Microbial diversity in soils suppressive to Fusarium diseases

Fusarium species are cosmopolitan soil phytopathogens from the division Ascomycota, which produce mycotoxins and cause significant economic losses of crop plants. However, soils suppressive to Fusarium diseases are known to occur, and recent knowledge on microbial diversity in these soils has shed new lights on phytoprotection effects. In this review, we synthesize current knowledge on soils suppressive to Fusarium diseases and the role of their rhizosphere microbiota in phytoprotection. This is an important issue, as disease does not develop significantly in suppressive soils even though pathogenic Fusarium and susceptible host plant are present, and weather conditions are suitable for disease. Soils suppressive to Fusarium diseases are documented in different regions of the world. They contain biocontrol microorganisms, which act by inducing plants’ resistance to the pathogen, competing with or inhibiting the pathogen, or parasitizing the pathogen. In particular, some of the Bacillus, Pseudomonas, Paenibacillus and Streptomyces species are involved in plant protection from Fusarium diseases. Besides specific bacterial populations involved in disease suppression, next-generation sequencing and ecological networks have largely contributed to the understanding of microbial communities in soils suppressive or not to Fusarium diseases, revealing different microbial community patterns and differences for a notable number of taxa, according to the Fusarium pathosystem, the host plant and the origin of the soil. Agricultural practices can significantly influence soil suppressiveness to Fusarium diseases by influencing soil microbiota ecology. Research on microbial modes of action and diversity in suppressive soils should help guide the development of effective farming practices for Fusarium disease management in sustainable agriculture.


Introduction
The fungal genus Fusarium encompasses several plantpathogenic species, which are among the most destructive phytopathogens world-wide, causing diseases on many agricultural crops (Burgess and Bryden, 2012).They are ubiquitous in parts of the world where cereals and other crops are grown and they produce a wide variety of mycotoxins, which may be present in feed and food products (Babadoost, 2018;Moretti et al., 2018;Chen et al., 2019).Consumption of products that are contaminated with mycotoxins may cause acute or chronic effects in both animals and humans, and could result in immune-suppressive or carcinogenic effects (Jard et al., 2011).By producing mycotoxins and by inducing necrosis and wilting in plants, Fusarium fungi are causing huge economic losses of cereal crops throughout the world (Khan et al., 2017).Their broad distribution has been attributed to their ability to develop on different substrates and plant species, and to produce spores that enable efficient propagation (Desjardins, 2006;Arie, 2019).They are typical soil-borne microorganisms, routinely found in plantassociated fungal communities (Reyes Gaige et al., 2020).
Efficient management of plant diseases caused by Fusarium is important to limit crop losses and to reduce mycotoxin production in alimentary products (Babadoost, 2018).Because mycotoxin synthesis can occur not only after harvesting but also before, one of the best ways to reduce its presence in food and feed products is to prevent its formation in the crop (Jard et al., 2011).Over the years, different methods, such as the use of resistant cultivars and chemical fungicides, have been undertaken in order to control or prevent crop diseases (Willocquet et al., 2021).In spite of that, Fusarium continues to cause huge crop losses, up to 70% in South America, 54% in the United States and 50% in Europe in the case of Fusarium head blight (FHB) disease of wheat (Scott et al., 2021).
Alternative control methods, based on plant-protection effects of beneficial microorganisms, have also been investigated (Janvier et al., 2007;Nguyen et al., 2018).Farming practices greatly influence these effects by shaping the rhizosphere microbial community (Campos et al., 2016), stimulating the activity of beneficial rhizosphere microorganisms and restricting the activity of soilborne Fusarium pathogens (Janvier et al., 2007).Indeed, crop rotation, tillage and addition of organic amendments may provide some control of soil-borne pathogens, through different microbial direct and indirect mechanisms (Janvier et al., 2007).The effect of plant-protecting soil microbiota on plant health is of particular interest in the case of disease-suppressive soils, which were defined by Baker and Cook (1974) as "soils in which the pathogen does not establish or persist, establishes but causes little or no damage, or establishes and causes disease for a while but thereafter the disease is less important, although the pathogen may persist in the soil".Suppressive soils represent a reservoir of beneficial microorganisms, which may confer effective plant protection against various soilborne diseases (Goḿez Expośito et al., 2017).This biocontrol potential of suppressive soils is of great importance when considering phytopathogens like Fusarium spp., which are causing increasing damage to crops in the on-going climate change context (Babadoost, 2018).Insight into the time and space microbial dynamics of soils suppressive to Fusarium diseases, together with the understanding of microbial modes of action and agricultural practices applied, is needed in order to develop safe, effective, and stable tools for disease management (Goḿez Expośito et al., 2017).
By selecting their rhizosphere microbiome (Tkacz et al., 2015;Gruet et al., 2023), plants may contribute themselves to suppressiveness (Almario et al., 2014;Goḿez Expośito et al., 2017).Soil represents the richest known reservoir of microbial biodiversity (Curtis et al., 2002;Wang et al., 2016) and displays several compartments, i.e. the bulk soil containing microorganisms that are not affected by the roots, the rhizosphere where soil microorganisms are under the influence of roots (and roots exudates), the rhizoplane with root-adhering microorganisms, and the endosphere for root tissues colonized by microorganisms (Sańchez-Cañizares et al., 2017).The rhizosphere and rhizoplane harbor an abundant community of bacteria, archaea, oomycetes and fungi, whose individual members can have beneficial, deleterious or neutral effects on the plant.The collective genome of this microbial community is larger than that of the plant itself, and is often referred to as the plant's second genome (Berendsen et al., 2012).Thus, this alliance of the plant and its associated microorganisms represents a holobiont, which has interdependent, fine-tuned and complex functioning (Berendsen et al., 2012;Vandenkoornhuyse et al., 2015;Sańchez-Cañizares et al., 2017).In this system, the plant is a key player, as nearly 40% of all photosynthates are released directly by roots into the rhizosphere, serving as a fuel for microbial communities, thus recruiting and shaping this microbiome (Berendsen et al., 2012;Tkacz and Poole, 2015).These photosynthates are conditioned by the plant genotype, developmental stage, metabolism, immune system and its ability to exudate (Sańchez-Cañizares et al., 2017).In this context, suppressiveness will depend on microbiome diversity and functioning.
This review deals with recent knowledge on soils suppressive to Fusarium diseases, which sheds new lights on molecular and ecological mechanisms underpinning phytoprotection effects and highlights the importance of microbial diversity in the functioning of these suppressive soils.To this end, we summarize current knowledge on Fusarium taxonomy and ecology, and their mechanisms of plant infection.In addition, we review our understanding of biocontrol agents against Fusarium and their modes of action.Finally, we focus on soils suppressive to Fusarium diseases and the importance of farming and environmental factors modulating suppressiveness, with an emphasis on the particularities of the different Fusarium pathosystems.(Smith, 2007;Summerell, 2019).Some can also be found in caves (Bastian et al., 2010) or in man-made water systems (Sautour et al., 2012).Fusarium species are mostly known as phytopathogens, but some of them have been evidenced as contaminants in industrial processes, indoor environments, or pharmaceutical and food products (Abdel-Azeem et al., 2019), whereas others behave as opportunistic human/animal pathogens (Al-Hatmi et al., 2019;da Silva Santos et al., 2020) or are fungicolous (Torbati et al., 2021).
Focusing on plant-interacting Fusarium species, their saprophytic potential enables them to survive the winter in the crop debris, in the form of mycelium or spores that serve as plantinfecting propagules in the spring (Figure 1A) (Leslie and Summerell, 2006).Fusarium species vary in reproduction strategies, and they produce sexual spores as well as three types of asexual spores, i.e. (i) microconidia, which are typically produced under all environmental conditions, (ii) macroconidia, which are often found on the surface of diseased plants, and (iii) chlamydospores (survival structures), which are thick walled and produced from macroconidia or older mycelium (Ajmal et al., 2023).More than 80% of Fusarium species propagate using asexual spores, but not all of them produce all three types of spores, while sexual reproduction can involve self-fertility or outcrossing (Rana et al., 2017).Additionally, some species produce sclerotia, which promote survival in soil (Leslie and Summerell, 2006).

Taxonomy of Fusarium spp.
The Fusarium genus exhibits high level of variability in terms of morphological, physiological and ecological properties, which represents a difficulty in establishing a consistent taxonomy of these species (Burgess et al., 1996).An additional difficulty for classification is the existence of both asexual (anamorph) and sexual (teleomorph) phases in their life cycle (Summerell, 2019).Based on the most widely used classification, the anamorph state of the genus Fusarium is classified in the family Nectriaceae, order Hypocreales and division Ascomycota (Crous et al., 2021).Several teleomorphs have been related to Fusarium species, but not all Fusarium species have a known sexual state in their life cycle (Munkvold, 2017).Most of these teleomorphs are in the genus Gibberella, including the economically important pathogens, such as G. zeae (anamorph F. verticillioides) and G. moniliformis (anamorph F. verticillioides) (Keszthelyi et al., 2007).Other Fusarium teleomorphs are members of the genera Albonectria, Neocosmospora or Haematonectria.Teleomorphs are usually not observed in the field, but rather under lab conditions.The dual anamorph-teleomorph nomenclature for fungi has now been abolished, and the name Fusarium has been retained for these fungi (Geiser et al., 2013).
Over the past 100 years, the taxonomy of Fusarium has undergone many changes, but most classification procedures have been based on the size and shape of the macroconidia, the presence or absence of microconidia and chlamydospores, and the structure of the conidiophores (Ristic, 2012).Identification of Fusarium species based on morphological characteristics also included observations of colony pigmentation and type of aerial mycelium (Crous et al., 2021).The standard method now used to identify Fusarium isolates to a species level is to sequence one (or more) of the following genes: translocation elongation factor-1a (tef-1a), RNA polymerase 1 and 2 (rpb1 and rpb2), b-tubulin (tub), histone (his), ATP citrate lyase (acl1) or calmodulin (CaM) (Herron et al., 2015;Summerell, 2019;Crous et al., 2021;Laraba et al., 2021;Yilmaz et al., 2021).The tef-1a gene is a first-choice marker as it has good resolution power for the majority of Fusarium species, while sequencing the gene rpb2 allows differentiation of close species.The other genetic markers mentioned have variable resolution power and are often used together with tef-1a or rpb2 (Crous et al., 2021).The internal transcribed spacer regions of the ribosomal gene (ITS), which are common barcodes to identify fungi, are not recommended for Fusarium identification, as they are not sufficiently informative for a significant number of Fusarium species (Summerell, 2019).

Mechanisms of Fusarium infection, symptoms and etiology
Before infecting the host plant tissues, soil-borne pathogens may grow in the rhizosphere or on the host as saprophytes, managing to escape the rhizosphere battlefield (Raaijmakers et al., 2009).The outcome is directly influenced by host and microbial defense mechanisms, at the level of the holobiont (Berendsen et al., 2012;Vandenkoornhuyse et al., 2015).During their life cycle, plants are exposed to numerous phytopathogens, and they have developed different adaptive strategies.Upon pathogen attack, both composition and quantity of root metabolites may change (Rolfe et al., 2019), which can be useful for direct defense against the pathogens (Rizaludin et al., 2021), for signaling the impending Phylogenetic relationship between pathogenic Fusarium species and 15 different species complexes.The distance-method tree (1000 bootstrap replicates) was inferred from the rpb1 (RNA Polymerase 1) data set, using the SeaView multiplatform (Gouy et al., 2010).The tree was visualized using iTol (Letunic and Bork, 2021).Sphaerostilbella aureonitens NRRL 13992 was used as an outgroup.Species complexes delimitation is based on the phylogeny published in Summerell (2019).Todorovic ´et al. 10.3389/fpls.2023.1228749Frontiers in Plant Science frontiersin.orgthreat to the neighboring plants (Peĺissier et al., 2021), or for recruiting beneficial microorganisms with biocontrol capabilities.The latter phenomenon is referred to as the 'a cry for help' strategy (Rizaludin et al., 2021).
If the pathogen manages to escape from the rhizosphere battlefield, the infection cycle can proceed.Plant infection by Fusarium occurs in a few successive stages (Figure 1A), which differs according to Fusarium species.Seeds infected with Fusarium in the previous season can also serve as disease initiators (Jimeńez-Dıáz et al., 2015).F. graminearum grows saprophytically on crop debris, which is the overwintering reservoir of the pathogen (Brown et al., 2010).The fungus may infect roots and cause damage to the collar (Ares et al., 2004).During the crop anthesis and under warm and humid weather conditions, asexual conidia, sexual ascospores or chlamydospores are dispersed by rain or wind and reach the outer anthers and outer glumes of the plant.After spore germination, hyphae penetrate the host plant through the cracked anthers, followed by inter-and intracellular mycelial growth, resulting in damage to host tissues and especially head blight disease (Brown et al., 2010).Unlike F. graminearum, F. culmorum produces only asexual conidia and chlamydospores, which are also dispersed by rain and wind, reaching plant heads and infecting the ears during the anthesis.Subsequently, conidia germinate on the lemma and palea, followed by inter-and intracellular mycelial growth (Wagacha and Muthomi, 2007).In contrast, the infection cycle of F. oxysporum begins when mycelia, germinating asexual conidia or chlamydspores enter the healthy plant through the root tip, lateral roots or root wounds.The fungus progresses intracellularly, entering the xylem sap flow and being transported to the aerial parts of the plant where it forms infection structures.The infection structures that form close the vascular vessels, disrupt nutrient translocation, leading to stomatal closure, leaf wilting and plant death (Banerjee and Mittra, 2018;Redkar et al., 2022a;Redkar et al., 2022b).In the case of F. verticillioides, infection starts when mycelia, asexual conidia or sexual ascospores are carried inside the seed or on the seed surface and later develop inside the growing plant, moving from the roots up to the maize kernels (Oren et al., 2003;Gai et al., 2018).Sometimes, the fungus colonizes and grows along the veins of the plant root, while sometimes it manages to penetrate the plant cells and form internal hyphae, therefore causing damage (Lei et al., 2011;Blacutt et al., 2018).Finally, for F. solani, the attachment of mycelia, asexual conidia, sexual ascospores or chlamydospores to the susceptible host is the first step in disease development, after which the fungus enters the host through stomata or the epidermis.Following penetration, F. solani is able to spread through the xylem, ultimately causing wilting of the host plant (Coleman, 2016).
It is reported that mycotoxins play a key role in pathogenesis, and that the aggressiveness of Fusarium depends on its toxin-producing capacity (Mesterhaźy, 2002;Xia et al., 2019;Laraba et al., 2021;Senatore et al., 2021;Yilmaz et al., 2021).Several mycotoxins are produced by Fusarium species, including the trichothecenes deoxynivalenol (DON) and nivalenol (NIV), zearalenone (ZEA), the cyclodepsipeptides beauvericin (BEA) and enniatins (ENN), and fusaric acid (Wagacha and Muthomi, 2007;Munkvold et al., 2021).The biosynthesis of these toxins is encoded by the tri, pks, bea and fus genes, respectively (Dhanti et al., 2017).However, not every species has the ability of producing all of the abovementioned mycotoxins.For example, DON and NIV are commonly produced by F. graminearum and F. culmorum, while ZEA and fusaric acid are often produced by some members of the F. sambucinum species complex (i.e.F. graminearum, F. culmorum), the F. fujikuroi complex (F.verticillioides) and the F. incarnatum-equiseti complex (Nesǐcé t al., 2014;Munkvold et al., 2021), and BEA and ENN are produced by certain F. oxysporum and members of the F. tricinctum species complex (Munkvold et al., 2021;Senatore et al., 2021).DON production by F. graminearum is reported to be essential for disease development in wheat spikes (Cuzick et al., 2008).Spikes treated with DON or NIV led to yield losses even in the absence of the pathogen, indicating a strong negative effect of these trichothecenes on wheat growth (Ittu et al., 1995).In addition to DON, fusaric acid is also a virulence factor involved in programmed cell death (Loṕez-Dıáz et al., 2018).It was shown that alkaline pH and low nitrogen and iron availabilities lead to increased fusaric acid production in F. oxysporum (Palmieri et al., 2023).Besides mycotoxins, there are other metabolites produced by Fusarium species that play a role in disease pathogenesis.Deletion of the F. graminearum gene cluster responsible for the synthesis of fusaoctaxin A abolished the fungal ability to colonize wheat coleoptiles (Jia et al., 2019).Extracellular lipases secreted by F. graminearum affected the plant's defense responses by inhibiting callose synthase activity (Blümke et al., 2014).Diseases caused by Fusarium species include blights, wilts and rots of various crops in natural environments and in agroecosystems (Nelson et al., 1994;Ma et al., 2013).Fusarium Head Blight (FHB) or 'scab' is a disease caused primarily by the F. graminearum species complex.It is the fourth-ranked fungal phytopathogen in term of economic importance (Dean et al., 2012;Legrand et al., 2017), causing yield losses of 20% to 70% (Bai and Shaner, 1994).F. graminearum is responsible for kernel damage and mycotoxin production (Ma et al., 2013) in cereals like wheat, barley, rice and oats (Goswami and Kistler, 2004).Typical symptoms of FHB begin soon after flowering, as diseased spikelets gradually bleach, leading to bleaching of the entire head.After this stage, black spherical structures called perithecia may appear on the surface of diseased spikelets.Later, as the disease becomes more severe, the fungus begins to attack the kernels inside the head, causing them to wrinkle and shrink (Schmale and Bergstrom, 2003).FHB can also be caused by F. culmorum, which is dominant in cooler regions of Europe (Wagacha and Muthomi, 2007).Vascular wilt is responsible for severe losses in crops such as melon, tomato, cotton, bean and banana.It is caused by Fusarium oxysporum, the fifth most economically important fungal phytopathogen (Michielse and Rep, 2009;Dean et al., 2012;Husaini et al., 2018).Symptoms of vascular wilt are first observed on the older leaves, as they begin to droop, followed by defoliation and yellowing of the younger leaves and eventually, plant death (Britannica, 2017;Redkar et al., 2022a).Root, stem and foot rots of various non-grain host plants are often caused by Fusarium solani, and the disease symptoms depend on the host plant and the particular forma specialis (Voigt, 2002;Coleman, 2016).However, typical symptoms of root, stem and foot rots include brown lesions on the affected plant organs.Fusarium verticillioides causes ear and stalk rot in hosts such as maize, sorghum and rice (Murillo-Williams and Munkvold, 2008;Dastjerdi and Karlovsky, 2015), whereas F. graminearum is responsible for causing Fusarium ear and stalk rot in maize (Goswami and Kistler, 2004).Fusarium ear rot is characterized by discoloration of single or multiple kernels in different areas of the ear, while early signs of stalk rot include lodging and discoloration of the stem.
3 Biocontrol agents against Fusarium and their modes of action Plant-beneficial microorganisms present in the rhizosphere may protect plants from Fusarium pathogens, through different modes of action including (i) induction of resistance in the plant, (ii) competition with the pathogens for space and nutrients, (iii) amensalism based on the production of different metabolites or lytic enzymes, or (iv) parasitism (Figure 1B) (Nguvo and Gao, 2019;Morimura et al., 2020).Some of them are also able to inhibit mycotoxin synthesis or to enhance their detoxification (Legrand et al., 2017;Morimura et al., 2020).Certain biocontrol microorganisms have multiple modes of action, which may be expressed simultaneously or sequentially (Legrand et al., 2017).

Induced systemic resistance
Induced Systemic Resistance (ISR) is the phenomenon whereby a plant, once appropriately stimulated by biological or chemical inducers, exhibits enhanced resistance when challenged by a pathogen (Walters et al., 2013).ISR involves (i) the plant perception of inducing signals, (ii) signal transduction by plant tissues, and (iii) expression of plant mechanisms inhibiting penetration of the pathogen into the host tissues (Magotra et al., 2016).A wide variety of microorganisms, including the bacteria Pseudomonas, Bacillus, Streptomyces and the fungi Trichoderma and non-pathogenic F. oxysporum can induce ISR (Fuchs et al., 1997;Choudhary et al., 2007;Zhao et al., 2014;Galletti et al., 2020) in plants against Fusarium (Table 1).ISR in the plant-Fusarium system is based on microbial induction of the activity of various defense-related enzymes in plants, such as chitinase (Amer et al., 2014), lipoxygenase (Aydi Ben Abdallah et al., 2017), polyphenol oxidase (Akram et al., 2013), peroxidase, phenylalanine ammonialyase (Zhao et al., 2012), b-1,3-glucanase, catalase (Sundaramoorthy et al., 2012), and also the accumulation of phytoalexins, defense metabolites against fungi (Kuc, 1995).Cyclic lipopeptide antibiotics, e.g.fusaricidin (Li and Chen, 2019) and external cell components, e.g.lipopolysaccharides (LPS) (Leeman et al., 1995) can also trigger ISR.Some biocontrol agents can lead to ISR in different plant species, while other biocontrol agents show plant species specificity, suggesting specific recognition between microorganisms and receptors on the root surface (Choudhary et al., 2007).
Bacillus amyloliquefaciens subsp.plantarum strain SV65 was assessed on tomato plants infected or not with F. oxysporum f. sp.Sundaramoorthy et al., 2012).Inoculation of chickpea plants with a combination of Bacillus sp., Brevibacillus brevis and Mesorhizobium ciceri lead to the accumulation of peroxidase, polyphenol oxidase, phenylalanine ammonia lyase and phenols in plants as well as resistance to F. oxysporum (Kumari and Khanna, 2019).Paenibacillus polymyxa WLY78 controls Fusarium wilt, caused by Fusarium oxysporum f. sp.cucumerinum, through the production of fusaricidin, which can induce ISR in cucumber via the salicylic acid pathway (Li and Chen, 2019).Tomato showed increased catalase and peroxidase activities when treated with either Streptomyces sp.IC10 and Y28, or with Y28 alone, respectively, outlining a strain-specific ISR in tomato against Fusarium wilt mediated by FOL (Abbasi et al., 2019).
Streptomyces bikiniensis increased the activities of peroxidase, phenylalanine ammonia-lyase and b-1,3-glucanase in cucumber leaves (Zhao et al., 2012).Nonpathogenic Fusarium oxysporum Fo47 can triger induced resistance to FOL and protects tomato from Fusarium wilt (Fuchs et al., 1999).Trichoderma gamsii IMO5 increased transcript levels of ISR-marker genes ZmLOX10, ZmAOS and ZmHPL in maize leaves, thereby protecting the plant from the pink ear rot pathogen F. verticillioides (Galletti et al., 2020).An important determinant of biocontrol efficacy is the population density of ISR-triggering microorganisms.For example, ~10 5 CFU of Pseudomonas defensor (ex fluorescens) WCS374 per g of root are required for significant suppression of Fusarium wilt of radish (Raaijmakers et al., 1995).Another important feature of ISR in plants is that its effects are not only expressed at the site of induction but also in plant parts that are distant from the site of induction (Pieterse et al., 2014).For example, root-colonizing Pseudomonas simiae (ex fluorescens) WCS417r induced resistance in carnation, with phytoalexin accumulation in stems, and protected shoots from F. oxysporum (Van Peer et al., 1991).Priming of barley heads with P. fluorescens MKB158 led to changes in the levels of 1203 transcripts (including some involved in host defense responses), upon inoculation with pathogenic F. culmorum (Petti et al., 2010).

Competition for space and nutrients
In the case of competition, biocontrol of pathogens occurs when another microorganism is able to colonize the environment faster and use nutrient sources more efficiently than the pathogen itself, especially under limited conditions (Maheshwari et al., 2013;Legrand et al., 2017).Bacteria and fungi have the ability to compete with Fusarium, but the underlying mechanism of competition is sometimes unclear.For example, competition between non-pathogenic F. oxysporum strains and pathogenic F. oxysporum has been described, reducing disease incidence (Eparvier and Alabouvette, 1994;Fuchs et al., 1999).Similarly, a nonaflatoxigenic Aspergillus flavus strain was found to outcompete a mycotoxin-producing F. verticillioides during colonization of maize (Reis et al., 2020).Competition may involve bacteria such as Pseudomonas capeferrum (ex putida) strain WCS358, which suppresses Fusarium wilt of radish (Lemanceau et al., 1993).
In some cases, traits involved in competition have been identified.In P. putida (Trevisan) Migula isolate Corvallis, competition for root colonization entails plant's production of agglutinin, and P. putida mutants lacking the ability to agglutinate with this plant glycoprotein showed reduced levels of rhizosphere colonization and suppression of Fusarium wilt of cucumber (Tari and Anderson, 1988).P. capeferrum WCS358 suppresses Fusarium wilt of radish by competing for iron through the production of its pseudobactin siderophore (Lemanceau et al., 1993).In addition to bacteria, the fungus Trichoderma asperellum strain T34 can control the disease caused by F. oxysporum f. sp.lycopersici on tomato plants by competing for iron (Segarra et al., 2010).

Parasitism
Mycoparasitism is an ancient lifestyle, during which one fungus parasitizes another fungus (Kubicek et al., 2011).It involves direct physical contact with the host mycelium (Pal and McSpadden Gardener, 2006), secretion of cell wall-degrading enzymes and subsequent hyphal penetration (Viterbo et al., 2002).Mycoparasitic relationships can be biotrophic, where the host remains alive and the mycoparasitic fungus obtains nutrients from the mycelium of its partner, or necrotrophic, where the parasite contacts and penetrates the host, resulting in the death of the host and allowing the mycoparasite to use the remains of the host as a nutrient source (Jeffries, 1995).Several species of fungi are mycoparasitic, of which Trichoderma is the best described.Contact between the mycoparasitic fungi Gliocladium roseum, Penicillium frequentans, T. atroviride, T. longibrachiatum or T. harzianum and their phytopathogenic targets F. culmorum, F. graminearum and F. nivale triggers the formation of various mycoparasitic structures, such as hooks and pincers, which lead to cell disruption in the phytopathogens (Pisi et al., 2001).When T. asperellum and T. harzianum were grown in the presence of F. solani cell wall, they secreted several cell wall-degrading enzymes, such as b-1,3glucanase, N-acetylglucosaminidases, chitinase, acid phosphatase, acid proteases and alginate lyase (Qualhato et al., 2013), and similarly, Clonostachys rosea produced chitinase and b-1,3glucanase in the presence of F. oxysporum cell wall (Chatterton and Punja, 2009).Sphaerodes mycoparasitica is a biotrophic fungus that parasitizes F. avenaceum, F. oxysporum and F. graminearum hyphae and forms hooks as parasitic structures (Vujanovićand Goh, 2009).However, the direct contribution of mycoparasitism to biological control is difficult to quantify as mycoparasitic fungi typically exhibit a number of different biocontrol mechanisms (Pal and McSpadden Gardener, 2006).

Inhibition and detoxification of mycotoxins
Biocontrol research often focuses on pathogen inhibition, and effects on mycotoxin synthesis or detoxification are often neglected (Pellan et al., 2020).It can be expected that Fusarium inhibition will diminish mycotoxin synthesis, but one comprehensive study found that B. amyloliquefaciens FZB42 inhibited F. graminearum but at the same time stimulated biosynthesis of DON toxin (Gu et al., 2017).Conversely, DON production of F. graminearum (on wheat kernels) was reduced by more than 80% with B. amyloliquefaciens WPS4-1 and WPP9 (Shi et al., 2014), and Paenibacillus polymyxa W1-14-3 and C1-8-b (He et al., 2009), whereas Pseudomonas strains MKB158 and MKB249 significantly reduced DON production in F. culmorum-infected wheat seeds (Khan and Doohan, 2009).Pseudomonas sp.MKB158 lowered expression of the gene coding for trichodiene synthase (an enzyme involved in the production of trichothecene mycotoxins in Fusarium) by 33%, in wheat treated with F. culmorum (Khan et al., 2006).DON production in both F. graminearum and F. verticillioides was also inhibited by the fungus T. asperellum TV1 and the oomycete Pythium oligandrum M1/ATCC (Pellan et al., 2020).Other mycotoxins may be targeted, as Trichoderma harzianum Q710613, T. atroviride Q710251 and T. asperellum Q710682 decreased ZEA production in a dual-culture assay with F. graminearum (Tian et al., 2018), and Streptomyces sp.XY006 lowered the synthesis of fusaric acid in Fusarium oxysporum f. sp.cubense (Wang et al., 2023).
4 Soils suppressive to Fusarium diseases 4.1 General suppressiveness Soils that are suppressive to soil-borne diseases have been known for more than 70 years (Vasudeva and Roy, 1950), and disease suppression is associated primarily with the activity of beneficial microorganisms (Schlatter et al., 2017).These microorganisms interact with phytopathogens, thus affecting their survival, development or infection of the plant (Weller et al., 2002;Raaijmakers et al., 2009).Two types of soil suppressiveness have been described, i.e. general (microbial community-based) suppressiveness and specific (microbial population-based) suppressiveness (Schlatter et al., 2017).General suppressiveness is dependent on the entire soil microbial biomass, which causes pathogen inhibition through various mechanisms, especially competition and the microbial release of inhibitors (Garbeva et al., 2011;De Boer et al., 2019), and it cannot be transferred experimentally between the soils (Weller et al., 2002).Hence, all soils may present some level of general suppressiveness to soil-borne diseases, and this level depends on soil type, agricultural practices and total microbial activity (Janvier et al., 2007;Raaijmakers et al., 2009).
General suppressiveness typically results in the inability of the pathogen to survive and proliferate in soil, and is termed fungistasis in the case of fungal phytopathogens.Fungistasis can affect Fusarium pathogens (De Boer et al., 2019;Legrand et al., 2019), but its significance in relation to different Fusarium species or formae speciales needs clarification.Legrand et al. (2019) determined the soil fungistasis status of 31 wheat fields in the case of F. graminearum, highlighting higher bacterial diversity, a higher prevalence of Pseudomonas and Bacillus species and a denser network of co-occurring bacterial taxa in soils with fungistasis.It suggests the importance of cooperations within diversified bacterial communities (including with antagonistic taxa) to control F. graminearum in soil (Legrand et al., 2019).Accordingly, both bacterial and fungal communities differed between Fusarium wiltdiseased soils vs healthy (presumably suppressive) soils taken from from eight countries and grown with different crop plants (Yuan et al., 2020).

Specific suppressiveness to Fusarium diseases
Besides general suppressiveness, there is also specific suppression to certain diseases, which relies on the activity of a few plant-protecting microbial groups (Weller et al., 2007;Almario et al., 2014;Mousa and Raizada, 2016).Specific suppressiveness may be conferred to non-suppressive soils (i.e.conducive soils) by inoculating them with 0.1% -10% of suppressive soil (Garbeva et al., 2004;Raaijmakers et al., 2009).Although abiotic factors, such as soil physicochemical properties, may contribute to the control of a given pathogen, specific suppressiveness is essentially a phenomenon mediated by beneficial soil microorganisms, since sterilization processes convert suppressive into conducive soils (Garbeva et al., 2004).It is expected that specific suppressiveness entails the contribution of a few plant-protecting microbial groups (Weller et al., 2007), but microbial community comparison of suppressive vs conducive soils may evidence significant differences for a large number of taxa (Kyselkováet al., 2009;Legrand et al., 2019;Ossowicki et al., 2020;Yuan et al., 2020;Lv et al., 2023).

Natural and induced specific suppressiveness to Fusarium diseases
Specific suppressiveness is sometimes an intrinsic property of the soil and persists over years, despite changing ecological conditions related to crop rotation.This natural/long-term suppressiveness is well documented for several pathosystems, for instance in Swiss soils suppressive to tobacco black root rot (T.basicola) near Morens (Stutz et al., 1986).Suppressive and conducive soils may be located at small geographic distances in the landscape, and differences in plant disease incidence between neighbouring fields that share similar climatic conditions and agronomic practices are attributed by the differences in the resident microbiota in these soils (Almario et al., 2014).Natural suppressiveness has also been extensively studied in the case of Fusarium diseases, in particular with the Fusarium wilt suppressive soils of Salinas Valley (California) or Chateaurenard (France).In these soils, Fusarium wilt disease remains minor despite the long history of cultivation of different crops, and the introduction of small amount of these soils to sterilized suppressive soil or conducive soil significantly decreased Fusarium wilt disease incidence (Scher and Baker, 1980;Alabouvette, 1986).In both locations, the small level of disease in plants cannot be attributed to the absence of Fusarium in the soil, but rather to plant protection by the soil microbiota (Sneh et al., 1984;Alabouvette et al., 1985;Siegel-Hertz et al., 2018), as found in later investigations (Bautista et al., 2023).Specific disease suppressiveness can also result from particular farming practices leading to the built-up of a plant-protecting microbiota.Often, this takes place following crop monoculture, typically after early disease outbreak, and is examplified by take-all decline of wheat (Weller et al., 2002;Sanguin et al., 2009) and barley (Schreiner et al., 2010).Induced suppressiveness is initiated and maintained by monoculture, in the presence of the pathogen Gaeumannomyces graminis var.tritici (Weller et al., 2002).Soil suppressiveness to Fusarium diseases is usually natural, but cases of induced suppressiveness are also documented.Thus, soils found in Hainan island (China) that were grown for years with banana in confontration with pathogenic F. oxysporum displayed rhizosphere enrichment in microbial taxa conferring protection from banana wilt (termed banana Panama disease) (Shen et al., 2022), watermelon monoculture in Florida induced suppressiveness to wilt caused by F. oxysporum f. sp.niveum (Larkin et al., 1993), and 15 years of strawberry monoculture in Korea triggered suppressiveness to wilt caused by F. oxysporum f. sp.fragariae (Cha et al., 2016).Soil addition of wild rocket residues resulted in suppressiveness to cucumber crown and root rot (F.oxysporum f. sp.radicis-cucumerinum) in Israel (Klein et al., 2013), whereas suppressiveness to Fusarium wilt can also be induced by microbial biofertilizer inoculants reshaping the soil microbiome (Xiong et al., 2017).Thus, organic fertilizer containing B. amyloliquefaciens W19 enhanced levels of indigenous Pseudomonas spp.and provided suppression of Fusarium wilt of banana (Tao et al., 2020).The combined action of B. amyloliquefaciens W19 and Pseudomonas spp. is thought to cause a decrease in Fusarium density in the root zone of banana.Organic fertilizers inoculated with Erythrobacter sp.YH-07 controlled Fusarium wilt in tomato, as a direct result of the bacteria and indirectly by altering the composition of the microbial community (Tang et al., 2023).Organic fertilizer amended with Bacillus and Trichoderma resulted in an increase in indigenous Lysobacter spp., thus indirectly inducing suppression of Fusarium wilt of vanilla (Xiong et al., 2017).Geographic locations of the main field sites with soils documented to be suppressive to Fusarium diseases, in Europe including France (Noirmoutier Island, Chateaurenard in Southeast France, and Brittany), Denmark, The Netherlands, Germany, Italy (Albenga, Carmagnola, and Santena), Gran Canaria Island (Spain, located in the Atlantic Ocean), and Serbia, in North America (California and Florida), Central America (Honduras, Costa Rica, Panama, and Guatemala), South America (Colombia), Asia (Korea, China, India, Israel, and Indonesia), and Africa (Morocco).Each location is marked with the corresponding pathogen: F. oxysporum (indicated by a red dot), F. culmorum (green triangle), F. graminearum (blue square), and F. udum (black pentagon).Many biocontrol strains originate from suppressive soils, and they were investigated as a mean to understand disease suppressiveness.In the case of Fusarium diseases, examples include Pseudomonas sp.Q2-87 (P.corrugata subgroup) (Weller et al., 2007), isolated from wheat in take-all decline soils but that protects tomato from F. oxysporum f. sp.radicis-lycopersici, Pseudomonas sp.C7 (P.corrugata subgroup) (Lemanceau and Alabouvette, 1991) isolated from soil suppressive to Fusarium wilt of tomato, and non-pathogenic F. oxysporum strains Fo47 (Fuchs et al., 1997;Duijff et al., 1998;Fuchs et al., 1999), CAV 255 (Sajeena et al., 2020) and Ro-3 (Bubici et al., 2019).Based on the biocontrol traits thus identified, the corresponding microbial functional groups have been characterized in suppressive vs conducive soils, using isolate collections, molecular fingerprints or sequencing.Fluorescent Pseudomonas bacteria, especially those producing the antifungal metabolite 2,4-diacetylphloroglucinol, have been extensively targeted in take-all-decline soils (Cook and Rovira, 1976;Weller et al., 2002;Weller et al., 2007) and soils suppressive to black root rot (Stutz et al., 1986;Laville et al., 1992;Kyselkovaá nd Moënne-Loccoz, 2012), whereas studies on soils suppressive to R. solani diseases have focused on Pseudomonas spp.producing antifungal lipopeptides (Mendes et al., 2011), Streptomyces spp.producing volatile metabolites (Cordovez et al., 2015) and Paraburkholderia graminis producing sulfurous volatile compounds (Carrioń et al., 2018).In the case of soils suppressive to Fusarium diseases, competition with pathogenic Fusarium species is considered important, involving the entire soil microbiota or more specifically non-pathogenic Fusarium strains in Chateaurenard soils (Louvet et al., 1976;Alabouvette, 1986), or fluorescent Pseudomonas (iron competition; Scher and Baker, 1980;Sneh et al., 1984) in soils of Salinas Valley (California) or Chateaurenard (France).The role of extracellular lytic enzymes can be significant, as soil microbiota may protect barley from Fusarium culmorum via a more efficient cellulolytic activity than the pathogen, which consequently is outcompeted for nutrients (Rasmussen et al., 2002).Suppressiveness may result in part from chitinolytic effects of the soil microbiota against the pathogen, based on inhibition of Fusarium fungi by chitinases in vitro and effective protection of plant by chitinase-producing inoculants (Veliz et al., 2017).Other modes of action evidenced include the production of antifungal secondary metabolites in wilt-suppressive soils, such as a new thiopeptide by Streptomyces (Cha et al., 2016) and phenazines by Pseudomonas spp.(Mazurier et al., 2009), and immunity stimulation in banana (induction of the jasmonate and salicylic acid pathways) by fluorescent Pseudomonas (Lv et al., 2023).

Microbial diversity in soils suppressive to Fusarium diseases
Specific disease suppressiveness is attributed to the contribution of a few plant-benefical populations, but comparison of suppressive vs conducive soils has evidenced differences in the occurrence or prevalence of multiple taxa, in the case of suppressiveness to take all (Sanguin et al., 2009;Schreiner et al., 2010;Chng et al., 2015), black root rot (Kyselkováet al., 2009), R. solani-mediated damping-off (Mendes et al., 2011), or potato common scab (Rosenzweig et al., 2012).Similar findings were made with soils suppressive to Fusarium diseases.No single phylum was uniquely associated with F. oxysporum wilt suppressiveness in Korean soils, even though Actinomycetota (formerly Actinobacteria) was identified as the most prevalent bacterial taxa colonizing strawberry in suppressive soils (Cha et al., 2016).Likewise, the bacterial genera Devosia, Flavobacterium and Pseudomonas were more abundant (and the pathogen less abundant) in Chinese soils suppressive to banana wilt than in conducive soils, and Pseudomonas inoculants isolated from suppressive could control the disease (Lv et al., 2023).Compared with conducive soil, Fusarium wilt suppressive soil from Châteaurenard displayed higher relative abundance of Adhaeribacter, Arthrobacter, Amycolatopsis, Geobacter, Massilia, Microvirga, Paenibacillus, Rhizobium, Rhizobacter, Rubrobacter and Stenotrophomonas (but not Pseudomonas) (Siegel-Hertz et al., 2018).However, differences were also found in the fungal community, with several fungal genera (Acremonium, Ceratobasidium, Chaetomium, Cladosporium, Clonostachys, Mortierella, Penicillium, Scytalidium, Verticillium, but also Fusarium) detected exclusively in the wilt suppressive soil (Siegel-Hertz et al., 2018).Data also pointed to a greater degree of microbial complexity in suppressive soils, with particular co-occurrence networks of taxa (Bakker et al., 2014;Lv et al., 2023).In German and Dutch soils, co-occurrence networks showed that the suppressive soil microbiota involves a guild of bacteria that probably function together, and in two of the suppressive soils this guild is dominated by Acidobacteriota (formerly Acidobacteria) (Ossowicki et al., 2020).
Many studies focused on a few, geographically-close soils, which does not provide a global view on the importance of microbial diversity.However, two studies have considered geographically diverse agricultural soils suppressive to Fusarium wilt.Various Chinese soils suppressive to banana wilt mediated by F. oxysporum were shown to share a common core microbiota, specific to suppressive soils, which included the genus Pseudomonas (Shen et al., 2022).In a wider range of soils from the Netherlands and Germany, soils suppressive to F. culmorummediated wilt of wheat did not display a specific bacterial species that correlated with suppressiveness (Ossowicki et al., 2020).There was no relation either with soil physicochemical composition (i.e.soil type, pH, contents in C, N, or bioavailable Fe, K, Mg, P) or field history, yet suppressiveness was microbial in nature, as suppressive soils made them become conducive.This suggests that each suppressive soil may harbor its own set of phytobeneficial bacteria, supporting the notion of functional redundancy between microbiomes, meaning that different microbiomes may share common functionalities despite taxonomic differences in the microbial actors involved (Lemanceau et al., 2017).Taken together, this might be explained by the fact that protection of wheat from F. culmorum-mediated wilt corresponds to a case of natural suppressiveness (Ossowicki et al., 2020), where biogeographic patterns are probably important, whereas soils suppressive to Fusarium wilt of banana are induced by monoculture (Wang et al., 2019;Shen et al., 2022), with convergent effects resulting from similar banana recruitment across different soil types.
To go beyond individual analyses considered separately, we reanalyzed sequence data from five investigations comparing diseasesuppressive and conducive soils of cultivated plants (flax, watermelon, bananas, and wheat) infected by different Fusarium species (F.oxysporum or F. culmorum).At the level of bacterial phyla, fluctuations among Chateaurenard (flax-F.oxysporum; Siegel-Hertz et al., 2018), Hainan (banana-F.oxysporum; Shen et al., 2022) (Figure S1A) and Dutch/German (wheat-F.culmorum; Ossowicki et al., 2020) (Figure S1B) suppressive soils were important, as were those among their conducive counterparts, and the comparison between suppressive and conducive soils at these locations was not fruitful.In another study, fluctuations among other Hainan (banana-F.oxysporum; Zhou et al., 2019) suppressive or conducive soils were of less magnitude, but again the comparison was not insightful (Figure S1B).In contrast, Jiangsu (watermelon-F.oxysporum; Wang et al., 2015) suppressive soils displayed a higher relative abundance of Acidobacteriota and Pseudomonadota than in conducive soils (Figure S1B), but this property was not relevant when considering the other locations/ plant species/Fusarium species.Based on heatmap comparisons (Figure S2), the main finding was the lower prevalence of the Bacillota phylum in the Jiangsu (watermelon-F.oxysporum) suppressive vs conducive soils, which was restricted to the case of these soils.
At the level of bacterial genera, the comparison of Châteaurenard (flax-F.oxysporum), Hainan (banana-F.oxysporum) or Dutch/German (wheat-F.culmorum) soils did not lead to the identification of indicator taxa (Figures 4, S3), but at Jiangsu (watermelon-F.oxysporum) the genera Bacillus, Dongia, Rhodoplanes and Terrimonas were less prevalent and the genera Ferruginibacter, Flavobacterium, Pseudomonas and Sphingomonas more prevalent in suppressive soils than in conducive soils (Figure S3A).Therefore, the comparison between suppressive and conducive soils was sometimes meaningful at the local scale, but typically not when considering a wider range of geographic or biological (plant and Fusarium species) conditions together.In other words, the information available so far points that suppressiveness to Fusarium diseases relies on microbial selection processes by roots that depend on local conditions, i.e. probably related to microbial biogeography, soil type, plant species, Fusarium genotype and most likely other local factors as well.
6 Variability and management of soil suppressiveness to Fusarium diseases 6.1 Environmental factors influencing soil suppressiveness to Fusarium diseases Environmental conditions in soil may influence Fusarium autecology, the composition and activity of the soil microbial community, the tripartite interactions between this microbiota, Fusarium pathogens and the plant, and ultimately the level of disease suppressiveness (Marshall and Alexander, 1960;Amir and Alabouvette, 1993;Mazzola, 2002;Czembor et al., 2015).Key environmental factors in this regard include soil physicochemical properties and weather conditions (Weber and Kita, 2010).
Early work on the suppressiveness of soils to vascular Fusarium diseases drew attention to the positive role of certain abiotic factors and, in particular, montmorillonite-type clays (Stover, 1956;Stotzky and Torrence Martin, 1963).In addition, higher clay contents may contribute to reduced infestation by Fusarium (Kurek and Jaroszuk-S ́ciseł, 2003;Deltour et al., 2017), by altering oxygen diffusion, pH buffering and nutrient availability (Orr and Nelson, 2018).Höper et al. (1995) showed that the level of suppressiveness to Fusarium wilt of flax increased in soils amended with montmorillonite, kaolinite or illite at pH 7. A negative correlation between soil pH and Fusarium disease severity was reported in experiments with flax (Senechkin et al., 2014), strawberry (Fang et al., 2012) and banana (Shen et al., 2015a).However, the correlation between pH and Fusarium wilt incidence was positive in studies on banana (Peng et al., 1999) and watermelon (Cao et al., 2016).Certain experiments acidified soil originally at pH 8.0 (Peng et al., 1999) or 7.4 (Cao et al., 2016), whereas others limed an acidic soil (Fang et al., 2012;Senechkin et al., 2014;Shen et al., 2015a).Inconsistencies may relate to the complexity of pH effects on Fusarium pathogens and diseases, and possible interactions with soil properties, Fusarium and plant genotypes, or other experimental conditions.In addition, soil suppressiveness to Fusarium wilt necessitates sufficient levels of nitrogen, as disease incidence negatively correlates with the NH 4 + and NO 3 -contents in the soil (Li et al., 2016;Meng et al., 2019).
Moreover, the addition of calcium to the soils suppressed Fusarium wilt in several soil type × plant conditions (Spiegel et al., 1987;Peng et al., 1999;Gatch and du Toit, 2017).In Brittany, F. graminearum growth positively correlated with manganese and iron contents in the soil (Legrand et al., 2019).A positive correlation was found between hemicellulose concentration and suppression of Fusarium wilt in tomato and carnation (Castaño et al., 2011), as well as cellulose concentration and suppression of Fusarium seedling blight of barley (Rasmussen et al., 2002).This is attributed to the activity of cellulolytic microorganisms that limit Fusarium growth, as lower organic matter content (following decomposition) would reduce resources supporting this microbiota and disease suppression (Orr and Nelson, 2018).Climatic conditions, notably temperature and precipitation may strongly affect the incidence of Fusarium diseases (Orr and Nelson, 2018).Phytopathogenic species F. oxysporum, F. solani, F. verticillioides, F. graminearum and F. culmorum develop best under humid conditions, at water activity above 0.86 (Table S1) (Thrane, Severity of Fusarium wilt in lettuce (Scott et al., 2009;Ferrocino et al., 2013) and FHB in wheat was positively correlated with soil temperature (Xu et al., 2007;Nazari et al., 2018).For example, Fusarium wilt incidence significantly increased when lettuce was grown at 22-26°C instead of 18-22°C (Ferrocino et al., 2013).Similarly, in both conducive and suppressive soils, severity of Fusarium wilt of banana was significantly increased when temperature was raised from 24°C to 34°C (Peng et al., 1999).

Farming practices and the management of soil suppressiveness to Fusarium diseases
As many other soil-inhabiting pathogenic fungi, the Fusarium spp.can overwinter as mycelium in plant debris or dormant structures in the soil, which leads to cause the initial infection of plants in the following season (Nelson et al., 1994;Janvier et al., 2007;Leplat et al., 2013;Xu et al., 2021).Therefore, cultural practices removing the primary inoculum of the pathogen from overwintering soils are useful to prevent future infection (Voigt, 2002).However, farming practices also influence soil suppressiveness by shaping the rhizosphere microbial community (Campos et al., 2016) and stimulating the activity of beneficial rhizosphere microorganisms (Janvier et al., 2007).In this context, various agricultural practices, such as crop rotation/monocropping, tillage, organic amendments and fertilisers, are important to consider to develop suppressiveness-based control methods in farm fields (Janvier et al., 2007;Fu et al., 2016).
Except in the few cases where monoculture induces suppressiveness to Fusarium diseases (Larkin et al., 1993;Shen et al., 2022), cropping systems based on rotation of different plant species result in reduced survival of soil-borne pathogen propagules Heatmap of the major bacterial genera detected in the rhizosphere of plants grown in soils suppressive or conducive to different Fusarium diseases, based on analysis (File S1) of selected studies (Shen et al., 2015b;Siegel-Hertz et al., 2018;Wang et al., 2019;Zhou et al., 2019;Ossowicki et al., 2020).(A) The 20 most abundant genera in soils conducive or suppressive to diseases caused by Fusarium oxysporum.In Siegel-Hertz et al. (2018), suppressive soils were assessed after Fusarium inoculation or without.(B) The 20 most abundant genera in soils conducive or suppressive to diseases caused by Fusarium culmorum.The color intensity in each cell indicates the relative abundance (%) of a genus in each study for each plant type.When relevant, dotted lines are used to separate pathogen-inoculated samples from non-inoculated samples (in Chateaurenard) or samples from different fields.More details on individual conditions are available in Table S2.
over the short term (Winter et al., 2014).Crop may reduce severity and incidence of diseases caused by Fusarium spp.(Wang et al., 2015;Khemir et al., 2020).For example, compared with the tomato monoculture, soil management under wheat -tomato rotation changes soil microbial composition by increasing the abundance of microbial taxa such as Bacillus, Paenibacillus, Pseudomonas, Streptomyces, Aspergillus, Penicillium and Mortierella, which may control Fusarium wilt of tomato (De Corato et al., 2020).Reduced incidence of F. pseudograminearum and F. culmorum in the soils under cereallegumes rotation management may be due to the non-host character of the legumes (Evans et al., 2010).However, not all crop rotations lead to reduced disease pressure (Ranzi et al., 2017).In the case of the FHB, it was advocated to rotate wheat and corn with crops like soybean, until it was shown that F. graminearum can also cause disease in soybean, as it has a wide range of hosts (Marburger et al., 2015).This suggests that there is no common rule regarding the relationship between crop rotation and Fusarium disease incidence.
Crop residues of high cellulose content promoted the activity of beneficial cellulolytic microorganisms and limited the development of Fusarium culmorum (Rasmussen et al., 2002), as organic amendments represent a favorable environment for beneficial microorganisms that are able to combat phytopathogenic Fusarium species (Maher et al., 2008;Cuesta et al., 2012).Accordingly, organic amendments like animal manure, solid wastes and different composts are often used to improve soil health by delivering nutrients to the soil and also by stimulating beneficial microbiota (Fu et al., 2016;Mousa and Raizada, 2016).Thus, soils with added organic amendments exhibited inhibitory effects against F. verticillioides by reducing the production of a fungal pigment and sporulation, consequently disabling fungal spread (Nguyen et al., 2018).Addition of vermicompost reduced tomato infection by F. oxysporum f. sp.lycopersici (Szczech, 1999) and mulched straw contributed to the suppression of seedling blight caused by F. culmorum (Knudsen et al., 1999).Soils supplemented with coffee residue compost or rapeseed meal exhibited suppressiveness to F. oxysporum-mediated wilt, and microorganisms isolated from supplemented soils inhibited F. oxysporum growth on agar plates (Mitsuboshi et al., 2018).Carbon addition to soil influenced the soil microbiome, enhancing Fusarium-inhibitory populations from the Streptomyces genus (Dundore-Arias et al., 2020).However, increasing organic matter content may promote Fusarium survival in certain (rare) cases.One study tested the effects of 18 composts (made from different mixtures of manure, domestic biowaste and green waste) on Fusarium wilt disease suppression, caused by F. oxysporum f. sp.lini, and it was shown that only one compost did not positively affect the disease suppression (Termorshuizen et al., 2006).The efficiency of organic amendments in controlling plant diseases is determined by the pathosystem, the application rate, the kind of amendment and the level of maturity of composts or disintegration phase of crop residues (Janvier et al., 2007).
Tillage, which is one factor influencing organic matter decomposition, appears to have contrasting effects on soil suppressiveness.Under conventional tillage, tillage depth appears to play a crucial role in soil survival of Fusarium, such that the deeper the tillage, the lower the abundance of Fusarium species (Steinkellner and Langer, 2004).This can be partly explained by the fact that the pathogen is displaced from its niche, reducing its ability to survive (Bailey and Lazarovits, 2003), and the rate of decomposition of buried residues is faster than at the soil surface (Leplat et al., 2013).The carbon released during these decomposition processes increases the activity of the soil microbiota, thereby improving the overall functioning of the soil (Bailey and Lazarovits, 2003).Under conservation tillage, surface residues persist and can act as a long-term source of inoculum for plant infection by F. verticillioides, F. proliferatum and F. subglutinans, as they can colonise crop residues and produce overwintering spores that often survive the period when plants are absent from the agrosystem (Bockus and Shroyer, 1998;Cotten and Munkvold, 1998;Pereyra et al., 2004).This is consistent with results suggesting that conservation tillage and leaving crop residues in situ increase Fusarium abundance (Govaerts et al., 2008;Wang et al., 2020).For example, spores of Fusarium species could be recovered from plant residues more than two years after harvest (Pereyra et al., 2004).In certain cases, lower occurrence of plant infection by F. culmorum, F. equiseti (Weber et al., 2001) and F. pseudograminearum (Theron et al., 2023) was found under conservation tillage compared with conventional tillage.These contrasting results might be due to differences in environmental factors, cropping patterns and soil types, which could modulate interactions between soil conditions, Fusarium ecology and plant physiology (Sturz and Carter, 1995).The use of simplified tillage practices was proposed to reduce F. culmorum abundance, by mixing crop residues with the topsoil layer to promote the growth of beneficial straw-decomposing microorganisms (Weber and Kita, 2010).
Different fertilizers have different effects on phytopathogenic Fusarium spp.On one hand, the development of FHB caused by F. culmorum and F. graminearum increased with inorganic nitrogen fertilization (Lemmens et al., 2004), and on the other hand, nitrite could reduce the population of F. oxysporum (Löffler et al., 1986).Besides, higher doses of nitrogen may contribute to higher accumulation of Fusarium mycotoxins (Podolska et al., 2017).The addition of phosphorus fertilizer, in the form of P 2 O 5 , significantly reduced Fusarium-caused wilting in chickpea, lentil and lupine, in both greenhouse and field conditions (Elhassan et al., 2010).Organic fertilizers can lead to an increase in indigenous microbial populations, thus contributing to suppression of Fusarium wilt disease (Montalba et al., 2010;Raza et al., 2015).When grown with the addition of organic N fertilizer, highbush blueberry exhibited increased tolerance to F. solani, in parallel to increased soil microbial activity and mycorrhizal colonization (Montalba et al., 2010).

Conclusion and outlook
Disease-suppressiveness of soils is a useful model to understand microbial phytoprotection and develop sustainable plant protection strategies for soils devoid of this property.In this review, we summarized the current knowledge on Fusarium phytopathogens, the available control methods and soils to Fusarium diseases, with the underlying mechanisms involved in the suppression.On one hand, extensive information is available on environmental and microbial properties responsible for suppressiveness to Fusarium diseases.One prominent feature is the diversity of Fusarium-based pathosystems for which suppressive soils are documented, in terms of Fusarium species (often F. oxysporum, but not only), host plants (both monocots and dicots), types of disease (often wilt, but not only), geographic locations of soil and farming conditions, and types of suppressiveness (i.e.natural suppressiveness to Fusarium diseases, but also monocultureinduced suppressiveness as well as fungistasis towards Fusarium pathogens).This diversity is paralleled by differences in microbiota composition and diversity associated with disease control in the different cases of suppressiveness.On the other hand, despite the fact that soils suppressive to Fusarium diseases have been studied for decades, they are still poorly understood in terms of microbiota functioning, and knowledge remains fragmented.On this basis, additional research is needed to integrate further the scientific approaches used to decipher suppressiveness to Fusarium diseases.First, by combining complementary assessment methodology with current next-generation sequencing and ecological networks research, and incorporating experimental strategies to manipulate and transplant rhizosphere microbiome (or single microorganisms) of plants grown in suppressive soils to those in conducive soils to go beyond correlative work, as started recently (Ye et al., 2020;Jiang et al., 2022).Second, by extending the range of soil conditions investigated, and develop meta-analyses to estimate key microbiota differences between suppressive and conducive soils, as pioneered by Yuan et al. (2020).Third, by considering a wider range of biological actors, including beneficial fungi (often neglected), soil fauna (likely to influence microbial communities, Fusarium vectorisation, and plant health; e.g.Dita et al., 2018;Wagner et al., 2022).Fourth, by taking into account plant genetics, behavior and physiological responses to Fusarium pathogens (e.g.Liu et al., 2019).Therefore, there is a need for a more multidisciplinary approach to understand microbiota functioning in soils suppressive to Fusarium diseases.

TABLE 1
Biocontrol agents, plant-Fusarium systems and ISR mechanisms.

TABLE 2
Biocontrol agents, plant-Fusarium systems and biocontrol enzymes and metabolites.

TABLE 3
List of locations with soils suppressive to Fusarium diseases known to date, with a pathosystem, disease and the underlying suppression mechanism.
and Roy, 1950), and F. graminearum on wheat (in Serbia;