The Notorious Soilborne Pathogenic Fungus Sclerotinia sclerotiorum: An Update on Genes Studied with Mutant Analysis

Ascomycete Sclerotinia sclerotiorum (Lib.) de Bary is one of the most damaging soilborne fungal pathogens affecting hundreds of plant hosts, including many economically important crops. Its genomic sequence has been available for less than a decade, and it was recently updated with higher completion and better gene annotation. Here, we review key molecular findings on the unique biology and pathogenesis process of S. sclerotiorum, focusing on genes that have been studied in depth using mutant analysis. Analyses of these genes have revealed critical players in the basic biological processes of this unique pathogen, including mycelial growth, appressorium establishment, sclerotial formation, apothecial and ascospore development, and virulence. Additionally, the synthesis has uncovered gaps in the current knowledge regarding this fungus. We hope that this review will serve to build a better current understanding of the biology of this under-studied notorious soilborne pathogenic fungus.


Introduction
Ascomycete Sclerotinia sclerotiorum (Lib.) de Bary is a soilborne pathogenic fungus, which was first described as Peziza sclerotiorum by Libert [1]. In 1945, Whetzel [2] established S. sclerotiorum (Lib.) de Bary as the major species of Sclerotinia, as it is the best-known and most economically important pathogen of the genus. It was proposed as the representative species for the genus Sclerotinia in the 1970s [3,4]. The Special Committee on Fungi and Lichens recommended the generic name Sclerotinia, and so did the General Committee of the International Association of Plant Taxonomists later on [5].
S. sclerotiorum belongs to the family Sclerotiniaceae [2] of the class Leotiomycetes. Sclerotiniaceae also includes two other very closely related species, S. minor and S. trifoliorum. All three species produce vegetative resting structures, termed sclerotia, and mushroom-like reproductive structures, named apothecia, during their life cycles. However, S. minor forms small sesame-sized sclerotia, whereas S. sclerotiorum and S. trifoliorum can form large sclerotia as big as peas. They all lack an obvious conidial stage. S. sclerotiorum has the broadest host range, which includes both herbaceous and succulent plants, and even some woody ornamentals and monocots. S. trifoliorum was isolated and succulent plants, and even some woody ornamentals and monocots. S. trifoliorum was isolated from vegetable legumes, while S. minor has been found on sunflowers, tomatoes, carrots, peanuts, and lettuce.
S. sclerotiorum is a highly damaging pathogen with diverse infection modes and a double feeding lifestyle of both biotroph and necrotroph ( Figure 1) [6,7]. It attacks host plants either by means of ascospores that can be discharged forcibly upwards from apothecia into the air, or by mycelium arising from infected tissue or from germinated sclerotia [8]. When ascospores land on susceptible host tissue, they can germinate under favorable conditions and start a new cycle of infection. To date, the microconidia of S. sclerotiorum have been observed amongst the mycelium and on the surfaces of sclerotia. However, no evidence of their functionality has been identified. Under moist and cool conditions, this fungus rapidly grows inside the infected host tissues and develops symptoms of browning, water-soaking, and a white, cotton-like mycelium, which leads to necrosis, stunting, premature ripening, and wilting of the host [8]. Therefore, the diseases that it causes have been given names, including stem rot, drop, crown rot, cottony rot, watery soft rot, blossom blight, and Sclerotinia blight. Upon killing the host, the fungus can saprophytically grow on the dead plant tissue. Sclerotia are later abundantly formed on the host surface and cavities, in plant debris, and in soil, where they are able to remain dormant for up to 10 years [9]. S. sclerotiorum has an extremely broad host range, which consists of more than 600 plant species, including almost all dicotyledonous and some monocotyledonous plants [10,11]. It can infect many economically important crops, such as canola; legumes such as soybean and peanut; sunflower; various vegetables such as lettuce and tomato; and monocotyledonous plants such as tulip and onion [10,12]. It is therefore not surprising that it causes significant economic losses globally every year [13]. For example, in China, oilseed rape yield losses that are caused by SSR (Sclerotinia stem rot) usually S. sclerotiorum has an extremely broad host range, which consists of more than 600 plant species, including almost all dicotyledonous and some monocotyledonous plants [10,11]. It can infect many economically important crops, such as canola; legumes such as soybean and peanut; sunflower; various vegetables such as lettuce and tomato; and monocotyledonous plants such as tulip and onion [10,12]. It is therefore not surprising that it causes significant economic losses globally every year [13]. For example, in China, oilseed rape yield losses that are caused by SSR (Sclerotinia stem rot) usually range from 10 to 20% and they may be up to 80% for severe SSR outbreak seasons [14][15][16]. In The United States, annual losses that are caused by this pathogen have exceeded $200 million [12].
Pathogens 2020, 9,27 3 of 22 As with most other fungal pathogens, chemical control is the most commonly used method for controlling the diseases that are caused by S. sclerotiorum, although it is environmentally unfriendly and can induce resistance from the pathogen. Chemical control also becomes less effective once the sclerotia are formed. Obviously, the incidence of disease would be greatly reduced if sclerotia could be either prevented from forming or destroyed in soil and plant debris. Flooding and crop rotation are good options that accelerate the decay of sclerotia [17]. However, flooding is impractical in most areas, and crop rotation has been found to be less effective as a control measure [18,19], because of the long survival periods of sclerotia in soil. Measures, like widening row spacing, and using wire trellis supports to raise foliage from the ground have been reported to be helpful [18,20]. However, factors, such as type of crop, method of cultivation, and environmental conditions, always affect the effectiveness of each control method.
For major crop diseases, the most welcoming and environmentally friendly control method is through the use of resistant cultivars. However, strong host single-gene resistance has not been found against S. sclerotiorum [12], which makes it difficult to improve resistance using classical breeding methods [7]. Even with mild resistance, the traits seem to be multigenic. Future development of integrated approaches may help to control the spread of the fungus with better understanding of the biology of this notorious pathogen and better QTL mapping strategies.
In this review, we focus on recent key molecular findings on the unique biology and pathogenesis process of S. sclerotiorum. A brief summary of the genome of the pathogen will also be discussed. The genomic sequences of S. sclerotiorum have been available for less than a decade [21]. However, it was recently updated with higher completion and better gene annotation [22]. We hope that this review will serve to build a better current understanding of the biology of this notorious soilborne pathogenic fungus.

Genome Sequences
The genome of S. sclerotiorum strain 1980, originally isolated from beans in western Nebraska USA, was first sequenced in 2011 using Sanger technology with 9.1x coverage. As approximately 1.6 Mb of sequences were not covered by the 38.0 Mb scaffolds, and thus missed from the assembly, the final size of S. sclerotiorum genome was estimated to be approximately 38.3 Mb, which consists of 16 linkage groups likely corresponding to 16 chromosomes (GenBank accession numbers: AAGT01000000) [21]. Most of the uncovered regions are located in the middle of chromosomes, which probably correspond to centromeres. The genome was predicted to contain 14,522 genes, with average GC contents of 41.8%. After removing genes encoding small proteins that are less than 100 amino acids in length or without evidence of expression (from ESTs and/or microarray signals), 11,860 predicted proteins with high confidence were deduced [21].
Initial analyses of the genome sequences revealed a large number of virulence related genes that were possessed by the S. sclerotiorum 1980 strain, including genes encoding cell wall degrading enzymes (CWDEs) and biosynthesis genes of phytotoxins and other secondary metabolites [21]. Amselem et al. also identified a significant amount of repetitive transposable elements (TEs), which comprises approximately 7.7% of the whole genome. This led to a postulation that the S. sclerotiorum genome experienced a recent major remodeling that was associated with a dramatic expansion of TEs. The TE expansion in S. sclerotiorum might also have an impact on its genome organization and function of gene inactivation, modification, or expression regulation [21].
Derbyshire et al. recently sequenced the genome sequence of the same S. sclerotiorum 1980 strain to near completion in 2017 using PacBio technology. It was annotated using extensive RNA-Seq data and manual curation. Therefore, this version of the whole-genome sequence became the choice of standard for Sclerotinia. They identified 70 candidate effector genes, and found a significant association between the positions of these secreted proteins and regions with a high relative RIP (repeat-induced point mutation) index, which suggested that S. sclerotiorum exhibits a subtly enhanced mutation rate of secreted proteins in specific genomic compartments as a result of transposition and RIP activity [23].
More recently, Derbyshire et al. sequenced the genomes of 25 field isolates of S. sclerotiorum that were collected from four different continents-Australia, Africa (north and south), Europe, and North America (Canada and the Northern United States). They conducted SNP-based analyses on population structure and selective sweeps [24,25]. These 25 isolates can be grouped into two major populations, where population 1 consists of 11 isolates from Canada, the USA, and France, and population 2 includes nine isolates from Australia and one from Morocco. A single candidate selective sweep was identified in the Australian and Moroccan group, which covers less than 0.001% of the genome. It is reasonable to speculate that a slow evolution rate and extensive self-fertilization of S. sclerotiorum may have led to a striking absence of strong selective sweeps, as there is only one gene within this 10 Kb region, which encodes a major facilitator superfamily transporter that was only negligibly expressed at late stages of Brassica napus infection [22]. In support of such genome conservation, S. sclerotiorum strains from even a tropical and a temperate region, although segregated in the populations, still exhibit similar genetic structure [26].

Transcriptomic and Secretomic Analysis
The availability of the high-quality genome sequences facilitated comparative transcriptomic and secretomic analysis of this notorious pathogen, which in turn facilitated biological investigations. Seifbarghi et al. (2017) was able to delineate gene expression patterns that signified transitions between pathogenic phases of S. sclerotiorum, namely, host penetration, ramification, and necrotrophic stages, through RNA-Seq analysis focusing on events occurring through the early (1 h) to the middle (48 h) stages of infection. These expression data provide evidence for the occurrence of a brief biotrophic phase soon after host penetration [27]. 432 proteins were identified in S. sclerotiorum in an interspecies comparative analysis of the predicted secretomes of S. sclerotiorum and Botrytis cinerea. Among them, 16% of the encoding genes reside in small gene clusters that are distributed over 13 of the 16 predicted chromosomes [28]. These candidate genes provide a reservoir for future reverse genetics analyses to delineate the pathogenesis mechanisms in the S. sclerotiorum-host plant interactions.
Mutant analysis is paramount to establish a causal relationship between a gene and a studied biological process, as omics experiments only provide vague association clues. Hence, the rest of this review will summarize all of the genes studied in S. sclerotiorum so far using such convincing mutant analysis methods. A Venn diagram of the functions of these encoded proteins is provided for readers' overview ( Figure 2). We also constructed a chromosomal map for these genes to detect possible clustering events that are often associated with virulence factors (Figure 3). However, little clustering is observed, except for genes contributing to mating and melanisation, reflecting the under-studied nature of this fungus. Table 1. List of all Sclerotinia sclerotiorum genes studied so far using mutant analysis. The genes are ordered in regards to their chromosomal positions ( Figure 3).    Table 1).

Figure 2.
A Venn diagram summary of the genes that have been studies with mutant analysis (detailed in Table 1). Pathogens 2020, 9, x FOR PEER REVIEW 6 of 22

Molecular Dissection of S. sclerotiorum Biology
Here we adopt the gene/mutant/protein nomenclature that is most commonly used by Sclerotinia researchers, as there are many discrepancies in S. sclerotiorum literature. As an example, Abc1 protein is encoded by wild-type gene that is italicized as Abc1. Mutant is denoted as abc1. Exceptions are specifically explained. Another consideration to keep in mind is the presence of huge variations in the field isolates and local hosts used in these studies (Table 1), which likely contribute to some of the discrepancies that were observed in the phenotypes of mutants of the same genes.

Regulation of Mycelial Growth and Virulence
During the last two decades, the availability of whole genome sequences and improved transformation as well as knockdown/knockout methods have greatly facilitated the molecular study of S. sclerotiorum. Most of the genes studied so far that affect growth are relatively conserved genes that also affect differentiation, such as sclerotial and apothecial development (Figure 2), which will be discussed in the next sections. In this part, we specifically discuss genes only affecting hyphal growth rate and virulence, but not differentiation, during development, as mycelial growth precedes processes of virulence and differentiation.
It is not surprising that the mutants here with major growth defects often exhibit virulence attenuation, although there are certainly exceptions such as mat1-2-1, which exhibits slower growth, aberrant apothecial morphogenesis, but normal virulence [29] (discussed more later).

Molecular Dissection of S. sclerotiorum Biology
Here we adopt the gene/mutant/protein nomenclature that is most commonly used by Sclerotinia researchers, as there are many discrepancies in S. sclerotiorum literature. As an example, Abc1 protein is encoded by wild-type gene that is italicized as Abc1. Mutant is denoted as abc1. Exceptions are specifically explained. Another consideration to keep in mind is the presence of huge variations in the field isolates and local hosts used in these studies (Table 1), which likely contribute to some of the discrepancies that were observed in the phenotypes of mutants of the same genes.

Regulation of Mycelial Growth and Virulence
During the last two decades, the availability of whole genome sequences and improved transformation as well as knockdown/knockout methods have greatly facilitated the molecular study of S. sclerotiorum. Most of the genes studied so far that affect growth are relatively conserved genes that also affect differentiation, such as sclerotial and apothecial development (Figure 2), which will be discussed in the next sections. In this part, we specifically discuss genes only affecting hyphal growth rate and virulence, but not differentiation, during development, as mycelial growth precedes processes of virulence and differentiation.
It is not surprising that the mutants here with major growth defects often exhibit virulence attenuation, although there are certainly exceptions such as mat1-2-1, which exhibits slower growth, aberrant apothecial morphogenesis, but normal virulence [29] (discussed more later).
The RNAi knockdown of highly conserved transcription factors Ams2 or Mads lead to reduced mycelial growth rate and virulence, which suggested that they are early growth regulators. Ams2 is a GATA-box domain transcription factor. Ams2 is required for proper expression of histone and cell cycle related genes, similar as in Schizosaccharomyces pombe [55]. Mads is orthologous to yeast Mcm1 with a MADS-box domain, which is essential for viability, cell cycle, mating, mini-chromosome maintenance, recombination, and stress tolerance [61]. Although mutants of these early growth factors severely affect growth, they can still form sclerotia, which indicated that the later stages of fungal differentiation pathways, such as sclerotial formation and apothecial development, might be less dependent on these early growth factors.
BAX inhibitor-1 (Bi1) is a highly conserved protein that is found in both eukaryotes and prokaryotes, exhibiting apoptosis-inhibiting activity. RNAi knockdown mutants in S. sclerotiorum showed more aerial hyphal growth and reduced virulence [48]. The mutants also exhibit reduced tolerance to chemical stress, as its gene expression is induced by various stresses. Bi1 is likely a stress specific regulator dedicated to both biotic and abiotic stress responses, due to its minor roles in growth and development.
Unlike the previously described genes, PemG1 serves an opposite function in regards to growth and virulence. The RNAi mutants of PemG1 exhibit faster mycelial growth and enhanced virulence, indicating that it is a negative regulator in these processes. Consistently, more infection cushions, more CWDE activities, and less susceptibility to salt stress were observed [54]. However, the exact mechanism on how this fungal specific protein negatively regulates growth and virulence is unclear.

Signaling Events Leading to Sclerotial Formation
Sclerotia of S. sclerotiorum initiate with small primordia and rapidly develop into white compact hyphal masses. They mature with dehydration and pigmentation once they cease growth. The melanised outer surface of the sclerotium is mostly responsible for resisting adverse conditions [8]. We will first discuss the genes affecting both growth and sclerotial formation, which are likely very upstream components that are shared by both processes. As expected for broad regulators, these mutants show not only defective growth and sclerotial malformation, but also almost always loss of virulence and other pleiotropic defects ( Table 1).
The melanisation of sclerotia is an important process for the long-term survival of Sclerotinia species in nature. The deletion of melanin biosynthesis genes scytalone dehydratase (Scd1) or trihydroxynaphthalene reductase (Thr1) resulted in slower mycelial growth, reduced number, size, and pigmentation of sclerotia [39], which suggested that melanisation is critical for sclerotial maturation and perhaps feedback promotion of growth. However, unlike in other pathogenic fungi, the virulence of these knockout strains is not affected, suggesting that melanin does not contribute to pathogenesis in Sclerotinia as it does for other fungi, like Magnaporthe species.
Knocking out some other conserved genes in S. sclerotiorum can also lead to diverse defects. For example, the peroxisomal carnitine acetyl transferase (CAT) encoding gene in S. sclerotiorum is Pth2. The CAT activity is required for acetyl-CoA transport, which is needed for fatty acid metabolism. The pth2 knockout mutants show aberrant growth, non-sclerotial formation, and reduced oxalic acid (OA) levels and virulence [41]. In another instance, the silencing of integrin-like gene Itl leads to altered colony morphology, smaller and irregular shaped sclerotia, and hypovirulence [62]. Although Cvnh homologs seem to only exist in Sclerotinia and Botrytis, the CVNH domain is a conserved domain. The silencing of this gene lead to reduced growth rate, non-sclerotial formation and lower virulence [45].
Phosphorylation and de-phosphorylation events are critical for almost all signal transductions. Therefore, it is not surprising that a number of kinases and phosphatases are involved in both growth and sclerotial formation. Mutation of two-component histidine kinase Shk1 leads to slower and altered hyphal growth and failed sclerotial formation [35]. Although the mutant is also sensitive to osmotic stress and shows heightened resistance to fungicides, it exhibits normal virulence. Shk1 likely works upstream of the MAPK cascade to control these processes. Smk1 and Smk3 are two MAPK kinases that have been studied in S. sclerotiorum so far. Mutants of these two kinases show similar growth defects and non-sclerotial formation [60,71], suggesting that they may function downstream of Shk1. cAMP could act upstream of the MAPKs, as exogenous application of cAMP inhibits Smk1 expression and MAPK phosphorylation [71]. Type 2A phosphoprotein phosphatase Pph1 possibly functions downstream of the MAPK. The mutants of pph1 show arrested growth [67]. Consistently, PP2A R2 B regulatory subunit rgb1 mutants show aberrant sclerotial formation and failed penetration during infection [67].
Reactive oxygen species (ROS) play diverse roles in development and virulence in fungi. This is supported by the mutant analysis of type I catalase Scat1. The deletion of Scat1 lead to slower radial growth, higher number of small sclerotia that are not properly melanised, and hypovirulence [44]. Calcium is another common factor that is involved in signaling. The study of Cna1 and Caf1 supports its contribution in S. sclerotiorum biology. Calcineurin Ser/Thr protein phosphatase Cna1 is conserved in eukaryotes, whose activity is dependent on calcium and calmodulin. Antisense cna1 mutants show altered growth, sclerotial development, and hypovirulence [32]. Caf1 is a secreted protein with a putative Ca 2+ -binding EF-hand motif, which seems to be conserved in fungi. caf1 T-DNA mutants exhibit slightly slower growth, deformed sclerotia, and hypovirulence [43].
Conserved transcription factors are another large group of genes that are responsible for signaling. Sfh1 is a GATA transcription factor, but with an SNF5 domain. It is a member of the housekeeping RSC (Remodels Structure of Chromatin) complex, which is an ATP-dependent chromatin remodeler that is essential for cell cycle [34]. Its diverse functions explain its involvement in growth, sclerotial development, and virulence. Transcription factor Ste12 likely acts downstream of the MAP kinase cascades. In ste12 RNAi lines, slow growth, smaller sclerotia, and fewer appressoria, leading to hypovirulence, were observed [13]. Forkhead (FKH) box (FOX) family transcription factors are important regulators of primary metabolism, cell cycle, and morphogenesis in animals and fungi. fkh1 RNAi mutants show slower growth, non-sclerotial formation, hypovirulence, and defective stress responses [52], suggesting that it is a transcriptional regulator that is shared by multiple pathways.
CWDEs have been speculated to play key roles in fungal pathogenesis. In support of this, the deletion of Xyl1 that encoded an endo-β-1, 4-xylanase leads to a loss of pathogenicity. Interesting, growth retardation, less sclerotial formation, and dense hyphal branching were also observed in the mutant [69], suggestive of a role of xylanase in development as well.
Opsins are conserved light-sensitive proteins involved in circadian rhythms. sop1 RNAi lines show reduced growth, non-sclerotial formation, hypovirulence, and heightened sensitivities to osmotic and fungicide stresses [33]. These observations suggest that light may have diverse roles in various processes in Sclerotinia through opsins.
Rearrangement hotspot (Rhs) repeat-containing proteins are widely distributed in bacteria and eukaryotes. Rhs1 is a secreted protein and its orthologs seem to only be present in Sclerotinia and Botrytis species. Rhs1 RNAi mutants show slightly slower colony growth, fewer and larger sclerotial formation, and hypovirulence [51]. These data suggest that Rhs1is likely a key virulence factor for Sclerotinia and it also slightly contributes to growth and sclerotial development. Cerato-platanins are fungal secreted elicitors that can trigger plant immunity and cell death. Sm1 can cause host hypersensitive response. Interestingly, besides virulence, RNAi mutants of Sm1 also exhibit slower growth, impaired sclerotial development, and heightened sensitivities to different stresses [73], suggesting a broader role of the elicitor. However, these defects were not detected in a separate study in deletion mutant cp1 mutating the same gene [72]. Future careful transgenic complementation experiments are needed to solve these discrepancies.
The existence of mutants that have normal mycelial growth, but do not form sclerotia, suggest that later stages of sclerotial development are separated from growth. Although most of these mutants still show attenuated virulence, some exhibit normal pathogenicity, such as sl2 and ggt1 (7-glutamyl transpeptidase). RNAi knockdown of either Sl2, GAPDH (Gpd), or Hex1 lead to altered sclerotial formation or melanisation, similar to its interacting proteins glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and Hex [46]. However, the virulence of sl2 is not affected, suggesting that sclerotial formation and virulence can be separately controlled by distinct downstream factors. As a fungal specific secreted protein localizing to fungal cell wall, how Sl2 affects sclerotial development is unclear. Ggt is a highly conserved enzyme that contributes to redox homeostasis. The deletion mutants of ggt1 show much higher GSH+GSSG and hydrogen peroxide accumulation, and separation of the cortex layer of mature sclerotia from the medulla [63]. In addition, the mutants are defective in compound appressoria production, only affecting the penetration stage of infection. These mutant phenotypes suggest that GSH, GSSG, or GSH/GSSG ratio contribute to the regulation of sclerotia maturation and appressoria development.
The rest of the genes studied so far that affect sclerotial formation, but not growth, all show attenuated virulence. They likely act upstream of Sl2 and Ggt. ROS increase was observed in sclerotial initials and infection cushions in S. sclerotiorum, likely due to the combined activities of NADPH oxidases Nox1 and Nox2. Similar to exogenous application of NADPH inhibitor or ROS scavenger, the RNAi mutants of either Nox1 or Nox2 exhibit failed sclerotial development [50], indicating that ROS is required for sclerotial initiation. Interestingly, reduced OA production and virulence was only observed in nox1, but not nox2, which suggests that these two genes are not fully redundant, likely due to their different expression patterns. Consistently, when the thioredoxin reductase (Trr1) gene that is partly responsible for ROS detoxification is silenced, less and larger sclerotia are formed [47]. However, virulence is attenuated in Trr1 silenced strains, suggesting that an optimum ROS is needed for virulence.
Low pH was known to promote sclerotial proliferation. The pH-sensitive transcription factor Pac1 is responsible for ambient pH-induced gene expression. Consistently, the deletion of Pac1 caused aberrant sclerotial formation and reduced OA levels, thus leading to hypovirulence [53]. As higher levels of OA reduce pH and may affect ROS homeostasis, proper balance among ROS, OA, and pH is likely paramount for the biology of this fungal pathogen.

Control of Apothecial and Ascospore Development
The apothecium consisting of a stalk (stipe) and a saucer-shaped disc is characteristic of the Sclerotinia genus (Figure 1). Asci develop on the upper surface of apothecia and when they mature, the ascospores are dispersed [8]. As mating-type (Mat) genes have generally been established as key regulators of sexual development of ascomycetes, it is not surprising that they affect apothecial development in the homothallic S. sclerotiorum. Similar to other fungi, the S. sclerotiorum Mat locus contains a cluster of four Mat genes encoding transcription factors, with MAT1-1s encoding an alpha-domain and Mat1-2s encoding high mobility group (HMG) proteins. By deletion knockout analysis [29], each of the four Mat genes were analyzed in regards to their contributions to regular mycelial growth, virulence, sclerotial formation, and apothecial development. While these Mats contribute little to vegetative growth, development, and pathogenesis, Mat1-1-5, Mat1-1-1, and Mat1-2-1 are essential for early, and Mat1-2-4 is critical for late sexual stages, including apothecial initiation and ascospore development. As transcription factors, it is also not surprising that these Mat gene mutants exhibit altered expression of the Mat locus and putative pheromone and pheromone receptor genes, explaining their potential regulatory mechanism and specific phenotypes.
Similar to the Mats, another gene that specifically contributes to apothecial development is FoxE2 [49]. FoxE2 is a member of the forkhead box (FOX) transcription factor (TF) family. FOX proteins play diverse roles in morphogenesis, development, pathogenicity, and stress responses in animals and fungi. Knockout analysis of FoxE2 in S. sclerotiorum showed that it does not affect mycelial growth, sclerotial formation, or virulence. The striking mutant phenotypic similarity between the Mats and FoxE2 brings the question on the relationship among these transcription factors. It could be possible that a transcriptional network that includes these transcription factors needs to be orchestrated for proper sexual stage development in S. sclerotiorum.
As a UV-A photoreceptor, Cry1 was suspected to be involved in apothecial development, as its expression was induced by UV during this specific developmental phase. However, no obvious defects in apothecial development were detected in the cry1 deletion mutant [58], suggesting that it is not a major regulator, or there exists redundancy masking its effects on photomorphogenesis.
All other mutants in the literature with defects in apothecial development show pleiotropic defects, including sac1 (Adenylate cyclase) [70] and nsd1 (GATA-type IVb zinc-finger transcription factor) [75], both encoding conserved proteins. cAMP is a well-known molecule with diverse roles for many biological processes in plant fungal pathogens. Although protein kinase A catalytic subunit gene (Pka1) was hypothesized to mediate cAMP signaling, the pka1 single mutant exhibits non-detectable defects, likely due to genetic redundancy of Pka2 [56]. In comparison, with dramatic cAMP level reduction, the sac1 mutant displays slower growth, abolished pathogenicity, and failed apothecial development [70]. In other fungi, Nsd1 serves as transcription factor for development and environmental responses. In nsd1 knockout mutants, altered hyphal growth, smaller sclerotia, failed appressorium formation leading to failed pathogenicity, and failed apothecial initiation were observed [75]. These indicate that Nsd1 is likely a general transcription factor that regulates many processes in S. sclerotiorum, or a specific regulator that regulates a general downstream factor that is shared by many pathways. Recently, the NO homeostasis regulating glutathione-dependent formaldehyde dehydrogenases (Fdh1) encoding gene was revealed to be one of the targets of Nsd1 [74]. The deletion mutants of fdh1 exhibit slower mycelial growth, smaller sclerotia, and defective compound appressoria, leading to failed penetration into host tissue, being suggestive of the keys roles NO plays in these processes. The high similarity between the fdh1 and nsd1 mutant phenotypes indicates that Fdh1 could be the main target of Nsd1 during growth and development.

Regulation of Fungal Pathogenesis of the Fungus
Following a short biotrophic phase after infection, S. sclerotiorum causes tissue maceration and necrosis leading to rapid cell death and host cell wall degradation during fungus colonization. Toxins and CWDEs are believed to play critical roles in promoting these processes. The study of axp mutants deleting a secreted arabinofuranosidase/β-xylosidase precursor reflects the importance of CWDEs in the virulence of S. sclerotiorum. Attenuated virulence was observed in canola. Similarly, the critical contribution of ROS in virulence is corroborated in sod1 mutants affecting Cu/Zn superoxide dismutase [36,37]. OA is another key factor that has long been hypothesized to contribute to the pathogenicity of this pathogen. Readers are referred to two excellent reviews that have summarized the historic and comprehensive studies of the pathogenicity and virulence of S. sclerotiorum [65,66].
During infection, S. sclerotiorum produces large amounts of OA. Earlier physiological and pharmacological studies have highlighted its importance. In addition to virulence, UV mutants where the exact mutations were not identified could not produce OA and failed to form sclerotia [76], also suggesting a link between OA and development. It was not until the availability of genetically defined OA deficient mutants that the contribution of OA towards Sclerotinia biology was partly settled. In odc2 RNAi strain where the gene encoding oxalate decarboxylase in charge of catabolism of OA is silenced, less and non-functional appressoria were formed, leading to hypovirulence [64]. These data support the involvement of developmentally regulated OA accumulation in penetration-dependent infection. In recently generated oah (oxaloacetate acetylhydrolase) mutants through the deletion or CRISPR, where OA biosynthesis is blocked, failed appresorium formation, hypovirulence, and delayed sclerotial formation was observed [40]. However, the deletion or T-DNA mutants of oah in a separate study show pH-dependent virulence alteration and normal growth and development. These discrepancies can likely be resolved through more careful examination of other mutants that are deficient in OA biosynthesis.
Many pathogen effectors are secreted into the host to dampen host immunity and/or promote virulence. In recent years, effector biology has been focusing on the study of secreted proteins, as these effector candidates can be relatively easily identified through secretome and RNA-seq analysis. Many secreted proteins in S. sclerotiorum have diverse roles in both development and virulence (Table 1 and discussions in the previous sections). Examples of effectors only affecting virulence in S. sclerotiorum include Ssvp1, V263 and Qdo (quercetin dioxygenase). Less conserved Ssvp1 [30] and V263 [38] are both small secreted proteins that contribute to virulence. On the other hand, more conserved Qdo catalyzes the cleavage of the flavonol carbon skeleton [57], revealing the critical contributions of degrading host anti-microbial flavonol in fungal virulence.
Besides secreted proteins, there are also non-secreted regulators that are solely dedicated to pathogen penetration and virulence regulation. The examples in S. sclerotiorum include Svf1 (Survival factor 1) [31], Pks13 (polyketide synthase 13) [40], and Nacα (nascent-polypeptide-associated complex alpha subunit) [59]. In yeast, Svf1 promotes survival under oxidative stress conditions. Therefore, it likely assists in coping with ROS during infection in Sclerotinia. Pks13 is likely involved in melanin biosynthesis in appressoria, as its CRISPR mutant exhibits albino compound appressoria without affecting virulence. Nacα is a highly conserved protein with transcriptional co-activator activity. Interesting, Nacα negatively impacts the expression of polygalacturonase-encoding genes, which likely contributes to the enhanced virulence phenotypes of the mutant.

Summary and Future Perspectives
Over the decade, many genes that have been involved in pathogen development and pathogenesis have been characterized in S. sclerotiorum (Table 1). Much has been unveiled for the pathogen growth, development, and differentiation signaling pathways. Accordingly, more evidence is supporting the two-phase infection model, where the pathogen suppresses host basal defense prior to killing and degrading host cells [11]. In the short biotrophic phase, the pathogen uses well-orchestrated strategies to overcome host immunity. It might spatially achieve this via the production of OA, ROS, CWDEs, and effectors in compound appressoria or primary invasive hyphae.
Among almost 15,000 genes encoded in its genome, less than 60 have been studied so far using careful mutant analysis. This explains the under-studied nature of this notorious pathogen and the lack of connections in the signaling pathways of its different special biological processes. Besides the traditional one-gene-at-a-time reverse genetics analysis, more robust methods, such as genome-wide CRISPR, should be developed to enhance the study of this economically important fungus.
Due to the short biotrophic period, as well as the switching to necrotrophic phase, identifying and assigning the temporal, spatial, and tissue-specific functions to virulence factors are real challenges. Future development of cell biology and biochemical tools tailored for S. sclerotiorum will offer opportunities for solving many mysteries of the fungus. Considering this specific form of pathogen-host interaction system, challenges also exist ahead in applying quantitative disease resistance breeding and phenotypic screens for stage-specific defenses. More collaboration among the molecular pathologists and breeders will be critical to solve the host resistance difficulties of the pathogen.