Due to its wide host range, ubiquitous global distribution, and increasing impact on agriculture, M. phaseolina has been dubbed a “global crop destroyer”. Disease caused by M. phaseolina is most often referred to as charcoal rot, but this pathogen is also known to cause wilting, stem canker, seedling blight, rot, and damping-off (Singh et al., 1990; Kaur et al., 2012b). These contribute to yield loss, reduction in seed quality, and viability of infected plants (Mughogho and Pande, 1983). Moreover, the severity of symptoms is correlated with high temperatures and low soil water potential conditions. Thus, plants growing in (sub)-tropical arid regions are at higher risk of succumbing to disease caused by this pathogen (Mayek-Pérez et al., 2002; Chamorro et al., 2015; Lodha and Mawar, 2020). Pandey and Basandrai reviewed the current literature on factors related to climate change that could potentially worsen M. phaseolina infestation (Pandey and Basandrai, 2021). Both the visible traits and genetic architecture of this pathogen, which are further described below, show that M. phaseolina as a species is well adapted to diverse climates and equipped to withstand environmental change.

Many important crops including grains and legumes are included among the wide host range of this pathogen (Pandey and Basandrai, 2021). Known hosts of M. phaseolina such as alfalfa, maize, canola, soybean, and sunflower have vast global economic implications, while others like cassava, chickpea, and mungbean are critical for regional food security (Ammon et al., 1974; Bashir and Malik, 1988; Kaiser and Das, 1988; Msikita et al., 1998; Pratt et al., 1998; Gaetán et al., 2006; Weems et al., 2011; Lakhran et al., 2018). The combined estimated loss of soybean yields in ten countries due to charcoal rot exceeded 2 million metric tons during 1998, ranking third among the most common soybean diseases (Wrather et al., 2001). Furthermore, charcoal rot ranked second for economic losses per hectare in the U.S. across two decades and combined with soybean cyst nematode accounted for 33% of the total economic losses (Bandara et al., 2020).

M. phaseolina also poses a great threat to specialty crops, such as strawberry, pistachio, and sunflower (Mertely et al., 2005; Bokor, 2007; de los Santos et al., 2019; Nouri et al., 2020). Research on specialty crops is often underfunded and disproportionately lacking in comparison to the steady rise in value of production, despite the inherently high risk involved in marketing fresh produce (Alston and Pardey, 2008; Neill and Morgan, 2021). For example, strawberries ranked fourth among the top California crops produced in 2017, with a value of $3.10 billion (California Department of Food and Agriculture; Koike et al., 2016). Strawberries are highly susceptible to M. phaseolina. Fumigation with methylbromide and chloropicrin served as a common method of pathogen control. In 2005, the agricultural use of methylbromide was phased out. Since then, the relationship between yield and area of strawberry production reversed, as yields stagnated and new incidences of charcoal rot increased (Holmes et al., 2020).

M. phaseolina is most often characterized by its production of microsclerotia, or small sclerotial bodies 200–600 μm in diameter, which appear as dark specs in infected plant tissue (Jackson and Jaronski, 2009). Microsclerotia are spherical clusters of hyphal cells, each with the ability to germinate and propagate when appropriate environmental conditions are met. They can remain dormant in dead host tissue or soil for an extended period of time, becoming the primary inoculum for the next growing season (Short et al., 1980). On potato dextrose agar (PDA), microsclerotium formation can be identified in hosts as early as 24–30 hours post-inoculation (Wyllie and Brown, 1970). Pycnidia, which are asexual reproductive structures that form spores, have also been documented for this fungus, but are rarely observed in vitro (Ma et al., 2010).

Microsclerotia germinate at a temperature range of 28–35°C, as the emerging mycelium extends radially until it identifies host tissue (Ammon et al., 1974; Mihail and Taylor, 1995). Appressoria-like structures, or hyphopodia, form when hyphal tips recognize host root cells, often penetrating the plant tissue between epidermal cells, before colonization is established in the middle lamella and intercellular space (Ammon et al., 1974). Early stages of infection remain mostly intercellular, but during host cell penetration the hyphae can invaginate the host plasma membrane creating an interface similar to that of haustoria of biotrophic pathogens (Figure 1A). Moreover, formation of intracellular vesicles, which is a characteristic of early hemibiotrophic infection, has been observed in the cortical tissue of infected soybean (Chowdhury et al., 2017). Production of pectolytic and cellulolytic enzymes by the pathogen has been observed both in vivo and in vitro, but it is not certain during which phases of the infection process this occurs (Radha, 1953; Chan and Sackston, 1969; Chan and Sackston, 1970). Development of thin, filamentous hyphae is thought to be an indicator of a biotrophy to necrotrophy switch (BNS), as subsequently, tissue begins to necrose (Chowdhury et al., 2017). M. phaseolina mycelia eventually colonize the host’s vascular tissue, and the formation of microsclerotia in the xylem vessels leads to blockage and consequently wilting of the host (Abawi and Corrales, 1990; Mayek-Pérez et al., 2002; Figure 1B). A deeper understanding of the initial stages of M. phaseolina infection may provide us with clues as to which modes of plant defense are effective against this pathogen.

M. phaseolina is adapted to survive in diverse environmental conditions, including sodium salt concentrations of 1 to 8%, and pH ranges of 3.5 to 10 (Islam et al., 2012). Its microsclerotia can persist in soil for up to 15 years. There is no universal, economically, and environmentally reasonable protocol for eradicating this pathogen from an infected field. What is known is that there is a clear relationship between inoculum concentration, disease occurrence, and yield loss. Numerous strategies to suppress charcoal rot have been tested, and the results have been thoroughly reviewed by authors such as Lodha and Mawar (2020) and Marquez et al. (2021). Current approaches to controlling this disease can be grouped into four categories: (1) Agronomic practices, (2) Chemical control by using fungicides or herbicides, (3) Host genetic resistance, and (4) Biological control (Marquez et al., 2021).

While certain combinations of agronomic practices showed some promise, none of these measures proved to be fully sufficient. The situation seems to be similar for fungicides against charcoal rot. Although highly effective in controlling M. phaseolina, methylbromide has been phased out throughout the world per the “Montreal Protocol Guidelines on Substances that Deplete the Ozone Layer”, released in 1991 (Chamorro et al., 2016; Guthman, 2017). Fumigation of strawberry fields with other chemicals continues in California, and in fact, chloropicrin use has increased significantly in correlation to the drop in methylbromide use, as it is less effective alone (Guthman, 2017). Although alternatives to methylbromide, including chloropicrin, have been shown to reduce M. phaseolina density and plant mortality in the field, crop production cost and chemical use have only increased (Chamorro et al., 2016; Guthman, 2017).

Unlike resistance against many foliar pathogens, host resistance against M. phaseolina seems not to be mediated by individual disease resistance (R)-genes, and multiple genetic loci seem to make combined quantitative contributions in protecting plants against this root pathogen (Marquez et al., 2021). Only gradual differences in quantitative resistance against M. phaseolina have been observed among different varieties of certain crop species, such as cowpea, soybean, or strawberry (Muchero et al., 2011; Reznikov et al., 2019; Gomez et al., 2020). While the underlying quantitative trait loci can be useful for breeding, genetic engineering can also result in crop varieties with enhanced protection against M. phaseolina. For example, transgenic jute lines expressing the rice chitinase (chi11) gene showed significantly smaller lesion sizes and improved fiber yield in comparison to the wild type when infected with M. phaseolina (Majumder et al., 2018). However, efforts to combat M. phaseolina by transgenic crop lines have been scarce. This approach has been used more so to identify genes and pathways important for host resistance (Lygin et al., 2013; Dabi et al., 2020).

Biological control strategies have shown some promise in protecting crops against M. phaseolina. Dhingra et al. (1976) found that increasing organic matter in soil reduces M. phaseolina colonization of soybean and maize stems, likely due to the increased presence of other microbes. The implication that M. phaseolina has low competitive saprophytic ability has been tested and confirmed with multiple fungal and bacterial biocontrol agents (BCAs). For example, several species of Bacillus have been reported to antagonize M. phaseolina growth both in vitro and in vivo and to suppress disease symptoms in soybean, common bean and cowpea (Yasmin et al., 2020; Bojórquez-Armenta et al., 2021; Rangel-Montoya et al., 2022). Other rhizospheric bacteria such as fluorescent Pseudomonas spp. have also been identified as M. phaseolina BCAs (Gupta et al., 2002; Das et al., 2008). Many of these rhizobacterial species also have plant growth-promoting properties.

The genus Trichoderma is a well-known fungal BCA (Lorito et al., 2010). Indeed, examples like T. viride and T. harzianum successfully inhibit the growth of M. phaseolina and further have positive effects on plant growth (Sankar and Sharma, 2001; Khaledi and Taheri, 2016; Khalili et al., 2016). Interestingly, it has been shown that the combined application of T. harzianum and arbuscular mycorrhizal fungi (AMF) allows for AMF colonization in non-host Brassica plants, including Arabidopsis thaliana, and increases silique numbers (Poveda et al., 2019). Additionally, AMF co-treatment with M. phaseolina resulted in a much lower number of up-regulated genes in comparison to the application of the pathogen alone. Pre-mycorrhized plants also exhibited greater expression of serine carboxipeptidase-like (SCPL) and lectin genes, which have been suggested to prime the plant for pathogen attack (Marquez et al., 2018; Marquez et al., 2021).

Despite the growing interest in reducing agrochemical inputs, the biopesticide market constitutes only a small fraction (6.4% in 2018) of the total pesticide market (Pal and McSpadden Gardener, 2006; Regnault-Roger, 2020). Obstacles that remain to be addressed, such as storage, lack of commercialized products, narrow target application, and limited user awareness have hindered widespread adoption of biological control strategies.

Comparisons between M. phaseolina isolates across host species and geographical distances have revealed high levels of morphological, physiological, pathogenic, and genetic variation. Despite this, the genus Macrophomina is monospecific, meaning that it constitutes a single species. Classification of M. phaseolina isolates based on chlorate sensitivity was previously proposed by Pearson et al. (1986). Chlorate is an analogue of nitrate and can be toxic to plants and fungi when reduced to chlorite via the nitrate reductase pathway. Since M. phaseolina isolates from corn were chlorate resistant and those from soybean were chlorate sensitive, it was suggested that the presence or absence of the nitrate reductase pathway could be a marker for host-specific strains (Pearson et al., 1986; Pearson et al., 1987). However, subsequent reports using this approach have had mixed results. Mihail and Taylor (1995) were unable to confirm relationships between the chlorate phenotype and origin host species when comparing 114 isolates. Su et al. (2001) found that both host species and location (root or soil) of a given isolate had effects on chlorate sensitivity, albeit at varying levels.

Random amplified polymorphic DNA (RAPD) analysis can also be used to genetically differentiate M. phaseolina isolates by host species (Su et al., 2001). Jana et al. (2005) also successfully used two different kinds of markers derived from repetitive sequences to fingerprint and differentiate isolates from soybean, cotton, and chickpea. Baird et al. (2010) surveyed 109 isolates across the U.S. using 12 simple sequence repeats and found them to group into clusters based on both the location and host from which these isolates were found, but exceptions were also not uncommon. There is no evidence of sexual reproduction in M. phaseolina (Marquez et al., 2021). In a study that examined isolates across host plant families and geographical distances, hyphal fusions were observed frequently enough (64.3%) for the authors to conclude that there are no genetic barriers to ​​nonsexual genetic exchange amongst M. phaseolina isolates (Mihail and Taylor, 1995). The potential for gene flow across populations and the movement of agricultural products including soil likely contribute to the variability seen in this species. As nearly 4% of the annotated M. phaseolina genome encodes transposable elements, genetic transposition is another potential avenue for diversity due to the introduction of novel mutations, gene duplications, and horizontal gene transfers (Islam et al., 2012). Understanding M. phaseolina variance is important for developing resistant crop genotypes and highlights the importance of experimental design in studying this pathogen.

There are many conflicting reports regarding whether this heterogenous pathogen is adapted to infecting particular host species. To what degree the variability described above contributes to host specialization or preference is still unclear. For some examples of generalist necrotrophic fungi such as Botrytis cinerea, there is evidence suggesting that selective pressure maintains moderate virulence against a broad spectrum of host species (Caseys et al., 2021). In a study by Koike et al. (2016), M. phaseolina isolates from melon, thyme, and apple were tested against strawberry and failed to cause disease. Moreover, five cover crop species tested against strawberry-derived isolates also did not develop symptoms. However, results from Su et al. (2001) indicate that no specialization occurs for isolates from sorghum, cotton, or soybean, and while corn isolates seemed to colonize corn roots better than other isolates, it also was the most virulent against soybean. Similar cross-host species pathogenicity experiments have been carried out testing hosts against isolates collected across species and geographic distances, but no isolate had restricted virulence to a single host (Mayék-Pérez et al., 2001; Reyes-Franco et al., 2006). Although there is some evidence to suggest that host preference exists for certain isolates, it is not certain whether this is the result of evolutionary specialization or an artifact of environmental factors, which may alter the virulence of the pathogen against specific hosts. Recently, a large-scale analysis of three M. phaseolina isolates from three different hosts revealed structural chromosomal rearrangements and SNPs, an indicator of genomic flexibility, although no pattern was necessarily associated with a particular host (Burkhardt et al., 2019). However, ten genes were identified to be exclusive to strawberry pathogenic isolates (Burkhardt et al., 2019). Whether these genes (alleles)? are the causal factor for the host preference of these isolates has not been reported.

In 2012, Islam et al. reported on the first draft M. phaseolina genome, from an isolate of infected jute. Cross-species homology analysis revealed its close synteny with another soil-borne generalist fungus, Fusarium oxysporum, which shares 54.10% of its predicted genes with M. phaseolina. Additionally, the authors identified 537 predicted proteins in the M. phaseolina genome that are included in the pathogen–host interaction database (PHI-base), a collection of experimentally verified virulence effectors from bacterial, fungal, and oomycete pathogens. So far, genomes of M. phaseolina isolates from a variety of hosts including jute, strawberry, alfalfa, sorghum, and castor have been sequenced (Islam et al., 2012; Verma et al., 2018; Burkhardt et al., 2019; Hossen et al., 2019; Purushotham et al., 2020; Marquez et al., 2021). Sequencing and assembly M. phaseolina genomes has led to comparative studies across isolates and geographic locations, and the development of new methods to study plant–M. phaseolina interactions to elucidate biological processes involved in pathogenicity.

Liu et al. (2022) used transcriptomic analysis to study the microsclerotia formation stages of M. phaseolina, revealing an enrichment of reactive oxygen species (ROS)-related genes among up-regulated differentially expressed genes.

The first documented proteomic analysis on M. phaseolina published by Zaman et al. (2020) compared the expression of proteins with and without challenge of the fungus by the jute endophytic bacterium Burkholderia contaminans NZ. A large number of hydrolyzing enzymes predicted in the M. phaseolina genome were confirmed to be expressed through this study. M. phaseolina co-cultured with this bacterial antagonist showed signs of decreased virulence, morphological changes, and altered metabolic processes switching to a more dormant state that prioritizes survival over growth. Characterization of the M. phaseolina mycelial proteome by Arafat et al. (2022) revealed a network of proteins that could be classified by their function including growth and reproduction, stress tolerance and virulence, nutrient synthesis, transcription and translation regulation, and more. Sinha et al. (2022) identified 117 proteins in the M. phaseolina secretome isolated from wheat bran samples, including cell wall degrading enzymes (CWDEs) and proteases which are vital for breaching plant defense mechanisms. Recently, Pineda-Fretez et al. (2023) identified 250 proteins secreted by M. phaseolina grown in liquid media supplemented with a soybean leaf infusion. Among these proteins were numerous CWDEs, and peptidases, as well as 54 putative effectors (Pineda-Fretez et al., 2023). These proteomic studies provide the groundwork for potential functional studies of specific genes or gene groups involved in virulence.

A notable characteristic of the M. phaseolina genome is the abundance of genes involved in cell wall degradation. Numerous carbohydrate-active enzymes (CAZymes) including glycoside hydrolases (GH), glycosyltransferases (GT), carbohydrate esterases (CE), carbohydrate-binding modules (CBM), and polysaccharide lyases (PL), were predicted within the genome. The M. phaseolina genome possesses high levels of GHs (four times more than GTs), and CEs (Islam et al., 2012), which is consistent with previous observations that this pathogen has greater cellulolytic activity than most other fungi (Kaur et al., 2012a). A β-1,4-endoglucanase produced by this pathogen was previously characterized and shown to be similar to plant-derived endoglucanases (Wang and Jones, 1995). While the significance of this is unclear, use of a CWDE that is similar to a plant-derived enzyme may allow for surreptitious modification of the host cell wall without eliciting plant defense responses. A repertoire of genes associated with lignin depolymerization, including laccases, lignin peroxidases, galactose oxidases, chloroperoxidase, haloperoxidases, and heme peroxidases were also identified. The M. phaseolina genome encodes the highest number of laccases in comparison to seven other fungal species (Islam et al., 2012). Production of these CWDEs has also been observed in M. phaseolina grown in vitro (Ahmad et al., 2006; Ramos et al., 2016). The activity of phosphatidases, which hydrolyze phosphatide lipids and alter the permeability of cellular membranes, was also observed during Brassica juncea infection with M. phaseolina isolates of various levels of virulence. A strong correlation between host susceptibility, pathogen virulence, and phosphatidase activity was described, which peaked at 18 days post-inoculation (Srivastava and Dhawan, 1982). The extensive collection of genes associated with cell wall and membrane disruption suggests that this pathogen is well equipped to breach diverse wall polysaccharide and lipid bilayer compositions, which differ between plant lineages (Kubicek et al., 2014). This perhaps explains, at least in part, M. phaseolina’s wide host range.

An unusually high number of amino acid transporters, which are likely useful for accessing host protein degradation products, were predicted in the M. phaseolina genome. In addition, enzymes involved in amino acid metabolism are enriched in the mycelial proteome (Arafat et al., 2022). Uptake of host-derived metabolites, such as amino acids, is likely to occur via hyphopodia of M. phaseolina formed during its biotrophic infection phase (Figure 1A) as well as from leaking host cells with degraded cell walls and plasma membranes during its necrotrophic phase.

The M. phaseolina genome encodes a multitude of protein kinases and G-protein coupled receptors, some of which are known to facilitate host recognition in other pathogens (Islam et al., 2012). The abundance of such receptors and signal transduction components may contribute to the ability of M. phaseolina to recognize and utilize a great variety of plant species as hosts.

In fungi, redox signaling is essential for the regulation of developmental processes, crosstalk with plant hosts, and degradation of host lignocellulose (Breitenbach et al., 2015). M. phaseolina infection causes nitric oxide (NO) accumulation in jute (Corchorus capsularis), but the fungus itself is also capable of NO production in vitro (Sarkar et al., 2014). NO is a key signaling factor in the early stages of the host hypersensitive response (HR). In plants pathogen-induced NO production is accompanied by a burst of reactive oxygen species (ROS), such as superoxide (O₂⁻) and hydrogen peroxide (H₂O₂) (Van Baarlen et al., 2004; Asai and Yoshioka, 2009). Balanced accumulation of both NO and H₂O₂ seems to be required to trigger HR-associated cell death (Delledonne et al., 1998; Delledonne et al., 2001), although there are somewhat conflicting reports as to whether NO and ROS act synergistically (Delledonne et al., 2001) or antagonistically (Asai and Yoshioka, 2009; Asai et al., 2010) in this process.

Interestingly, genes involved in ROS-related functions were found to be differentially expressed in M. phaseolina during microsclerotia formation, suggesting that ROS, specifically O₂−, stimulates microsclerotia formation (Liu et al., 2022). Moreover, studies across multiple filamentous fungi have shown that ROS have functions in cell differentiation and development in fungi (Breitenbach et al., 2015). In vitro treatment of cultures with ROS further supported this hypothesis. A fascinating possibility is that ROS produced by M. phaseolina may interfere with ROS/NO-controlled processes in its hosts and vice versa. Pathogen-derived ROS and NO may contribute to HR in the host, which may increase its virulence during necrotrophy, while the defense-linked oxidative burst in the host may contribute to M. phaseolina microsclerotia formation.

Fungi produce diverse bioactive natural products, some of which enhance their virulence, such as mycotoxins (Chooi and Tang, 2012). A total of 75 putative secondary metabolite-related genes were identified in the M. phaseolina genome, which is more than double of what can be found in each of the plant pathogenic fungi Magnaporthe grisea, B. cinerea, Sclerotinia sclerotiorum, and Fusarium graminearum. These include genes encoding polyketide synthases (PKS), non-ribosomal peptide synthases (NPRS) and PKS-NPRS hybrids, which synthesize precursors to various secondary metabolites including those involved in virulence (Islam et al., 2012). Additional expression analyses and functional studies are required to confirm the role of these genes. Examples of secondary metabolites produced by M. phaseolina include phaseolinone (Siddiqui et al., 1979; Dhar et al., 1982), phaseolinic acid (Mahato et al., 1987), (−)-botryodiplodin (Ramezani et al., 2007; Abbas et al., 2019), succinic acid, tyrosol, (R)-mellein, cis-(3R,4R)-4-hydroxymellein, azelaic acid (Salvatore et al., 2020), kojic acid, moniliformin, orsellinic acid (Khambhati et al., 2020), macrollins A–C (Gao et al., 2021), phaseocyclopentenones A and B, and guignardone A (Masi et al., 2021). However production of such metabolites seems to be dependent on a number of factors including isolate variability and the presence of host substrates. To what extent virulence is affected by these metabolites is largely unknown.

Phaseolinone was described to be a novel phytotoxin when it was first identified in M. phaseolina and has been documented to affect plant seed germination, wilting, and leaf necrosis (Dhar et al., 1982; Bhattacharya et al., 1992; Bhattacharya et al., 1994). The compound (−)-botryodiplodin is a well-documented toxic metabolite prevalent in ascomycetes such as M. phaseolina. When soybean plants were infected with M. phaseolina, this compound was isolated exclusively from root tissues displaying charcoal rot symptoms (Ramezani et al., 2007; Abbas et al., 2019), suggesting that it is produced during pathogenesis. In these assays, however, phaseolinone could not be detected in any soybean tissue examined despite previous reports describing it as a metabolite unique to M. phaseolina. Analysis by Shier et al. (2007) provides several possible reasons for this discrepancy, one being that Dhar et al. mistakenly characterized (−)-botryodiplodin as phaseolinone. Another possible explanation is that different isolates of M. phaseolina produce different “cocktails” of toxins. This is supported by the fact that mellein was identified in some, but not all, isolates of this pathogen recovered from symptomatic soybean plants collected in Mississippi, U.S. (Khambhati et al., 2020). The abundance of phytotoxic secondary metabolites potentially contributing to M. phaseolina pathogenicity is mirrored by the induced expression of numerous detoxification-related genes in A. thaliana after infection with this pathogen (Schroeder et al., 2019).

Eight orthologs of the Phytophthora parasitica cellulose-binding elicitor lectin (CBEL) gene, which has a role in the perception of host cell wall components and is a known PAMP, were found in the M. phaseolina genome (Gaulin et al., 2002; Islam et al., 2012). The 13 amino acid comprising peptide Pep-13 is a known PAMP motif in cell wall transglutaminases from Phytophthora sojae (Brunner et al., 2002). Three homologs containing the Pep-13 motif are found in M. phaseolina (Islam et al., 2012). As several M. phaseolina-derived PAMPs have been identified thus far, it is likely that plants recognize the presence of this pathogen through mechanisms similar to those of other pathosystems.