Downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere

Downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere Jelle Spooren, Yadong Shao, Tilda Tarrant, Hannah Ploehmacher, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6693507/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Plants can respond to pathogen attack by assembling disease-suppressive soil microbiomes. In Arabidopsis thaliana , infection by the obligate foliar downy mildew pathogen Hyaloperonospora arabidopsidis (Hpa) consistently led to the formation of a soil microbial community, termed the soilborne legacy (SBL), that enhanced resistance in subsequent plant populations grown in the same soil. Previous work identified an enrichment of specific Hpa-associated microbiota (HAM) in the phyllospheres of infected plants, which suppressed pathogen proliferation. However, the relationship between rhizosphere and phyllosphere microbiota in generating the SBL and assembling protective HAM remained unclear. Here, we identified a community of 25 core-HAM that consistently dominated the phyllospheres of 14 sets of distinct Hpa-infected plant populations across six independent experiments. Using HAM-free, gnotobiotic Hpa spores, the infection-driven assembly of the core-HAM member Sphingobium ASV ed6be was recapitulated, showing de novo and progressive accumulation under sustained disease pressure. Although HAM transmission in SBL occurred via soil, these bacteria were shown to be phyllosphere specialists, accumulating more abundantly on aboveground than belowground tissues. Moreover, leaf wash-offs from plant populations that inherited SBL, effectively suppressed downy mildew disease when applied to leaves of plants grown in unconditioned soil. These findings reveal that downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere, highlighting a crucial link between belowground and aboveground plant-driven microbiome assembly processes. Paradoxically, the phyllosphere thus emerges as a central hub for the accumulation of disease-suppressive soil microbiomes. Biological sciences/Plant sciences/Plant symbiosis/Parasitism Biological sciences/Microbiology Biological sciences/Ecology/Microbial ecology Biological sciences/Plant sciences/Plant immunity/Microbe Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction Plants host diverse and complex microbiomes comprised of pathogenic, commensal, and beneficial microbes 1 . Different plant tissues provide unique niches for microbial colonization 2 , 3 . In the rhizosphere, the zone of soil surrounding roots, plants exude metabolites that selectively stimulate or inhibit distinct microbes 4 , creating a nutrient-rich yet selective environment 5 . Some rhizosphere microbes may enter the root endosphere, but must overcome additional selective pressures from root metabolites, structural root barriers, and plant immune responses 3 , 6 – 8 . Aboveground, the phyllosphere presents a relatively nutrient-poor habitat where microbes cluster in hotspot sites that provide moisture, nutrients, or shelter from environmental stressors 3 , 9 – 14 . Despite the distinct nature of these microbiome compartments, there is reciprocal exchange of microbes occurring between aboveground, belowground and inner plant tissues 10 , 15 , 16 . The soil, acting as a diverse microbial reservoir, exerts a major influence on the composition of plant-associated microbiomes in all compartments 16 – 19 . Plants can selectively steer microbial colonization to enhance their health by activating immune responses to stop pathogen infection or by stimulating protective microbiota that can control growth of invading pathogens 14 , 20 – 22 . In response to pathogen attack, plants can selectively enrich their rhizosphere microbiome with disease-suppressive microbiota, which help reduce disease progression 14 , 20 , 23 – 26 . This phenomenon is evident in so-called disease-suppressive soils, where plants stay relatively healthy despite the presence of virulent pathogens. Disease-suppressive soils typically emerge after an initial severe outbreak of disease, demonstrating the plant's capacity to promote protective microbial communities in the soil 21 , 27 – 29 . Prime examples include take-all decline in wheat and Rhizoctonia -suppressive soils in sugar beet, where disease suppression is linked to the buildup of disease-suppressive microbiomes in the rhizosphere and root endosphere, respectively 29 – 32 . The phyllosphere likewise harbors microbiota with disease-suppressive properties that can boost plant immunity 33 , 34 or act through direct microbial antagonism 33 , 35 . As in soil, phyllosphere disease-suppressive microbiomes may emerge in response to pathogen challenge 36 , 37 . We previously discovered that phyllospheres of Arabidopsis thaliana (Arabidopsis) plants infected by the foliar obligate biotrophic downy mildew pathogen Hyaloperonospora arabidopsidis (Hpa) are enriched with a specific group of Hpa-associated microbiota (HAM) 38 . These HAM increase in abundance in the Hpa-infected phyllosphere environment and suppress Hpa spore production, effectively functioning as a phyllosphere disease-suppressive microbiome 38 . It was further demonstrated that Hpa-infected Arabidopsis plants, referred to as a conditioning population, can condition the soil in which they grow, creating a disease-suppressive soilborne legacy (SBL). As result, subsequent “response populations” Arabidopsis plants grown in SBL soil exhibit greater resistance to Hpa compared to plants grown in soil conditioned by healthy plants (control soil) 20 , 38 – 40 . Notably, disease-suppressive HAM are enriched in the rhizosphere and phyllosphere of response population plants growing in SBL soil, even if these plants were not themselves infected by Hpa 38 . Despite these insights, the interplay between the rhizosphere and phyllosphere in plant-driven microbiome assembly, particularly in response to foliar pathogen attack, remains poorly understood. In this study, disease-induced shifts in rhizosphere and phyllosphere microbiome composition were investigated in detail, with a focus on the assembly of HAM in response to the selective regime imposed by downy mildew infection and its role in shaping the SBL. While disease-suppressive soils evidently are often attributed to microbiota in the rhizosphere 27 , 28 , our results demonstrate that the phyllosphere can function as a critical hub where soil-transmitted microbiota accumulate and contribute to the suppression of foliar disease. Results A live soil microbiome is required for the creation of a soilborne legacy Previous work showed that a disease-suppressive soilborne legacy (SBL) is consistently established in soils where wild-type Arabidopsis Col-0 plants, infected with the foliar pathogen Hpa, were grown. These plants had been inoculated with either standard Hpa spore suspensions containing Hpa-associated microbiota (HAM), or with gnotobiotic, HAM-free spore suspensions (gnoHpa) 38 . In contrast, inoculation of transgenic Col-0 RPP5 plants, resistant to Hpa infection, with the same spores did not result in a disease-suppressive SBL. Similarly, mutant Arabidopsis plants deficient in MYB72 and F6’H1 , both essential for coumarin biosynthesis and rhizobacteria-mediated induced systemic resistance 41 , also failed to create a suppressive SBL 40 . These results indicate that both a successful downy mildew infection and subsequent disease-induced plant responses are necessary for the creation of a disease-suppressive SBL, while the mere co-inoculation of HAM bacteria without disease induction is insufficient. The disease-suppressive SBL is thus believed to result from plant-driven infection-induced changes in the soil microbiome, but it has not been formally demonstrated that the disease suppressiveness of these SBL soils is caused by living microbes. To test this, Col-0 plants were grown in three soil types: 1) non-sterile field soil (100% live soil) collected at the Reijerscamp nature reserve which supports an abundant endemic Arabidopsis population 20 , 2) the same soil sterilized by gamma irradiation, and 3) a 1:9 mix of live and sterilized field soil (10% live soil). Fourteen-day-old plants were inoculated with gnoHpa spores or mock-treated and then cultivated to condition the soil. Seven days post inoculation (dpi), when downy mildew disease had visibly manifested through the emergence of sporangiophores, aboveground plant parts were removed and a second population of Col-0 plants (response population) was sown directly in the conditioned soil. These plants were again inoculated with gnoHpa at 14 days, and spore production was quantified 7 dpi. Spore production was significantly reduced in the response population grown in live field soil conditioned by gnoHpa-inoculated plants, compared to those grown in live field soil conditioned by healthy plants (Fig. 1 A). This disease-suppressive effect was abolished in sterilized soil, but restored by supplementation with 10% live soil, confirming a microbial origin for the SBL. These findings demonstrate that foliar Hpa infection and the resulting plant response give rise to a persistent microbial community in the soil that enhances resistance in subsequent plant populations. A core set of HAM is consistently enriched in the Hpa-infected phyllosphere We previously also showed that disease-suppressive HAM are significantly enriched in response plant populations germinating and growing in SBL soil, even when the conditioning plant population had been inoculated with HAM-free gnoHpa spores 38 . Based on this, we hypothesized that HAM, originally identified in the phyllosphere 38 , are causative agents of disease-suppression in the SBL. We reasoned that these microbiota could accumulate around the root systems of (gno)Hpa-infected conditioning population plants and be subsequently acquired by new plantings that germinate in the same soil. However, not all HAM that are identified as significantly enriched in Hpa-infected phyllospheres are consistently detected across experiments, suggesting some degree of context-dependent selection. To identify core-HAM that are reproducibly enriched in the Hpa-infected phyllosphere, phyllosphere-derived amplicon sequencing data of 14 Hpa-infected plant populations were analyzed. These Arabidopsis plant populations had been inoculated with Hpa strains Noco2 or Cala2 and grown in a river sand-potting soil mixture, or in live Reijerscamp field soil for which the creation of an Hpa - suppressive SBL had been confirmed. These 14 plant populations were part of six independent experiments performed over a span of five years 38 . A total of 25 amplicon sequence variants (ASVs) representing distinct HAM were significantly enriched ( P < 0.05, DESeq2, Table S1 ) in more than 8 of the 14 Hpa-infected phyllospheres and were designated as ‘core-HAM’. These 25 core-HAM ASVs accounted for 46% − 75% of the bacterial abundance in the Hpa-inoculated phyllospheres, but were generally low in abundance or undetectable in healthy plants (Fig. 1 B, Fig. S1 A). Among them, Xanthomonas ASV a0e1a and Acidovorax ASV a4065 were enriched in all tested Hpa-infected phyllospheres, occupying up to 27% relative abundance of the phyllosphere bacterial community (Fig. 1 B). Xanthomonas ASV a0e1a was the most abundant core-HAM ASV, averaging 10% of bacterial reads from Hpa-infected leaves. Additionally, Chryseobacterium ASV 3a0c1, Flavobacterium ASV ef66d and Methylophilus ASV e50db were enriched in 13 out of 14 Hpa-infected phyllospheres, while Agromyces ASV efbd0, Pedobacter ASV f2a1b and Sphingobium ASV ed6be were enriched in 12 out of 14 Hpa - infected phyllospheres, respectively (Table S1 , Fig. S1 A). These eight highly consistent ASVs together accounted for 49% relative abundance in infected leaves, while remaining low abundant (< 1%) in healthy phyllospheres (Fig. S1 A). These results suggest that core-HAM are natural leaf colonizers strongly promoted by the Hpa-infected phyllosphere environment. The 25 ASVs representing the core-HAM cover a broad taxonomic diversity (Fig. S1 B), yet their abundance and taxonomic distribution remained remarkably stable across independent experiments conducted over a span of more than five years (Fig. 1 C). This highlights the robustness of core-HAM recruitment and persistence in Hpa-infected leaves. Downy mildew-infected phyllospheres selectively accumulate core-HAM Given the consistent enrichment of core-HAM in Hpa-infected phyllospheres, we hypothesized that these bacteria are selectively promoted during Hpa infection and gradually accumulate across successive infected plant populations. To test this, an initial population of Arabidopsis Col-0 plants was grown in live Reijerscamp field soil and inoculated with HAM-free gnoHpa spores. Leaf wash-offs, containing spores and associated microbiota, were then passaged to newly grown Col-0 plants in fresh live Reijerscamp soil. This procedure was repeated every seven days, coinciding with visible Hpa sporulation, to generate five consecutive plant populations (gnoHpa lineages; Fig. S2). Control lineages were started by spraying the first population of plants with regular HAM-containing Hpa spore suspensions (Hpa lineages) or with a mock treatment of sterile water (uninfected lineage). For each population, phyllosphere material was sampled from the passaged lineages and from newly grown untreated plants, and DNA was extracted for 16S amplicon sequencing. Phyllosphere microbiome compositions were analyzed using Principal Coordinate Analysis (PCoA) based on Bray-Curtis dissimilarities (Fig. 2 A). Consistent with previous findings 38 , no significant differences in phyllosphere microbiome composition were observed between untreated, uninfected, and gnoHpa-inoculated plants in the first population. However, plants inoculated with HAM-containing Hpa spore suspensions exhibited a significantly distinct microbiome compared to all other lineages (PERMANOVA results are detailed in Fig. S3 and Table S2). In later populations, however, the phyllosphere microbiomes of the passaged uninfected, gnoHpa, and Hpa lineages all showed significant differentiation from those of untreated plants (Fig. S3, Table S2, indicating that the transfer of phyllosphere microbial communities across successive plant populations influenced microbiome composition. By the third population, gnoHpa and uninfected lineages also diverged significantly, with differences becoming more pronounced in the fourth and fifth populations (Fig. 2 A-B, Table S2). These findings indicate that downy mildew-infected plants progressively assemble distinct phyllosphere microbiomes over time. To identify ASVs selectively enriched in the gnoHpa lineages, we applied three filters: 1) consistent enrichment in the gnoHpa versus uninfected lineages (75 ASVs, P adj < 0.05, DESeq2 or ANCOM-BC, Table S3), 2) accumulation within the gnoHpa lineages over time but not in uninfected lineages (30 ASVs, P adj < 0.05, DESeq2 or ANCOM-BC, Table S4), and 3) positive correlation with pathogen levels, as quantified by qPCR (23 ASVs, P adj < 0.05, spearman, Table S5). We identified 12 ASVs that satisfied all three selection criteria (Fig. 2 C, Tables S3-5). While these promoted ASVs were sporadically detected in uninfected samples, they persisted and remained abundant only in the phyllospheres of downy mildew-infected lineages after their initial appearance (Fig. S4). Their combined relative abundance increased progressively during the passages in the gnoHpa lineages and accumulated to a significantly higher cumulative relative abundance in the gnoHpa lineages compared to all other lineages (Fig. 2 D). To assess whether these changes in relative abundance reflected shifts in absolute bacterial load, DNA from Salinibacter ruber , typically absent from soil and plant samples 42 , was spiked into samples for normalization. Collectively, the 12 ASVs enriched in gnoHpa lineages were already present in relatively high numbers in population 1 of the Hpa lineages (Fig. S5), suggesting that here they constitute HAM members introduced via Hpa inoculation. Remarkably, their absolute abundance in the gnoHpa lineages rose over successive passages, reaching levels comparable to those in the Hpa lineages and remaining significantly higher than in uninfected lineages (Fig. S5). These 12 downy mildew-associated ASVs thus appear to be sporadically occurring phyllosphere colonizers that specifically benefit from the downy mildew infected environment and progressively build up in the phyllospheres of infected plant populations. Interestingly, all 12 ASVs that accumulated in the gnoHpa lineages match genera that were previously demonstrated to be downy mildew associated 38 , 43 . At the ASV level, Sphingomonas ASV f359d and Brevundimonas ASV 2cd30 have previously been found enriched in Peronospora effusa -infected spinach leaves 43 . Notably, the most abundant of the 12 downy mildew-associated ASVs, Sphingobium ASV ed6be, is part of the core-HAM (Fig. 1 B-C) and had also been found associated with P. effusa 43 . This ASV increased from 0.01–8% relative abundance between population 1 and 5 of the gnoHpa lineages (Fig. 2 D). We further quantified the absolute abundance of Sphingobium ASV ed6be as 16S rRNA gene copies per gram of leaf tissue (Fig. 2 E). In the Hpa lineages, ASV ed6be was abundant from the start (2.7 × 10⁷ copies/g) and increased steadily with successive passages, reaching 1.5 × 10⁸ copies/g in population 5. In the gnoHpa lineages, the prevalence of ASV ed6be rose across passages, being detected in 2, 3, 7, 8, and 10 out of 12 lineages from population 1 through 5, respectively. Its average abundance increased 2,556-fold, from 3.3 × 10⁴ to 8.5 × 10⁷ copies/g. In contrast, ASV ed6be was only sporadically detected in uninfected lineages (1–4 out of 12 lineages per population) and remained low in abundance. Thus, while this ASV can occur in uninfected phyllospheres, it persists and proliferates specifically under the selective regime imposed by downy mildew infection. Collectively, these data demonstrate that among the diverse microbiota capable of colonizing plants grown in live Reijerscamp field soil, distinct downy mildew-associated bacteria including core-HAM become consistently assembled in the phyllosphere of downy mildew-infected plants. Core-HAM are phyllosphere specialists that are inherited as soilborne legacy A live soil microbiome is required for the creation of a SBL by Hpa-infected plants (Fig. 1 A), and previous findings have suggested that HAM bacteria are assembled in both the rhizosphere and phyllosphere of response population plants grown in SBL soil 38 . This supports the idea that core-HAM originate from the microbial community in the SBL soil conditioned by (gno)Hpa-infected plants and subsequently form a disease-suppressive microbiome on a next generation of plant hosts. The root endosphere may serve as a conduit for microbial migration between belowground and aboveground habitats 15 . However, the connection between core-HAM colonization in rhizosphere, root endosphere, and phyllosphere of Hpa-infected plants and their establishment in successive plant populations grown in SBL soil remains unclear. To trace the origin and spatiotemporal colonization of core-HAM in Col-0 plants grown in the disease-suppressive SBL soil, we generated a detailed map of the microbiota communities in unplanted bulk soil, rhizosphere, root endosphere, and phyllosphere in both conditioning and response plant populations in a SBL experiment (Fig. S6). Bacterial community compositions in all microbiome compartments were characterized using 16S amplicon sequencing, enabling comparisons of identity and abundance across spatial compartments and time. Microbiomes in bulk soil, rhizosphere, root endosphere, and phyllosphere were significantly distinct, confirming successful separation of these microbiome compartments (PERMANOVA results are detailed in Tables S6-S7). In total, we detected 9953 different ASVs (Fig. S7), of which 9731 were detected in unplanted bulk soil samples. Only 222 ASVs were either absent or remained below the detection limit in bulk soil but appeared in plant-associated compartments. Of the ASVs detected in rhizosphere, root endosphere and phyllosphere, respectively, 99%, 97%, and 94%, were also detected in unplanted bulk soil. These ASVs cumulatively accounted for over 98% of the relative abundance in the rhizosphere and root endosphere, and 83% in the phyllosphere. Remarkably, even in the phyllosphere, 98% of the community’s relative abundance consisted of bacteria also detected belowground in the bulk soil, rhizosphere, or root endosphere. This finding supports the notion that the soil acts as a microbial reservoir for the assembly of plant-associated microbial communities, including those in the aboveground phyllosphere environment 16 , 17 , 19 . We then assigned each of the 9953 detected ASVs to the compartment in which it reached its highest relative abundance, resulting in 3936 “bulk soil” ASVs, 4119 “rhizosphere” ASVs, 1553 “root endosphere” ASVs, and 345 “phyllosphere” ASVs. As expected, the ASVs with high relative abundance in the bulk soil and rhizosphere categories were numerous and taxonomically diverse, with seven bacterial phyla representing more than 1% relative abundance (Fig. S8). In contrast, only a select number of ASVs in the root endosphere and phyllosphere reached the same threshold of 1% relative abundance, suggesting strong niche specialization. In this light, Fig. 3 A shows that the root endosphere, and especially the phyllosphere compartment, favor the growth of a select group of microbiota. For example, the 345 “phyllosphere” ASVs, which had the highest relative abundance in the phyllosphere microbiome compartment, represent only 12% of the total number of ASVs detected in the phyllosphere but accounted for 85% of the total relative abundance in this compartment. In contrast, the cumulative relative abundance of all “phyllosphere” ASVs combined in the bulk soil was below 1%. These results demonstrate that microbiota originating from the bulk soil thrive best in distinct ecological niches provided by the plant, with strong selective effects observed in the root endosphere and the phyllosphere. We next investigated whether the effect of downy mildew infection on microbiome composition was also compartment specific (Fig. 3 B, Table S8). Consistent with earlier findings, Hpa infection significantly altered rhizosphere 40 and phyllosphere 38 communities in conditioning population plants (Fig. 2 ). No significant effects were observed in bulk soil or root endosphere. Also in response populations, non-infected plants grown in disease-suppressive SBL soil exhibited distinct rhizosphere and phyllosphere microbiomes compared to those grown in control soil conditioned by mock-treated plants. The effect of SBL on microbiome composition was modest in the rhizosphere of response population plants ( R 2 = 0.081 in PERMANOVA) but more pronounced in the phyllosphere ( R 2 = 0.35 in PERMANOVA). This paradoxically suggests that a belowground microbial legacy most strongly affects the aboveground microbiome of a subsequent planting. Of the 25 core-HAM ASVs identified earlier (Fig. 1 B-C), 21 were also detected in the experiment described in Fig. 3 . These 21 ASVs were all detected in the phyllosphere and 16 were also detected in the belowground plant compartments but only 8 were detected in bulk soil (Table S9). This either suggests that the majority of core-HAMs have abundances below the detection limit in bulk soil and that their competitive colonization is particularly favored on or within the plant host, or that they originate from alternative sources. The collective abundance of the 21 core-HAM ASVs was low in the bulk soil, rhizosphere, and root endosphere (< 1.0%). Cumulative core-HAM ASV relative abundances significantly increased in both the rhizospheres of Hpa-infected conditioning population plants and of mock-treated response population plants grown in SBL soil, but not in the bulk soil or root endosphere of these plants (Fig. 3 C). All 21 core-HAM ASVs reached the highest relative abundances in the phyllosphere. Consistent with previous findings, they accounted for a high relative abundance (73%) in the Hpa-inoculated phyllosphere of conditioning population plants compared to mock-inoculated control plants (9%, Fig. 3 C). Interestingly, even in healthy response plants grown in SBL soil, the relative abundance of the 21 core-HAM ASVs was significantly higher in the phyllosphere of plants grown in disease-suppressive SBL soil (61%) compared to plants grown in control soil conditioned by healthy plants (27%; Fig. 3 C). This shows that the microbial SBL created by Hpa-infected plants drives a robust shift in the phyllosphere microbiome of subsequent plantings, that become dominated by core-HAM. To determine whether this enrichment reflected increased bacterial population densities or displacement of phyllosphere resident microbiota, we quantified absolute abundances in each microbiome compartment using spiked-in S. ruber DNA. Hpa infection of conditioning population plants significantly increased total bacterial load in both phyllosphere and rhizosphere (Fig. S9). Remarkably, also response population plants grown in SBL soil had significantly higher bacterial loads than those in control soil, but only in the phyllosphere. This increase corresponded with elevated absolute abundance of core-HAM, while the absolute abundance of other ASVs remained unchanged (Fig. S9-S10). These results confirm that plants grown in SBL soil are primarily affected aboveground, where their phyllospheres become more densely colonized, especially by disease-suppressive HAM. Downy mildew infection enriches a soilborne core-HAM isolate specifically in the phyllosphere Previous work showed that HAM ASVs are enriched in both the rhizosphere and phyllosphere of plants grown in disease-suppressive SBL soil and that their assembly is promoted by downy mildew infection 38 . However, although these ASVs originate from soil and contribute to the disease-suppressive legacy, they paradoxically accumulate most prominently in the phyllosphere of plants grown in SBL soil (Fig. 3 ). This observation led us to hypothesize that the disease-induced assembly of core-HAM, which appears to be causal to the creation of downy mildew disease-suppressive soils, is initiated in the phyllosphere rather than the rhizosphere. Downy mildew infection coincides with an increased bacterial load in the phyllosphere (Fig. 2 E, Fig. S9). To confirm this, we quantified bacterial colony forming units (CFU) in the phyllosphere of Arabidopsis Col-0 plants grown in live Reijerscamp field soil that were either mock-treated, or inoculated with HAM-containing Hpa or HAM-free gnoHpa spore suspensions. In both Hpa- and gnoHpa-inoculated plants, bacterial densities were significantly higher than in mock-treated controls (Fig. 4 A), indicating that downy mildew infection, and not just the co-inoculation of HAM, promotes the proliferation of phyllosphere-associated bacteria. To investigate whether core-HAM are specifically promoted in the phyllosphere, we selected Xanthomonas sp. WCS2014-23 which was isolated from the roots of Hpa - infected Arabidopsis plants 20 and is represented by the most abundant and robust core-HAM Xanthomonas ASV a0e1a. Rifampicin-resistant WCS2014-23 was mixed into live Reijerscamp field soil at 10 6 CFU/g soil and Col-0 plants were grown in this soil. Two-weeks-old plant were mock treated or inoculated with gnoHpa, and at 7 dpi densities of WCS2014-23 were quantified in both the rhizosphere and phyllosphere. Despite being introduced via the soil, WCS2014-23 population densities were approximately 1000-fold higher in the phyllosphere than in the rhizosphere (Fig. 4 B-C). Rhizosphere colonization was unaffected by infection status, whereas phyllosphere colonization was significantly higher in gnoHpa-infected plants (Fig. 4 B-C). When co-inoculated directly into the phyllosphere with gnoHpa, WCS2014-23 populations reached densities of 10 8.8 CFU per gram of diseased leaf tissue, compared to 10 7.1 CFU per gram of leaf tissue of healthy plants (Fig. S11). These results demonstrate that while core-HAM Xanthomonas WCS2014-23 originates from the soil, it preferentially colonizes the phyllosphere, where its growth is strongly promoted by downy mildew infection. Accumulation of HAM in the phyllosphere suppresses downy mildew disease Previous work demonstrated that co-inoculation of HAM with gnoHpa suppresses downy mildew spore formation 38 . We thus wondered whether the HAM community that assembles in the phyllosphere of healthy plants grown in disease-suppressive SBL soil provides protection against a subsequent downy mildew infections. To test this, we used our standard SBL setup (Fig. S6), conditioning Reijerscamp field soil with either Hpa- or mock-inoculated Col-0 plants. Healthy Col-0 response populations were then grown in the conditioned soils, allowing HAM-enriched phyllosphere communities to assemble on plants in SBL soil. Microbial leaf wash-offs were collected from both groups, mixed with HAM-free gnoHpa spores, and used to inoculate a third set of Col-0 plants grown in unconditioned live soil. Disease severity was quantified as spore production 7 dpi (Fig. 5 A). Bacterial leaf wash-offs from plants grown in control soil had no impact on downy mildew disease development, as the spore production on plants inoculated with gnoHpa mixed with control leaf wash-off was similar to plants inoculated with gnoHpa spores suspended in sterile water (Fig. 5 B). In contrast, when gnoHpa spores were mixed with the HAM-enriched bacterial leaf wash-offs from plants grown in disease-suppressive SBL soil, the spore production was significantly reduced. This demonstrates that the phyllosphere microbiome assembled in healthy plants grown in disease-suppressive SBL soil confers protection against downy mildew. These findings indicate that the suppressiveness of SBL soil indeed results from the soilborne inheritance of HAM. Remarkably it is the assembly in the phyllosphere of successive plant populations that provides effective protection against downy mildew disease. Discussion Both rhizosphere and phyllosphere microbiomes are crucial to sustain plant health 44 , 45 . It is well established that the assembly of beneficial microbiota belowground can give rise to disease-suppressive soils 14 , 27 , 28 . In such soils, plants are protected against pathogens either by rhizobacteria-mediated production of antibiotics 21 , 30 , competition over scarce resources e.g. iron 46 , 47 , or via the induction of systemic resistance 48 , 49 . While research on disease-suppressive soils has largely focused on their effects against root pathogens 27 , 29 , 32 , disease suppressive soils can also protect from pathogens in aboveground plant parts 14 , 15 , 34 , 38 . Arabidopsis plants infected with the foliar pathogen Hpa create disease-suppressive soil by forming an SBL that enhances resistance to Hpa in successive plant populations grown in the same soil 20 . However, the extent to which aboveground and belowground microbiome compartments respectively contribute to the assembly of a soilborne protective legacy that suppresses foliar downy mildew disease has remained unclear. Here, we demonstrated that downy mildew-infected plants assemble a protective core microbiome in the phyllosphere, and that these disease-suppressive microbiota subsequently form a SBL that is transmitted to the phyllosphere of successive plant populations grown in the same soil. To the best of our knowledge, this is the first indication thatthat the phyllosphere can act as an assembly hub for disease-suppressive soilborne microbiomes. We identified specific core-HAM that were consistently abundant in 14 distinct sets of Hpa-infected Arabidopsis phyllospheres, across six independent experiments conducted over a span of more than five years. Notably, we successfully recapitulated the recruitment of the core-HAM Sphingobium ASV ed6be by inoculating Arabidopsis plants grown in field soil with HAM-free gnoHpa spores and monitoring its buildup during serial passaging of phyllosphere leaf wash-offs. This indicates that downy mildew infection selects for specific and reproducible core-HAM enrichment. Interestingly, in a commercial field, spinach leaves naturally infected by the downy mildew P. effusa were found to be enriched for ASVs identical to the downy mildew-associated ASVs in this study, including core-HAM Sphingobium ASV ed6be 43 . In addition, we showed that although originating from soil, Xanthomonas WCS2014-23, which is representative of the most consistent and abundant core-HAM Xanthomonas ASV a0e1a, predominantly colonizes the phyllosphere where it is boosted upon downy mildew infection. Together, these data suggest that the selective regime imposed by downy mildew infection facilitates the consistent and specific assembly of core-HAM in the phyllosphere. The immensely diverse soil microbiome serves as a key microbial reservoir from which the phyllosphere microbiome is assembled 16 , 18 , 19 . Our results confirm that the majority of phyllosphere microbiota are also present in the soil and are likely to originate from it. However, despite the significant taxonomical and functional overlap between rhizosphere and phyllosphere microbiota 17 , our data suggests that the soilborne phyllosphere microbiota are specialized to thrive in the aboveground microbial habitat. Among these phyllosphere specialists are the core-HAM, which seem particularly well adapted to the selective environment shaped by downy mildew infections. Future studies could focus on characterizing the bacterial traits that determine phyllosphere competence and that are selectively enriched during pathogen attack. Understanding these traits could shed light on the mechanisms by which phyllosphere microbiota are assembled, establish and persist, particularly in pathogen-stressed environments, potentially offering new strategies for biocontrol and sustainable crop protection. Although the core-HAM appear to be initially assembled in the phyllosphere of downy mildew-infected plants (Fig. 2 , Fig. 4 ), we found that these plants create a disease-suppressive soil that transmits the protective core-HAM to the phyllosphere of successive plant populations grown in the same soil. The routes by which core-HAM, initially assembled in the phyllosphere, buildup in the soil and migrate back to the phyllosphere from the disease-suppressive SBL soil remain to be elucidated. The endosphere has been identified as a potential route for bidirectional microbial migration between the rhizosphere and the phyllosphere 15 and an important compartment for microbial legacies 50 . However, our data suggests that the endosphere is of minor importance for the migration of core-HAM. Firstly, only a subset of core-HAM colonizes the root endosphere. Secondly, core-HAM abundances did not increase significantly in either the root endosphere of Hpa-infected plants nor plants grown in the disease-suppressive SBL soil. Thirdly, certain core-HAM that accumulated in the phyllosphere of plants grown in SBL soil were not detected in the endophytic compartment. This suggests that core-HAM likely migrate through alternative routes. The previous findings that fully resistant Col-0 RPP5 plants 38 and plants disturbed in the biosynthesis of coumarins 40 do not create the disease-suppressive SBL when inoculated with HAM-containing Hpa spore suspensions indicate that disease-induced plant responses control the assembly, migration or persistence of the core-HAM as SBL. Whether the accumulation of core-HAM in the phyllosphere of plants grown in SBL soils is similarly driven by the plant, or depends on priority effects 51 – 53 of the soilborne core-HAM that lift on and preferentially colonize the new plants shoots as they emerge 54 , remains to be investigated. While the infection-induced assembly of disease-suppressive microbiota has been separately documented to occur in both the rhizosphere and phyllosphere 20 , 21 , 23 , 24 , 36 , 37 , research on disease-suppressive soils has evidently predominantly concentrated on the rhizosphere due to its direct interface with plant roots. However, our results provide the first evidence of a critical link between belowground and aboveground disease-suppressive microbiome assembly processes with a crucial role of phyllosphere microbiomes in the functioning of downy mildew disease-suppressive soils. Based on our data, we propose that the initial infection-induced assembly of disease-suppressive core-HAM in the phyllosphere is followed by their buildup in soil. These core-HAM are subsequently transmitted through the SBL to the phyllosphere of successive plant populations that are germinated and grown in the disease-suppressive SBL soil. Successive infections and spread of core-HAM, that are easily washed off from leaves 38 , to new plant populations 55 could further enhance core-HAM population densities. We propose that this creates a feed-forward loop leading to core-HAM accumulation and the progressive suppression of downy mildew disease. Thus, the phyllosphere might serve as a crucial distribution hub from which disease-suppressive microbiomes - both rhizosphere and phyllosphere-associated - can disseminate throughout plant populations leading to fieldwide disease suppressiveness. This revelation raises an important question: could the phyllosphere be involved in other types of disease-suppressive soils, including those that are of agricultural relevance? Evidence supports the possibility of taxonomical and functional overlap between microbes that suppress both soilborne and foliar pathogens. For example, beneficial pseudomonads, which are known to antagonize the soilborne pathogen Gaeumannomyces tritici in Take-all decline soils of wheat 30 , 31 , have also been implicated in the suppression of the foliar pathogen Zymoseptoria tritici in the wheat phyllosphere 35 . Similarly, a beneficial Streptomyces sp., initially identified for its ability to suppress Fusarium wilt disease in the rhizosphere of strawberry plants 56 , can migrate bidirectionally throughout the plant endosphere and vasculature bundles, increasing resistance against Botrytis cinerea infections in the strawberry phyllosphere 15 . Thus, the phyllosphere microbiome may indeed be a crucial component of disease-suppressive soils of agricultural relevance that has been largely overlooked in past research and warrants further investigation. Materials and methods Soil preparation and plant growth conditions Field soil was collected at the Reijerscamp nature reserve in the Netherlands where an endemic population of Arabidopsis has been found (52.0107° N, 5.7825° E) 20 . The soil was air-dried and sieved (1 x 1 cm 2 ) to remove rocks and plant debris. Arabidopsis accession Col-0 seeds were suspended in 0.2% (w/v) agar solution and stratified in dark conditions at 5°C for 2–5 days prior to sowing. On the day of sowing, soil was watered in a 1:10 v/w ratio, 60-mL pots were filled with 120 g of moist soil (± 2.5 g) and placed in 60-mm Petri dishes. For soilborne legacy experiments that include two growth cycles (conditioning population and response population) in the same soil, the soil surface was covered with circular cutouts of plastic micropipette-tip holders (Greiner Bio-one, 0.5–10 µL, item number 771280) to prevent algal growth and ensure consistent spatial sowing in both generations. The circular cutout was used as a sowing template and two Arabidopsis seeds were pipetted into each of 16 holes equally distributed across the soil surface. Pots were randomized in trays with closed transparent lids and incubated in a growth chamber (21°C, 70% relative humidity, 10 h light and 14 h dark, light intensity 100 µmol m − 2 s − 1 ). The soil was watered from the bottom two times a week with 3-mL tap water. One-week after sowing, closed lids were replaced by mash-lids to reduce humidity and plants were once watered with 5-mL ½ strength Hoagland nutrient solution 57 . Hpa culture maintainance The (gno)Hpa cultures of isolate Noco2 58 were routinely maintained on Col-0 plants, but additionally weekly inoculated onto hypersusceptible eds1 59,60 plants to proliferate pathogenic spores and resistant Col-0 RPP5 58 plants to check for contamination. For the Noco2 Hpa-culture, Col-0, eds1 and Col-0 RPP5 seeds were sown on Primasta© potting soil saturated with tap water. For the Noco2 gnoHpa-culture, Col-0 and eds1 seeds were vapour-phase sterilized 61 and sown on Murashige and Skoog (MS) 62 agar-solidified medium without sucrose. After a stratification period of 2–5 days at 5°C, plants were incubated in a growth chamber (21°C, 70% relative humidity, 16 h light and 8 h dark, light intensity 100 µmol m − 2 s − 1 ) for 10 days. The Hpa culture was weekly passed from diseased Col-0 plants onto newly grown Col-0, eds1 and Col-0 RPP5 plants by spray inoculation with an Hpa spore suspension. The gnoHpa-culture was passed by gently touching leaves of infected Col-0 plants to leaves of healthy Col-0 and eds1 plants in axenic conditions. Inoculated plants were incubated in a separate growth chamber (16°C, 10 h light and 14 h dark, light intensity 100 µmol m − 2 s − 1 ). Plant inoculation (gno)Hpa spore suspensions were prepared by collecting shoot material of the culture maintenances, that were inoculated 7–14 days prior to usage, into autoclaved tap water. Tubes were vigorously shaken to loosen the spores, plant material was filtered out with Miracloth (22–25 µM pore size) and spore density was quantified by counting three separate 1-µL droplets using a transmitted-light microscope (Carl Zeiss Microscopy, Standard 25 International Classification for Standards, item number 450815.9902). Spore suspensions between 50–100 spores/µL were directly spray-inoculated onto plants using an airbrush until clear droplet formation could be observed on the leaves. Plants were airdried for 1 h, randomized in trays and incubated in the growth chamber with closed lids that were sprinkled with water on the inside to ensure high humidity. Disease quantification and sampling Seven days post inoculation, infected Arabidopsis phyllosphere material was collected for disease quantification in 15-mL Greiner tubes filled with 3–6 mL water, depending on observed fresh weight and sporulation. Shoot fresh weight was quantified, Greiner tubes were hand shaken for 15 s, spores were counted in three 1-µL droplets using a transmitted-light microscope (Carl Zeiss Microscopy, Standard 25 International Classification for Standards, item number 450815.9902) and the average spore count was normalized by shoot fresh weight. For sequencing, phyllosphere material was collected by cutting the shoots with surface-sterilized razors, carefully avoiding the sampling of root or soil. The rhizosphere was sampled by picking roots and closely adhering soil with surface sterilized tweezers. Unplanted bulk soil samples were taken from the center of the pot after removing the top soil layer (approximately 2 cm). All samples were collected in 2-mL Eppendorf tubes, snap-frozen in liquid nitrogen and stored at -80°C until further processing. Sample compartmentalization and genomic DNA extractions Two 3-mm glass beads were added to frozen phyllosphere samples and samples were mechanically lysed using the Tissuelyser II® (Qiagen) for four cycles of 60 s at 30 Hz, snap freezing in between cycles. Rhizosphere soil and the root endosphere were separated based on Lundberg et al . (2012) 63 with minor adaptations. Roots with adhering soil were washed in 1 mL phosphate-buffered saline (PBS) buffer by gently vortexing for 5 s. Next, tubes were centrifuged for 1 min at 2350 g to spin down the soil while keeping the roots floating. Root material was transferred to a new tube and this cycle was repeated a total of 5 times per sample. Clean roots were then sonicated in PBS buffer for 5 min with 5 s pauses every 30 s. Roots were dried on sterile Miracloth, snap-frozen in liquid nitrogen, and lysed with Tissuelyser II for four cycles of 60 s at 30 Hz. These samples were considered root endosphere. The rhizosphere soil that was washed-off from roots was pooled in 15-mL Greiner tubes, vigorously vortexed, centrifuged at 4700 g for 5 min. Hereafter, the supernatant was removed without disturbing the soil pellet and tubes were frozen at -80°C. Unplanted bulk soil samples remained in -80°C unprocessed until DNA extraction. All DNA was extracted using the Qiagen MagAttract PowerSoil DNA KF Kit and a ThermoFisher KingFisher® (Waltham, USA). Unplanted bulk soil, rhizosphere soil and lysed root endosphere and phyllosphere material were suspended in 750 µL PowerMag Bead solution and spiked with S. ruber DNA at a concentration of 1% of the expected microbial DNA yield, determined by quantitative real-time PCR (qPCR), for unplanted bulk soil, rhizosphere and 0.1% for root endosphere, respectively. Phyllosphere samples were spiked with 1% S. ruber DNA for the compartment experiment and with 0.33% for the passaging experiment. Suspended samples were added to the PowerMag Bead 96-well plate and DNA was extracted according to the manufacturer’s instructions. All DNA concentrations were quantified using a NanoDrop2000®. Hpa quantification by qPCR Hpa levels were quantified from gDNA extracted from the phyllosphere of (gno)Hpa, uninfected or untreated plants by qPCR 64 . Two-step quantitative real-time PCRs were performed in optical 96-well plates using a BioRad OPUS384 qPCR system, iTaq SYBR Green PCR Supermix (BioRad) and Arabidopsis and Hpa actin primers: 5’ AATCACAGCACTTGCACCA 3’ (AtActFwd), 5’ GAGGGAAGCAAGAATGGAAC 3’ (AtActRv), 5’ GTGTCGCACACTGTACCCATTTAT 3’ (HpaActFwd), 5’ ATCTTCATCATGTAGTCGGTCAAGT 3’ (HpaActRv). A standard thermal profile was used: 50°C for 2 min, 95°C for 10 min, 40 cycles of 95°C for 15 s and 60°C for 1 min. Amplicon dissociation curves were recorded after cycle 40 by heating from 60°C to 95°C with a ramp speed of 1.0°C − 1 . (gno)Hpa abundance was calculated by 2 −(CtHpaACTIN − CtArabidopsisACTIN) , in which CtHpaACTIN and CtArabidopsisACTIN are the cycle treshold (Ct-) values obtained within samples for the ACTIN PCR products for (gno)Hpa and Arabidopsis, respectively, as previously described 38 . 16S rDNA amplicon library preparation and sequencing analysis For 16S rDNA amplicon sequencing of gDNA samples, library preparations were performed by Genome Quebec (Quebec, Montreal, Canada) using NextSeq chemistry (2x 300 base pairs paired-end sequencing, Fig. 2 ) or NovaSeq chemistry (2x 250 base pairs paired-end sequencing, Fig. 3 ). The 16S variable regions 3 and 4 were amplified using the primers 16S-B341F (5′-CCTACGGGNGGCWGCAG) and 16S-B806 (5′-GACTACHVGGGTATCTAATCC) according to Genome Quebec’s standard operating protocols. Plastid- and mitochondrial-blocking peptide nucleic acids (pPNA, 5′-GGCTCAACCCTGGACAG, and mPNA, 5′- GGCAAGTGTTCTTCGGA, respectively) were used in the PCRs to prevent amplification of plant-derived sequences. 16Sr DNA amplicon sequencing datasets included in Fig. 1 B-C were generated as previously described 38 . Preprocessing of sequencing data was performed in the Qiime2 environment (version 2022.11) 65 , and executed similar as described by Goossens et al. (2023) 38 . Removal of primer sequences was performed using Cutadapt 66 , and quality filtering, error correction, chimaera removal and dereplication to ASVs was performed using DADA2 67 . Optimal DADA2 truncation and maximum expected error parameters were determined using FIGARO 68 . ASV identifiers were assigned based on the sequence-specific MD5-sums using the–p-hashed-feature-ids parameter, of which we used the first five characters in the text above to designate the individual ASVs. Taxonomic assignment of ASVs was performed using the VSEARCH plugin and the SILVA database (QIIME-compatible 132-release, 99% clustering identity, seven-level Ribosomal Database Project-compatible consensus taxonomies). ASVs with unassigned taxonomies, or that were annotated as ‘D_0__Archaea’ at the kingdom level, were removed. Moreover, plant-derived sequences were identified and removed based on the annotation of ‘D_4__Mitochondria’ at the family level, and ‘D_3__Chloroplast’ or ‘D_2__Chloroplast’ at the order or class level, respectively. ASVs that were relatively less abundant than 0.005% (comprising the lowest ~ 8% of total cumulative abundance, Fig. 2 ) or 0.00075% (comprising the lowest ~ 3% of total cumulative abundance, Fig. 3 ), or that were detected in fewer than N *0.5 samples ( N < 6 for Fig. 2 , N < 5 for Fig. 3 ), were removed. This resulted in datasets comprising 8,801,409 reads from 439 ASVs in 250 samples (Fig. 2 ) and 53,383,924 reads from 9954 ASVs in 200 samples (Fig. 3 ), including the spiked-in S. ruber . For the passaging experiment, samples ‘M2w5’, ‘M4w5’, ‘M9w5’, ‘M12w5’, ‘M11w4’ and ‘M11w5’ were excluded from the analysis as qPCR showed that they were contaminated with (gno)Hpa, and sample ‘U2w1’ was removed because of a labeling error. For the microbiome compartment experiment, samples ‘B-M-G2-1’, ‘P-M-G1-4’, ‘P-M-G1-3’,’E-H-G2-10’, ‘WR-H-G2-2’, ‘E-H-G2-2’, ‘E-M-G2-1’, ‘B-M-G1-4’, ‘E-M-G2-7’, ‘E-H-G2-7’ and ‘E-H-G2-6’ were excluded from data analysis as they only had limited number of reads and as their read counts did not exceed those of blank DNA extraction controls. All alpha- and beta-diversity related calculations, graphs and differential abundance analyses were performed in R (version 3.6.3) using the phyloseq package (version 1.30.0). All PCoA ordinations and PERMANOVA tests were performed on Bray-Curtis dissimilarity matrices calculated for relative abundance data, using the vegan package (version 2.5.7) or vegan functionalities embedded in phyloseq. For the compartment experiment, samples ‘P-M-G1-5’, ‘P-M-G2-4’ and ‘P-M-G2-2’ were obvious outliers in the PCoA and were removed from downstream analysis. PERMANOVA tests involving multiple comparisons were performed using the pairwiseAdonis package (version 0.0.4). Differential abundance testing was performed with DESeq2 69 (package version 1.26.0) and ANCOM-BC 70 (microbiome package version 1.8.0 and nloptr package version 1.2.2.2). As we used DESeq2 to detect differences in abundance between prevalent ASVs, we used an additional prevalence filter set at 0.5*N. For ANCOM-BC no additional filter step was used to enable the detection of structural zero’s. ASVs associated to gnoHpa lineages in the passaging experiment were selected based on three criteria: (1) ASVs that are enriched in gnoHpa lineages compared to uninfected lineages in at least 2 passages from population 2 untill population 5 (DESeq2 or ANCOM-BC). (2) ASVs that are enriched in consecutive passages (populations 2, 3, 4, 5; 3, 4, 5 or 4, 5) of gnoHpa lineages compared to population 1, but not in uninfected lineages (DESeq2 or ANCOM-BC). (3) ASVs of which the relative abundance correlates to the amount of downy mildew, as quantified by qPCR, in gnoHpa and uninfected lineages (Spearman). Absolute abundances were calculated by transforming the reads of each ASV relative to the number of reads from S. ruber , multiplied by the amount of spiked-in DNA (ng) and the number of S. ruber cells expected per ng of S. ruber DNA (2.46*10 5 cells/ng). Absolute abundances were corrected for plant fresh weight for the samples from the passaging experiment (Fig. 2 ). For the microbiome compartment experiment (Fig. 3 ), due to the sample processing, sample fresh weight could not be obtained and uncorrected absolute abundances were used. Graphs were made using the ggplot2 (version 3.3.5), ggpubr (version 0.4.0), UpSetR (version 1.4.0) and cowplot (version 1.1.1) packages. Statistical analyses were performed using the stats (version 3.6.3) and multcompView package (version 0.1.8). Data wrangling was done with packages from the Tidyverse suite. Gamma-irradiated soil experiment For the gamma-irradiated (GI) soil experiment, air-dried live Reijerscamp field soil was wrapped in two autoclave bags and tightly sealed. GI was performed by Steris Applied Sterilization Technologies. Similarly to live field soil, bags with GI-soil were stored at room temperature untill usage. Sterility of GI soil was confirmed by suspending 10 g of soil in 90 mL 10 mM MgSO 4 and plating on one-tenth-strength tryptic soy agar (1/10th TSA) medium amended with 100 mg/L cycloheximide and potato dextrose agar medium (PDA) amended with 13 mg/L chloramphenicol and 150 mg/L rose bengal before usage. Live soil, GI soil and a 9:1 mix of live and GI soil were mixed, watered, potted, sown and incubated in a growth chamber as previously described, carefully avoiding any contact between the GI soil and the live soil. From this point onward, the GI soil was not kept in sterile conditions, but was considered to have a completely diminished microbiome as it had a sterile starting point. The conditioning population of Arabidopsis Col-0 plants was mock- or gnoHpa-inoculated (50 spores/µL) and the response population of Arabidopsis Col-0 plants was gnoHpa-inoculated (67 spores/µL). gnoHpa was used to avoid the co-inoculation of Hpa-associated microbiota. Disease was quantified as previously described. Core-HAM characterization To identify the core-HAM that were consistently enriched upon Hpa infection, we analyzed 16S amplicon sequencing datasets from 14 distinct Hpa-infected phyllospheres from six independent experiments. Experiments 1–5 were previously reported by Goossens et al. 2023 38 (represented by Extended data Fig. 1 , Main Fig. 1 , Main Fig. 2 , Main Fig. 4 and main Fig. 6, respectively). Experiment 6 refers to the microbiome compartment experiment presented in Fig. 3 of this study. Exp. 1–4 were performed in sand-potting soil mixture, whereas Exp. 5 and 6 were performed in live Reijerscamp soil. Differentially abundant ASVs were identified per experiment using DESeq2 69 and ASVs that were enriched in over approximately two-thirds (> 8 out of 14) of distinct Hpa-infected phyllospheres tested were considered core-HAM. The passaging experiment Arabidopsis Col-0 plants were grown in live Reijerscamp field soil as previously described and two-week-old plants were either inoculated with sterile tap water (uninfected), Hpa (80 spores/µL), gnoHpa (80 spores/µL) or remained untreated. Individual pots were placed in Eco2Boxes labeled with a unique number and incubated in a growth chamber (16°C, 10 h light and 14 h dark, light intensity 100 µmol m − 2 s − 1 ). One week post inoculation, half of the phyllosphere, equally distributed throughout the pot, was sampled for sequencing. The other half of the plants was cut-off with a razor and suspended in 1.6 mL 10 mM MgSO 4 . Leaf wash-offs were prepared by vigorously vortexing for 15 s. The leaf wash-offs were transferred to clean 2-mL perfume spraying bottles and spray-inoculated onto a new set of 2-week-old plants. During this process, we ensured that the time between obtaining the leaf wash-offs and spraying was similar between treatments and replicates, carefully avoiding any cross-contamination. Pots were airdried and incubated in a clean set of Eco2Boxes that were labeled accordingly, so that experimental passaging-lines are maintained completely separate. This enables us to directly link the phyllosphere microbiome composition per replicate lineages between passages. This process was repeated for a total of 5 successive populations. In population 5, a (gno)Hpa contamination was spotted on uninfected plants after which the experiment was terminated. For every population, an untreated set of plants was sampled for reference. gDNA was extracted as previously described and used for (gno)Hpa disease-quantification through qPCR and sequencing. The microbiome compartment experiment Live Reijerscamp field soil was conditioned by mock- and Hpa-inoculated (50 spores/µL) plants that were grown and inoculated as previously described. At the end of the conditioning population, all above-ground plant biomass was removed with a razor. Directly after, a response plant population was sown on the same soil using the circular cutouts of plastic micropipette-tip holders as template to ensure that the response plant population grows in the exact same location as the conditioning plant population. The response plant population was grown as previously described and two weeks post sowing, all plants were mock-inoculated. From the conditioning and response population, the unplanted bulk soil, roots and phyllosphere were sampled and processed to separate the rhizosphere and root endosphere as previously described. DNA was extracted and samples were send for 16S rDNA amplicon sequencing. A schematic overview of this experimental setup is presented in Fig. S6 Bacterial densities on healthy and downy mildew-infected plants For determining phyllosphere bacterial densities on healthy and downy mildew-infected plants, plants were grown in live Reijerscamp field soil as previously described and mock-, Hpa- or gnoHpa-inoculated (50 spores/µL). The mock-, Hpa- and gnoHpa inoculums were plated in serial dilutions on 1/10th TSA medium amended with cycloheximide (100 mg/L) to prevent fungal growth and incubated at room temperature. Four days post inoculation, the number of bacterial CFU was quantified. Seven days post inoculation, phyllosphere material was collected and submerged in 3 mL 10 mM MgSO 4 amended with 0.02% Silwet L77. Tubes were incubated shaking at 180 rpm for 1 h. A dilution series up to 10 7 was prepared using 10 mM MgSO 4 and of the 10 3 -10 7 dilutions, 100 µL was plated on 1/10th TSA medium amended with cycloheximide and plates were incubated at RT. CFU-numbers were quantified following 2, 4 and 6 days of incubation at room temperature and normalized to shoot fresh weight. After 4 days, CFU number no longer increased. Xanthomonas sp. WCS2014-23 inoculation experiments Rifampicin resistant Xanthomonas WCS2014-23 was cultured from − 80°C glycerol stocks on Luria-Bertani (LB) agar medium supplemented with rifampicin (50 ng/µL) at 28°C for 2–4 days. Single bacterial colonies were transferred to LB broth medium supplemented with rifampicin (50 ng/µL) and cultured at 28°C for 1–3 days. 100 mL bacterial culture was pelleted and bacterial cells were washed with 50 mL 10 mM MgSO 4 three times. Optical density of the bacterial suspension was determined at 600 nm and Xanthoomonas was inoculated in live Reijerscamp field soil at a concentration of 10 6 CFU/g. The soil was vigorously mixed before being potted, sown and incubated in a growth chamber as previously described. Two weeks after sowing, plants were mock-inoculated with sterile tap water or inoculated with gnoHpa (50 spores/µL). For the experiment in which Xanthomonas was co-inoculated on the leaves with gnoHpa, plants were spray-inoculated with 1 mL of bacterial suspension (OD 600 = 0.3), airdried and directly after spray-inoculated with gnoHpa (50 spores/µL). One week after inoculation, phyllosphere material was collected in 3 mL 10 mM MgSO 4 amended with 0.02% silwet, incubated shaking (180 rpm) at RT for 1 h. A dilution series was plated on 1/10th TSA medium amended with 100 ng/µL rifampicin and 100 mg/L Delvocid and incubated at RT. Bacterial CFU were quantified 3 days after plating. Soilborne legacy leaf wash-off experiment A conditioning population of Arabidopsis Col-0 plants was grown as previously described and mock- or Hpa-inoculated (75 spores/µL). On the same soil, a response population of Col-0 plants was grown and mock-inoculated. Microbial leaf wash-offs were obtained by collecting all phyllosphere material in 5 mL sterilized tap water and vortexing for 15 s. Of these wash-offs or a sterile water control, 1 mL was then inoculated onto two-week old plant populations grown as previously described. Directly after, all pots were spray-inoculated with gnoHpa (30 spores/µL). One week post inoculation, disease was quantified as previously described. Declarations Data and code availability The experimental data and the post-processing amplicon sequencing data that support the findings of this study are available at https://github.com/JelleSpooren/Spooren-et-al-2025, together with the code used to analyze the data and generate figures. Raw amplicon sequence data generated by this study are available at https://www.ncbi.nlm.nih.gov/bioproject/1262419 Acknowledgements This study was sponsored by the Dutch Research Council (NWO) through the XL program “Unwiring beneficial functions and regulatory networks in the plant endosphere” (grant no. OCENW.GROOT.2019.063), and through the Gravitation program MiCRop (grant no. 024.004.014). Author contributions J.S., C.M.J.P., & R.L.B. designed the experiments and wrote the manuscript. J.S., T.T., H.P., S.H., U.Y., and H.D. performed the experiments. J.S. and Y.S. performed the microbiome data analysis. R.Q. provided technical support in maintenance of Hpa- and gnoHpa cultures and P.G., R.Q., and S.C.M.W. provided valuable input on experimental design and execution. C.M.J.P. and R.L.B. supervised the project. References Berendsen, R. L., Pieterse, C. M. J. & Bakker, P. A. H. M. The rhizosphere microbiome and plant health. Trends Plant Sci. 17 , 478-486 (2012). Trivedi, P., Leach, J. E., Tringe, S. G., Sa, T. & Singh, B. K. Plant–microbiome interactions: from community assembly to plant health. Nat. Rev. Microbiol. 18 , 607-621 (2020). Bulgarelli, D., Schlaeppi, K., Spaepen, S., Van Themaat, E. V. L. & Schulze-Lefert, P. Structure and functions of the bacterial microbiota of plants. Annu. Rev. Plant Biol. 64 , 807-838 (2013). Wang, X., Zhang, J., Lu, X., Bai, Y. & Wang, G. Two diversities meet in the rhizosphere: root specialized metabolites and microbiome. J. Genet. Genom. 51 , 467-478 (2023). Sasse, J., Martinoia, E. & Northen, T. Feed your friends: do plant exudates shape the root microbiome? Trends Plant Sci. 23 , 25-41 (2018). Compant, S. et al. The plant endosphere world–bacterial life within plants. Environ. Microbio l. 23 , 1812-1829 (2021). Reinhold-Hurek, B. & Hurek, T. Living inside plants: bacterial endophytes. Curr. Opin. Plant Biol. 14 , 435-443 (2011). Yu, K., Pieterse, l.C. M. J., Bakker, P. A. H. M. & Berendsen, R. L. Beneficial microbes going underground of root immunity. Plant Cell Environ. 42 , 2860-2870 (2019). Doan, H. K. et al. Topography-driven shape, spread, and retention of leaf surface water impacts microbial dispersion and activity in the phyllosphere. Phytobiomes 4 , 268-280 (2020). Paauw, M. et al. Hydathode immunity protects the Arabidopsis leaf vasculature against colonization by bacterial pathogens. Curr. Biol. 33 , 697-710 (2023). Vorholt, J. A. Microbial life in the phyllosphere. Nat. Rev. Microbiol. 10 , 828-840 (2012). Remus-Emsermann, M. N., Tecon, R., Kowalchuk, G. A. & Leveau, J. H. J. Variation in local carrying capacity and the individual fate of bacterial colonizers in the phyllosphere. ISME J. 6 , 756-765 (2012). Kusstatscher, P. et al. Trichomes form genotype-specific microbial hotspots in the phyllosphere of tomato. Environ. Microbiome 15 , 1-10 (2020). Spooren, J. et al. Plant-driven assembly of disease-suppressive soil microbiomes. Annu. Rev. Phytopathol. 62 (2024). Kim, D.-R. et al. A mutualistic interaction between Streptomyces bacteria, strawberry plants and pollinating bees. Nat. Commun. 10 , 4802 (2019). Tkacz, A., Bestion, E., Bo, Z., Hortala, M. & Poole, P. S. Influence of plant fraction, soil, and plant species on microbiota: a multikingdom comparison. MBio 11 , 10-1128 (2020). Bai, Y. et al. Functional overlap of the Arabidopsis leaf and root microbiota. Nature 528 , 364-369 (2015). Zhou, S. Y. et al. Microbial flow within an air-phyllosphere-soil continuum. Front. Microbiol. 11 , 615481 (2020). Massoni, J., Bortfeld-Miller, M., Widmer, A. & Vorholt, J. A. Capacity of soil bacteria to reach the phyllosphere and convergence of floral communities despite soil microbiota variation. Proc.Natl. Acad. Sci. U. S. A. 118 , e2100150118 (2021). Berendsen, R. L. et al. Disease-induced assemblage of a plant-beneficial bacterial consortium. ISME J. 12 , 1496-1507 (2018). Weller, D. M., Raaijmakers, J. M., Gardener, B. B. M. & Thomashow, L. S. Microbial populations responsible for specific soil suppressiveness to plant pathogens. Annu. Rev. Phytopathol. 40 , 309-348 (2002). Pfeilmeier, S. et al. The plant NADPH oxidase RBOHD is required for microbiota homeostasis in leaves. Nat. Microbiol. 6 , 852-864 (2021). Gao, M. et al. Disease-induced changes in plant microbiome assembly and functional adaptation. Microbiome 9 , 187 (2021). Liu, H. et al. Evidence for the plant recruitment of beneficial microbes to suppress soil-borne pathogens. New Phytol. 229 , 2873-2885 (2021). Rolfe, S. A., Griffiths, J. & Ton, J. Crying out for help with root exudates: adaptive mechanisms by which stressed plants assemble health-promoting soil microbiomes. Curr. Opin. Microbiol. 49 , 73-82 (2019). Rizaludin, M. S., Stopnisek, N., Raaijmakers, J. M. & Garbeva, P. The chemistry of stress: understanding the ‘cry for help’of plant roots. Metabolites 11 , 357 (2021). Schlatter, D., Kinkel, L., Thomashow, L., Weller, D. & Paulitz, T. Disease suppressive soils: new insights from the soil microbiome. Phytopathology 107 , 1284-1297 (2017). Gómez Expósito, R., De Bruijn, I., Postma, J. & Raaijmakers, J. M. Current insights into the role of rhizosphere bacteria in disease suppressive soils. Front. Microbiol. 8 , 2529 (2017). Carrión, V. J. et al. Pathogen-induced activation of disease-suppressive functions in the endophytic root microbiome. Science 366 , 606-612 (2019). Raaijmakers, J. M. & Weller, D. M. Natural plant protection by 2, 4-diacetylphloroglucinol-producing Pseudomonas spp. in take-all decline soils. Mol. Plant Microbe Interact. 11 , 144-152 (1998). Weller, D. M. et al. Role of 2, 4-diacetylphloroglucinol-producing fluorescent Pseudomonas spp. in the defense of plant roots. Plant Biol. 9 , 4-20 (2007). Mendes, R. et al. Deciphering the rhizosphere microbiome for disease-suppressive bacteria. Science 332 , 1097-1100 (2011). Vogel, C. M., Potthoff, D. B., Schäfer, M., Barandun, N. & Vorholt, J. A. Protective role of the Arabidopsis leaf microbiota against a bacterial pathogen. Nat. Microbiol. 6 , 1537-1548 (2021). Liu, X. et al. Phyllosphere microbiome induces host metabolic defence against rice false-smut disease. Nat. Microbiol. 8 , 1419-1433 (2023). Francisco, C. S. et al. The apoplastic space of two wheat genotypes provide highly different environment for pathogen colonization: Insights from proteome and microbiome profiling. bioRxiv , 543792 (2023). Ehau-Taumaunu, H. & Hockett, K. L. Passaging phyllosphere microbial communities develop suppression towards bacterial speck disease in tomato. Phytobiomes J.l 7 , 233-243 (2023). Li, P.-D. et al. The phyllosphere microbiome shifts toward combating melanose pathogen. Microbiome 10 , 56 (2022). Goossens, P. et al. Obligate biotroph downy mildew consistently induces near-identical protective microbiomes in Arabidopsis thaliana . Nat. Microbiol. 8 , 2349-2364 (2023). Bakker, P. A. H. M., Pieterse, C. M. J., de Jonge, R. & Berendsen, R. L. The soil-borne legacy. Cell 172 , 1178-1180 (2018). Vismans, G. et al. Coumarin biosynthesis genes are required after foliar pathogen infection for the creation of a microbial soil-borne legacy that primes plants for SA-dependent defenses. Sci. Rep. 12 , 22473 (2022). Stringlis, I. A., De Jonge, R. & Pieterse, C. M. J. The age of coumarins in plant–microbe interactions. Plant Cell Physiol. 60 , 1405-1419 (2019). Stämmler, F. et al. Adjusting microbiome profiles for differences in microbial load by spike-in bacteria. Microbiome 4 , 1-13 (2016). Goossens, P. et al. Selective enrichment of specific bacterial taxa in downy mildew-affected spinach: Comparative analysis in laboratory and field conditions. BioRxiv , 609345 (2024). Kwak, M.-J. et al. Rhizosphere microbiome structure alters to enable wilt resistance in tomato. Nat. Biotechnol. 36 , 1100-1109 (2018). de Sousa, L. P. & Mondego, J. M. C. Leaf surface microbiota transplantation confers resistance to coffee leaf rust in susceptible Coffea arabica . FEMS Microbiol. Ecol. 100 , fiae049 (2024). Gu, S. et al. Competition for iron drives phytopathogen control by natural rhizosphere microbiomes. Nat. Microbiol. 5 , 1002-1010 (2020). Höfte, M. & Bakker, P. A. H. M. in Microbial siderophores (eds A Varma & S. B. Chincholkar) 121-133 (Springer, 2007). Weller, D. M. et al. Disease-suppressive soils induce systemic resistance in Arabidopsis thaliana against Pseudomonas syringae pv. tomato. PhytoFront. 4 , 515-523 (2024). Pieterse, C. M. J. et al. Induced systemic resistance by beneficial microbes. Annu. Rev. Phytopathol. 52 , 347-375 (2014). Hannula, S. E. et al. Persistence of plant-mediated microbial soil legacy effects in soil and inside roots. Nat. Commun. 12 , 5686 (2021). Debray, R., Conover, A., Zhang, X., Dewald-Wang, E. A. & Koskella, B. Within-host adaptation alters priority effects within the tomato phyllosphere microbiome. Nat. Ecol. & Evol. 7 , 725-731 (2023). Debray, R. et al. Priority effects in microbiome assembly. Nat. Rev. Microbiol. 20 , 109-121 (2022). Carlström, C. I. et al. Synthetic microbiota reveal priority effects and keystone strains in the Arabidopsis phyllosphere. Nat. Ecol. & Evol. 3 , 1445-1454 (2019). Raaijmakers, J. M., Van Der Sluis, I., Van Den Hout, M., Bakker, P. A. H. M. & Schippers, B. Dispersal of wild-type and genetically-modified Pseudomonas spp from treated seeds or soil to aerial parts of radish plants. Soil Biol. Biochem. 27 , 1473-1478 (1995). Cevallos-Cevallos, J. M., Danyluk, M. D., Gu, G., Vallad, G. E. & van Bruggen, A. H. Dispersal of Salmonella Typhimurium by rain splash onto tomato plants. J. Food Prot. 75 , 472-479 (2012). Cha, J.-Y. et al. Microbial and biochemical basis of a Fusarium wilt-suppressive soil. ISME J. 10 , 119-129 (2016). Pieterse, C. M. J., Van Wees, S. C., Hoffland, E., Van Pelt, J. A. & Van Loon, L. C. Systemic resistance in Arabidopsis induced by biocontrol bacteria is independent of salicylic acid accumulation and pathogenesis-related gene expression. Plant Cell 8 , 1225-1237 (1996). Parker, J. E. et al. Phenotypic characterization and molecular mapping of the Arabidopsis thaliana locus RPP5, determining disease resistance to Peronospora parasitica . Plant J. 4 , 821-831 (1993). Parker, J. E. et al. Characterization of eds1 , a mutation in Arabidopsis suppressing resistance to Peronospora parasitica specified by several different RPP genes. Plant Cell 8 , 2033-2046 (1996). Aarts, N. et al. Different requirements for EDS1 and NDR1 by disease resistance genes define at least two R gene-mediated signaling pathways in Arabidopsis . Proc.Natl. Acad. Sci. U. S. A 95 , 10306-10311 (1998). Lindsey III, B. E., Rivero, L., Calhoun, C. S., Grotewold, E. & Brkljacic, J. Standardized method for high-throughput sterilization of Arabidopsis seeds. J. Vis. Exp. , e56587 (2017). Murashige, T. & Skoog, F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant. 15 , 473-497 (1962). Lundberg, D. S. et al. Defining the core Arabidopsis thaliana root microbiome. Nature 488 , 86-90 (2012). Anderson, R. G. & McDowell, J. M. A PCR assay for the quantification of growth of the oomycete pathogen Hyaloperonospora arabidopsidis in Arabidopsis thaliana . Mol. Plant Pathol. 16 , 893-898 (2015). Bolyen, E. et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat. Biotechnol. 37 , 852-857 (2019). Martin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet. J. 17 , 10-12 (2011). Callahan, B. J. et al. DADA2: High-resolution sample inference from Illumina amplicon data. Nat. Methods 13 , 581-583 (2016). Weinstein, M. M., Prem, A., Jin, M., Tang, S. & Bhasin, J. M. FIGARO: An efficient and objective tool for optimizing microbiome rRNA gene trimming parameters. BioRxiv , 610394 (2019). Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15 , 1-21 (2014). Lin, H. & Peddada, S. D. Analysis of compositions of microbiomes with bias correction. Nat. Commun. 11 , 3514 (2020). Additional Declarations There is NO Competing Interest. Supplementary Files SupplementalSpoorenetal2025NaturePlantsMay182025.docx Supporting Information Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6693507","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":469925619,"identity":"90411580-9026-4630-ad4b-1a7b9b634ede","order_by":0,"name":"Jelle Spooren","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Jelle","middleName":"","lastName":"Spooren","suffix":""},{"id":469925620,"identity":"7c555c04-1ad6-4be9-a1a0-8fd657073c80","order_by":1,"name":"Yadong Shao","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Yadong","middleName":"","lastName":"Shao","suffix":""},{"id":469925621,"identity":"4b921956-63e5-4b21-81d4-edeade2abafa","order_by":2,"name":"Tilda Tarrant","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Tilda","middleName":"","lastName":"Tarrant","suffix":""},{"id":469925622,"identity":"b984ac9d-4fb7-4ddf-b0f6-db4174bd86b9","order_by":3,"name":"Hannah Ploehmacher","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Hannah","middleName":"","lastName":"Ploehmacher","suffix":""},{"id":469925623,"identity":"66b925e9-1f6f-4949-b08f-e0e303f6d841","order_by":4,"name":"Run Qi","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Run","middleName":"","lastName":"Qi","suffix":""},{"id":469925629,"identity":"f4fd59ab-a862-4c55-9fbc-101bc5bb2bf5","order_by":5,"name":"Syb Hopkoper","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Syb","middleName":"","lastName":"Hopkoper","suffix":""},{"id":469925624,"identity":"22ee15f2-fedf-46d7-8245-15c7daa925bf","order_by":6,"name":"Umut Yüce","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Umut","middleName":"","lastName":"Yüce","suffix":""},{"id":469925625,"identity":"40711ab7-360c-42c5-9079-a4f304ee0a4a","order_by":7,"name":"Hangyu Dong","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Hangyu","middleName":"","lastName":"Dong","suffix":""},{"id":469925626,"identity":"af9fac82-aad2-420c-8cf7-1a7d45e5667f","order_by":8,"name":"Pim Goosssens","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Pim","middleName":"","lastName":"Goosssens","suffix":""},{"id":469925627,"identity":"be414bb7-b66c-4693-8e64-7c1ea69c5ad5","order_by":9,"name":"Saskia C.M. Van Wees","email":"","orcid":"","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Saskia","middleName":"C.M. Van","lastName":"Wees","suffix":""},{"id":469925628,"identity":"4fb73552-4f52-4669-8d75-c571527b5b9e","order_by":10,"name":"Corné M.J. Pieterse","email":"","orcid":"https://orcid.org/0000-0002-5473-4646","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":false,"prefix":"","firstName":"Corné","middleName":"M.J.","lastName":"Pieterse","suffix":""},{"id":469925618,"identity":"63539099-e062-4592-bd35-323fbbc80d7a","order_by":11,"name":"Roeland L. Berendsen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwklEQVRIiWNgGAWjYFAC5gZmCCMBiCsgIgS0MCJrOQMRIUELYxsRWswbgFoKKu7JmbOnP3xcOO9wYoN0I34tMgeAWmacKTa27HljbDxzG1CLzEH8WiRADuNtS0jccCOHTZoXpEUikRgt/xLqN9xIf/6bdw7RWhoSEgxuJJgBGcRoYWZsODzjWILhhjNvjKV5jqUbtxHUwt588HFBTYK8wfH0h595aqxl+yWSD+DVwgCMFFQVbPjVj4JRMApGwSggBgAA/NFEIn1kijEAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0003-2707-8919","institution":"Plant-Microbe Interactions, Institute of Environmental Biology, Department of Biology, Science4Life, Utrecht University, 3584 CH Utrecht, the Netherlands","correspondingAuthor":true,"prefix":"","firstName":"Roeland","middleName":"L.","lastName":"Berendsen","suffix":""}],"badges":[],"createdAt":"2025-05-18 20:05:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6693507/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6693507/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":84463062,"identity":"c8118921-1814-4ba5-a28f-70e18b9d5843","added_by":"auto","created_at":"2025-06-12 09:12:19","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":572517,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMicrobial origin of the disease-suppressive SBL and relative abundance of core-HAM in Hpa-infected phyllosphere microbiomes. (A) \u003c/strong\u003eSpore production of gnotobiotic Hpa (gnoHpa) in response (R) populations of Arabidopsis Col-0 plants growing in soil conditioned (C) by populations of Col-0 plants that were either mock treated or inoculated with gnoHpa. Plants were grown in live Reijerscamp field soil (100% live soil), field soil sterilized by gamma-irradiation (sterilized soil), or a 1:9 mix of live and sterilized field soil (10% live soil). Asterisks indicate significance level in one-sided Student’s \u003cem\u003et\u003c/em\u003e-test, from left to right: **\u003cem\u003eP \u003c/em\u003e= 0.0068; NS, not significant; **\u003cem\u003eP\u003c/em\u003e = 0.0046. Bars and error bars indicate the average and standard error, respectively, of between 7 and 11 biological replicates.\u003cstrong\u003e (B)\u003c/strong\u003e Barplot showing the relative abundance (%) of the 25 core-HAM that are significantly enriched in more than 8 out of 14 datasets of Hpa-infected Arabidopsis phyllospheres from six independent experiments conducted over a timespan of more than five years. Core-HAM abundances were cumulated based on the number of Hpa-infected phyllospheres in which they were enriched (colors). The number of ASVs that belong to each category is indicated in parentheses. Experiments are ordered chronologically from left to right. \u003cstrong\u003e(C)\u003c/strong\u003e The contribution of each single core-HAM ASV to the cumulative relative abundance (%) of the 25-member core-HAM community, colored by the taxonomy of single HAM ASVs, except for the genera \u003cem\u003eMethylophilus, Rhizobium \u003c/em\u003eand\u003cem\u003e Sphingobacterium, \u003c/em\u003ewhich are represented by 2 or more ASVs.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/47e5cb29bfb62316fd1ffe83.png"},{"id":84463063,"identity":"0e4085ae-10a6-4a46-8e0b-7bde88be5b17","added_by":"auto","created_at":"2025-06-12 09:12:19","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1061347,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eChanges in microbiome composition and enrichment of specific bacteria with persistent selective pressure in the downy mildew-infected phyllosphere. (A) \u003c/strong\u003ePrincipal coordinate analysis ordination plots based on Bray-Curtis dissimilartiy of phyllosphere bacterial community composition of Arabidopsis Col-0 plant populations grown in live Reijerscamp field soil. Treatments include newly grown untreated plants (black) from every population, and uninfected (blue), gnoHpa (orange) and Hpa (red) lineages in which leaf wash-offs were successivley passaged from their initial inoculation (population 1) untill population 5. \u003cstrong\u003e(B) \u003c/strong\u003eEffect size of the observed changes in microbiome composition between uninfected and gnoHpa lineages, indicated by the \u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e-value in PERMANOVA analysis. Asterisks indicate significance level (FDR-corrected) in PERMANOVA: (from left to right) *\u003cem\u003eP\u003c/em\u003e = 0.046; **\u003cem\u003eP\u003c/em\u003e = 0.0075; **\u003cem\u003eP\u003c/em\u003e = 0.0094.\u003cstrong\u003e (C)\u003c/strong\u003e Venn diagram showing the number of ASVs that meet each of three criteria defined to detect ASVs that were selectively promoted in gnoHpa lineages: (1) consistently enriched ASVs in gnoHpa compared to uninfected lineages in at least 2 out of 4 passages (detected by DESeq2 or ANCOM-BC), (2) ASVs that accumulate in the gnoHpa lineages but not in the uninfected lineages (detected by DESeq2 or ANCOM-BC), and (3) ASVs of which the relative abundances correlate with disease quantification in gnoHpa and uninfected lineages (Spearman correlations).\u003cstrong\u003e (D) \u003c/strong\u003eBarplots showing the cumulative relative abundance of the 12 downy mildew-associated ASVs that are identified by all three selection criteria. Relative abundances were plotted in the untreated control and uninfected, gnoHpa\u003cem\u003e,\u003c/em\u003e and Hpa lineages across populations. Colors represent individual ASVs. Letters indicate significant differences (\u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05, ANOVA with Tukey’s post-hoc test) in the cumulative relative abundance of all 12 ASVs tested across all populations and lineages. \u003cstrong\u003e(E)\u003c/strong\u003e Violin plots showing absolute abundance of core-HAM \u003cem\u003eSphingobium \u003c/em\u003eASV\u003cem\u003e \u003c/em\u003eed6be, calculated based on spiked-in \u003cem\u003eS. ruber\u003c/em\u003e DNA and represented as Log-10 transformed 16S copy numbers per gram of shoot fresh weight. Letters indicate significant differences (\u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05, ANOVA with Tukey’s post-hoc test) tested between untreated plants and uninfected, gnoHpa\u003cem\u003e, \u003c/em\u003eand Hpa lineages across populations. For all panels, each dot and bar\u003cstrong\u003e \u003c/strong\u003erepresents one of, or the average of, 12 independent biological replicates, except for untreated population 1 (\u003cem\u003eN\u003c/em\u003e = 11), uninfected population 4 (\u003cem\u003eN\u003c/em\u003e = 11) and uninfected population 5 (\u003cem\u003eN\u003c/em\u003e = 7).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/6bb39c4baecf37073a08bd82.png"},{"id":84463446,"identity":"05d9a09a-9d54-4928-b1f0-a16dda3ef835","added_by":"auto","created_at":"2025-06-12 09:20:19","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":843184,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpatiotemporal dynamics of microbiota in distinct microbiome compartments from Hpa-infected\u003c/strong\u003e\u003cem\u003e\u003cstrong\u003e \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eplants and healthy plants grown in disease-suppressive SBL soil. \u003c/strong\u003eArabidopsis Col-0 plants were grown as mock- or Hpa-inoculated conditioning plant population (C), or as mock-inoculated response plant population (R) grown in disease-suppressive SBL soil conditioned by Hpa-inoculated plants or control soil conditioned by mock-inoculated plants (Fig. S6). Microbiome composition in the unplanted bulk soil (black), rhizosphere (orange), root endosphere (gold), and phyllosphere (dark green) microbiome compartments of these plants were analyzed with 16S amplicon sequencing. All data is based on 6-10 biological replicates per microbiome compartment per plant population. \u003cstrong\u003e(A) \u003c/strong\u003eSankey plot showing the contribution of ASVs with the highest relative abundance in each microbiome compartment relative to the total relative abundance within each microbiome compartment. Light green dashed lines indicate the relative abundance of the 21 core-HAM ASVs across the microbiome compartments in the experiment.\u003cstrong\u003e (B) \u003c/strong\u003ePrincipal coordinate analysis ordination plot based on Bray-Curtis dissimilarities showing bacterial community composition in the microbiome compartments of mock- (triangles) or Hpa-inoculated (circles) plants. The conditioning population (C) plants were either or mock- or Hpa\u003cem\u003e-\u003c/em\u003einoculated, and the response population (R) plants were mock inoculated but grown in soil conditioned by either mock- or Hpa-inoculated plants. Bars indicate effect size (\u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e) based on PERMANOVA analysis, representing the impact of Hpa infection on microbiome community composition in the conditioning population plants or the corresponding SBL effect in the response population plants. Asterisks indicate significance level (FDR-corrected) in PERMANOVA: NS, not significant; *\u003cem\u003eP\u003c/em\u003e = 0.011; ***\u003cem\u003eP\u003c/em\u003e \u0026lt; 0.001. \u003cstrong\u003e(C) \u003c/strong\u003eBarplots representing the average cumulative relative abundances (%) of the 21 core-HAM ASVs in each microbiome compartment per plant population. For belowground microbiome compartments, inner panels show 100x magnification of the outer panel, with relative abundances on the y-axis scaled from 0.0 to 1.0%. Colors represent individual core-HAM ASVs. Asterisks indicate significance level of the Hpa infection on core-HAM relative abundance in the conditioning population plants or on the corresponding SBL effect in the response population plants in FDR-corrected one-sided Student’s \u003cem\u003et-\u003c/em\u003etest (left to right): NS, not significant; ***\u003cem\u003eP\u003c/em\u003e = 8.4 10\u003csup\u003e-4\u003c/sup\u003e; *\u003cem\u003eP\u003c/em\u003e = 0.046; ***\u003cem\u003eP \u003c/em\u003e= 2.1 x 10\u003csup\u003e-10\u003c/sup\u003e; ***\u003cem\u003eP\u003c/em\u003e = 1.4 x 10\u003csup\u003e-6\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/fae50d885d3a99a7c0d1dbba.png"},{"id":84463453,"identity":"215ad4c8-60d5-4d74-b42d-c27e0152b89a","added_by":"auto","created_at":"2025-06-12 09:20:20","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":177940,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eTotal bacterial abundances and core-HAM \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eXanthomonas \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003esp. WCS2014-23 population densities on downy mildew-infected plants. (A)\u003c/strong\u003e Boxplots showing phyllosphere bacterial population densities (10-log transformed CFU/g shoot fresh weight) of mock-, Hpa- or gnoHpa-inoculated Arabidopsis Col-0 plant populations grown in live Reijerscamp field soil. Letters indicate significance based on the average of 10 biological replicates (\u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05 in ANOVA with Tukey’s post-hoc test). \u003cstrong\u003e(B-C) \u003c/strong\u003eBoxplots showing population densities (10-log transformed CFU/g fresh weight) of \u003cem\u003eXanthomonas \u003c/em\u003esp. WCS2014-23, which was inoculated into live Reijerscamp filed soil, in the \u003cstrong\u003e(B) \u003c/strong\u003erhizosphere and \u003cstrong\u003e(C) \u003c/strong\u003ephyllosphere of mock- and gnoHpa-inoculated Arabidopsis Col-0 plant populations. Asterisks indicate significance level based on one-sided Student’s \u003cem\u003et\u003c/em\u003e-test of 20 (rhizosphere) and 9 (phyllosphere) biological replicates: NS, not significant; ***\u003cem\u003eP\u003c/em\u003e = 1.0 x 10\u003csup\u003e-4\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/78f16fab114cfaea4950c8ca.png"},{"id":84464662,"identity":"8b55e68d-419e-4405-a446-9766667129db","added_by":"auto","created_at":"2025-06-12 09:28:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":199641,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSpore production on Arabidopsis Col-0 plants inoculated with gnoHpa spores in presence of microbial leaf wash-offs from plants grown in disease-suppressive SBL soil or control soil. (A)\u003c/strong\u003e Schematic representation of the experimental setup used to test the effect of phyllosphere microbiomes on subsequent downy mildew infection. Conditioning populations of Arabidopsis Col-0 plants were grown in live Reijerscamp field soil and either mock- or Hpa-inoculated. After one week, the conditioning population plants were removed, and new Col-0 plants (response population) were sown and grown in the conditioned soils. Microbial leaf wash-offs were collected from healthy plants grown in the control or the disease-suppressive SBL soils and mixed with gnoHpa\u003cem\u003e \u003c/em\u003espores before being used to inoculate a new set of Col-0 plants grown in unconditioned soil. GnoHpa spores suspended in sterile water was used as control inoculum. Spore production was quantified 7 days post inoculation. \u003cstrong\u003e(B) \u003c/strong\u003eBarplot showing the average spore production on a population of Arabidopsis Col-0 plants grown in unconditioned live Reijerscamp field soil (U) inoculated with a suspension of gnoHpa spores in either sterile water (-), or microbial leaf wash-offs (+) from a healthy response plant population (R) grown in soil conditioned (C) by mock (control soil) or Hpa-inoculated (SBL soil) plant populations. Error bars represent standard error based on 11-12 biological replicates, letters indicate significant differences (\u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05 in ANOVA with Tukey’s post-hoc test).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/1d941289a0464d0b70c1f519.png"},{"id":87222050,"identity":"47b1ea40-28fd-466a-aba3-3766c326efc3","added_by":"auto","created_at":"2025-07-21 16:24:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4460268,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/ea1a2ab9-413c-40ca-b499-35451fb955b0.pdf"},{"id":84463066,"identity":"7ace76e9-476f-489a-b499-da6ca4dc7f76","added_by":"auto","created_at":"2025-06-12 09:12:19","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":6038784,"visible":true,"origin":"","legend":"Supporting Information","description":"","filename":"SupplementalSpoorenetal2025NaturePlantsMay182025.docx","url":"https://assets-eu.researchsquare.com/files/rs-6693507/v1/88a10cdfe6a90c8c0c52154e.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePlants host diverse and complex microbiomes comprised of pathogenic, commensal, and beneficial microbes\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e. Different plant tissues provide unique niches for microbial colonization\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e. In the rhizosphere, the zone of soil surrounding roots, plants exude metabolites that selectively stimulate or inhibit distinct microbes\u003csup\u003e\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u003c/sup\u003e, creating a nutrient-rich yet selective environment\u003csup\u003e\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Some rhizosphere microbes may enter the root endosphere, but must overcome additional selective pressures from root metabolites, structural root barriers, and plant immune responses\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e. Aboveground, the phyllosphere presents a relatively nutrient-poor habitat where microbes cluster in hotspot sites that provide moisture, nutrients, or shelter from environmental stressors\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e. Despite the distinct nature of these microbiome compartments, there is reciprocal exchange of microbes occurring between aboveground, belowground and inner plant tissues\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u003c/sup\u003e. The soil, acting as a diverse microbial reservoir, exerts a major influence on the composition of plant-associated microbiomes in all compartments\u003csup\u003e\u003cspan additionalcitationids=\"CR17 CR18\" citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePlants can selectively steer microbial colonization to enhance their health by activating immune responses to stop pathogen infection or by stimulating protective microbiota that can control growth of invading pathogens\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan additionalcitationids=\"CR21\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e. In response to pathogen attack, plants can selectively enrich their rhizosphere microbiome with disease-suppressive microbiota, which help reduce disease progression\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan additionalcitationids=\"CR24 CR25\" citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. This phenomenon is evident in so-called disease-suppressive soils, where plants stay relatively healthy despite the presence of virulent pathogens. Disease-suppressive soils typically emerge after an initial severe outbreak of disease, demonstrating the plant's capacity to promote protective microbial communities in the soil\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan additionalcitationids=\"CR28\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. Prime examples include take-all decline in wheat and \u003cem\u003eRhizoctonia\u003c/em\u003e-suppressive soils in sugar beet, where disease suppression is linked to the buildup of disease-suppressive microbiomes in the rhizosphere and root endosphere, respectively\u003csup\u003e\u003cspan additionalcitationids=\"CR30 CR31\" citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe phyllosphere likewise harbors microbiota with disease-suppressive properties that can boost plant immunity\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e or act through direct microbial antagonism\u003csup\u003e\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. As in soil, phyllosphere disease-suppressive microbiomes may emerge in response to pathogen challenge\u003csup\u003e\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e. We previously discovered that phyllospheres of \u003cem\u003eArabidopsis thaliana\u003c/em\u003e (Arabidopsis) plants infected by the foliar obligate biotrophic downy mildew pathogen \u003cem\u003eHyaloperonospora arabidopsidis\u003c/em\u003e (Hpa) are enriched with a specific group of Hpa-associated microbiota (HAM)\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. These HAM increase in abundance in the Hpa-infected phyllosphere environment and suppress Hpa spore production, effectively functioning as a phyllosphere disease-suppressive microbiome\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIt was further demonstrated that Hpa-infected Arabidopsis plants, referred to as a conditioning population, can condition the soil in which they grow, creating a disease-suppressive soilborne legacy (SBL). As result, subsequent \u0026ldquo;response populations\u0026rdquo; Arabidopsis plants grown in SBL soil exhibit greater resistance to Hpa compared to plants grown in soil conditioned by healthy plants (control soil) \u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. Notably, disease-suppressive HAM are enriched in the rhizosphere and phyllosphere of response population plants growing in SBL soil, even if these plants were not themselves infected by Hpa\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eDespite these insights, the interplay between the rhizosphere and phyllosphere in plant-driven microbiome assembly, particularly in response to foliar pathogen attack, remains poorly understood. In this study, disease-induced shifts in rhizosphere and phyllosphere microbiome composition were investigated in detail, with a focus on the assembly of HAM in response to the selective regime imposed by downy mildew infection and its role in shaping the SBL. While disease-suppressive soils evidently are often attributed to microbiota in the rhizosphere\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e, our results demonstrate that the phyllosphere can function as a critical hub where soil-transmitted microbiota accumulate and contribute to the suppression of foliar disease.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eA live soil microbiome is required for the creation of a soilborne legacy\u003c/h2\u003e \u003cp\u003ePrevious work showed that a disease-suppressive soilborne legacy (SBL) is consistently established in soils where wild-type Arabidopsis Col-0 plants, infected with the foliar pathogen Hpa, were grown. These plants had been inoculated with either standard Hpa spore suspensions containing Hpa-associated microbiota (HAM), or with gnotobiotic, HAM-free spore suspensions (gnoHpa)\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. In contrast, inoculation of transgenic Col-0 \u003cem\u003eRPP5\u003c/em\u003e plants, resistant to Hpa infection, with the same spores did not result in a disease-suppressive SBL. Similarly, mutant Arabidopsis plants deficient in \u003cem\u003eMYB72\u003c/em\u003e and \u003cem\u003eF6\u0026rsquo;H1\u003c/em\u003e, both essential for coumarin biosynthesis and rhizobacteria-mediated induced systemic resistance\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e, also failed to create a suppressive SBL\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e. These results indicate that both a successful downy mildew infection and subsequent disease-induced plant responses are necessary for the creation of a disease-suppressive SBL, while the mere co-inoculation of HAM bacteria without disease induction is insufficient.\u003c/p\u003e \u003cp\u003eThe disease-suppressive SBL is thus believed to result from plant-driven infection-induced changes in the soil microbiome, but it has not been formally demonstrated that the disease suppressiveness of these SBL soils is caused by living microbes. To test this, Col-0 plants were grown in three soil types: 1) non-sterile field soil (100% live soil) collected at the Reijerscamp nature reserve which supports an abundant endemic Arabidopsis population\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e, 2) the same soil sterilized by gamma irradiation, and 3) a 1:9 mix of live and sterilized field soil (10% live soil). Fourteen-day-old plants were inoculated with gnoHpa spores or mock-treated and then cultivated to condition the soil. Seven days post inoculation (dpi), when downy mildew disease had visibly manifested through the emergence of sporangiophores, aboveground plant parts were removed and a second population of Col-0 plants (response population) was sown directly in the conditioned soil. These plants were again inoculated with gnoHpa at 14 days, and spore production was quantified 7 dpi.\u003c/p\u003e \u003cp\u003eSpore production was significantly reduced in the response population grown in live field soil conditioned by gnoHpa-inoculated plants, compared to those grown in live field soil conditioned by healthy plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). This disease-suppressive effect was abolished in sterilized soil, but restored by supplementation with 10% live soil, confirming a microbial origin for the SBL. These findings demonstrate that foliar Hpa infection and the resulting plant response give rise to a persistent microbial community in the soil that enhances resistance in subsequent plant populations.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eA core set of HAM is consistently enriched in the Hpa-infected phyllosphere\u003c/h3\u003e\n\u003cp\u003eWe previously also showed that disease-suppressive HAM are significantly enriched in response plant populations germinating and growing in SBL soil, even when the conditioning plant population had been inoculated with HAM-free gnoHpa spores\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Based on this, we hypothesized that HAM, originally identified in the phyllosphere\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, are causative agents of disease-suppression in the SBL. We reasoned that these microbiota could accumulate around the root systems of (gno)Hpa-infected conditioning population plants and be subsequently acquired by new plantings that germinate in the same soil. However, not all HAM that are identified as significantly enriched in Hpa-infected phyllospheres are consistently detected across experiments, suggesting some degree of context-dependent selection.\u003c/p\u003e \u003cp\u003eTo identify core-HAM that are reproducibly enriched in the Hpa-infected phyllosphere, phyllosphere-derived amplicon sequencing data of 14 Hpa-infected plant populations were analyzed. These Arabidopsis plant populations had been inoculated with Hpa strains Noco2 or Cala2 and grown in a river sand-potting soil mixture, or in live Reijerscamp field soil for which the creation of an Hpa\u003cem\u003e-\u003c/em\u003esuppressive SBL had been confirmed. These 14 plant populations were part of six independent experiments performed over a span of five years\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eA total of 25 amplicon sequence variants (ASVs) representing distinct HAM were significantly enriched (\u003cem\u003eP\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05, DESeq2, Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) in more than 8 of the 14 Hpa-infected phyllospheres and were designated as \u0026lsquo;core-HAM\u0026rsquo;. These 25 core-HAM ASVs accounted for 46% \u0026minus;\u0026thinsp;75% of the bacterial abundance in the Hpa-inoculated phyllospheres, but were generally low in abundance or undetectable in healthy plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA). Among them, \u003cem\u003eXanthomonas\u003c/em\u003e ASV a0e1a and \u003cem\u003eAcidovorax\u003c/em\u003e ASV a4065 were enriched in all tested Hpa-infected phyllospheres, occupying up to 27% relative abundance of the phyllosphere bacterial community (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). \u003cem\u003eXanthomonas\u003c/em\u003e ASV a0e1a was the most abundant core-HAM ASV, averaging 10% of bacterial reads from Hpa-infected leaves. Additionally, \u003cem\u003eChryseobacterium\u003c/em\u003e ASV 3a0c1, \u003cem\u003eFlavobacterium\u003c/em\u003e ASV ef66d and \u003cem\u003eMethylophilus\u003c/em\u003e ASV e50db were enriched in 13 out of 14 Hpa-infected phyllospheres, while \u003cem\u003eAgromyces\u003c/em\u003e ASV efbd0, \u003cem\u003ePedobacter\u003c/em\u003e ASV f2a1b and \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be were enriched in 12 out of 14 Hpa\u003cem\u003e-\u003c/em\u003einfected phyllospheres, respectively (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA). These eight highly consistent ASVs together accounted for 49% relative abundance in infected leaves, while remaining low abundant (\u0026lt;\u0026thinsp;1%) in healthy phyllospheres (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eA). These results suggest that core-HAM are natural leaf colonizers strongly promoted by the Hpa-infected phyllosphere environment. The 25 ASVs representing the core-HAM cover a broad taxonomic diversity (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB), yet their abundance and taxonomic distribution remained remarkably stable across independent experiments conducted over a span of more than five years (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). This highlights the robustness of core-HAM recruitment and persistence in Hpa-infected leaves.\u003c/p\u003e\n\u003ch3\u003eDowny mildew-infected phyllospheres selectively accumulate core-HAM\u003c/h3\u003e\n\u003cp\u003eGiven the consistent enrichment of core-HAM in Hpa-infected phyllospheres, we hypothesized that these bacteria are selectively promoted during Hpa infection and gradually accumulate across successive infected plant populations. To test this, an initial population of Arabidopsis Col-0 plants was grown in live Reijerscamp field soil and inoculated with HAM-free gnoHpa spores. Leaf wash-offs, containing spores and associated microbiota, were then passaged to newly grown Col-0 plants in fresh live Reijerscamp soil. This procedure was repeated every seven days, coinciding with visible Hpa sporulation, to generate five consecutive plant populations (gnoHpa lineages; Fig. S2). Control lineages were started by spraying the first population of plants with regular HAM-containing Hpa spore suspensions (Hpa lineages) or with a mock treatment of sterile water (uninfected lineage). For each population, phyllosphere material was sampled from the passaged lineages and from newly grown untreated plants, and DNA was extracted for 16S amplicon sequencing.\u003c/p\u003e \u003cp\u003ePhyllosphere microbiome compositions were analyzed using Principal Coordinate Analysis (PCoA) based on Bray-Curtis dissimilarities (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Consistent with previous findings\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, no significant differences in phyllosphere microbiome composition were observed between untreated, uninfected, and gnoHpa-inoculated plants in the first population. However, plants inoculated with HAM-containing Hpa spore suspensions exhibited a significantly distinct microbiome compared to all other lineages (PERMANOVA results are detailed in Fig. S3 and Table S2). In later populations, however, the phyllosphere microbiomes of the passaged uninfected, gnoHpa, and Hpa lineages all showed significant differentiation from those of untreated plants (Fig. S3, Table S2, indicating that the transfer of phyllosphere microbial communities across successive plant populations influenced microbiome composition. By the third population, gnoHpa and uninfected lineages also diverged significantly, with differences becoming more pronounced in the fourth and fifth populations (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-B, Table S2). These findings indicate that downy mildew-infected plants progressively assemble distinct phyllosphere microbiomes over time.\u003c/p\u003e \u003cp\u003eTo identify ASVs selectively enriched in the gnoHpa lineages, we applied three filters: 1) consistent enrichment in the gnoHpa versus uninfected lineages (75 ASVs, \u003cem\u003eP\u003c/em\u003e\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.05, DESeq2 or ANCOM-BC, Table S3), 2) accumulation within the gnoHpa lineages over time but not in uninfected lineages (30 ASVs, \u003cem\u003eP\u003c/em\u003e\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.05, DESeq2 or ANCOM-BC, Table S4), and 3) positive correlation with pathogen levels, as quantified by qPCR (23 ASVs, \u003cem\u003eP\u003c/em\u003e\u003csub\u003eadj\u003c/sub\u003e \u0026lt; 0.05, spearman, Table S5). We identified 12 ASVs that satisfied all three selection criteria (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC, Tables S3-5). While these promoted ASVs were sporadically detected in uninfected samples, they persisted and remained abundant only in the phyllospheres of downy mildew-infected lineages after their initial appearance (Fig. S4). Their combined relative abundance increased progressively during the passages in the gnoHpa lineages and accumulated to a significantly higher cumulative relative abundance in the gnoHpa lineages compared to all other lineages (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eTo assess whether these changes in relative abundance reflected shifts in absolute bacterial load, DNA from \u003cem\u003eSalinibacter ruber\u003c/em\u003e, typically absent from soil and plant samples\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e, was spiked into samples for normalization. Collectively, the 12 ASVs enriched in gnoHpa lineages were already present in relatively high numbers in population 1 of the Hpa lineages (Fig. S5), suggesting that here they constitute HAM members introduced via Hpa inoculation. Remarkably, their absolute abundance in the gnoHpa lineages rose over successive passages, reaching levels comparable to those in the Hpa lineages and remaining significantly higher than in uninfected lineages (Fig. S5). These 12 downy mildew-associated ASVs thus appear to be sporadically occurring phyllosphere colonizers that specifically benefit from the downy mildew infected environment and progressively build up in the phyllospheres of infected plant populations.\u003c/p\u003e \u003cp\u003eInterestingly, all 12 ASVs that accumulated in the gnoHpa lineages match genera that were previously demonstrated to be downy mildew associated\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. At the ASV level, \u003cem\u003eSphingomonas\u003c/em\u003e ASV f359d and \u003cem\u003eBrevundimonas\u003c/em\u003e ASV 2cd30 have previously been found enriched in \u003cem\u003ePeronospora effusa\u003c/em\u003e-infected spinach leaves\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. Notably, the most abundant of the 12 downy mildew-associated ASVs, \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be, is part of the core-HAM (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB-C) and had also been found associated with \u003cem\u003eP. effusa\u003c/em\u003e\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. This ASV increased from 0.01\u0026ndash;8% relative abundance between population 1 and 5 of the gnoHpa lineages (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD).\u003c/p\u003e \u003cp\u003eWe further quantified the absolute abundance of \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be as 16S rRNA gene copies per gram of leaf tissue (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). In the Hpa lineages, ASV ed6be was abundant from the start (2.7 \u0026times; 10⁷ copies/g) and increased steadily with successive passages, reaching 1.5 \u0026times; 10⁸ copies/g in population 5. In the gnoHpa lineages, the prevalence of ASV ed6be rose across passages, being detected in 2, 3, 7, 8, and 10 out of 12 lineages from population 1 through 5, respectively. Its average abundance increased 2,556-fold, from 3.3 \u0026times; 10⁴ to 8.5 \u0026times; 10⁷ copies/g. In contrast, ASV ed6be was only sporadically detected in uninfected lineages (1\u0026ndash;4 out of 12 lineages per population) and remained low in abundance. Thus, while this ASV can occur in uninfected phyllospheres, it persists and proliferates specifically under the selective regime imposed by downy mildew infection.\u003c/p\u003e \u003cp\u003eCollectively, these data demonstrate that among the diverse microbiota capable of colonizing plants grown in live Reijerscamp field soil, distinct downy mildew-associated bacteria including core-HAM become consistently assembled in the phyllosphere of downy mildew-infected plants.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eCore-HAM are phyllosphere specialists that are inherited as soilborne legacy\u003c/h3\u003e\n\u003cp\u003eA live soil microbiome is required for the creation of a SBL by Hpa-infected plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA), and previous findings have suggested that HAM bacteria are assembled in both the rhizosphere and phyllosphere of response population plants grown in SBL soil\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. This supports the idea that core-HAM originate from the microbial community in the SBL soil conditioned by (gno)Hpa-infected plants and subsequently form a disease-suppressive microbiome on a next generation of plant hosts. The root endosphere may serve as a conduit for microbial migration between belowground and aboveground habitats\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. However, the connection between core-HAM colonization in rhizosphere, root endosphere, and phyllosphere of Hpa-infected plants and their establishment in successive plant populations grown in SBL soil remains unclear.\u003c/p\u003e \u003cp\u003eTo trace the origin and spatiotemporal colonization of core-HAM in Col-0 plants grown in the disease-suppressive SBL soil, we generated a detailed map of the microbiota communities in unplanted bulk soil, rhizosphere, root endosphere, and phyllosphere in both conditioning and response plant populations in a SBL experiment (Fig. S6). Bacterial community compositions in all microbiome compartments were characterized using 16S amplicon sequencing, enabling comparisons of identity and abundance across spatial compartments and time. Microbiomes in bulk soil, rhizosphere, root endosphere, and phyllosphere were significantly distinct, confirming successful separation of these microbiome compartments (PERMANOVA results are detailed in Tables S6-S7).\u003c/p\u003e \u003cp\u003eIn total, we detected 9953 different ASVs (Fig. S7), of which 9731 were detected in unplanted bulk soil samples. Only 222 ASVs were either absent or remained below the detection limit in bulk soil but appeared in plant-associated compartments. Of the ASVs detected in rhizosphere, root endosphere and phyllosphere, respectively, 99%, 97%, and 94%, were also detected in unplanted bulk soil. These ASVs cumulatively accounted for over 98% of the relative abundance in the rhizosphere and root endosphere, and 83% in the phyllosphere. Remarkably, even in the phyllosphere, 98% of the community\u0026rsquo;s relative abundance consisted of bacteria also detected belowground in the bulk soil, rhizosphere, or root endosphere. This finding supports the notion that the soil acts as a microbial reservoir for the assembly of plant-associated microbial communities, including those in the aboveground phyllosphere environment\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe then assigned each of the 9953 detected ASVs to the compartment in which it reached its highest relative abundance, resulting in 3936 \u0026ldquo;bulk soil\u0026rdquo; ASVs, 4119 \u0026ldquo;rhizosphere\u0026rdquo; ASVs, 1553 \u0026ldquo;root endosphere\u0026rdquo; ASVs, and 345 \u0026ldquo;phyllosphere\u0026rdquo; ASVs. As expected, the ASVs with high relative abundance in the bulk soil and rhizosphere categories were numerous and taxonomically diverse, with seven bacterial phyla representing more than 1% relative abundance (Fig. S8). In contrast, only a select number of ASVs in the root endosphere and phyllosphere reached the same threshold of 1% relative abundance, suggesting strong niche specialization. In this light, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA shows that the root endosphere, and especially the phyllosphere compartment, favor the growth of a select group of microbiota. For example, the 345 \u0026ldquo;phyllosphere\u0026rdquo; ASVs, which had the highest relative abundance in the phyllosphere microbiome compartment, represent only 12% of the total number of ASVs detected in the phyllosphere but accounted for 85% of the total relative abundance in this compartment. In contrast, the cumulative relative abundance of all \u0026ldquo;phyllosphere\u0026rdquo; ASVs combined in the bulk soil was below 1%. These results demonstrate that microbiota originating from the bulk soil thrive best in distinct ecological niches provided by the plant, with strong selective effects observed in the root endosphere and the phyllosphere.\u003c/p\u003e \u003cp\u003eWe next investigated whether the effect of downy mildew infection on microbiome composition was also compartment specific (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, Table S8). Consistent with earlier findings, Hpa infection significantly altered rhizosphere\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e and phyllosphere\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e communities in conditioning population plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). No significant effects were observed in bulk soil or root endosphere. Also in response populations, non-infected plants grown in disease-suppressive SBL soil exhibited distinct rhizosphere and phyllosphere microbiomes compared to those grown in control soil conditioned by mock-treated plants. The effect of SBL on microbiome composition was modest in the rhizosphere of response population plants (\u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.081 in PERMANOVA) but more pronounced in the phyllosphere (\u003cem\u003eR\u003c/em\u003e\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;=\u0026thinsp;0.35 in PERMANOVA). This paradoxically suggests that a belowground microbial legacy most strongly affects the aboveground microbiome of a subsequent planting.\u003c/p\u003e \u003cp\u003eOf the 25 core-HAM ASVs identified earlier (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB-C), 21 were also detected in the experiment described in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. These 21 ASVs were all detected in the phyllosphere and 16 were also detected in the belowground plant compartments but only 8 were detected in bulk soil (Table S9). This either suggests that the majority of core-HAMs have abundances below the detection limit in bulk soil and that their competitive colonization is particularly favored on or within the plant host, or that they originate from alternative sources. The collective abundance of the 21 core-HAM ASVs was low in the bulk soil, rhizosphere, and root endosphere (\u0026lt;\u0026thinsp;1.0%). Cumulative core-HAM ASV relative abundances significantly increased in both the rhizospheres of Hpa-infected conditioning population plants and of mock-treated response population plants grown in SBL soil, but not in the bulk soil or root endosphere of these plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eAll 21 core-HAM ASVs reached the highest relative abundances in the phyllosphere. Consistent with previous findings, they accounted for a high relative abundance (73%) in the Hpa-inoculated phyllosphere of conditioning population plants compared to mock-inoculated control plants (9%, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). Interestingly, even in healthy response plants grown in SBL soil, the relative abundance of the 21 core-HAM ASVs was significantly higher in the phyllosphere of plants grown in disease-suppressive SBL soil (61%) compared to plants grown in control soil conditioned by healthy plants (27%; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC). This shows that the microbial SBL created by Hpa-infected plants drives a robust shift in the phyllosphere microbiome of subsequent plantings, that become dominated by core-HAM.\u003c/p\u003e \u003cp\u003eTo determine whether this enrichment reflected increased bacterial population densities or displacement of phyllosphere resident microbiota, we quantified absolute abundances in each microbiome compartment using spiked-in \u003cem\u003eS. ruber\u003c/em\u003e DNA. Hpa infection of conditioning population plants significantly increased total bacterial load in both phyllosphere and rhizosphere (Fig. S9). Remarkably, also response population plants grown in SBL soil had significantly higher bacterial loads than those in control soil, but only in the phyllosphere. This increase corresponded with elevated absolute abundance of core-HAM, while the absolute abundance of other ASVs remained unchanged (Fig. S9-S10). These results confirm that plants grown in SBL soil are primarily affected aboveground, where their phyllospheres become more densely colonized, especially by disease-suppressive HAM.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eDowny mildew infection enriches a soilborne core-HAM isolate specifically in the phyllosphere\u003c/h3\u003e\n\u003cp\u003ePrevious work showed that HAM ASVs are enriched in both the rhizosphere and phyllosphere of plants grown in disease-suppressive SBL soil and that their assembly is promoted by downy mildew infection\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. However, although these ASVs originate from soil and contribute to the disease-suppressive legacy, they paradoxically accumulate most prominently in the phyllosphere of plants grown in SBL soil (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). This observation led us to hypothesize that the disease-induced assembly of core-HAM, which appears to be causal to the creation of downy mildew disease-suppressive soils, is initiated in the phyllosphere rather than the rhizosphere.\u003c/p\u003e \u003cp\u003eDowny mildew infection coincides with an increased bacterial load in the phyllosphere (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE, Fig. S9). To confirm this, we quantified bacterial colony forming units (CFU) in the phyllosphere of Arabidopsis Col-0 plants grown in live Reijerscamp field soil that were either mock-treated, or inoculated with HAM-containing Hpa or HAM-free gnoHpa spore suspensions. In both Hpa- and gnoHpa-inoculated plants, bacterial densities were significantly higher than in mock-treated controls (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), indicating that downy mildew infection, and not just the co-inoculation of HAM, promotes the proliferation of phyllosphere-associated bacteria.\u003c/p\u003e \u003cp\u003eTo investigate whether core-HAM are specifically promoted in the phyllosphere, we selected \u003cem\u003eXanthomonas\u003c/em\u003e sp. WCS2014-23 which was isolated from the roots of Hpa\u003cem\u003e-\u003c/em\u003einfected Arabidopsis plants\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e and is represented by the most abundant and robust core-HAM \u003cem\u003eXanthomonas\u003c/em\u003e ASV a0e1a. Rifampicin-resistant WCS2014-23 was mixed into live Reijerscamp field soil at 10\u003csup\u003e6\u003c/sup\u003e CFU/g soil and Col-0 plants were grown in this soil. Two-weeks-old plant were mock treated or inoculated with gnoHpa, and at 7 dpi densities of WCS2014-23 were quantified in both the rhizosphere and phyllosphere. Despite being introduced via the soil, WCS2014-23 population densities were approximately 1000-fold higher in the phyllosphere than in the rhizosphere (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-C). Rhizosphere colonization was unaffected by infection status, whereas phyllosphere colonization was significantly higher in gnoHpa-infected plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-C). When co-inoculated directly into the phyllosphere with gnoHpa, WCS2014-23 populations reached densities of 10\u003csup\u003e8.8\u003c/sup\u003e CFU per gram of diseased leaf tissue, compared to 10\u003csup\u003e7.1\u003c/sup\u003e CFU per gram of leaf tissue of healthy plants (Fig. S11). These results demonstrate that while core-HAM \u003cem\u003eXanthomonas\u003c/em\u003e WCS2014-23 originates from the soil, it preferentially colonizes the phyllosphere, where its growth is strongly promoted by downy mildew infection.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eAccumulation of HAM in the phyllosphere suppresses downy mildew disease\u003c/h2\u003e \u003cp\u003ePrevious work demonstrated that co-inoculation of HAM with gnoHpa suppresses downy mildew spore formation\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. We thus wondered whether the HAM community that assembles in the phyllosphere of healthy plants grown in disease-suppressive SBL soil provides protection against a subsequent downy mildew infections. To test this, we used our standard SBL setup (Fig. S6), conditioning Reijerscamp field soil with either Hpa- or mock-inoculated Col-0 plants. Healthy Col-0 response populations were then grown in the conditioned soils, allowing HAM-enriched phyllosphere communities to assemble on plants in SBL soil. Microbial leaf wash-offs were collected from both groups, mixed with HAM-free gnoHpa spores, and used to inoculate a third set of Col-0 plants grown in unconditioned live soil. Disease severity was quantified as spore production 7 dpi (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA).\u003c/p\u003e \u003cp\u003eBacterial leaf wash-offs from plants grown in control soil had no impact on downy mildew disease development, as the spore production on plants inoculated with gnoHpa mixed with control leaf wash-off was similar to plants inoculated with gnoHpa spores suspended in sterile water (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). In contrast, when gnoHpa spores were mixed with the HAM-enriched bacterial leaf wash-offs from plants grown in disease-suppressive SBL soil, the spore production was significantly reduced. This demonstrates that the phyllosphere microbiome assembled in healthy plants grown in disease-suppressive SBL soil confers protection against downy mildew. These findings indicate that the suppressiveness of SBL soil indeed results from the soilborne inheritance of HAM. Remarkably it is the assembly in the phyllosphere of successive plant populations that provides effective protection against downy mildew disease.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eBoth rhizosphere and phyllosphere microbiomes are crucial to sustain plant health\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e\u003c/sup\u003e. It is well established that the assembly of beneficial microbiota belowground can give rise to disease-suppressive soils\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e. In such soils, plants are protected against pathogens either by rhizobacteria-mediated production of antibiotics\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e, competition over scarce resources e.g. iron\u003csup\u003e\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e, or via the induction of systemic resistance\u003csup\u003e\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e\u003c/sup\u003e. While research on disease-suppressive soils has largely focused on their effects against root pathogens\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e, disease suppressive soils can also protect from pathogens in aboveground plant parts\u003csup\u003e\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e,\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eArabidopsis plants infected with the foliar pathogen Hpa create disease-suppressive soil by forming an SBL that enhances resistance to Hpa in successive plant populations grown in the same soil\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. However, the extent to which aboveground and belowground microbiome compartments respectively contribute to the assembly of a soilborne protective legacy that suppresses foliar downy mildew disease has remained unclear. Here, we demonstrated that downy mildew-infected plants assemble a protective core microbiome in the phyllosphere, and that these disease-suppressive microbiota subsequently form a SBL that is transmitted to the phyllosphere of successive plant populations grown in the same soil. To the best of our knowledge, this is the first indication thatthat the phyllosphere can act as an assembly hub for disease-suppressive soilborne microbiomes.\u003c/p\u003e \u003cp\u003eWe identified specific core-HAM that were consistently abundant in 14 distinct sets of Hpa-infected Arabidopsis phyllospheres, across six independent experiments conducted over a span of more than five years. Notably, we successfully recapitulated the recruitment of the core-HAM \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be by inoculating Arabidopsis plants grown in field soil with HAM-free gnoHpa spores and monitoring its buildup during serial passaging of phyllosphere leaf wash-offs. This indicates that downy mildew infection selects for specific and reproducible core-HAM enrichment. Interestingly, in a commercial field, spinach leaves naturally infected by the downy mildew \u003cem\u003eP. effusa\u003c/em\u003e were found to be enriched for ASVs identical to the downy mildew-associated ASVs in this study, including core-HAM \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e. In addition, we showed that although originating from soil, \u003cem\u003eXanthomonas\u003c/em\u003e WCS2014-23, which is representative of the most consistent and abundant core-HAM \u003cem\u003eXanthomonas\u003c/em\u003e ASV a0e1a, predominantly colonizes the phyllosphere where it is boosted upon downy mildew infection. Together, these data suggest that the selective regime imposed by downy mildew infection facilitates the consistent and specific assembly of core-HAM in the phyllosphere.\u003c/p\u003e \u003cp\u003eThe immensely diverse soil microbiome serves as a key microbial reservoir from which the phyllosphere microbiome is assembled\u003csup\u003e\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e,\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e. Our results confirm that the majority of phyllosphere microbiota are also present in the soil and are likely to originate from it. However, despite the significant taxonomical and functional overlap between rhizosphere and phyllosphere microbiota\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, our data suggests that the soilborne phyllosphere microbiota are specialized to thrive in the aboveground microbial habitat. Among these phyllosphere specialists are the core-HAM, which seem particularly well adapted to the selective environment shaped by downy mildew infections. Future studies could focus on characterizing the bacterial traits that determine phyllosphere competence and that are selectively enriched during pathogen attack. Understanding these traits could shed light on the mechanisms by which phyllosphere microbiota are assembled, establish and persist, particularly in pathogen-stressed environments, potentially offering new strategies for biocontrol and sustainable crop protection.\u003c/p\u003e \u003cp\u003eAlthough the core-HAM appear to be initially assembled in the phyllosphere of downy mildew-infected plants (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e), we found that these plants create a disease-suppressive soil that transmits the protective core-HAM to the phyllosphere of successive plant populations grown in the same soil. The routes by which core-HAM, initially assembled in the phyllosphere, buildup in the soil and migrate back to the phyllosphere from the disease-suppressive SBL soil remain to be elucidated. The endosphere has been identified as a potential route for bidirectional microbial migration between the rhizosphere and the phyllosphere\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e and an important compartment for microbial legacies\u003csup\u003e\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e. However, our data suggests that the endosphere is of minor importance for the migration of core-HAM. Firstly, only a subset of core-HAM colonizes the root endosphere. Secondly, core-HAM abundances did not increase significantly in either the root endosphere of Hpa-infected plants nor plants grown in the disease-suppressive SBL soil. Thirdly, certain core-HAM that accumulated in the phyllosphere of plants grown in SBL soil were not detected in the endophytic compartment. This suggests that core-HAM likely migrate through alternative routes. The previous findings that fully resistant Col-0 \u003cem\u003eRPP5\u003c/em\u003e plants\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e and plants disturbed in the biosynthesis of coumarins\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e do not create the disease-suppressive SBL when inoculated with HAM-containing Hpa spore suspensions indicate that disease-induced plant responses control the assembly, migration or persistence of the core-HAM as SBL. Whether the accumulation of core-HAM in the phyllosphere of plants grown in SBL soils is similarly driven by the plant, or depends on priority effects\u003csup\u003e\u003cspan additionalcitationids=\"CR52\" citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e of the soilborne core-HAM that lift on and preferentially colonize the new plants shoots as they emerge\u003csup\u003e\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e, remains to be investigated.\u003c/p\u003e \u003cp\u003eWhile the infection-induced assembly of disease-suppressive microbiota has been separately documented to occur in both the rhizosphere and phyllosphere\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e, research on disease-suppressive soils has evidently predominantly concentrated on the rhizosphere due to its direct interface with plant roots. However, our results provide the first evidence of a critical link between belowground and aboveground disease-suppressive microbiome assembly processes with a crucial role of phyllosphere microbiomes in the functioning of downy mildew disease-suppressive soils. Based on our data, we propose that the initial infection-induced assembly of disease-suppressive core-HAM in the phyllosphere is followed by their buildup in soil. These core-HAM are subsequently transmitted through the SBL to the phyllosphere of successive plant populations that are germinated and grown in the disease-suppressive SBL soil. Successive infections and spread of core-HAM, that are easily washed off from leaves\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e, to new plant populations\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e\u003c/sup\u003e could further enhance core-HAM population densities. We propose that this creates a feed-forward loop leading to core-HAM accumulation and the progressive suppression of downy mildew disease. Thus, the phyllosphere might serve as a crucial distribution hub from which disease-suppressive microbiomes - both rhizosphere and phyllosphere-associated - can disseminate throughout plant populations leading to fieldwide disease suppressiveness.\u003c/p\u003e \u003cp\u003eThis revelation raises an important question: could the phyllosphere be involved in other types of disease-suppressive soils, including those that are of agricultural relevance? Evidence supports the possibility of taxonomical and functional overlap between microbes that suppress both soilborne and foliar pathogens. For example, beneficial pseudomonads, which are known to antagonize the soilborne pathogen \u003cem\u003eGaeumannomyces tritici\u003c/em\u003e in Take-all decline soils of wheat\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e, have also been implicated in the suppression of the foliar pathogen \u003cem\u003eZymoseptoria tritici\u003c/em\u003e in the wheat phyllosphere\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e. Similarly, a beneficial \u003cem\u003eStreptomyces\u003c/em\u003e sp., initially identified for its ability to suppress Fusarium wilt disease in the rhizosphere of strawberry plants\u003csup\u003e\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e, can migrate bidirectionally throughout the plant endosphere and vasculature bundles, increasing resistance against \u003cem\u003eBotrytis cinerea\u003c/em\u003e infections in the strawberry phyllosphere\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e. Thus, the phyllosphere microbiome may indeed be a crucial component of disease-suppressive soils of agricultural relevance that has been largely overlooked in past research and warrants further investigation.\u003c/p\u003e"},{"header":"Materials and methods","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eSoil preparation and plant growth conditions\u003c/h2\u003e \u003cp\u003eField soil was collected at the Reijerscamp nature reserve in the Netherlands where an endemic population of Arabidopsis has been found (52.0107° N, 5.7825° E)\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e. The soil was air-dried and sieved (1 x 1 cm\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e) to remove rocks and plant debris. Arabidopsis accession Col-0 seeds were suspended in 0.2% (w/v) agar solution and stratified in dark conditions at 5°C for 2–5 days prior to sowing. On the day of sowing, soil was watered in a 1:10 v/w ratio, 60-mL pots were filled with 120 g of moist soil (± 2.5 g) and placed in 60-mm Petri dishes. For soilborne legacy experiments that include two growth cycles (conditioning population and response population) in the same soil, the soil surface was covered with circular cutouts of plastic micropipette-tip holders (Greiner Bio-one, 0.5–10 µL, item number 771280) to prevent algal growth and ensure consistent spatial sowing in both generations. The circular cutout was used as a sowing template and two Arabidopsis seeds were pipetted into each of 16 holes equally distributed across the soil surface. Pots were randomized in trays with closed transparent lids and incubated in a growth chamber (21°C, 70% relative humidity, 10 h light and 14 h dark, light intensity 100 µmol m\u003csup\u003e− 2\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e). The soil was watered from the bottom two times a week with 3-mL tap water. One-week after sowing, closed lids were replaced by mash-lids to reduce humidity and plants were once watered with 5-mL ½ strength Hoagland nutrient solution\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eHpa culture maintainance\u003c/h2\u003e \u003cp\u003eThe (gno)Hpa cultures of isolate Noco2\u003csup\u003e58\u003c/sup\u003e were routinely maintained on Col-0 plants, but additionally weekly inoculated onto hypersusceptible \u003cem\u003eeds1\u003c/em\u003e\u003csup\u003e59,60\u003c/sup\u003e plants to proliferate pathogenic spores and resistant Col-0 \u003cem\u003eRPP5\u003c/em\u003e\u003csup\u003e58\u003c/sup\u003e plants to check for contamination. For the Noco2 Hpa-culture, Col-0, \u003cem\u003eeds1\u003c/em\u003e and Col-0 \u003cem\u003eRPP5\u003c/em\u003e seeds were sown on Primasta© potting soil saturated with tap water. For the Noco2 gnoHpa-culture, Col-0 and \u003cem\u003eeds1\u003c/em\u003e seeds were vapour-phase sterilized\u003csup\u003e\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e\u003c/sup\u003e and sown on Murashige and Skoog (MS) \u003csup\u003e\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e\u003c/sup\u003e agar-solidified medium without sucrose. After a stratification period of 2–5 days at 5°C, plants were incubated in a growth chamber (21°C, 70% relative humidity, 16 h light and 8 h dark, light intensity 100 µmol m\u003csup\u003e− 2\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e) for 10 days. The Hpa culture was weekly passed from diseased Col-0 plants onto newly grown Col-0, \u003cem\u003eeds1\u003c/em\u003e and Col-0 \u003cem\u003eRPP5\u003c/em\u003e plants by spray inoculation with an Hpa spore suspension. The gnoHpa-culture was passed by gently touching leaves of infected Col-0 plants to leaves of healthy Col-0 and \u003cem\u003eeds1\u003c/em\u003e plants in axenic conditions. Inoculated plants were incubated in a separate growth chamber (16°C, 10 h light and 14 h dark, light intensity 100 µmol m\u003csup\u003e− 2\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003ePlant inoculation\u003c/h2\u003e \u003cp\u003e(gno)Hpa spore suspensions were prepared by collecting shoot material of the culture maintenances, that were inoculated 7–14 days prior to usage, into autoclaved tap water. Tubes were vigorously shaken to loosen the spores, plant material was filtered out with Miracloth (22–25 µM pore size) and spore density was quantified by counting three separate 1-µL droplets using a transmitted-light microscope (Carl Zeiss Microscopy, Standard 25 International Classification for Standards, item number 450815.9902). Spore suspensions between 50–100 spores/µL were directly spray-inoculated onto plants using an airbrush until clear droplet formation could be observed on the leaves. Plants were airdried for 1 h, randomized in trays and incubated in the growth chamber with closed lids that were sprinkled with water on the inside to ensure high humidity.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eDisease quantification and sampling\u003c/h2\u003e \u003cp\u003eSeven days post inoculation, infected Arabidopsis phyllosphere material was collected for disease quantification in 15-mL Greiner tubes filled with 3–6 mL water, depending on observed fresh weight and sporulation. Shoot fresh weight was quantified, Greiner tubes were hand shaken for 15 s, spores were counted in three 1-µL droplets using a transmitted-light microscope (Carl Zeiss Microscopy, Standard 25 International Classification for Standards, item number 450815.9902) and the average spore count was normalized by shoot fresh weight.\u003c/p\u003e \u003cp\u003eFor sequencing, phyllosphere material was collected by cutting the shoots with surface-sterilized razors, carefully avoiding the sampling of root or soil. The rhizosphere was sampled by picking roots and closely adhering soil with surface sterilized tweezers. Unplanted bulk soil samples were taken from the center of the pot after removing the top soil layer (approximately 2 cm). All samples were collected in 2-mL Eppendorf tubes, snap-frozen in liquid nitrogen and stored at -80°C until further processing.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eSample compartmentalization and genomic DNA extractions\u003c/h2\u003e \u003cp\u003eTwo 3-mm glass beads were added to frozen phyllosphere samples and samples were mechanically lysed using the Tissuelyser II® (Qiagen) for four cycles of 60 s at 30 Hz, snap freezing in between cycles. Rhizosphere soil and the root endosphere were separated based on Lundberg \u003cem\u003eet al\u003c/em\u003e. (2012)\u003csup\u003e\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e\u003c/sup\u003e with minor adaptations. Roots with adhering soil were washed in 1 mL phosphate-buffered saline (PBS) buffer by gently vortexing for 5 s. Next, tubes were centrifuged for 1 min at 2350 g to spin down the soil while keeping the roots floating. Root material was transferred to a new tube and this cycle was repeated a total of 5 times per sample. Clean roots were then sonicated in PBS buffer for 5 min with 5 s pauses every 30 s. Roots were dried on sterile Miracloth, snap-frozen in liquid nitrogen, and lysed with Tissuelyser II for four cycles of 60 s at 30 Hz. These samples were considered root endosphere. The rhizosphere soil that was washed-off from roots was pooled in 15-mL Greiner tubes, vigorously vortexed, centrifuged at 4700 g for 5 min. Hereafter, the supernatant was removed without disturbing the soil pellet and tubes were frozen at -80°C. Unplanted bulk soil samples remained in -80°C unprocessed until DNA extraction. All DNA was extracted using the Qiagen MagAttract PowerSoil DNA KF Kit and a ThermoFisher KingFisher® (Waltham, USA). Unplanted bulk soil, rhizosphere soil and lysed root endosphere and phyllosphere material were suspended in 750 µL PowerMag Bead solution and spiked with \u003cem\u003eS. ruber\u003c/em\u003e DNA at a concentration of 1% of the expected microbial DNA yield, determined by quantitative real-time PCR (qPCR), for unplanted bulk soil, rhizosphere and 0.1% for root endosphere, respectively. Phyllosphere samples were spiked with 1% \u003cem\u003eS. ruber\u003c/em\u003e DNA for the compartment experiment and with 0.33% for the passaging experiment. Suspended samples were added to the PowerMag Bead 96-well plate and DNA was extracted according to the manufacturer’s instructions. All DNA concentrations were quantified using a NanoDrop2000®.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eHpa quantification by qPCR\u003c/h2\u003e \u003cp\u003eHpa levels were quantified from gDNA extracted from the phyllosphere of (gno)Hpa, uninfected or untreated plants by qPCR\u003csup\u003e\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e\u003c/sup\u003e. Two-step quantitative real-time PCRs were performed in optical 96-well plates using a BioRad OPUS384 qPCR system, iTaq SYBR Green PCR Supermix (BioRad) and \u003cem\u003eArabidopsis\u003c/em\u003e and Hpa actin primers: 5’ AATCACAGCACTTGCACCA 3’ (AtActFwd), 5’ GAGGGAAGCAAGAATGGAAC 3’ (AtActRv), 5’ GTGTCGCACACTGTACCCATTTAT 3’ (HpaActFwd), 5’ ATCTTCATCATGTAGTCGGTCAAGT 3’ (HpaActRv). A standard thermal profile was used: 50°C for 2 min, 95°C for 10 min, 40 cycles of 95°C for 15 s and 60°C for 1 min. Amplicon dissociation curves were recorded after cycle 40 by heating from 60°C to 95°C with a ramp speed of 1.0°C\u003csup\u003e− 1\u003c/sup\u003e. (gno)Hpa abundance was calculated by 2\u003csup\u003e−(CtHpaACTIN − CtArabidopsisACTIN)\u003c/sup\u003e, in which CtHpaACTIN and CtArabidopsisACTIN are the cycle treshold (Ct-) values obtained within samples for the \u003cem\u003eACTIN\u003c/em\u003e PCR products for (gno)Hpa and Arabidopsis, respectively, as previously described\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e16S rDNA amplicon library preparation and sequencing analysis\u003c/h2\u003e \u003cp\u003eFor 16S rDNA amplicon sequencing of gDNA samples, library preparations were performed by Genome Quebec (Quebec, Montreal, Canada) using NextSeq chemistry (2x 300 base pairs paired-end sequencing, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) or NovaSeq chemistry (2x 250 base pairs paired-end sequencing, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The 16S variable regions 3 and 4 were amplified using the primers 16S-B341F (5′-CCTACGGGNGGCWGCAG) and 16S-B806 (5′-GACTACHVGGGTATCTAATCC) according to Genome Quebec’s standard operating protocols. Plastid- and mitochondrial-blocking peptide nucleic acids (pPNA, 5′-GGCTCAACCCTGGACAG, and mPNA, 5′- GGCAAGTGTTCTTCGGA, respectively) were used in the PCRs to prevent amplification of plant-derived sequences. 16Sr DNA amplicon sequencing datasets included in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB-C were generated as previously described\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003ePreprocessing of sequencing data was performed in the Qiime2 environment (version 2022.11)\u003csup\u003e\u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e\u003c/sup\u003e, and executed similar as described by Goossens \u003cem\u003eet al.\u003c/em\u003e (2023)\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e. Removal of primer sequences was performed using Cutadapt\u003csup\u003e\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e\u003c/sup\u003e, and quality filtering, error correction, chimaera removal and dereplication to ASVs was performed using DADA2\u003csup\u003e67\u003c/sup\u003e. Optimal DADA2 truncation and maximum expected error parameters were determined using FIGARO\u003csup\u003e\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e\u003c/sup\u003e. ASV identifiers were assigned based on the sequence-specific MD5-sums using the–p-hashed-feature-ids parameter, of which we used the first five characters in the text above to designate the individual ASVs. Taxonomic assignment of ASVs was performed using the VSEARCH plugin and the SILVA database (QIIME-compatible 132-release, 99% clustering identity, seven-level Ribosomal Database Project-compatible consensus taxonomies). ASVs with unassigned taxonomies, or that were annotated as ‘D_0__Archaea’ at the kingdom level, were removed. Moreover, plant-derived sequences were identified and removed based on the annotation of ‘D_4__Mitochondria’ at the family level, and ‘D_3__Chloroplast’ or ‘D_2__Chloroplast’ at the order or class level, respectively. ASVs that were relatively less abundant than 0.005% (comprising the lowest ~ 8% of total cumulative abundance, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) or 0.00075% (comprising the lowest ~ 3% of total cumulative abundance, Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), or that were detected in fewer than \u003cem\u003eN\u003c/em\u003e*0.5 samples (\u003cem\u003eN\u003c/em\u003e \u0026lt; 6 for Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, N \u0026lt; 5 for Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), were removed. This resulted in datasets comprising 8,801,409 reads from 439 ASVs in 250 samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e) and 53,383,924 reads from 9954 ASVs in 200 samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), including the spiked-in \u003cem\u003eS. ruber\u003c/em\u003e. For the passaging experiment, samples ‘M2w5’, ‘M4w5’, ‘M9w5’, ‘M12w5’, ‘M11w4’ and ‘M11w5’ were excluded from the analysis as qPCR showed that they were contaminated with (gno)Hpa, and sample ‘U2w1’ was removed because of a labeling error. For the microbiome compartment experiment, samples ‘B-M-G2-1’, ‘P-M-G1-4’, ‘P-M-G1-3’,’E-H-G2-10’, ‘WR-H-G2-2’, ‘E-H-G2-2’, ‘E-M-G2-1’, ‘B-M-G1-4’, ‘E-M-G2-7’, ‘E-H-G2-7’ and ‘E-H-G2-6’ were excluded from data analysis as they only had limited number of reads and as their read counts did not exceed those of blank DNA extraction controls.\u003c/p\u003e \u003cp\u003eAll alpha- and beta-diversity related calculations, graphs and differential abundance analyses were performed in R (version 3.6.3) using the phyloseq package (version 1.30.0). All PCoA ordinations and PERMANOVA tests were performed on Bray-Curtis dissimilarity matrices calculated for relative abundance data, using the vegan package (version 2.5.7) or vegan functionalities embedded in phyloseq.\u0026nbsp;For the compartment experiment, samples ‘P-M-G1-5’, ‘P-M-G2-4’ and ‘P-M-G2-2’ were obvious outliers in the PCoA and were removed from downstream analysis. PERMANOVA tests involving multiple comparisons were performed using the pairwiseAdonis package (version 0.0.4). Differential abundance testing was performed with DESeq2\u003csup\u003e69\u003c/sup\u003e (package version 1.26.0) and ANCOM-BC\u003csup\u003e\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e\u003c/sup\u003e (microbiome package version 1.8.0 and nloptr package version 1.2.2.2). As we used DESeq2 to detect differences in abundance between prevalent ASVs, we used an additional prevalence filter set at 0.5*N. For ANCOM-BC no additional filter step was used to enable the detection of structural zero’s. ASVs associated to gnoHpa lineages in the passaging experiment were selected based on three criteria: (1) ASVs that are enriched in gnoHpa lineages compared to uninfected lineages in at least 2 passages from population 2 untill population 5 (DESeq2 or ANCOM-BC). (2) ASVs that are enriched in consecutive passages (populations 2, 3, 4, 5; 3, 4, 5 or 4, 5) of gnoHpa lineages compared to population 1, but not in uninfected lineages (DESeq2 or ANCOM-BC). (3) ASVs of which the relative abundance correlates to the amount of downy mildew, as quantified by qPCR, in gnoHpa and uninfected lineages (Spearman). Absolute abundances were calculated by transforming the reads of each ASV relative to the number of reads from \u003cem\u003eS. ruber\u003c/em\u003e, multiplied by the amount of spiked-in DNA (ng) and the number of \u003cem\u003eS. ruber\u003c/em\u003e cells expected per ng of \u003cem\u003eS. ruber\u003c/em\u003e DNA (2.46*10\u003csup\u003e5\u003c/sup\u003e cells/ng). Absolute abundances were corrected for plant fresh weight for the samples from the passaging experiment (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). For the microbiome compartment experiment (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e), due to the sample processing, sample fresh weight could not be obtained and uncorrected absolute abundances were used. Graphs were made using the ggplot2 (version 3.3.5), ggpubr (version 0.4.0), UpSetR (version 1.4.0) and cowplot (version 1.1.1) packages. Statistical analyses were performed using the stats (version 3.6.3) and multcompView package (version 0.1.8). Data wrangling was done with packages from the Tidyverse suite.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eGamma-irradiated soil experiment\u003c/h2\u003e \u003cp\u003eFor the gamma-irradiated (GI) soil experiment, air-dried live Reijerscamp field soil was wrapped in two autoclave bags and tightly sealed. GI was performed by Steris Applied Sterilization Technologies. Similarly to live field soil, bags with GI-soil were stored at room temperature untill usage. Sterility of GI soil was confirmed by suspending 10 g of soil in 90 mL 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e and plating on one-tenth-strength tryptic soy agar (1/10th TSA) medium amended with 100 mg/L cycloheximide and potato dextrose agar medium (PDA) amended with 13 mg/L chloramphenicol and 150 mg/L rose bengal before usage. Live soil, GI soil and a 9:1 mix of live and GI soil were mixed, watered, potted, sown and incubated in a growth chamber as previously described, carefully avoiding any contact between the GI soil and the live soil. From this point onward, the GI soil was not kept in sterile conditions, but was considered to have a completely diminished microbiome as it had a sterile starting point. The conditioning population of Arabidopsis Col-0 plants was mock- or gnoHpa-inoculated (50 spores/µL) and the response population of Arabidopsis Col-0 plants was gnoHpa-inoculated (67 spores/µL). gnoHpa was used to avoid the co-inoculation of Hpa-associated microbiota. Disease was quantified as previously described.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eCore-HAM characterization\u003c/h2\u003e \u003cp\u003eTo identify the core-HAM that were consistently enriched upon Hpa infection, we analyzed 16S amplicon sequencing datasets from 14 distinct Hpa-infected phyllospheres from six independent experiments. Experiments 1–5 were previously reported by Goossens \u003cem\u003eet al.\u003c/em\u003e 2023\u003csup\u003e38\u003c/sup\u003e (represented by Extended data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Main Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Main Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Main Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e and main Fig.\u0026nbsp;6, respectively). Experiment 6 refers to the microbiome compartment experiment presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e of this study. Exp. 1–4 were performed in sand-potting soil mixture, whereas Exp. 5 and 6 were performed in live Reijerscamp soil. Differentially abundant ASVs were identified per experiment using DESeq2\u003csup\u003e69\u003c/sup\u003e and ASVs that were enriched in over approximately two-thirds (\u0026gt; 8 out of 14) of distinct Hpa-infected phyllospheres tested were considered core-HAM.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eThe passaging experiment\u003c/h2\u003e \u003cp\u003eArabidopsis Col-0 plants were grown in live Reijerscamp field soil as previously described and two-week-old plants were either inoculated with sterile tap water (uninfected), Hpa (80 spores/µL), gnoHpa (80 spores/µL) or remained untreated. Individual pots were placed in Eco2Boxes labeled with a unique number and incubated in a growth chamber (16°C, 10 h light and 14 h dark, light intensity 100 µmol m\u003csup\u003e− 2\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e). One week post inoculation, half of the phyllosphere, equally distributed throughout the pot, was sampled for sequencing. The other half of the plants was cut-off with a razor and suspended in 1.6 mL 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e. Leaf wash-offs were prepared by vigorously vortexing for 15 s. The leaf wash-offs were transferred to clean 2-mL perfume spraying bottles and spray-inoculated onto a new set of 2-week-old plants. During this process, we ensured that the time between obtaining the leaf wash-offs and spraying was similar between treatments and replicates, carefully avoiding any cross-contamination. Pots were airdried and incubated in a clean set of Eco2Boxes that were labeled accordingly, so that experimental passaging-lines are maintained completely separate. This enables us to directly link the phyllosphere microbiome composition per replicate lineages between passages. This process was repeated for a total of 5 successive populations. In population 5, a (gno)Hpa contamination was spotted on uninfected plants after which the experiment was terminated. For every population, an untreated set of plants was sampled for reference. gDNA was extracted as previously described and used for (gno)Hpa disease-quantification through qPCR and sequencing.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eThe microbiome compartment experiment\u003c/h2\u003e \u003cp\u003eLive Reijerscamp field soil was conditioned by mock- and Hpa-inoculated (50 spores/µL) plants that were grown and inoculated as previously described. At the end of the conditioning population, all above-ground plant biomass was removed with a razor. Directly after, a response plant population was sown on the same soil using the circular cutouts of plastic micropipette-tip holders as template to ensure that the response plant population grows in the exact same location as the conditioning plant population. The response plant population was grown as previously described and two weeks post sowing, all plants were mock-inoculated. From the conditioning and response population, the unplanted bulk soil, roots and phyllosphere were sampled and processed to separate the rhizosphere and root endosphere as previously described. DNA was extracted and samples were send for 16S rDNA amplicon sequencing. A schematic overview of this experimental setup is presented in Fig. S6\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eBacterial densities on healthy and downy mildew-infected plants\u003c/h2\u003e \u003cp\u003eFor determining phyllosphere bacterial densities on healthy and downy mildew-infected plants, plants were grown in live Reijerscamp field soil as previously described and mock-, Hpa- or gnoHpa-inoculated (50 spores/µL). The mock-, Hpa- and gnoHpa inoculums were plated in serial dilutions on 1/10th TSA medium amended with cycloheximide (100 mg/L) to prevent fungal growth and incubated at room temperature. Four days post inoculation, the number of bacterial CFU was quantified. Seven days post inoculation, phyllosphere material was collected and submerged in 3 mL 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e amended with 0.02% Silwet L77. Tubes were incubated shaking at 180 rpm for 1 h. A dilution series up to 10\u003csup\u003e7\u003c/sup\u003e was prepared using 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e and of the 10\u003csup\u003e3\u003c/sup\u003e-10\u003csup\u003e7\u003c/sup\u003e dilutions, 100 µL was plated on 1/10th TSA medium amended with cycloheximide and plates were incubated at RT. CFU-numbers were quantified following 2, 4 and 6 days of incubation at room temperature and normalized to shoot fresh weight. After 4 days, CFU number no longer increased.\u003c/p\u003e \u003cp\u003e \u003cb\u003eXanthomonas\u003c/b\u003e \u003cb\u003esp. WCS2014-23 inoculation experiments\u003c/b\u003e\u003c/p\u003e \u003cp\u003eRifampicin resistant \u003cem\u003eXanthomonas\u003c/em\u003e WCS2014-23 was cultured from − 80°C glycerol stocks on Luria-Bertani (LB) agar medium supplemented with rifampicin (50 ng/µL) at 28°C for 2–4 days. Single bacterial colonies were transferred to LB broth medium supplemented with rifampicin (50 ng/µL) and cultured at 28°C for 1–3 days. 100 mL bacterial culture was pelleted and bacterial cells were washed with 50 mL 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e three times. Optical density of the bacterial suspension was determined at 600 nm and \u003cem\u003eXanthoomonas\u003c/em\u003e was inoculated in live Reijerscamp field soil at a concentration of 10\u003csup\u003e6\u003c/sup\u003e CFU/g. The soil was vigorously mixed before being potted, sown and incubated in a growth chamber as previously described. Two weeks after sowing, plants were mock-inoculated with sterile tap water or inoculated with gnoHpa (50 spores/µL). For the experiment in which \u003cem\u003eXanthomonas\u003c/em\u003e was co-inoculated on the leaves with gnoHpa, plants were spray-inoculated with 1 mL of bacterial suspension (OD\u003csub\u003e600\u003c/sub\u003e = 0.3), airdried and directly after spray-inoculated with gnoHpa (50 spores/µL). One week after inoculation, phyllosphere material was collected in 3 mL 10 mM MgSO\u003csub\u003e4\u003c/sub\u003e amended with 0.02% silwet, incubated shaking (180 rpm) at RT for 1 h. A dilution series was plated on 1/10th TSA medium amended with 100 ng/µL rifampicin and 100 mg/L Delvocid and incubated at RT. Bacterial CFU were quantified 3 days after plating.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eSoilborne legacy leaf wash-off experiment\u003c/h2\u003e \u003cp\u003eA conditioning population of Arabidopsis Col-0 plants was grown as previously described and mock- or Hpa-inoculated (75 spores/µL). On the same soil, a response population of Col-0 plants was grown and mock-inoculated. Microbial leaf wash-offs were obtained by collecting all phyllosphere material in 5 mL sterilized tap water and vortexing for 15 s. Of these wash-offs or a sterile water control, 1 mL was then inoculated onto two-week old plant populations grown as previously described. Directly after, all pots were spray-inoculated with gnoHpa (30 spores/µL). One week post inoculation, disease was quantified as previously described.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData and code availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe experimental data and the post-processing amplicon sequencing data that support the findings of this study are available at https://github.com/JelleSpooren/Spooren-et-al-2025, together with the code used to analyze the data and generate figures. Raw amplicon sequence data generated by this study are available at https://www.ncbi.nlm.nih.gov/bioproject/1262419\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was sponsored by the Dutch Research Council (NWO) through the XL program “Unwiring beneficial functions and regulatory networks in the plant endosphere” (grant no. OCENW.GROOT.2019.063), and through the Gravitation program MiCRop (grant no. 024.004.014).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJ.S., C.M.J.P., \u0026amp; R.L.B. designed the experiments and wrote the manuscript. J.S., T.T., H.P., S.H., U.Y., and H.D. performed the experiments. J.S. and Y.S. performed the microbiome data analysis. R.Q. provided technical support in maintenance of Hpa- and gnoHpa cultures and P.G., R.Q., and S.C.M.W. provided valuable input on experimental design and execution. C.M.J.P. and R.L.B. supervised the project.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBerendsen, R. L., Pieterse, C. M. J. \u0026amp; Bakker, P. A. H. M. The rhizosphere microbiome and plant health. \u003cem\u003eTrends Plant Sci.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 478-486 (2012).\u003c/li\u003e\n\u003cli\u003eTrivedi, P., Leach, J. E., Tringe, S. G., Sa, T. \u0026amp; Singh, B. K. Plant\u0026ndash;microbiome interactions: from community assembly to plant health. \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 607-621 (2020).\u003c/li\u003e\n\u003cli\u003eBulgarelli, D., Schlaeppi, K., Spaepen, S., Van Themaat, E. V. L. \u0026amp; Schulze-Lefert, P. Structure and functions of the bacterial microbiota of plants. \u003cem\u003eAnnu. Rev. Plant Biol.\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, 807-838 (2013).\u003c/li\u003e\n\u003cli\u003eWang, X., Zhang, J., Lu, X., Bai, Y. \u0026amp; Wang, G. Two diversities meet in the rhizosphere: root specialized metabolites and microbiome. \u003cem\u003eJ. Genet. Genom.\u003c/em\u003e \u003cstrong\u003e51\u003c/strong\u003e, 467-478 (2023).\u003c/li\u003e\n\u003cli\u003eSasse, J., Martinoia, E. \u0026amp; Northen, T. Feed your friends: do plant exudates shape the root microbiome? \u003cem\u003eTrends Plant Sci.\u003c/em\u003e \u003cstrong\u003e23\u003c/strong\u003e, 25-41 (2018).\u003c/li\u003e\n\u003cli\u003eCompant, S.\u003cem\u003e et al.\u003c/em\u003e The plant endosphere world\u0026ndash;bacterial life within plants. \u003cem\u003eEnviron. Microbio\u003c/em\u003el. \u003cstrong\u003e23\u003c/strong\u003e, 1812-1829 (2021).\u003c/li\u003e\n\u003cli\u003eReinhold-Hurek, B. \u0026amp; Hurek, T. Living inside plants: bacterial endophytes. \u003cem\u003eCurr. Opin. Plant Biol.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 435-443 (2011).\u003c/li\u003e\n\u003cli\u003eYu, K., Pieterse, l.C. M. J., Bakker, P. A. H. M. \u0026amp; Berendsen, R. L. Beneficial microbes going underground of root immunity. \u003cem\u003ePlant Cell Environ.\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 2860-2870 (2019).\u003c/li\u003e\n\u003cli\u003eDoan, H. K.\u003cem\u003e et al.\u003c/em\u003e Topography-driven shape, spread, and retention of leaf surface water impacts microbial dispersion and activity in the phyllosphere. \u003cem\u003ePhytobiomes\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 268-280 (2020).\u003c/li\u003e\n\u003cli\u003ePaauw, M.\u003cem\u003e et al.\u003c/em\u003e Hydathode immunity protects the \u003cem\u003eArabidopsis\u003c/em\u003e leaf vasculature against colonization by bacterial pathogens. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 697-710 (2023).\u003c/li\u003e\n\u003cli\u003eVorholt, J. A. Microbial life in the phyllosphere. \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 828-840 (2012).\u003c/li\u003e\n\u003cli\u003eRemus-Emsermann, M. N., Tecon, R., Kowalchuk, G. A. \u0026amp; Leveau, J. H. J. Variation in local carrying capacity and the individual fate of bacterial colonizers in the phyllosphere. \u003cem\u003eISME J.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 756-765 (2012).\u003c/li\u003e\n\u003cli\u003eKusstatscher, P.\u003cem\u003e et al.\u003c/em\u003e Trichomes form genotype-specific microbial hotspots in the phyllosphere of tomato. \u003cem\u003eEnviron. Microbiome\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1-10 (2020).\u003c/li\u003e\n\u003cli\u003eSpooren, J.\u003cem\u003e et al.\u003c/em\u003e Plant-driven assembly of disease-suppressive soil microbiomes. \u003cem\u003eAnnu. Rev. Phytopathol.\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e (2024).\u003c/li\u003e\n\u003cli\u003eKim, D.-R.\u003cem\u003e et al.\u003c/em\u003e A mutualistic interaction between Streptomyces bacteria, strawberry plants and pollinating bees. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 4802 (2019).\u003c/li\u003e\n\u003cli\u003eTkacz, A., Bestion, E., Bo, Z., Hortala, M. \u0026amp; Poole, P. S. Influence of plant fraction, soil, and plant species on microbiota: a multikingdom comparison. \u003cem\u003eMBio\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 10-1128 (2020).\u003c/li\u003e\n\u003cli\u003eBai, Y.\u003cem\u003e et al.\u003c/em\u003e Functional overlap of the \u003cem\u003eArabidopsis\u003c/em\u003e leaf and root microbiota. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e528\u003c/strong\u003e, 364-369 (2015).\u003c/li\u003e\n\u003cli\u003eZhou, S. Y.\u003cem\u003e et al.\u003c/em\u003e Microbial flow within an air-phyllosphere-soil continuum. \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 615481 (2020).\u003c/li\u003e\n\u003cli\u003eMassoni, J., Bortfeld-Miller, M., Widmer, A. \u0026amp; Vorholt, J. A. Capacity of soil bacteria to reach the phyllosphere and convergence of floral communities despite soil microbiota variation. \u003cem\u003eProc.Natl. Acad. Sci. U. S. A.\u003c/em\u003e \u003cstrong\u003e118\u003c/strong\u003e, e2100150118 (2021).\u003c/li\u003e\n\u003cli\u003eBerendsen, R. L.\u003cem\u003e et al.\u003c/em\u003e Disease-induced assemblage of a plant-beneficial bacterial consortium. \u003cem\u003eISME J.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 1496-1507 (2018).\u003c/li\u003e\n\u003cli\u003eWeller, D. M., Raaijmakers, J. M., Gardener, B. B. M. \u0026amp; Thomashow, L. S. Microbial populations responsible for specific soil suppressiveness to plant pathogens. \u003cem\u003eAnnu. Rev. Phytopathol.\u003c/em\u003e \u003cstrong\u003e40\u003c/strong\u003e, 309-348 (2002).\u003c/li\u003e\n\u003cli\u003ePfeilmeier, S.\u003cem\u003e et al.\u003c/em\u003e The plant NADPH oxidase RBOHD is required for microbiota homeostasis in leaves. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 852-864 (2021).\u003c/li\u003e\n\u003cli\u003eGao, M.\u003cem\u003e et al.\u003c/em\u003e Disease-induced changes in plant microbiome assembly and functional adaptation. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 187 (2021).\u003c/li\u003e\n\u003cli\u003eLiu, H.\u003cem\u003e et al.\u003c/em\u003e Evidence for the plant recruitment of beneficial microbes to suppress soil-borne pathogens. \u003cem\u003eNew Phytol.\u003c/em\u003e \u003cstrong\u003e229\u003c/strong\u003e, 2873-2885 (2021).\u003c/li\u003e\n\u003cli\u003eRolfe, S. A., Griffiths, J. \u0026amp; Ton, J. Crying out for help with root exudates: adaptive mechanisms by which stressed plants assemble health-promoting soil microbiomes. \u003cem\u003eCurr. Opin. Microbiol.\u003c/em\u003e \u003cstrong\u003e49\u003c/strong\u003e, 73-82 (2019).\u003c/li\u003e\n\u003cli\u003eRizaludin, M. S., Stopnisek, N., Raaijmakers, J. M. \u0026amp; Garbeva, P. The chemistry of stress: understanding the \u0026lsquo;cry for help\u0026rsquo;of plant roots. \u003cem\u003eMetabolites\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 357 (2021).\u003c/li\u003e\n\u003cli\u003eSchlatter, D., Kinkel, L., Thomashow, L., Weller, D. \u0026amp; Paulitz, T. Disease suppressive soils: new insights from the soil microbiome. \u003cem\u003ePhytopathology\u003c/em\u003e \u003cstrong\u003e107\u003c/strong\u003e, 1284-1297 (2017).\u003c/li\u003e\n\u003cli\u003eG\u0026oacute;mez Exp\u0026oacute;sito, R., De Bruijn, I., Postma, J. \u0026amp; Raaijmakers, J. M. Current insights into the role of rhizosphere bacteria in disease suppressive soils. \u003cem\u003eFront. Microbiol.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2529 (2017).\u003c/li\u003e\n\u003cli\u003eCarri\u0026oacute;n, V. J.\u003cem\u003e et al.\u003c/em\u003e Pathogen-induced activation of disease-suppressive functions in the endophytic root microbiome. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e366\u003c/strong\u003e, 606-612 (2019).\u003c/li\u003e\n\u003cli\u003eRaaijmakers, J. M. \u0026amp; Weller, D. M. Natural plant protection by 2, 4-diacetylphloroglucinol-producing \u003cem\u003ePseudomonas \u003c/em\u003espp. in take-all decline soils. \u003cem\u003eMol. Plant Microbe Interact.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 144-152 (1998).\u003c/li\u003e\n\u003cli\u003eWeller, D. M.\u003cem\u003e et al.\u003c/em\u003e Role of 2, 4-diacetylphloroglucinol-producing fluorescent \u003cem\u003ePseudomonas\u003c/em\u003e spp. in the defense of plant roots. \u003cem\u003ePlant Biol. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 4-20 (2007).\u003c/li\u003e\n\u003cli\u003eMendes, R.\u003cem\u003e et al.\u003c/em\u003e Deciphering the rhizosphere microbiome for disease-suppressive bacteria. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e332\u003c/strong\u003e, 1097-1100 (2011).\u003c/li\u003e\n\u003cli\u003eVogel, C. M., Potthoff, D. B., Sch\u0026auml;fer, M., Barandun, N. \u0026amp; Vorholt, J. A. Protective role of the \u003cem\u003eArabidopsis\u003c/em\u003e leaf microbiota against a bacterial pathogen. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1537-1548 (2021).\u003c/li\u003e\n\u003cli\u003eLiu, X.\u003cem\u003e et al.\u003c/em\u003e Phyllosphere microbiome induces host metabolic defence against rice false-smut disease. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1419-1433 (2023).\u003c/li\u003e\n\u003cli\u003eFrancisco, C. S.\u003cem\u003e et al.\u003c/em\u003e The apoplastic space of two wheat genotypes provide highly different environment for pathogen colonization: Insights from proteome and microbiome profiling. \u003cem\u003ebioRxiv\u003c/em\u003e, 543792 (2023).\u003c/li\u003e\n\u003cli\u003eEhau-Taumaunu, H. \u0026amp; Hockett, K. L. Passaging phyllosphere microbial communities develop suppression towards bacterial speck disease in tomato. \u003cem\u003ePhytobiomes J.l\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 233-243 (2023).\u003c/li\u003e\n\u003cli\u003eLi, P.-D.\u003cem\u003e et al.\u003c/em\u003e The phyllosphere microbiome shifts toward combating melanose pathogen. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 56 (2022).\u003c/li\u003e\n\u003cli\u003eGoossens, P.\u003cem\u003e et al.\u003c/em\u003e Obligate biotroph downy mildew consistently induces near-identical protective microbiomes in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2349-2364 (2023). \u003c/li\u003e\n\u003cli\u003eBakker, P. A. H. M., Pieterse, C. M. J., de Jonge, R. \u0026amp; Berendsen, R. L. The soil-borne legacy. \u003cem\u003eCell\u003c/em\u003e \u003cstrong\u003e172\u003c/strong\u003e, 1178-1180 (2018).\u003c/li\u003e\n\u003cli\u003eVismans, G.\u003cem\u003e et al.\u003c/em\u003e Coumarin biosynthesis genes are required after foliar pathogen infection for the creation of a microbial soil-borne legacy that primes plants for SA-dependent defenses. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 22473 (2022).\u003c/li\u003e\n\u003cli\u003eStringlis, I. A., De Jonge, R. \u0026amp; Pieterse, C. M. J. The age of coumarins in plant\u0026ndash;microbe interactions. \u003cem\u003ePlant Cell Physiol.\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 1405-1419 (2019).\u003c/li\u003e\n\u003cli\u003eSt\u0026auml;mmler, F.\u003cem\u003e et al.\u003c/em\u003e Adjusting microbiome profiles for differences in microbial load by spike-in bacteria. \u003cem\u003eMicrobiome\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 1-13 (2016).\u003c/li\u003e\n\u003cli\u003eGoossens, P.\u003cem\u003e et al.\u003c/em\u003e Selective enrichment of specific bacterial taxa in downy mildew-affected spinach: Comparative analysis in laboratory and field conditions. \u003cem\u003eBioRxiv\u003c/em\u003e, 609345 (2024).\u003c/li\u003e\n\u003cli\u003eKwak, M.-J.\u003cem\u003e et al.\u003c/em\u003e Rhizosphere microbiome structure alters to enable wilt resistance in tomato. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 1100-1109 (2018).\u003c/li\u003e\n\u003cli\u003ede Sousa, L. P. \u0026amp; Mondego, J. M. C. Leaf surface microbiota transplantation confers resistance to coffee leaf rust in susceptible \u003cem\u003eCoffea arabica\u003c/em\u003e. \u003cem\u003eFEMS Microbiol. Ecol.\u003c/em\u003e \u003cstrong\u003e100\u003c/strong\u003e, fiae049 (2024).\u003c/li\u003e\n\u003cli\u003eGu, S.\u003cem\u003e et al.\u003c/em\u003e Competition for iron drives phytopathogen control by natural rhizosphere microbiomes. \u003cem\u003eNat. Microbiol.\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, 1002-1010 (2020). \u003c/li\u003e\n\u003cli\u003eH\u0026ouml;fte, M. \u0026amp; Bakker, P. A. H. M. in \u003cem\u003eMicrobial siderophores\u003c/em\u003e (eds A Varma \u0026amp; S. B. Chincholkar) 121-133 (Springer, 2007).\u003c/li\u003e\n\u003cli\u003eWeller, D. M.\u003cem\u003e et al.\u003c/em\u003e Disease-suppressive soils induce systemic resistance in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e against \u003cem\u003ePseudomonas syringae\u003c/em\u003e pv. tomato. \u003cem\u003ePhytoFront.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 515-523 (2024).\u003c/li\u003e\n\u003cli\u003ePieterse, C. M. J.\u003cem\u003e et al.\u003c/em\u003e Induced systemic resistance by beneficial microbes. \u003cem\u003eAnnu. Rev. Phytopathol.\u003c/em\u003e \u003cstrong\u003e52\u003c/strong\u003e, 347-375 (2014).\u003c/li\u003e\n\u003cli\u003eHannula, S. E.\u003cem\u003e et al.\u003c/em\u003e Persistence of plant-mediated microbial soil legacy effects in soil and inside roots. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 5686 (2021).\u003c/li\u003e\n\u003cli\u003eDebray, R., Conover, A., Zhang, X., Dewald-Wang, E. A. \u0026amp; Koskella, B. Within-host adaptation alters priority effects within the tomato phyllosphere microbiome. \u003cem\u003eNat. Ecol. \u0026amp; Evol.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 725-731 (2023).\u003c/li\u003e\n\u003cli\u003eDebray, R.\u003cem\u003e et al.\u003c/em\u003e Priority effects in microbiome assembly. \u003cem\u003eNat. Rev. Microbiol.\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 109-121 (2022). \u003c/li\u003e\n\u003cli\u003eCarlstr\u0026ouml;m, C. I.\u003cem\u003e et al.\u003c/em\u003e Synthetic microbiota reveal priority effects and keystone strains in the \u003cem\u003eArabidopsis \u003c/em\u003ephyllosphere. \u003cem\u003eNat. Ecol. \u0026amp; Evol.\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1445-1454 (2019).\u003c/li\u003e\n\u003cli\u003eRaaijmakers, J. M., Van Der Sluis, I., Van Den Hout, M., Bakker, P. A. H. M. \u0026amp; Schippers, B. Dispersal of wild-type and genetically-modified \u003cem\u003ePseudomonas\u003c/em\u003e spp from treated seeds or soil to aerial parts of radish plants. \u003cem\u003eSoil Biol. Biochem.\u003c/em\u003e \u003cstrong\u003e27\u003c/strong\u003e, 1473-1478 (1995).\u003c/li\u003e\n\u003cli\u003eCevallos-Cevallos, J. M., Danyluk, M. D., Gu, G., Vallad, G. E. \u0026amp; van Bruggen, A. H. Dispersal of \u003cem\u003eSalmonella\u003c/em\u003e Typhimurium by rain splash onto tomato plants. \u003cem\u003eJ. Food Prot.\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 472-479 (2012).\u003c/li\u003e\n\u003cli\u003eCha, J.-Y.\u003cem\u003e et al.\u003c/em\u003e Microbial and biochemical basis of a \u003cem\u003eFusarium\u003c/em\u003e wilt-suppressive soil. \u003cem\u003eISME J.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 119-129 (2016).\u003c/li\u003e\n\u003cli\u003ePieterse, C. M. J., Van Wees, S. C., Hoffland, E., Van Pelt, J. A. \u0026amp; Van Loon, L. C. Systemic resistance in \u003cem\u003eArabidopsis \u003c/em\u003einduced by biocontrol bacteria is independent of salicylic acid accumulation and pathogenesis-related gene expression. \u003cem\u003ePlant Cell\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 1225-1237 (1996).\u003c/li\u003e\n\u003cli\u003eParker, J. E.\u003cem\u003e et al.\u003c/em\u003e Phenotypic characterization and molecular mapping of the \u003cem\u003eArabidopsis thaliana \u003c/em\u003elocus RPP5, determining disease resistance to \u003cem\u003ePeronospora parasitica\u003c/em\u003e. \u003cem\u003ePlant J.\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 821-831 (1993).\u003c/li\u003e\n\u003cli\u003eParker, J. E.\u003cem\u003e et al.\u003c/em\u003e Characterization of \u003cem\u003eeds1\u003c/em\u003e, a mutation in \u003cem\u003eArabidopsis \u003c/em\u003esuppressing resistance to \u003cem\u003ePeronospora parasitica\u003c/em\u003e specified by several different \u003cem\u003eRPP\u003c/em\u003e genes. \u003cem\u003ePlant Cell\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2033-2046 (1996).\u003c/li\u003e\n\u003cli\u003eAarts, N.\u003cem\u003e et al.\u003c/em\u003e Different requirements for \u003cem\u003eEDS1\u003c/em\u003e and \u003cem\u003eNDR1\u003c/em\u003e by disease resistance genes define at least two R gene-mediated signaling pathways in \u003cem\u003eArabidopsis\u003c/em\u003e. \u003cem\u003eProc.Natl. Acad. Sci. U. S. A\u003c/em\u003e\u003cstrong\u003e 95\u003c/strong\u003e, 10306-10311 (1998).\u003c/li\u003e\n\u003cli\u003eLindsey III, B. E., Rivero, L., Calhoun, C. S., Grotewold, E. \u0026amp; Brkljacic, J. Standardized method for high-throughput sterilization of \u003cem\u003eArabidopsis\u003c/em\u003e seeds. \u003cem\u003eJ. Vis. Exp.\u003c/em\u003e, e56587 (2017).\u003c/li\u003e\n\u003cli\u003eMurashige, T. \u0026amp; Skoog, F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. \u003cem\u003ePhysiol. Plant.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 473-497 (1962).\u003c/li\u003e\n\u003cli\u003eLundberg, D. S.\u003cem\u003e et al.\u003c/em\u003e Defining the core \u003cem\u003eArabidopsis thaliana\u003c/em\u003e root microbiome. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e488\u003c/strong\u003e, 86-90 (2012).\u003c/li\u003e\n\u003cli\u003eAnderson, R. G. \u0026amp; McDowell, J. M. A PCR assay for the quantification of growth of the oomycete pathogen \u003cem\u003eHyaloperonospora arabidopsidis\u003c/em\u003e in \u003cem\u003eArabidopsis thaliana\u003c/em\u003e. \u003cem\u003eMol. Plant Pathol.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 893-898 (2015).\u003c/li\u003e\n\u003cli\u003eBolyen, E.\u003cem\u003e et al.\u003c/em\u003e Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 852-857 (2019).\u003c/li\u003e\n\u003cli\u003eMartin, M. Cutadapt removes adapter sequences from high-throughput sequencing reads. \u003cem\u003eEMBnet. J.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 10-12 (2011).\u003c/li\u003e\n\u003cli\u003eCallahan, B. J.\u003cem\u003e et al.\u003c/em\u003e DADA2: High-resolution sample inference from Illumina amplicon data. \u003cem\u003eNat. Methods\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 581-583 (2016).\u003c/li\u003e\n\u003cli\u003eWeinstein, M. M., Prem, A., Jin, M., Tang, S. \u0026amp; Bhasin, J. M. FIGARO: An efficient and objective tool for optimizing microbiome rRNA gene trimming parameters. \u003cem\u003eBioRxiv\u003c/em\u003e, 610394 (2019).\u003c/li\u003e\n\u003cli\u003eLove, M. I., Huber, W. \u0026amp; Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. \u003cem\u003eGenome Biol.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1-21 (2014).\u003c/li\u003e\n\u003cli\u003eLin, H. \u0026amp; Peddada, S. D. Analysis of compositions of microbiomes with bias correction. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 3514 (2020).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6693507/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6693507/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003ePlants can respond to pathogen attack by assembling disease-suppressive soil microbiomes. In \u003cem\u003eArabidopsis thaliana\u003c/em\u003e, infection by the obligate foliar downy mildew pathogen \u003cem\u003eHyaloperonospora arabidopsidis\u003c/em\u003e (Hpa) consistently led to the formation of a soil microbial community, termed the soilborne legacy (SBL), that enhanced resistance in subsequent plant populations grown in the same soil. Previous work identified an enrichment of specific Hpa-associated microbiota (HAM) in the phyllospheres of infected plants, which suppressed pathogen proliferation. However, the relationship between rhizosphere and phyllosphere microbiota in generating the SBL and assembling protective HAM remained unclear. Here, we identified a community of 25 core-HAM that consistently dominated the phyllospheres of 14 sets of distinct Hpa-infected plant populations across six independent experiments. Using HAM-free, gnotobiotic Hpa spores, the infection-driven assembly of the core-HAM member \u003cem\u003eSphingobium\u003c/em\u003e ASV ed6be was recapitulated, showing \u003cem\u003ede novo\u003c/em\u003e and progressive accumulation under sustained disease pressure. Although HAM transmission in SBL occurred via soil, these bacteria were shown to be phyllosphere specialists, accumulating more abundantly on aboveground than belowground tissues. Moreover, leaf wash-offs from plant populations that inherited SBL, effectively suppressed downy mildew disease when applied to leaves of plants grown in unconditioned soil. These findings reveal that downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere, highlighting a crucial link between belowground and aboveground plant-driven microbiome assembly processes. Paradoxically, the phyllosphere thus emerges as a central hub for the accumulation of disease-suppressive soil microbiomes.\u003c/p\u003e","manuscriptTitle":"Downy mildew disease-suppressive soils transmit a protective core microbiome to the phyllosphere","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-12 09:12:15","doi":"10.21203/rs.3.rs-6693507/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9df20085-6965-4f69-a720-92a4c6f22ecd","owner":[],"postedDate":"June 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":49900239,"name":"Biological sciences/Plant sciences/Plant symbiosis/Parasitism"},{"id":49900240,"name":"Biological sciences/Microbiology"},{"id":49900241,"name":"Biological sciences/Ecology/Microbial ecology"},{"id":49900242,"name":"Biological sciences/Plant sciences/Plant immunity/Microbe"}],"tags":[],"updatedAt":"2025-07-21T16:15:53+00:00","versionOfRecord":[],"versionCreatedAt":"2025-06-12 09:12:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6693507","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6693507","identity":"rs-6693507","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}