota among the three test fungal isolates with RStatix R package [45].

NovaSeq PE250 sequencing of the 16S rRNA fragment from the V3-V4 region yielded 6,272,775 raw reads from the 39 samples. After sub-sampling 50,000 reads per sample and processing through the modified Mothur pipeline resulted in 13,619 reads per sample classified as 1,201 OTUs. The hyphosphere bacterial composition was classified as 5 phyla and 21 families, with Proteobacteria being the dominant phyla (97%) followed by unclassified bacteria (1.16%). At the family level, Oxalobacteraceae (39.57%), Burkholderiaceae (27.72%), and Comamonadaceae (24.47%) were the majority of OTUs identified from all samples.

Alpha diversity indices Shannon (Fig. 3A), Chao (Fig. 3B), Simpson evenness and observed (Figure S3-4) were utilized to compare the overall diversity among the three FON2 isolates hyphosphere. Within Shannon diversity analysis, there is a significant difference among the diversity of the FON2 hyphosphere; when comparing COM and W2, W2 were significantly higher Shannon values (P < 0.05) whereas COM compared to HAR and HAR to W2 were not significantly different. This trend wasn’t consistent with the Chao richness estimate results where no significant differences were found among the FON2 isolates. To further evaluate this inconsistency, analysis of variance (ANOVA) and the Tukey HSD (honestly significant difference) test were utilized with the dominant OTU of each FON2 isolate (Fig. 3C). Interestingly, the relative abundance of the Otu001 (dominant OTU of HAR FON2 isolate) was significantly higher than Otu003 (dominant OTU of W2) (p^adj = 0.017). Whereas Otu002 (dominant OTU of COM) was not significantly different to either one. LDA was conducted with LefSe at the OTU level to identify unique individual bacteria to FON2 isolate (Fig. 4A). We expect a reduction in alpha-diversity with each generation until the fungal selection of bacterial community is stabilized. Hence, we compared alpha-diversity of the three generations. We found no significant difference in Shannon diversity, Simpson evenness and Chao species richness analysis (S5-7).

Histograms visualizing results from LDA analysis found Otu002 (Comamonadaceae) was highly correlated to COM, while W2 significantly correlated to Otu003 (Burkholderiaceae). Furthermore, Otu0007 (Oxalobacteraceae), and Otu001 (Oxalobacteraceae) were significantly correlated with HAR. Except for Otu007, that was significantly associated with HAR, the rest of the 3 significant OTUs were the largest by relative abundance for their respective hyphosphere communities (Fig. 4A). Another observation that we noted was that all four OTUs that showed significant in LDA (and hence all the dominant OTUs) belonged to the order Burkholderiales. Unweighted UniFrac analysis of the hyphosphere communities could not be clustered based on FON2 hosts nor by generations. Therefore, the community structure from the FON2 isolates is not dissimilar even though the fungal host background has temporal and spatial differences (Fig. 4B).

Of the ten orders that were identified, Pseudomonadles, unclassified proteobacteria, unclassified bacteria, Actinomycetales, Sphingobacteriales, and Burkholderiales were found within all of the three FON2 hyphosphere (Fig. 5A). The core microbiota of the FON2 hyphosphere resulted in a total of 15 OTUs, which made up 96% of the total number of OTUs. A ternary plot was used to visualize the distribution of the 15 OTUs among the three FON2 hosts (Fig. 5B). The result from the ternary plot demonstrates that Otu004 (Comamonadaceae), Otu005 (Oxalobacteraceae), Otu006 (Burkholderiaceae), Out008 (Micrococcaceae), Otu009 (Pseudomonadaceae), Otu010 (Propionibacteriaceae), Otu0015 (Bacteria unclassified), Otu0020 (Burkholderiales), Otu0027 (Oxalobacteraceae), and Otu033 (Burkholderiales) were clustered towards the center, suggesting they might be central to FON2 hyphosphere functions regardless of spatial and temporal differences. Whereas Otu0018 (Burkholderiales) and Otu0011 (Bacillaceae) are relatively higher in W2 host compared to COM and HAR. As mentioned earlier, outliers Otu001, Otu002 and Otu003 which are the most abundant OTUs and found within all the samples while they have a strong relationship with one host.

This strong relationship of a single OTU dominance with the FON2 host is visualized within the heatmap (Fig. 5C) and shown as a high relative abundance value, further validating results observed from the ternary plot. The heatmap displays more details of Otu001 (Oxalobacteraceae), Otu002 (Comamonadaceae), and Otu003 (Burkholderiaceae) presence within all samples and those of the other core OTUs. This again agrees with the LDA results, as Otu001 has a significant relationship with HAR, Otu002 with COM, and Otu003 with W2. Further LDA results of Otu007 indicate a significant relationship with HAR.

Individual FON2 isolate's microbiome was analyzed through co-occurrence networks to compare possible host-specific assemblies (Fig. 7). Within the COM hyphosphere network were 215 edges (33 negative and 182 positive edges), 32 nodes. HAR had 1 negative and 40 positives edges. W2 had a total of 732 edges (4 negative and 728 positives). Core OTU of individual isolates were identified to analyze host-specific assemblies further, these criteria identified 8 Otus within COM, 17 within HAR, and 10 with W2. The correlations among the individual FON2 core microbiota were compared, as shown in Fig. 6. Similar to results from the co-occurrence model with Spearman correlation there is a negative interaction among Otu001, Otu002, and Otu003. COM dominant Otu002 is negatively correlated to the other 6 Otus. W2 dominant Otu003 negatively correlates to 7 Otus and 1 positive (Otu0019). HAR dominant Otu001 has a negative correlation with 8 other Otus and a positive correlation with 4 OTUs (Otu0014, Otu007, Otu0026, and Otu0024. Within HAR there was a positive cluster of Otu0025, Otu008, Otu0021, and Otu0013 that is negatively correlated to the dominant Otu001; therefore, may be a different assembly within the hyphosphere. Although there is a robust negative correlation among the dominant OTUs from our correlation plot, most of the co-occurrence network was positively correlated, displaying a distinct community structure.

PICRUSt results

PICRUSt analysis from 16S rRNA sequencing referenced to the KEGG database resulted 5,380 predicted KEGG Orthology (KO) groups. Of these KOs, 779 were unique to COM, 150 to HAR, 69 unique to W2 and 4,137 were shared among all three as seen in Fig. 8A. The predicted categories of major functions within the hyphosphere, as shown in Fig. 8B, included Metabolism (COM at 44.99%, W2 at 48.22% and HAR at 42.71%), Environmental information processing (COM at 18.77%, W2 at 16.24% and HAR at 16.37%) and Unclassified (COM at 13.73%, W2 at 14.04% and HAR at 14.70%). In level 2 of predicted function, we chose to visualize the top 24 to compare among the three isolates hyphosphere. Within COM hyphosphere all 24 were significantly lower than HAR and W2. FON2 isolate HAR’s hyphosphere was found to have 8 predicted functions to be significantly higher compared to W2, these predicted functions were Replication and Repair, Energy Metabolism, Cellular Process and Signaling, Enzyme Families, Genetic Information Processing, Glycan Biosynthesis and Metabolism, Biosynthesis of other secondary metabolites and Neurogenerative diseases. Within W2, four functional genes were significantly higher than those found in HAR (Amino Acid Metabolism, Lipid Metabolism, Xenobiotics Biodegradation and Metabolism and Metabolism of Terpenoids and Polyketides. The overall results indicate within the hyphosphere of COM (the most recent isolate) is predicted to have less shared activity compared to HAR (the oldest isolate) and W2 but has more unique unshared functions. Within the predicted results associated with Terpenoids and Polyketides and Ubiquinone and other terpenoid-quinone biosynthesis functions results identified 137 reference enzyme commission pathways (EC). The predicted EC included Ubiquinone and other terpenoid-quinone biosynthesis, Sesquiterpenoid and triterpenoid biosynthesis, streptomycin biosynthesis, carotenoid biosynthesis, lysine degradation, caprolactam degradation and zeatin biosynthesis.

Microscopic results

The microscopic image shown in Fig. 2 depicts bacteria that congregate at the elongating hyphae. Out of the four cultured plates, one plate was dedicated to observing "hyphal riding." This interplay between bacteria and hyphae was consistently observed across all isolates and subsequent generations. The accompanying video (S8) portrays the motility of bacteria, freely navigating and traversing along the hyphal structure. These findings serve as visual confirmation of interkingdom interactions occurring within the hyphosphere between bacteria and the hyphae.

Pathobionts represent a major component of a disease, where along with the pathogen the pathobionts form the microbial constituents of a pathobiome. However, their precise contribution to virulence remains uncertain [46]. As a zone of influence of the fungal hypha, the hyphosphere represents the micro-niche that a soil-borne pathogen manipulates to enable its pathogenic lifestyle. Thus, the bacterial microbiota that are present in this zone, and attached to the fungal hypha represent an essential component of the pathogen’s lifecycle. In this study, we aimed to identify a core hyphosphere microbiota characterizing the pathobiome of FON2. This marks the first step towards unraveling the microbiome functions that includes the pathogen’s microbial assemblage.

We suggest that if the hyphosphere composition of FON2 remains consistent across successive subcultures and exhibits similarity among various strains, it is highly likely that these bacteria play a functional role within the assemblage. As the assemblage is that of a pathogen, this should represent a pathobiome. Therefore, we tested our hypothesis with three isolates of different spatial and temporal backgrounds. The three FON2 Isolates (COM from 2020, W2 from 2018, and HAR from 2016) inoculum of oatmeal and sand medium was added directly near the rhizosphere soil of watermelon seedlings. Unlike root dipping, this would cause lesser disturbances to the root, and potentially mimics the processes within the soil where the pathogen will migrate towards the host root. Once symptoms were visible, stem tissue was collected to re-culture FON2 and candidate pathobiome. It should be noted that in regular plant pathology protocols, the stem tissues are placed on media containing antibiotics (generally streptomycin) to prevent bacterial growth. However, we avoided the use of antibiotics to ensure that we do obtain the co-occurring bacteria that will ride the pathogen’s hypha. The diseased plant tissue is the major source of the pathogen in fusarium wilt situations [47] and hence this we recognize to represent the pathobiome relevant to the disease.

In our experiments, the pathogen’s hypha and the bacteria that ride them grew out from the infected tissues. Microscopic examination revealed that the bacterial microbiota had a preferential association towards the hyphal tip. Further results from 16S rRNA sequencing identified prominent OTUs that may have an underlying function within the hyphosphere microbiome. Simpson evenness (S3) and Shannon index (Fig. 2A) were significantly different among the different isolates whereas Chao diversity was relatively unaltered. We envisage that the differences in evenness is driven by the most abundant OTUs of each of the three hyphosphere, that make up more than 80% of the relative abundances. This was tested via ANOVA and Tukey HSD analysis revealing their abundance is significantly different within each isolate. This overabundance of particular OTU was seen to affect other aspects of our results, such as the co-occurrence network analysis and the correlation plots. The co-occurrence analysis visualized a negative correlation with the dominant OTUs from each isolate. This same interaction was reconfirmed in the correlation plot as the majority of bacteria had a negative correlation to the three major OTUs. Although this may be due to an overabundance of these OTUs in comparison to the rest of the microbiota this may be due to a positive correlation to the fusaria that we did not test.

These particular highly abundant OTUs are significantly associated with a single FON2 i.e COM with Otu002, W2 with Otu0003 and HAR with Otu0001 as also seen in our LefSe analyses (Figure.1C). These significant results suggest that there is a fungal-isolate to bacterial-strain interaction occurring. Further supporting results are from the ternary map visualizing a similar abundance of central core hyphosphere microbiota associated with the three isolates except for one dominant OTU strongly associated with a particular isolate. This pattern was demonstrated once more in the heatmap, as the most abundant OTUs were present within a particular FON2 isolates hyphosphere and was consistent throughout the isolate’s samples regardless of repetition from re-culturing and differences in host fungi’s genetic backgrounds [27].

The PiCRUSt tool predicted several potential pathways related to plant defense, development, and antimicrobial functions. Of the identified EC are pathways associated with plant defense against plant pathogens results identified ubiquinone and other terpenoid-quinone biosynthesis [48, 49], sesquiterpenoid and triterpenoid biosynthesis [50, 51] and carotenoid biosynthesis[52]. Other plant defense associated pathways identified within the hyphosphere included zeatin biosynthesis, this is of interest in plant pathology as recent study of Plasmodiophora brassicae to convert adenine to trans-zeatin to produce cytokinin to assist in gall formation and disease success [53, 54] how this may be beneficial within the pathobiome is unknown and should be researched further. Lysine degradation was predicted within the hyphosphere microbiome, previous literature associated the addition of cadaverine affected lysine degradation and reduced biofilm formation with the opportunistic pathogen Pseudomonas aeruginosa and significantly increase planktonic movement [55]. Furthermore, lysine degradation has been identified in fungal plant pathogen Ustilago maydis but the mechanism for pathogen success warrants further research [56, 57]. Streptomycin biosynthesis was also identified within the hyphosphere, this is of particular importance as previous research of the Fusarium oxysporum f.sp. cubense was studied to have endohyphal bacteria that were difficult to remove with antibiotics (ampicillin and chloramphenicol), therefore antimicrobial resistance may be occurring physically with a hyphal barrier or from genetic resistance [58].

Results found that the core microbiota was dominated by the order Burkholderiales (i.e OTU001 Oxalobacteraceae Naxibacter, OTU002 Comamndaceae Delftia and OTU003 Bur

kholderia Cupriavidus). Strains from this order have been previously shown to occur in hyphosphere as well. Burkholderia terrae BS001, from hyphosphere of Lyophyllum, was observed to move along with hyphal growth but also developed biofilms around the hyphae [7]. This strain was also found to use type III secretion system (T3SS) for hyphal adhesion [59]. The bacteria utilized this relationship to exchange nutrients and impart protection against antifungal agents [60]. It remains to be further explored whether the hyphal riding bacteria that are observed in our study can also impart fungicide resistance to fusaria. Indirect evidence exists where [61] a strain of Burkholderia terrae BS001 that moves along the hypha of two Fusarium oxysporum isolates (Fo47 and Foln3) can also impart fungicide resistance to these strains of Fusaria.

Other hyphal associated bacteria that were observed as part of the FON’s pathobiome was Oxalobacteraceae that has been studied to associate with arbuscular mycorrhizae and also colonize the hyphae of Pythium aphanidermatum[62, 63]. Especially, the authors noted the hyphosphere isolation was associated with a virulent phase of Pythium on seeds. Therefore, we suggest from our results and previous research support Oxalobacteraceae Naxibacter sp., Comamonadaceae Delftia sp. and Burkholderiaceae Cupriavidus sp. to be organisms associated with the hyphosphere, i.e hyphal riders.

Other than the most abundant OTUs (OTU001, OTU002 and OTU003), OTU004 (Comamonadaceae Curvibacter) was noted as the most abundant OTU (Fig. 5C) is one of the most abundant OTUs and is present in majority of our samples. This particular OTU in the network analysis (Fig. 7) was near the cluster of the dominant OTUs and therefore, it is likely that a primary group acts as the leading pathobiome community and a secondary group that is less abundant but would take over the primary group's functions, as expected with functional redundancy [64, 65]. In summary, to maintain the pathobiome function within the hyphosphere we propose OTU001, OTU002 and OTU003 is the primary group and a secondary group including OTU004. This furthermore indicates that the hyphosphere pathobiome is a functioning community that likely has a role in virulence, and therefore warrants more research.

Plant pathogens and their interactions within the pathobiome is a relatively new concept [17]. Although this project has identified the shared core microbiota within the hyphosphere, research is necessary to investigate if other FOSC also have a consistent hyphal-associated microbiota. Further research is needed to differentiate the pathobiome community from the plant host community under dysbiosis. Other aspects to consider are whether these interactions can be replicated in situ, through the isolation of the core bacteria and reintroducing them to the pathogen. This question of replicating the alliance within the pathobiome should be tested with other FON2 isolates of to observe if strain to isolate interactions results from this study are reproducible. Finally, these interactions should be profiled for their metabolites. Through metabolomic profiling the presence and absence of the hyphosphere bacterial microbiota may assist in describing their roles within the pathobiome and within the host infection process.

Our study establishes the core bacteriome associated with the hypha of pathogenic FON2. This pathobiome stays consistent through subcultures. The community structure by beta-diversity did not separate the communities obtained from the three different strains. All these points to that there is a defined hyphosphere pathobiome of FON2. The use of hyphosphere as a proxy to pathobiome provides future researchers with a tool to specifically bait out the pathobiome assemblage. Such a tool will enable understanding the pathogen’s lifestyle and the processes in soil microniches that lead to infection.

In this study, we characterize the core microbiota associated with the hyphal tip of plant pathogen FON2. Closely related members of our core microbiota groups have been previously researched as hyphal-associated and now present new members of Burkholderia, Bacili and Pseudomonas to be associated with hyphal interactions. As this specialized niche is the point of entry for root infection, this microbiota may contribute to essential functions of the pathobiome and may assist in disease success. Future studies that investigate the role of pathogen-associated microbiota associated with host infections are required. Results from our study emphasize the importance of the pathobiome paradigm, that pathogens are no longer characterized as lone invaders but instead recruit and manipulate the hosts' established microbiota to form a pathobiome.

Funding: United States Department of Agriculture HATCH 1021870 project number: TEX09714. United States Department of Agriculture -NIFA grant from the Agricultural Microbiomes program, Proposal 2022-11111, Grant 2023-67013-40174.

Data availability

Sequences were submitted to the SRA (bioproject PRJNA1032089).

Authors contribution

The research plan for this project was constructed based on several conversations led by Sanjay Antony-Babu (SAB) with Vanessa Thomas. SAB contributed as a mentor within the project and editing of the manuscript. Vanessa Thomas carried out the experiments, analyzed the data and wrote the manuscript.

USDA-NIFA grant 2022-11111 from the Agricultural Microbiomes program.

Acknoldgements

We would like to acknowledge Dr. Tom Isakeit for his assistance and mentorship in growing watermelon and all things related to Fusaria.

Deveau A, Bonito G, Uehling J, Paoletti M, Becker M, Bindschedler S, et al. Bacterial–fungal interactions: ecology, mechanisms and challenges. FEMS Microbiol Rev. 2018;42(3):335–52.
Meier IC, Pritchard SG, Brzostek ER, McCormack ML, Phillips RP. The rhizosphere and hyphosphere differ in their impacts on carbon and nitrogen cycling in forests exposed to elevated CO 2. New Phytol. 2015;205(3):1164–74.
Frey-Klett P, Burlinson P, Deveau A, Barret M, Tarkka M, Sarniguet A. Bacterial-fungal interactions: hyphens between agricultural, clinical, environmental, and food microbiologists. Microbiol Mol Biol Rev. 2011;75(4):583–609.
Miquel Guennoc C, Rose C, Labbé J, Deveau A. Bacterial biofilm formation on the hyphae of ectomycorrhizal fungi: a widespread ability under controls? FEMS Microbiol Ecol. 2018;94(7):fiy093.
Kohlmeier S, Smits TH, Ford RM, Keel C, Harms H, Wick LY. Taking the fungal highway: mobilization of pollutant-degrading bacteria by fungi. Environ Sci Technol. 2005;39(12):4640–6.
Jansa J, Hodge A. Swimming, gliding, or hyphal riding? On microbial migration along the arbuscular mycorrhizal hyphal highway and functional consequences thereof. New Phytol. 2021;230(1):14–6.
Warmink J, Van Elsas J. Migratory response of soil bacteria to Lyophyllum sp. strain Karsten in soil microcosms. Appl Environ Microbiol. 2009;75(9):2820–30.
Deveau A, Palin B, Delaruelle C, Peter M, Kohler A, Pierrat J-C, et al. The mycorrhiza helper Pseudomonas fluorescens BBc6R8 has a specific priming effect on the growth, morphology and gene expression of the ectomycorrhizal fungus Laccaria bicolor S238N. New Phytol. 2007;175(4):743–55.
Wang Y-H, Dai Y, Kong W-L, Zhu M-L, Wu X-Q. Improvement of Sphaeropsis Shoot Blight Disease Resistance by Applying the Ectomycorrhizal Fungus Hymenochaete sp. Rl and Mycorrhizal Helper Bacterium Bacillus pumilus HR10 to Pinus thunbergii. Phytopathology®. 2022;112(6):1226–34.
Antony-Babu S, Deveau A, Van Nostrand JD, Zhou J, Le Tacon F, Robin C, et al. Black truffle‐associated bacterial communities during the development and maturation of T uber melanosporum ascocarps and putative functional roles. Environ Microbiol. 2014;16(9):2831–47.
Deveau A, Antony-Babu S, Le Tacon F, Robin C, Frey-Klett P, Uroz S. Temporal changes of bacterial communities in the Tuber melanosporum ectomycorrhizosphere during ascocarp development. Mycorrhiza. 2016;26(5):389–99.
Barbieri E, Ceccaroli P, Saltarelli R, Guidi C, Potenza L, Basaglia M, et al. New evidence for nitrogen fixation within the Italian white truffle Tuber magnatum. Fungal biology. 2010;114(11–12):936–42.
Sun R-L, Jing Y-L, de Boer W, Guo R-J, Li S-D. Dominant hyphae-associated bacteria of Fusarium oxysporum f. sp. cucumerinum in different cropping systems and insight into their functions. Appl Soil Ecol. 2021;165:103977.
Newton A, Toth I. Helper bacteria and pathogenicity assessments. New Phytol. 1999;144(3):385–6.
Arora P, Riyaz-Ul-Hassan S. Endohyphal bacteria; the prokaryotic modulators of host fungal biology. Fungal Biology Reviews. 2019;33(1):72–81.
Obasa K, White FF, Fellers J, Kennelly M, Liu S, Katz B, et al. A dimorphic and virulence-enhancing endosymbiont bacterium discovered in Rhizoctonia solani. Phytobiomes. 2017;1(1):14–23.
Vayssier-Taussat M, Albina E, Citti C, Cosson J-F, Jacques M-A, Lebrun M-H, et al. Shifting the paradigm from pathogens to pathobiome: new concepts in the light of meta-omics. Front Cell Infect Microbiol. 2014;4:29.
Mannaa M, Seo Y-S. Plants under the attack of allies: Moving towards the plant pathobiome paradigm. Plants. 2021;10(1):125.
Brader G, Compant S, Vescio K, Mitter B, Trognitz F, Ma L-J, et al. Ecology and genomic insights into plant-pathogenic and plant-nonpathogenic endophytes. Annu Rev Phytopathol. 2017;55:61–83.
Nemchinov LG, Irish BM, Uschapovsky IV, Grinstead S, Shao J, Postnikova OA. Composition of the alfalfa pathobiome in commercial fields. Front Microbiol. 2023;14.
Martyn R. Fusarium wilt of watermelon: a historical review. In: Cucurbitaceae 2012 Proceedings of the Xth EUCARPIA Meeting on Genetics and Breeding of Cucurbitaceae, Antalya, Turkey, 15–18 October, 2012. University of Cukurova, Ziraat Fakultesi; 2012: 136 – 56.
Ma LJ, van der Does HC, Borkovich KA, Coleman JJ, Daboussi MJ, Di Pietro A, et al. Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature. 2010;464(7287):367–73. 10.1038/nature08850.
van Dam P, Fokkens L, Schmidt SM, Linmans JH, Kistler HC, Ma LJ, et al. Effector profiles distinguish formae speciales of Fusarium oxysporum. Environ Microbiol. 2016;18(11):4087–102. 10.1111/1462-2920.13445.
Amaradasa B, Beckham K, Dufault N, Sanchez T, Ertek T, Iriarte F, et al. First report of Fusarium oxysporum f. sp. niveum race 3 causing wilt of watermelon in Florida, USA. Plant Dis. 2018;102(5):1029.
Egel D, Harikrishnan R, Martyn R. First report of Fusarium oxysporum f. sp. niveum race 2 as causal agent of Fusarium wilt of watermelon in Indiana. Plant Dis. 2005;89(1):108.
Hoagland DR, Arnon DI. The water-culture method for growing plants without soil. Circular California agricultural experiment station. 1950;347(2nd edit).
Thomas VE, Isakeit T, Antony Babu S. Draft genomes of five Fusarium oxysporum f. sp. niveum strains isolated from infected watermelon from Texas with temporal and spatial differences. PhytoFrontiers. 2022;(ja).
White TJ, Bruns T, Lee S, Taylor J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. PCR protocols: a guide to methods and applications. 1990;18(1):315–22.
Lane D. Nucleic acid sequencing techniques in bacterial systematics. John Wiley and Sons; 1991.
Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, et al. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41(1):e1–e.
Schloss PD, Gevers D, Westcott SL. Reducing the effects of PCR amplification and sequencing artifacts on 16S rRNA-based studies. PLoS ONE. 2011;6(12):e27310.
Kozich JJ, Westcott SL, Baxter NT, Highlander SK, Schloss PD. Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Appl Environ Microbiol. 2013;79(17):5112–20.
Wickham H, Chang W, Wickham MH. Package ‘ggplot2’. Create elegant data visualisations using the grammar of graphics Version. 2016;2(1):1–189.
Graves S, Piepho H-P, Selzer ML. Package ‘multcompView’. Visualizations of paired comparisons. 2015.
Liu C, Cui Y, Li X, Yao M. microeco: an R package for data mining in microbial community ecology. FEMS Microbiol Ecol. 2021;97(2):fiaa255.
Bastian M, Heymann S, Jacomy M. Gephi: an open source software for exploring and manipulating networks. In: Proceedings of the international AAAI conference on web and social media. vol. 3; 2009: 361-2.
Tamura K, Stecher G, Kumar S. MEGA11: molecular evolutionary genetics analysis version 11. Mol Biol Evol. 2021;38(7):3022–7.
Letunic I, Bork P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res. 2021;49(W1):W293–W6.
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(16):4027–9.
Hamilton N, Hamilton MN. Package ‘ggtern’. 2022.
McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013;8(4):e61217.
Wei T, Simko V, Levy M, Xie Y, Jin Y, Zemla J. Package ‘corrplot’. Statistician. 2017;56(316):e24.
Douglas GM, Maffei VJ, Zaneveld J, Yurgel SN, Brown JR, Taylor CM et al. PICRUSt2: An improved and extensible approach for metagenome inference. BioRxiv. 2019:672295.
Kanehisa M, Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28(1):27–30.
Kassambara A. Pipe-friendly framework for basic statistical tests [R Package Rstatix Version 0.7. 0]. Boston, MA, USA: Free Software Foundation Inc; 2021.
Jochum L, Stecher B. Label or concept–what is a pathobiont? Trends Microbiol. 2020;28(10):789–92.
McMullen MP, Stack RW. Effects of isolation techniques and media on the differential isolation of Fusarium species. Phytopathology. 1983;73(3):458.
Zhao L, Shu Y, Liang L, Wang Y, Godana EA, Zhang X, et al. Integrated transcriptomic and metabonomic analysis reveal mechanisms of disease resistance in apples induced by Wickerhamomyces anomalus. Biol Control. 2022;173:105005.
He L, Wang M, Wang H, Zhao T, Cui K, Zhou L. iTRAQ proteomic analysis of the inhibitory effect of 1, 6-O, O-diacetylbritannilactone on the plant pathogenic oomycete Phytophthora capsici. Pestic Biochem Physiol. 2022;184:105125.
Cárdenas PD, Almeida A, Bak S. Evolution of structural diversity of triterpenoids. Front Plant Sci. 2019;10:1523.
Chadwick M, Trewin H, Gawthrop F, Wagstaff C. Sesquiterpenoids lactones: benefits to plants and people. Int J Mol Sci. 2013;14(6):12780–805.
Boba A, Kulma A, Kostyn K, Starzycki M, Starzycka E, Szopa J. The influence of carotenoid biosynthesis modification on the Fusarium culmorum and Fusarium oxysporum resistance in flax. Physiol Mol Plant Pathol. 2011;76(1):39–47.
Prerostova S, Dobrev PI, Konradyova V, Knirsch V, Gaudinova A, Kramna B, et al. Hormonal responses to Plasmodiophora brassicae infection in Brassica napus cultivars differing in their pathogen resistance. Int J Mol Sci. 2018;19(12):4024.
Dekhuijzen H. The occurrence of free and bound cytokinins in plasmodia of Plasmodiophora brassicae isolated from tissue cultures of clubroots. Plant Cell Rep. 1981;1(1):18–20.
Leggett A, Li D-W, Sindeldecker D, Staats A, Rigel N, Bruschweiler-Li L, et al. Cadaverine is a switch in the lysine degradation pathway in Pseudomonas aeruginosa biofilm identified by untargeted metabolomics. Front Cell Infect Microbiol. 2022;12:833269.
McCann MP, Snetselaar KM. A genome-based analysis of amino acid metabolism in the biotrophic plant pathogen Ustilago maydis. Fungal Genet Biol. 2008;45:77–S87.
Knorr S, Sinn M, Galetskiy D, Williams RM, Wang C, Müller N et al. Widespread bacterial lysine degradation proceeding via glutarate and L-2-hydroxyglutarate. Nature communications. 2018;9(1):5071.
Rahayu G, Maulana I. Endobacterial symbiont of Fusarium oxysporum f. sp. cubense and the pathogenicity of their symbiosis towards banana plantling. In: IOP Conference Series: Earth and Environmental Science. vol. 457: IOP Publishing; 2020: 012051.
Yang P, Zhang M, Warmink JA, Wang M, van Elsas JD. The type three secretion system facilitates migration of Burkholderia terrae BS001 in the mycosphere of two soil-borne fungi. Biol Fertil Soils. 2016;52:1037–46.
Nazir R. The ecological success of Burkholderia terrae BS001 and related strains in the mycosphere. 2012.
Nazir R, Tazetdinova DI, van Elsas JD. Burkholderia terrae BS001 migrates proficiently with diverse fungal hosts through soil and provides protection from antifungal agents. Front Microbiol. 2014;5:598.
Scheublin TR, Sanders IR, Keel C, Van Der Meer JR. Characterisation of microbial communities colonising the hyphal surfaces of arbuscular mycorrhizal fungi. ISME J. 2010;4(6):752–63.
Ofek M, Hadar Y, Minz D. Ecology of root colonizing Massilia (Oxalobacteraceae). PLoS ONE. 2012;7(7):e40117.
Rosenfeld JS. Functional redundancy in ecology and conservation. Oikos. 2002;98(1):156–62.
Lawton JH, Brown VK, Schulze E-D, Mooney HA. Biodiversity and ecosystem function. Biodivers Ecosyst Function. 1993:255–70.

No competing interests reported.

Supplementarymaterial.docx

Download PDF

Journal Publication

published 09 Mar, 2024

Read the published version in Environmental Microbiome →

Editorial decision: Revision requested
08 Jan, 2024
Reviews received at journal
27 Dec, 2023
Reviewers agreed at journal
16 Dec, 2023
Reviewers agreed at journal
07 Dec, 2023
Reviewers invited by journal
07 Dec, 2023
Editor assigned by journal
07 Dec, 2023
Submission checks completed at journal
02 Dec, 2023
First submitted to journal
01 Dec, 2023

You are reading this latest preprint version

Core hyphosphere microbiota of Fusarium oxysporum f. sp. niveum

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Methods

Seed and Soil Sterilization:

FON2 inoculum preparation:

Statistical Analysis

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1