Next Article in Journal
Genome-Wide Analysis of LEA Gene Family in Rosa chinensis ‘Old Blush’ and Cold-Induced Expression Patterns in Two Species
Next Article in Special Issue
Symbiotic Cultivation of Gastrodia elata: Armillaria Strain Selection Reprograms Carbon Allocation to Balance Tuber Yield and Phenolic Glycosides
Previous Article in Journal
An Improved Lightweight ConvNeXt for Peach Ripeness Classification
Previous Article in Special Issue
An Integrated Analysis of WRKY Genes in Autotetraploid Bupleurum chinense: Evolution, Stress Response, and Impact on Saikosaponin Biosynthesis
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Horticulturae
Volume 12
Issue 2
10.3390/horticulturae12020135
horticulturae-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Variation in the Number of Genes in the Secretomes of Isolates of Ilyonectria robusta and Ilyonectria mors-panacis Pathogenic to American Ginseng (Panax quinquefolius)

by
Paul H. Goodwin
,
Moez Valliani
and
Tom Hsiang
*
School of Environmental Sciences, University of Guelph, 50 Stone Road East, Guelph, ON N1G 2W1, Canada
*
Author to whom correspondence should be addressed.
Horticulturae 2026, 12(2), 135; https://doi.org/10.3390/horticulturae12020135
Submission received: 18 December 2025 / Revised: 19 January 2026 / Accepted: 22 January 2026 / Published: 25 January 2026
(This article belongs to the Special Issue Breeding, Cultivation and Metabolic Regulation of Medicinal Plants—2nd Edition)
Browse Figures
Versions Notes

Abstract

For 12 isolates of Ilyonectria mors-panacis and 4 isolates of Ilyonectria robusta, the number of genes in the secretome showed a negative correlation with growth rates in culture, especially for small secreted non-cysteine-rich and cysteine-rich proteins, and several proteases and lipases, while it was positively correlated with genes for six CAZyme classes/modules and other proteases and lipases. However, this significant correlation with growth rate was influenced by the I. robusta isolates mostly having faster growth rates than the I. mors-panacis isolates on PDA, indicating a species-level difference. The only significant relationship of gene number to virulence was a positive correlation with genes for secreted glycoside hydrolases in families 18 and 78, and this was related to differences between isolates, even if only I. mors-panacis isolates were examined, indicating a difference within species. Glycoside hydrolase family 18 includes chitinase-like proteins, endo-β-N-acetylglucosaminidases, lectins, and xylanase inhibitors, which could help suppress triggered immunity by the host and regulate fungal xylanase activity. Glycoside hydrolase family 78 contain α-L-rhamnosidases that can cleave flavonoid glycosides, saponins, and ginsenosides, which could degrade antimicrobial compounds produced as a host response during infection. These results indicate that the number of certain classes of secreted enzymes could be a factor in both growth rate in culture and virulence.

1. Introduction

Ilyonectria spp. are soil-borne fungi that can cause root rot in ginseng [1] as well as many other plants [2]. They typically cause red-to-dark brown lesions on ginseng roots, wilting with reddish aerial parts, and plant death [3]. Cabral et al. [2] proposed a reclassification of Ilyonectria spp. with 18 species in the Ilyonectria radicicola complex, with the clearest separation based on the histone H3 sequences, and among those species, I. mors-panacis, I. crassa, I. panacis, and I. robusta were described from ginseng roots. Ilyonectria mors-panacis was formerly classified as Cylindrocarpon destructans f. sp. panacis [4]. Full genome sequencing of 16 isolates of I. mors-panacis and I. robusta revealed that sequences of both histone H3 and total predicted gene sequences in the genomes could clearly differentiate isolates of I. mors-panacis from those of I. robusta. Furthermore, the total predicted genes in the genomes could divide the I. mors-panacis isolates into closely related types 1 and 2 [5]. These differences were also observed when comparing just the predicted secretomes of the isolates.
The secretomes of plant pathogenic fungi have been the focus of many studies of virulence genes because the secretome can act during infection to manipulate host cell physiology to obtain nutrients, suppress plant defense, and ultimately promote infection [6]. There is evidence that the secretomes of I. robusta and I. mors-panacis isolates may be important for virulence. For example, secreted resorcylic acid lactones of I. mors-panacis may suppress the defenses of ginseng roots allowing secreted siderophores, such as N,N′,N”-triacetylfusarine C, to be more effective in planta for pathogen invasion [7]. In addition, Rahman and Punja [8] reported that C. destructans isolates with higher pathogenicity to ginseng produced higher amounts of certain secreted enzymes, such as pectinase for plant cell-wall degradation and polyphenoloxidase for lignin degradation, compared to weakly virulent isolates.
One factor that can affect the traits of fungi is gene number. Differences in gene number among isolates can potentially be related to differences in gene expression because expression from multiple genes in a gene family can be much higher than from a single gene [9]. Variants in gene number can result from gene duplication, deletions, or a combination of both, and have been identified in a variety of fungal species [10]. For example, an examination of 132 isolates of Saccharomyces cerevisiae for genome-wide variation in gene number revealed that approximately 4% of the genome per isolate differed in number, with as few as 2 copies per genome and up to 57 copies per genome, such as for the S-adenosylmethionine-dependent lysine methyltransferase (YGR001C) gene [11]. They reported that the greatest variation in gene number was generally observed for genes associated with fermentation-related processes, such as copper resistance, flocculation, and glucose metabolism. This may be because they are expressed before or during the diauxic shift, when glucose becomes limiting, and the yeast needs to be able to quickly switch to other substrates, usually from glucose to ethanol and acetate [12]. Steenwyk and Rokas [11] suggested that isolates with higher numbers of genes related to the diauxic shift were able to increase production of proteins needed at that time for processes such as environmental stress responses. An examination of 71 isolates of Aspergillus fumigatus based on variation in predicted gene numbers showed that approximately 10% of the genome was composed of gene families with variable gene number, and variation in gene number was associated with functions such as cell surface transporters and hydrolases, possibly involved in responding to changes in the environment and pathogenicity, respectively [13]. Gene number variation has also been observed for plant pathogenic fungi. Isolates of the barley pathogen, Rhynchosporium commune, showed higher variation in the number for genes contributing to host infection (effectors and cell-wall degrading enzymes), which was associated with gene duplications and transposable elements [14]. Gene number variation was also observed for a gene of the wheat pathogen Zymoseptoria tritici, related to how effective host defenses were against the pathogen, indicating that it may be involved in avoiding or responding to such defenses [15].
In this study, the number for genes of the secretome was examined using the full draft genome sequences of 16 isolates of I. mors-panacis and I. robusta. The genes of the secretome under examination were related to small secreted non-cysteine-rich proteins (SSNPs), small secreted cysteine-rich proteins (SSCPs), secreted carbohydrate active enzymes (CAZymes), secreted proteases, and secreted lipases. They were chosen because of their potential roles in nutrient acquisition and plant pathogenicity, such as small secreted proteins (SSPs) inducing programmed cell death [16], secreted CAZymes degrading sugar polymers in the plant cell wall [17], secreted proteases degrading host defense proteins in the apoplast [18], and secreted lipases degrading plant cell wall membranes [19]. The goal was to examine variation in the number of genes for isolates within and between two species of Ilyonectria and their relationship to growth rate on culture media and virulence to roots of Panax quinquefolius.

2. Materials and Methods

2.1. Source of Isolates

Sixteen isolates of Ilyonectria spp. were obtained from diseased roots of P. quinquefolius in Ontario and British Columbia, Canada, except for IR.DAOM139398, which was isolated from roots of Prunus cerasus from British Columbia (Table 1). The isolates were grown on PDA (potato dextrose agar, Fisher Scientific, Mississauga, ON, Canada) for 4 weeks in dark at 22 °C, and conidia were stored in 10% sterile glycerol (Fisher Scientific) at −70 °C.

2.2. Isolate Growth Rate and Virulence

Isolates were grown on PDA for 12 days in the dark at 22 °C. Growth rate was determined from eight replicates with three plates per replication by measuring the diameter (cm) of each culture. To determine virulence, the isolates were first cultured on V8 agar (20% commercial V8 juice) for 4 weeks in the dark. Macroconidia were then harvested and suspended in sterile distilled water (dsH2O) to 1 × 106 spores/mL. Panax quinquefolius roots collected from ginseng gardens near the Simcoe area of Ontario were surface-sterilized with 75% ethanol for 10 min followed by 5% bleach for 5 min, and then thoroughly washed with dsH2O. Holes in the roots (1.5 mm wide, 9 mm deep) were made using a sterilized needle. To each hole, 15 µL of spore suspension was added, and the inoculated roots were incubated in parafilm-wrapped sterile Petri dishes at 22 °C. Control roots were wounded and treated with only dsH2O. Lesion areas were determined at 12 days post-inoculation (dpi) by tracing the lesion area on acetate sheets, and the areas were quantified using ImageJ software 1.54b [20]. Statistical analyses of growth rate and virulence were performed using Minitab version 16 with analysis of variance (ANOVA) and means comparisons performed using Fisher’s LSD Test with the level of significance set at α = 0.05. For the correlation analysis of growth rate or virulence to gene number, R2 was calculated using the fitted linear model function with the quadratic equation (lm(formula = y ~ poly(x, degree = 2, raw = TRUE))) using R 4.5.1 (2025-06-13 ucrt, https://cran.r-project.org/bin/windows/base/old/4.5.1, accessed on 21 January 2026) through RStudio (2025.05.1 Build 513, Mariposa Orchid (ab7c1bc7) release for Windows x64, https://posit.co/downloads, accessed on 1 June 2025). The p-value of the linear correlation between the number of genes and the dependent variable (growth rate or lesion size) was obtained using the function correl() from the R package agricolae 1.3-7 (https://www.rdocumentation.org/packages/agricolae/versions/1.3-7, accessed on 1 June 2025), which calls the cor.test() function, with the Pearson method and a two-sided t-test method with α = 0.05.

2.3. Genome Sequencing, Assembly, and Gene Prediction

DNA was extracted from 600 mg of mycelium per isolate cultured on PDA covered with sterilized cellophane, as per Edwards et al. [21]. The DNA was sent to Génome Québec and McGill University Innovation Centre (http://www.genomequebec.com, accessed on 21 January 2026, Montreal, QC, Canada) for 150 bp paired-end sequencing with the Illumina HiSeq X Ten platform (Illumina, San Diego, CA, USA). Assembly of raw sequenced reads into contigs and scaffolds was performed with Velvet (https://github.com/dzerbino/velvet, accessed on 14 May 2018), SOAPdenovo (http://soapdenovo2.sourceforge.net, accessed on 21 May 2018), and Abyss (http://www.bcgsc.ca/platform/bioinfo/software/abyss, accessed on 28 May 2018) with odd value k-mers ranging from 19 to 101. Assemblies were assessed for completeness using BUSCO 5.4.4 with the sordariomycetes_odb10 lineage dataset (https://busco.ezlab.org, accessed on 1 June 2018) (Table 2). Prediction of gene sets was performed with AUGUSTUS 3.3.1 (https://github.com/Gaius-Augustus/Augustus, accessed on 11 June 2018) using Fusarium graminearum as the gene model. Standalone BLASTN was performed for reciprocal comparisons of the genomes, using the BLAST v2.6.0+ suite (https://ftp.ncbi.nlm.nih.gov/blast/executables/blast+/LATEST, accessed on 15 June 2018), with an e-value cut-off of 1 × 10−3.

2.4. Secretome Analysis

To predict secreted proteins with a signal peptide (classical secretion), SignalP 4.1F (https://services.healthtech.dtu.dk/services/SignalP-4.1, accessed on 20 August 2018) with the default cut-off was used. To predict non-classical secreted proteins, SecretomeP 1.0H (https://services.healthtech.dtu.dk/services/SecretomeP-1.0, accessed on 20 August 2018) with a cut-off NN-score ≥ 0.50 was used. SSPs were defined as being predicted to be secreted and a sequence length ≤ 300 aa without a significant match to any longer protein through reciprocal BLASTP. SSPs were divided into either SSNPs (cysteine < 4) or SSCPs (cysteine ≥ 4). The number of cysteine residues was determined using Pepstats from EMBOSS 6.6.0 package (https://emboss.sourceforge.net/download, accessed on 27 August 2018). Identification of secreted CAZymes was performed by domain searches using the HMMer profile (release 7.0) from dbCAN2 (http://bcb.unl.edu/dbCAN2, accessed on 7 September 2018). Prediction of secreted peptidase and peptidase inhibitor proteins was performed with BLASTP against the MEROPS protein database release 12.0 (https://www.ebi.ac.uk/merops, accessed on 12 September 2018). Prediction of secreted lipase proteins was performed with HMMER profiles obtained using the Lipase Engineering Database release 3.0 (LED; https://led.biocatnet.de/, accessed on 15 October 2018). Gene number was determined when there was a difference of 3 or more in the number of a particular gene between the genomes of any of the I. mors-panacis and I. robusta isolates.
Visual depiction of the relatedness of predicted proteins of glycoside hydrolase families 18 (GH18) and 78 (GH78) was performed by aligning aa sequences using MUSCLE 3.8.31 (https://www.drive5.com/muscle/downloads_v3.htm, accessed on 1 November 2018) and generating an unrooted maximum likelihood dendrogram through RAxML (https://github.com/stamatak/standard-RAxML, accessed on 1 November 2018) and viewed in MEGA6 (https://megasoftware.net, accessed on 1 November 2018).

3. Results

3.1. Identification and Genome Characteristics of the Ilyonectria Isolates

The assembled genomes of I. robusta isolates with >98% completeness from BUSCO ranged from 56.1 Mbp to 63.9 Mbp with GC content ranging from 50.0% to 50.3% and the number of predicted genes ranging from 16,697 to 19,070 (Table 1). The number of predicted genes per Mbp ranged from 296.0 to 298.4. For type 1 I. mors-panacis isolates, genome sizes ranged from 64.8 Mb to 65.3 Mbp with GC content ranging from 48.93% to 48.98% and the number of predicted genes ranging from 18,149 to 18,263, with the number of predicted proteins encoded per Mbp ranging from 279.2 to 280.0. For type 2 I. mors-panacis isolates, genome sizes ranged from 64.9 Mb to 65.2 Mbp with GC content ranging from 48.97% to 49.02% and the number of predicted genes ranged from 18,173 to 18,451. The number of predicted proteins encoded per Mbp ranged from 281.4 to 283.7. Thus, both type 1 and 2 I. mors-panacis isolates had much less variability in genome size and number of genes in their genomes than the I. robusta isolates, with type 2 isolates of I. mors-panacis having slightly more predicted genes than the type 1 I. mors-panacis. All the I. robusta isolates examined had smaller genome sizes with higher GC contents than the I. mors-panacis isolates and a higher density of predicted genes. The density of predicted genes (genes/Mbp) was also slightly higher with all the type 2 than the type 1 isolates of I. mors-panacis.
For the I. robusta isolates, the total number of predicted genes encoding non-secreted proteins ranged from 8481 to 9440, while the total number of predicted genes encoding secreted proteins ranged from 8217 to 9631 (Table 2). For the type 1 I. mors-panacis isolates, it was much less variable, ranging from 8916 to 8977 for non-secreted protein genes, and 9208 to 9239 for secreted protein genes. For the type 2 I. mors-panacis isolates, it was more variable than type 1 isolates, ranging from 8842 to 9969 for non-secreted protein genes, and 8441 to 9610 for secreted protein genes. The secretome was divided into SSNPs, SSCPs, CAZymes, proteases, lipases, and other secreted proteins (Table 2). Overall, the variation in these categories among the I. robusta, type 1 I. mors-panacis, and type 2 I. mors-panacis isolates reflected the variation in the total secretome gene number.

3.2. Growth Rate and Relationship to Gene Number

There was a continuous distribution of growth rates of the isolates on PDA ranging from 0.25 to 0.63 cm/day (Figure 1). Among the four I. robusta isolates, three were the fastest growing and the fourth was the fifth-fastest growing. Thus, only the type 2 I. mors-panacis isolate, IMP.K112, and type 1 I. mors-panacis isolate, IMP.ND3P14-1, had growth rates not significantly different than the four I. robusta isolates. Growth rate per isolate was compared to the number of the genes for the annotated categories of the secretome. For total SSNPs and total SSCPs, there was a significant negative correlation to the growth rate of the isolates on PDA (Figure 2). However, the relationship between growth rate and SSNP or SSCP gene number was related to the I. robusta isolates having fewer genes for SSNPs and SSCPs and having faster growth rates. There were no significant correlations to growth rates when the I. mors-panacis and I. robusta isolates were analyzed separately.
The number of the genes for secreted CAZymes in each genome was not significantly related to the growth rate of isolates on PDA (Figure 3A). However, among the carbohydrate-binding modules and the five classes of secreted CAZymes (glycoside hydrolase, glycosyl transferase, polysaccharide lyase, carbohydrate esterase, and auxiliary activity), there was a significant positive correlation only for secreted glycosyl transferases and polysaccharide lyases (Figure 3B,C). However, no significant correlations to growth rates were observed when the I. mors-panacis and I. robusta isolates were analyzed separately. Among the families within the CAZyme classes, there was a positive significant correlation for secreted glycoside hydrolase (GH) families 2, 3, 16, 18, 64, and 109, and carbohydrate-binding modules (CBMs) 50 and 67 with the strongest correlations for GH18 and CBM50 (Figure S1). There were only significant correlations to growth rates for I. mors-panacis isolates for secreted GH3 (Figure S1B) when the isolates were analyzed separately for each CAZyme class.
The number of total secreted protease genes per genome was not significantly related to the growth rate of the isolates on PDA (Figure 4A). Among the nine classifications (six protease catalytic types (aspartic acid, cysteine, metallo, serine, threonine, and glutamic) and three other classifications (mixed, unknown, and protease inhibitor), only cysteine protease and protease inhibitor showed a significant relationship between gene number and growth rate, which were negative and positive relationships, respectively (Figure 4B,C). No significant correlations between gene number and growth rates were found when the I. mors-panacis and I. robusta isolates were analyzed separately. Among the protease families within each classification, there was a significant negative correlation between growth rate and the number of aspartic acid protease genes in family 2 or serine proteases in family 33, and a significant positive correlation to serine protease family 8 (Figure 5). However, there were no significant correlations to growth rates when the gene numbers of those protease families were analyzed separately for the I. mors-panacis and I. robusta isolates.
The number of total secreted lipases in each genome was significantly negatively correlated to the growth rate of the isolates on PDA (Figure S2). Among the 38 lipase superfamilies (carboxyesterase, Yarrowia lipolytica lipase-like, Candida rugosa lipase-like, Moraxella lipase-2-like, Moraxella lipase-3-like, cytosolic hydrolase, microsomal hydrolase, carboxylesterase, hydroxynitrile lyase, bacterial esterase, gastric lipase, Burkholderia lipase, Streptomyces lipase, Bacillus lipase, thioesterase, lipoprotein lipase, bacterial esterase, lysophospholipase, filamentous fungi lipase, Pseudomonas lipase, Moraxella lipase-1-like, deacetylase, dipeptidyl peptidase-IV-like, prolyl endopeptidase, cocaine esterase, dienlactone hydrolase, xylanase esterase, antigen 85, lysosomal protective protein-like, cutinase, and Candida antartica lipase-A-like), a significant positive correlation between gene number and growth rate was observed for nine of the superfamilies. These superfamilies were Yarrowia lipolytica lipase-like, Candida rugosa lipase-like, carboxylesterase, Streptomyces lipase, Bacillus lipase, Moraxella lipase-1-like, dienlactone hydrolase, xylanase esterase, and antigen (Figure S2). In contrast, there was a significant negative correlation between gene number and growth rate for genes for ten of the superfamilies. Those superfamilies were Moraxella lipase-2-like, cytosolic hydrolase, hydroxynitrile lyase, bacterial esterase, bacterial esterase, lysophospholipase, filamentous fungi lipase, dipeptidyl peptidase-IV-like, cutinase, and Candida antartica lipase-A-like (Figure S2). There were no significant correlations to growth rates when the gene numbers of those lipase superfamilies were analyzed separately for the I. mors-panacis and I. robusta isolates. For lipase families within the lipase superfamilies, there was a significant positive correlation of growth rate to gene number for acetylcholinesterase, Bacillus sphaericus lipase-like, dienlactone hydrolase abH31.01, and dienlactone hydrolase abH31.02, with notably strong correlations for Bacillus sphaericus lipase-like and dienlactone hydrolase abH31.02 (Figure S3). There was a significant negative correlation to growth rate for Moraxella lipase-2-like abH4.01 and abH4.02, and soluble mammalian epoxide hydrolase with the strongest correlation for Moraxella lipase-2-like abH4.02 (Figure S3). However, no significant correlations were observed between growth rate and those lipase families analyzed separately for the I. mors-panacis and I. robusta isolates.

3.3. Virulence and Relationship to Gene Number

There was a continuous range of lesion sizes produced by the isolates from 0.09 to 0.30 cm2 by 12 dpi (Figure 6). The I. robusta isolates were distributed among the isolates with mid-virulence, whereas the five most virulent isolates and three least virulent isolates were all I. mors-panacis. Virulence as measured by lesion size on roots was compared to the number of genes for the SSNPs, SSCPs, secreted CAZymes, secreted proteases, and secreted lipases. The SSNP or SSCP gene number per genome was not significantly correlated with lesion sizes produced by the isolates on roots of P. quinquefolius, whether analyzed for all isolates or else I. mors-panacis and I. robusta isolates separately by species (Figure S4). The number of total secreted CAZyme genes per genome was not significantly correlated to lesion size analyzed for all isolates, I. mors-panacis isolates only, or I. robusta isolates only (Figure 7A). Among the six CAZyme families, the gene number of members in the glycoside hydrolase family had a significant positive correlation with lesion size analyzed for all isolates and I. mors-panacis isolates only (Figure 7B). For members within the 14 CAZyme families, a significant positive correlation between gene number and lesion size was observed for GH18 and GH78 when analyzed for all isolates and I. mors-panacis isolates only (Figure 8). The number of total secreted protease genes per genome was not significantly correlated to lesion size (Figure S5), nor was there a significant correlation to lesion size with the number of secreted protease genes per genome divided by protease classification and family. Similarly, the number of genes for total secreted lipase per genome was not significantly correlated to lesion size (Figure S6), nor were there any significant correlations when lipases were divided by superfamily or family. The lack of a significant correlation was observed for total secreted protease and total secreted lipase genes whether analyzed for all isolates or else I. mors-panacis and I. robusta isolates separately (Figures S5 and S6).

4. Discussion

The growth rate of fungi in culture can be affected by a wide range of factors, such as the most quickly exhausted nutrient [22], and the trade-off between growth and the metabolic costs of production of enzymes for the degradation of recalcitrant carbon forms [23]. In this study, there was a continuous range of growth rates on PDA among the isolates tested, indicating that growth rate was not controlled by a single gene or a small number of genes. Although isolates of both species varied in growth rate, the growth rates of the I. robusta isolates were mostly higher than those of the I. mors-panacis isolates. Significant negative and positive correlations between growth rate on PDA and gene number of the isolates were observed for a wide range of genes for secreted proteins, including some for SSNPs, SSCPs, CAZymes, proteases, and lipases. However, the significant correlations with gene number can generally be explained by the I. robusta isolates having the highest or lowest number per genome, along with a faster growth rate compared to the I. mors-panacis isolates. This is supported by the observation that a significant correlation between growth rate and gene number was only observed for GH3 genes when analysis was performed independently for the two Ilyonectria species. Thus, difference in gene number may be related to growth rate, primarily at a species level.
For total small secreted proteins, significant negative correlations between growth rate on PDA and gene number was found for both SSNPs and SSCPs. Fewer copies of SSNPs and SSCPs could affect growth rate by altering access to nutrients in PDA. For example, fungal SSPs can help to depolymerize cellulose and induce degradative enzymes [24,25]. A lower number of genes for SSNPs and SSCPs in the genomes could make it less likely that complex polymer-degrading enzyme production is triggered, reducing the metabolic costs of producing enzymes not needed for growth on PDA. However, that may not be an advantage when an isolate encounters plant tissues, such as live or dead ginseng roots containing complex carbohydrate polymers.
While total CAZyme gene number was not related to the growth rate of the Ilyonectria isolates on PDA, the gene number for two families of CAZymes, glycosyl transferases and polysaccharide lyases, were both positively correlated with growth rate, as were the gene numbers for eight families of CAZymes. The highest positive correlations were with the gene numbers for polysaccharide lyases, which include enzymes such as alginate lyases, gellan lyases, and pectin lyases [26]. One possibility is that a higher number of some of these genes could result in higher enzymatic activity, allowing for greater degradation of certain relatively simple sugar polymers, such as pectin, that may be extracted from the boiled potatoes in the production of PDA [27].
The gene number for total secreted proteases did not correlate with fungal growth on PDA, but the number for two classifications, cysteine protease and protease inhibitor, were negatively and positively correlated with the growth rates, respectively. The highest correlation was the positive correlation with the gene number for protease inhibitor. Speculatively, better growth on PDA could be related to protease inhibitors regulating the activity of secreted proteases to more effectively degrade proteins in PDA [28]. At the protease family level, the highest correlation was with the S8 family of serine proteases (subtilases), which were positively related to fungal growth rate. This could be due to the role of subtilases in fungal nutrient acquisition, where they are the main broad-spectrum proteases for digesting proteins to obtain amino acids for fungal growth, and the number of subtilase genes can vary considerably among fungi due to extensive gene duplications and losses [29].
The gene number for total secreted lipases was negatively correlated with fungal growth rate, but there were both positive and negative correlations with growth rate and many of the superfamilies and families of lipases. One of the highest positive correlations for a lipase family was with dienlactone hydrolases. Dienlactone hydrolases can catalyze catechol degradation [30]. Catechol is present in potatoes [31], and thus it could be in the potato extract in PDA. Since catechol is an antimicrobial compound [32], dienlactone hydrolases may be degrading catechol or related compounds in PDA to eliminate inhibitory effects on fungal growth. The strongest negative correlation for a lipase family was with soluble mammalian epoxide hydrolase, which is also involved in detoxification, but has other roles, such as the turnover of signaling lipids [33]. The wide range of lipase superfamilies and families correlating with gene number could be related to the faster growth rate due to a wide variety of factors, including reducing negative effects of toxic compounds, greater access to lipids in PDA, or altered regulation of lipid signaling.
Virulence, based on lesion sizes of detached roots, showed a continuous range of variation among the I. robusta and I. mors-panacis isolates examined. While lesion sizes on detached ginseng roots can provide an accurate means of accessing virulence of Ilyonectria isolates, it does not include all factors, such as chemotaxis for fungal growth to the root and penetration, which normally occurs via wounds [3]. Unlike growth rate in culture, neither species dominated among the high- or low-virulent isolates. Most of the I. robusta isolates were in the mid-range of virulence. Therefore, comparisons with gene number should be less impacted by species-level differences. This is supported by the observation that all the significant correlations between gene number and virulence observed when all isolates were analyzed was also observed when only I. mors-panacis isolates were analyzed. Gene numbers for total SSNPs, SSCPs, secreted CAZymes, secreted proteases, and secreted lipases were not significantly related to the lesion size produced by the isolates, nor were the number of secreted CAZyme genes when examined by classes and family, secreted protease genes when examined by classification and family, or secreted lipase genes when examined at the superfamily and family level. The only significant relationship between gene number and lesion size was found for secreted CAZymes in the glycoside hydrolase family, and then within that family, for GH18 and GH78. For those families, the positive relationship between gene number and virulence was observed across isolates of both Ilyonectria species examined. For example, the least virulent isolate was IMP.RD3U14-8, an I. mors-panacis type 1 isolate, which had 21 genes for GH18 and 6 genes for GH78, and the most virulent isolate was IMP.ND3A16-1, also an I. mors-panacis type 1 isolate, which had 28 genes for GH18 and 14 genes for GH78.
Glycoside hydrolase family 18 members are catalytically active as chitinases and endo-β-N-acetylglucosaminidases as well as non-catalytically active as carbohydrate-binding-like lectins and xylanase inhibitors [34]. A possible role for them during the infection of ginseng roots could be as the chitinases and endo-β-N-acetylglucosaminidases catalyzing the degradation of glycoproteins with N-glycans, which act as signaling molecules in plants [35]. Chitinases of the powdery mildew fungus, Podosphaera xanthii, are released during host penetration to hydrolyse immunogenic chitin oligomers, which prevents the activation of chitin-triggered host immunity [36]. Fungal carbohydrate-binding lectins have been found in a number of plant pathogenic fungi and may have a role in virulence [37]. For example, a carbohydrate-binding lectin of Magnaporthe oryzae was found to be highly expressed during penetration of rice leaves and was presumably involved in virulence [38]. It also had similar carbohydrate-binding as the Ecp6 lectin of Cladosporium fulvum that prevents chitin-triggered host immunity in tomato [39]. As there is evidence for PAMP-triggered immunity during the infection of ginseng roots by Ilyonectria species [3], a higher number of GH18 genes might increase virulence by enhancing the ability of the fungus to produce proteins that suppress the host immune system during infection. For those GH18 members with xylanase inhibitor activity, xylanase inhibitors can regulate the activity of fungal xylanases [40], and xylanases can be important virulence factors of plant pathogenic fungi [41].
Glycoside hydrolase family 78 members are α-L-rhamnosidases that cleave L-rhamnose from flavonoid glycosides, such as the antifungal compounds naringenin, rutin, and hesperidin [42]. An α-L-rhamnosidase of the plant pathogen Stagonospora avenae is a saponin hydrolysing enzyme [43], and α-L-rhamnosidase can hydrolyse ginsenosides [44,45]. Ginsenosides are triggered by infection of ginseng roots by Ilyonectria species, and disease resistance of ginseng roots can be increased by triggering ginsenosides with silicon, which supports a role for ginsenosides in host defense [3]. Thus, a higher number of genes for α-L-rhamnosidases could lead to more of the enzyme to degrade more of the antimicrobial compounds during root infection, such as ginsenosides.
In conclusion, the saprophytic growth rate on PDA was related to the number of several classes of genes for secreted proteins of the I. robusta and I. mors-panacis isolates tested, possibly by enhancing access to nutrients, detoxifying compounds, or regulating other elements of saprophytic growth. The differences in gene number were mostly related to species differences in growth rate as the I. robusta isolates were generally faster-growing on PDA than the I. mors-panacis isolates. In contrast, differences in gene number and virulence to ginseng roots were not clearly related to Ilyonectria species, and indicated that members of GH18 and GH78, such as endo-β-N-acetylglucosaminidases, lectins, xylanase inhibitors, and α-L-rhamnosidases, may be potential virulence factors of both I. robusta and I. mors-panacis during the infection of ginseng roots. The benefit to virulence of having more genes of GH18 and GH78 may be primarily related to producing more proteins more quickly when needed to suppress host resistance and combat host defense compounds. This study has shown that an examination of the number of genes can help identify candidate genes that may be important in the biology of I. robusta and I. mors-panacis. Confirmation of the roles of these genes in growth and virulence is needed, such as by CRISPR-Cas technology to mutate individual genes, which has been applied to examine genes for virulence factors of other plant pathogenic fungi [46].

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/horticulturae12020135/s1, Figure S1: Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of CAZyme families with 3 or more copy number difference and a significant correlation with growth rate. The families are GH2 (glycoside hydrolase 2) (A), GH3 (glycoside hydrolase 3) (B), GH16 (glycoside hydrolase 16) (C), GH18 (glycoside hydrolase 18) (D), GH64 (glycoside hydrolase 64) (E), GH109 (glycoside hydrolase 109) (F), CBM50 (carbohydrate-binding modules 50) (G), and CBM67 (carbohydrate-binding modules 67) (H). Orange crosses, blue triangles, and blue circles indicate I. robusta, and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 30.5x2 − 19.8x + 12.1 (A), y = 140.2x2 − 108x + 59.4 (B), y = −72.7x2 + 46.7x + 29.5 (C), y = 54.9x2 − 68.6x + 39.8 (D), y = −29.8x2 + 18.2x + 3.6 (E), y = 85.6x2 − 59.1x + 37.7 (F), y = −55.8x2 + 18.6x + 23.9 (G), and y = 156.3x2 − 114.4x + 28.1 (H). Figure S2: Relationship between growth rate on PDA and gene copy number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total lipases (A) and lipase superfamilies with 3 or more copy number difference and a significant correlation with growth rate. The superfamilies are abH01 (Carboxylesterases) (A), abH02 (Yarrowia lipolytica lipase-like) (B), abH03 (Candida rugosa lipase-like) (C), abH04 (Moraxella lipase 2 like) (D), abH08 (cytosolic hydrolases) (E), abH11 (carboxylesterases) (F), abH12 (hydroxynitrile lyases) (G), abH13 (bacterial esterases) (H), abH16 (Streptomyces lipases) (I), abH18 (Bacillus lipases) (J), abH21 (bacterial esterases) (K), abH22 (lysophospholipase) (L), abH23 (filamentous fungi lipases) (M), abH25 (Moraxella lipase 1 like) (N), abH27 (dipeptidyl peptidase IV like) (O), abH31 (dienlactone hydrolases) (P), abH32 (xylanase esterases) (Q), abH3 (antigen 85) (R), abH34 (lysosomal protective protein like) (S), abH36 (cutinases) (T), and abH38 (Candida antartica lipase A like) (U). Orange crosses, blue triangles, and blue circles indicate I. robusta, and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −3480.8x2 + 2340.7x + 291.4 (A), y = 17.8x2 − 11.6x + 3.4 (B), y = 103.1x2 − 71.6x + 49 (C), y = −91.5x2 + 40.6x + 179.4 (D), y = −76.6x2 + 53x + 101 (E), y = 41.1x2 − 25.9x + 7 (F), y = −16.2x2 + 10.7x + 7.3 (G), y = −13.9x2 + 9.3x + 3.5 (H), y = 8x2 − 5x + 0.8 (I), y = 13.9x2 − 9.3x + 1.5 (J), y = −12.5x2 + 6.2x + 7.2 (K), y = −83.6x2 + 56x + 5.1 (L), y = −13.9x2 + 9.3x + 7.5 (M), y = 13.9x2 − 9.3x + 1.5 (N), y = −27.9x2 + 18.7x + 6 (O), y = 246x2 − 168.5x + 52.4 (P), y = 16.8x2 − 10.7x + 13.6 (Q), y = 16.2x2 − 11.7x + 3.1 (R), y = 6.5x2 − 4.2x + 15.5 (S), y = −20.6x2 + 11.3x + 19.4 (T), and y = −27.9x2 + 18.7x + 1 (U). Figure S3: Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of lipase families with 3 or more copy number difference and a significant correlation with growth rate. The families are abH1.04 (acetlycholinesterases) (A), abH4.01 (Moraxella lipase 2 like) (B), abH4.02 (Moraxella lipase 2 like) (C), abH4.04 (Bacillus sphaericus lipase-like) (D), abH8.03 (soluble mammalian epoxide hydrolases) (E), abH31.01 (dienlactone hydrolase) (F), and abH31.02 (dienlactone hydrolase) (G). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 72.4x2 − 52.1x + 15.8 (A), y = −35.5x2 + 17.9x + 63.9 (B), y = −179x2 + 117.1x + 94.2 (C), y = 83.9x2 − 57.7x + 12.4 (D), y = −33.5x2 + 20.1x + 10.3 (E), y = 95.6x2 − 64.7x + 12.4 (F), and y = 106.6x2 − 72.7x + 24.8 (G). Figure S4: Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total small secreted non-cysteine-rich proteins (SSNPs) (A) and total predicted small secreted cysteine-rich proteins (SSCPs) (B). Orange crosses, blue triangles, and blue circles indicate I. robusta, and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 3256.2x2 − 1109x + 1430.1 (A) and y = 209x2 − 38.7x + 694.6 (B). Figure S5: Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total proteases. Orange crosses, blue triangles, and blue circles indicate I. robusta, and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 3256.2x2 − 1109x + 1430.1 (A) and y = 209x2 − 38.7x + 694.6 (B). Figure S6: Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total lipases. Orange crosses, blue triangles, and blue circles indicate I. robusta, and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equation for the regression line with all isolates was y = 8123.5x2 − 2714x + 807.7.

Author Contributions

Conceptualization, T.H. and P.H.G.; methodology, T.H. and M.V.; validation, T.H., M.V., and P.H.G.; formal analysis, T.H., M.V., and P.H.G.; investigation, T.H., M.V., and P.H.G.; resources, T.H. and P.H.G.; writing—original draft preparation, T.H. and P.H.G.; writing—review and editing, T.H., M.V., and P.H.G.; visualization, T.H., M.V., and P.H.G. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

NCBI accessions to the assembled Ilyonectria genome scaffolds, whole genome shotgun (WGS) sequences, and BioSample entries are available in NCBI BioProject accession PRJNA885578. The data that support the findings of this study are available on request from the corresponding author (T.H). (Email address: thsiang@uoguelph.ca). The data are not publicly available due to their being used in an ongoing project.

Acknowledgments

We wish to thank Behrang Behdarvandi, whose thesis involved some of the material presented here: Behdarvandi B. The Relationship of Ilyonectria to Replant Disease of American Ginseng (Panax quinquefolius). Ph.D. Thesis, University of Guelph, Guelph, ON, Canada, 2020.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Abbreviations

The following abbreviations are used in this manuscript:
ANOVAAnalysis of variance
CAZymeCarbohydrate active enzyme
CBMCarbohydrate-binding module
dpiDays post-inoculation
dsH2ODistilled sterile water
GCGuanine-cytosine
GHGlycoside hydrolase
MbpMega base pairs
PDAPotato dextrose agar
SSCPSmall secreted cysteine-rich protein
SSNPSmall secreted non-cysteine-rich protein
SSPSmall secreted protein

References

  1. Seifert, K.A.; Axelrood, P.E. Cylindrocarpon destructans var. destructans. Can. J. Plant Pathol. 1998, 20, 115–117. [Google Scholar] [CrossRef]
  2. Cabral, A.; Groenewald, J.Z.; Rego, C.; Oliveira, H.; Crous, P.W. Cylindrocarpon root rot: Multi-gene analysis reveals novel species within the Ilyonectria radicicola species complex. Mycol. Prog. 2012, 11, 655–688. [Google Scholar] [CrossRef]
  3. Bischoff Nunes, I.; Goodwin, P.H. Interaction of ginseng with Ilyonectria root rot pathogens. Plants 2022, 11, 2152. [Google Scholar] [CrossRef]
  4. Matuo, T.; Miyazawa, Y. Scientific name of Cylindrocarpon sp. causing root rot of ginseng. Ann. Phytopathol. Soc. Jpn. 1984, 50, 649–652. [Google Scholar] [CrossRef]
  5. Behdarvandi, B.; Hsiang, T.; Valliani, M.; Goodwin, P.H. Differences in saprophytic growth, virulence, genomes, and secretomes of Ilyonectria robusta and I. mors-panacis isolates from roots of American ginseng (Panax quinquefolius). Horticulturae 2023, 9, 713. [Google Scholar] [CrossRef]
  6. Rafiqi, M.; Ellis, J.G.; Ludowici, V.A.; Hardham, A.R.; Dodds, P.N. Challenges and progress towards understanding the role of effectors in plant–fungal interactions. Curr. Opin. Plant Biol. 2012, 15, 477–482. [Google Scholar] [CrossRef]
  7. Walsh, J.P.; McMullin, D.R.; Yeung, K.K.C.; Sumarah, M.W. Resorcylic acid lactones from the ginseng pathogen Ilyonectria mors-panacis. Phytochem. Lett. 2022, 48, 94–99. [Google Scholar] [CrossRef]
  8. Rahman, M.; Punja, Z.K. Factors influencing development of root rot on ginseng caused by Cylindrocarpon destructans. Phytopathology 2005, 95, 1381–1390. [Google Scholar] [CrossRef]
  9. Cheng, L.; Wang, P.; Yang, S.; Yang, Y.; Zhang, Q.; Zhang, W.; Xiao, H.; Gao, H.; Zhang, Q. Identification of genes with a correlation between copy number and expression in gastric cancer. BMC Med. Genom. 2012, 5, 14. [Google Scholar] [CrossRef] [PubMed]
  10. Steenwyk, J.L.; Rokas, A. Copy number variation in fungi and its implications for wine yeast genetic diversity and adaptation. Front. Microbiol. 2018, 9, 288. [Google Scholar] [CrossRef] [PubMed]
  11. Steenwyk, J.; Rokas, A. Extensive copy number variation in fermentation-related genes among Saccharomyces cerevisiae wine strains. G3 2017, 7, 1475–1485. [Google Scholar] [CrossRef]
  12. Turcotte, B.; Liang, X.B.; Robert, F.; Soontorngun, N. Transcriptional regulation of nonfermentable carbon utilization in budding yeast. FEMS Yeast Res. 2010, 10, 2–13. [Google Scholar] [CrossRef] [PubMed]
  13. Zhao, S.; Gibbons, J.G. A population genomic characterization of copy number variation in the opportunistic fungal pathogen Aspergillus fumigatus. PLoS ONE 2018, 13, e0201611. [Google Scholar] [CrossRef] [PubMed]
  14. Stalder, L.; Oggenfuss, U.; Mohd-Assaad, N.; Croll, D. The population genetics of adaptation through copy number variation in a fungal plant pathogen. Mol. Ecol. 2023, 32, 2443–2460. [Google Scholar] [CrossRef]
  15. Tralamazza, S.M.; Gluck-Thaler, E.; Feurtey, A.; Croll, D. Copy number variation introduced by a massive mobile element facilitates global thermal adaptation in a fungal wheat pathogen. Nat. Commun. 2024, 15, 5728. [Google Scholar] [CrossRef]
  16. Liu, Z.; Zhang, Z.; Faris, J.D.; Oliver, R.P.; Syme, R.; McDonald, M.C.; McDonald, B.A.; Solomon, P.S.; Lu, S.; Shelver, W.L.; et al. The cysteine rich necrotrophic effector SnTox1 produced by Stagonospora nodorum triggers susceptibility of wheat lines harboring Snn1. PLoS Pathog. 2012, 8, e1002467. [Google Scholar] [CrossRef]
  17. Pereira, J.L.; Noronha, E.F.; Miller, R.N.G.; Franco, O.L. Novel insights in the use of hydrolytic enzymes secreted by fungi with biotechonological potential. Lett. Appl. Microbiol. 2007, 44, 573–581. [Google Scholar] [CrossRef]
  18. Olivieri, F.; Zanetti, M.E.; Oliva, R.C.; Covarrubias, A.A.; Claudia, A. Casalongue characterization of an extracellular serine protease of Fusarium eumartii and its action on pathogenesis related proteins. Eur. J. Plant Pathol. 2002, 108, 63–72. [Google Scholar] [CrossRef]
  19. Kikot, G.E.; Hours, P.A.; Alconada, T.M. Contribution of cell wall degrading enzymes to pathogenesis of Fusarium graminearum: A review. J. Basic Microbiol. 2009, 49, 231–241. [Google Scholar] [CrossRef]
  20. Schneider, C.A.; Rasband, W.S.; Eliceiri, K.W. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012, 9, 671–675. [Google Scholar] [CrossRef]
  21. Edwards, K.; Johnstone, C.; Thompson, C. A simple and rapid method for the preparation of plant genomic DNA for PCR analysis. Nucleic Acids Res. 1991, 19, 1349. [Google Scholar] [CrossRef] [PubMed]
  22. Vrabl, P.; Schinagl, C.W.; Artmann, D.J.; Heiss, B.; Burgstaller, W. Fungal growth in batch culture–what we could benefit if we start looking closer. Front. Microbiol. 2019, 10, 2391. [Google Scholar] [CrossRef] [PubMed]
  23. Zheng, W.; Lehmann, A.; Ryo, M.; Vályi, K.K.; Rillig, M.C. Growth rate trades off with enzymatic investment in soil filamentous fungi. Sci. Rep. 2020, 10, 11013. [Google Scholar] [CrossRef]
  24. Saloheimo, M.; Paloheimo, M.; Hakola, S.; Pere, J.; Swanson, B.; Nyyssönen, E.; Bhatia, A.; Ward, M.; Penttilä, M. Swollenin, a Trichoderma reesei protein with sequence similarity to the plant expansins, exhibits disruption activity on cellulosic materials. Eur. J. Biochem. 2002, 269, 4202–4211. [Google Scholar] [CrossRef]
  25. Feldman, D.; Kowbel, D.J.; Glass, N.L.; Yarden, O.; Hadar, Y. A role for small secreted proteins (SSPs) in a saprophytic fungal lifestyle: Ligninolytic enzyme regulation in Pleurotus ostreatus. Sci. Rep. 2017, 7, 14553. [Google Scholar] [CrossRef]
  26. Sutherland, J.R.; Sturrock, R.N.; Shrimpton, G.M. Diseases and Insects in British Columbia Forest Seedling Nurseries; FRDA Report 065; Forestry Canada: Victoria, BC, Canada, 1989. [Google Scholar]
  27. Ramasamy, U.R. Water Holding Capacity and Enzymatic Modification of Pressed Potato Fibres. Ph.D. Thesis, Wageningen University, Wageningen, The Netherlands, 2014. [Google Scholar]
  28. Sabotič, J.; Kos, J. Microbial and fungal protease inhibitors–current and potential applications. Appl. Microbiol. Biotechnol. 2012, 93, 1351–1375. [Google Scholar] [CrossRef]
  29. Li, J.; Gu, F.; Wu, R.; Yang, J.; Zhang, K.Q. Phylogenomic evolutionary surveys of subtilase superfamily genes in fungi. Sci. Rep. 2017, 7, 45456. [Google Scholar] [CrossRef]
  30. Bruckmann, M.; Blasco, R.; Timmis, K.N.; Pieper, D.H. Detoxification of protoanemonin by dienelactone hydrolase. J. Bacteriol. 1998, 180, 400–402. [Google Scholar] [CrossRef]
  31. Gerdemann, C.; Eicken, C.; Krebs, B. The crystal structure of catechol oxidase:  New insight into the function of type-3 copper proteins. Acc. Chem. Res. 2002, 35, 183–191. [Google Scholar] [CrossRef] [PubMed]
  32. Link, K.P.; Walker, J.C. The isolation of catechol from pigmented onion scales and its significance in relation to disease resistance in onions. J. Biol. Chem. 1933, 100, 379–383. [Google Scholar] [CrossRef]
  33. Decker, M.; Arand, M.; Cronin, A. Mammalian epoxide hydrolases in xenobiotic metabolism and signalling. Arch. Toxicol. 2009, 83, 297–318. [Google Scholar] [CrossRef]
  34. Funkhouser, J.D.; Aronson, N.N. Chitinase family GH18: Evolutionary insights from the genomic history of a diverse protein family. BMC Evol. Biol. 2007, 7, 96. [Google Scholar] [CrossRef] [PubMed]
  35. Kimura, Y.; Tokuda, T.; Ohno, A.; Tanaka, H.; Ishiguro, Y. Enzymatic properties of endo-β-N-acetylglucosaminidases from developing tomato fruits and soybean seeds: Substrate specificity of plant origin endoglycosidase. Biochim. Biophys. Acta 1998, 138, 27–36. [Google Scholar] [CrossRef]
  36. Martínez-Cruz, J.; Romero, D.; Hierrezuelo, J.; Thon, M.; de Vicente, A.; Pérez-García, A. Effectors with chitinase activity (EWCAs), a family of conserved, secreted fungal chitinases that suppress chitin-triggered immunity. Plant Cell 2021, 33, 1319–1340. [Google Scholar] [CrossRef]
  37. Varrot, A.; Basheer, S.M.; Imberty, A. Fungal lectins: Structure, function and potential applications. Curr. Opin. Struct. Biol. 2013, 23, 678–685. [Google Scholar] [CrossRef] [PubMed]
  38. Koharudin, L.M.; Viscomi, A.R.; Montanini, B.; Kershaw, M.J.; Talbot, N.J.; Ottonello, S.; Gronenborn, A.M. Structure-function analysis of a CVNH-LysM lectin expressed during plant infection by the rice blast fungus Magnaporthe oryzae. Structure 2011, 19, 662–674. [Google Scholar] [CrossRef]
  39. De Jonge, R.; van Esse, H.P.; Kombrink, A.; Shinya, T.; Desaki, Y.; Bours, R.; Van Der Krol, S.; Shibuya, N.; Joosten, M.H.; Thomma, B.P. Conserved fungal LysM effector Ecp6 prevents chitin-triggered immunity in plants. Science 2010, 329, 953–955. [Google Scholar] [CrossRef] [PubMed]
  40. Gusakov, A.V. Proteinaceous inhibitors of microbial xylanases. Biochemistry 2010, 75, 1185–1199. [Google Scholar] [CrossRef]
  41. Yu, C.; Li, T.; Shi, X.; Saleem, M.; Li, B.; Liang, W.; Wang, C. Deletion of endo-β-1,4-xylanase VmXyl1 impacts the virulence of Valsa mali in apple tree. Front. Plant Sci. 2018, 9, 663. [Google Scholar] [CrossRef]
  42. Oliveira, V.M.; Carraroa, E.; Aulerb, M.E.; Khalil, N.M. Quercetin and rutin as potential agents antifungal against Cryptococcus spp. Brazil J. Biol. 2016, 76, 1029–1034. [Google Scholar] [CrossRef]
  43. Hughes, H.B.; Morrissey, J.P.; Osbourn, A.E. Characterisation of the saponin hydrolysing enzyme avenacoside-α-l-rhamnosidase from the fungal pathogen of cereals, Stagonospora avenae. Eur. J. Plant Pathol. 2004, 110, 421–427. [Google Scholar] [CrossRef]
  44. Park, C.S.; Yoo, M.H.; Noh, K.H.; Oh, D.K. Biotransformation of ginsenosides by hydrolyzing the sugar moieties of ginsenosides using microbial glycosidases. Appl. Microbiol. Biotechnol. 2010, 87, 9–19. [Google Scholar] [CrossRef]
  45. Yu, H.; Gong, J.; Zhang, C.; Jin, F. Purification and characterization of ginsenoside-α-L-rhamnosidase. Chem. Pharm. Bull. 2002, 50, 175–178. [Google Scholar] [CrossRef] [PubMed]
  46. Gosavi, G.; Yan, F.; Ren, B.; Kuang, Y.; Yan, D.; Zhou, X.; Zhou, H. Applications of CRISPR technology in studying plant-pathogen interactions: Overview and perspective. Phytopathol. Res. 2020, 2, 21. [Google Scholar] [CrossRef]
Horticulturae 12 00135 g001
Figure 1. Growth rates of I. mors-panacis (IMP) type 1 (solid blue bars) and type 2 (dotted blue bars) and I. robusta (IR, orange bars) isolates on PDA over 12 days in dark at 22 °C. Letters in common over the bars indicate no significance difference at α = 0.05 as assessed using Fisher’s LSD.
Figure 1. Growth rates of I. mors-panacis (IMP) type 1 (solid blue bars) and type 2 (dotted blue bars) and I. robusta (IR, orange bars) isolates on PDA over 12 days in dark at 22 °C. Letters in common over the bars indicate no significance difference at α = 0.05 as assessed using Fisher’s LSD.
Horticulturae 12 00135 g001
Horticulturae 12 00135 g002
Figure 2. Relationship between growth rate on PDA and gene number in the genomes of I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total small secreted non-cysteine-rich proteins (SSNPs) (A), and total small secreted cysteine-rich proteins (SSCPs) (B) with 3 or more copy number difference. Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −1010.4x2 + 609.8x + 1283.2 (A) and y = −134.3x2 + 56.6x + 696.6 (B).
Figure 2. Relationship between growth rate on PDA and gene number in the genomes of I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total small secreted non-cysteine-rich proteins (SSNPs) (A), and total small secreted cysteine-rich proteins (SSCPs) (B) with 3 or more copy number difference. Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −1010.4x2 + 609.8x + 1283.2 (A) and y = −134.3x2 + 56.6x + 696.6 (B).
Horticulturae 12 00135 g002
Horticulturae 12 00135 g003
Figure 3. Relationship between growth rate on PDA and gene number in the genomes of I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total secreted CAZymes (A), and CAZyme classes with 3 or more copy number difference and a significant correlation with growth rate. The classes are glycosyl transferases (B) and polysaccharide lyases (C). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 132.4x2 − 90.8x + 681.7 (A), y = 44.5x2 − 29x + 129.4 (B), and y = 90x2 − 65.4x + 47.9 (C).
Figure 3. Relationship between growth rate on PDA and gene number in the genomes of I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total secreted CAZymes (A), and CAZyme classes with 3 or more copy number difference and a significant correlation with growth rate. The classes are glycosyl transferases (B) and polysaccharide lyases (C). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 132.4x2 − 90.8x + 681.7 (A), y = 44.5x2 − 29x + 129.4 (B), and y = 90x2 − 65.4x + 47.9 (C).
Horticulturae 12 00135 g003
Horticulturae 12 00135 g004
Figure 4. Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total proteases (A) and protease classifications with 3 or more copy number difference and a significant correlation with growth rate (cysteine protease (B) and protease inhibitor (C)). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −19.9x2 + 10.1x + 25.2 (A), y = 67.1x2 − 46.5x + 37.8 (B), and y = −119.9x2 + 80.4x + 78.3 (C).
Figure 4. Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total proteases (A) and protease classifications with 3 or more copy number difference and a significant correlation with growth rate (cysteine protease (B) and protease inhibitor (C)). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −19.9x2 + 10.1x + 25.2 (A), y = 67.1x2 − 46.5x + 37.8 (B), and y = −119.9x2 + 80.4x + 78.3 (C).
Horticulturae 12 00135 g004
Horticulturae 12 00135 g005
Figure 5. Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of protease families with 3 or more copy number difference and a significant correlation with growth rate. The families are aspartic acid 2 (A), serine 8 (B), and serine 3 (C) proteases. Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −19.9x2 + 10.1x + 25.2 (A), y = 67.1x2 − 46.5x + 37.8 (B), and y = −119.9x2 + 80.4x + 78.3 (C).
Figure 5. Relationship between growth rate on PDA and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of protease families with 3 or more copy number difference and a significant correlation with growth rate. The families are aspartic acid 2 (A), serine 8 (B), and serine 3 (C) proteases. Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = −19.9x2 + 10.1x + 25.2 (A), y = 67.1x2 − 46.5x + 37.8 (B), and y = −119.9x2 + 80.4x + 78.3 (C).
Horticulturae 12 00135 g005
Horticulturae 12 00135 g006
Figure 6. Ilyonectria mors-panacis (IMP) type 1 (solid blue bars) and type 2 (dotted blue bars) and I. robusta (IR, in orange bar) isolates lesion sizes produced on ginseng roots at 12 dpi (day post-inoculation) inoculated with 1 × 106 conidia/mL in dsH2O obtained from 4-week-old cultures grown on V8 media at 22 °C in the dark. Letters in common over the bars indicate no significance difference at α = 0.05 as assessed using Fisher’s LSD.
Figure 6. Ilyonectria mors-panacis (IMP) type 1 (solid blue bars) and type 2 (dotted blue bars) and I. robusta (IR, in orange bar) isolates lesion sizes produced on ginseng roots at 12 dpi (day post-inoculation) inoculated with 1 × 106 conidia/mL in dsH2O obtained from 4-week-old cultures grown on V8 media at 22 °C in the dark. Letters in common over the bars indicate no significance difference at α = 0.05 as assessed using Fisher’s LSD.
Horticulturae 12 00135 g006
Horticulturae 12 00135 g007
Figure 7. Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total CAZyme (A) and members of CAZyme families with 3 or more copy number difference and a significant correlation with lesion size. The CAZyme family is glycoside hydrolase (B). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 195.5x2 − 67.8x + 674.2 (A) and y = 16.3x2 + 142.2x + 362.6 (B).
Figure 7. Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of total CAZyme (A) and members of CAZyme families with 3 or more copy number difference and a significant correlation with lesion size. The CAZyme family is glycoside hydrolase (B). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 195.5x2 − 67.8x + 674.2 (A) and y = 16.3x2 + 142.2x + 362.6 (B).
Horticulturae 12 00135 g007
Horticulturae 12 00135 g008
Figure 8. Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of CAZyme families with 3 or more copy number difference and a significant correlation with lesion size. The families are GH18 (glycoside hydrolase 18) (A) and GH78 (glycoside hydrolase 78) (B). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 268x2 − 59.1x + 23.4 (A) and y = 35.7x2 + 20.7x + 4.2 (B).
Figure 8. Relationship between lesion size on P. quinquefolius roots and gene number in the genomes of the I. mors-panacis and I. robusta isolates (Table 1) for predicted proteins of CAZyme families with 3 or more copy number difference and a significant correlation with lesion size. The families are GH18 (glycoside hydrolase 18) (A) and GH78 (glycoside hydrolase 78) (B). Orange crosses, blue triangles, and blue circles indicate I. robusta and I. mors-panacis type 1 and 2 isolates, respectively, with the text color for the R2 and p-values matching the regression lines. The equations for the regression line with all isolates were y = 268x2 − 59.1x + 23.4 (A) and y = 35.7x2 + 20.7x + 4.2 (B).
Horticulturae 12 00135 g008
Table 1. Ilyonectria mors-panacis type 1 and 2, and I. robusta isolate information and genome characteristics. The isolates were from infected roots of P. quinquefolius, except IR.DAOM139398, which was isolated from Prunus cerasus cv. Montmorency. Species identification was determined from the histone H3 sequence as per Cabral et al. (2012) [2]. NCBI accessions to the assembled Ilyonectria genome scaffolds, whole genome shotgun (WGS) sequences, and BioSample entries are presented in NCBI BioProject accession PRJNA885578 [5].
Table 1. Ilyonectria mors-panacis type 1 and 2, and I. robusta isolate information and genome characteristics. The isolates were from infected roots of P. quinquefolius, except IR.DAOM139398, which was isolated from Prunus cerasus cv. Montmorency. Species identification was determined from the histone H3 sequence as per Cabral et al. (2012) [2]. NCBI accessions to the assembled Ilyonectria genome scaffolds, whole genome shotgun (WGS) sequences, and BioSample entries are presented in NCBI BioProject accession PRJNA885578 [5].
Species/TypeIsolateLocationGenome Size (Mbp)GC ContentPredicted GenesGenes/MbpAssemblerAssembly k-mer
I. robustaIR.NR1BC16-1 aSummerland, BC58.350.05%17,256296.0SOAPdenovo81
I. robustaIR.ND1BC16-2 aSummerland, BC63.950.27%19,070298.4SOAPdenovo85
I. robustaIR.NR2BC16-4 aSummerland, BC58.650.04%17,391296.8ABySS93
I. robustaIR.DAOM139398 bGeorgian Bay, ON56.150.08%16,697297.6SOAPdenovo85
I. mors-panacis type 1IMP.ND3P14-1A cSt. Williams, ON65.348.93%18,248279.4SOAPdenovo85
I. mors-panacis type 2IMP.ND3P14-3 cLynedoch, ON65.048.97%18,288281.4ABySS11
I. mors-panacis type 1IMP.ND3A16-1 cDelhi, ON64.948.94%18,149279.6SOAPdenovo85
I. mors-panacis type 2IMP.ND4Z15 cSimcoe, ON65.148.99%18,440283.3ABySS67
I. mors-panacis type 1IMP.ND3A16-2 cDelhi, ON65.048.93%18,160279.4SOAPdenovo81
I. mors-panacis type 1IMP.ND3P14-1 cLynedoch, ON65.348.93%18,251279.5SOAPdenovo85
I. mors-panacis type 2IMP.K112 dKamloops, BC65.048.97%18,263281.0ABySS11
I. mors-panacis type 1IMP.RD3U14-8 cScotland, ON64.948.95%18173280.0SOAPdenovo81
I. mors-panacis type 2IMP.RR3A14-1 cDelhi, ON64.949.02%18,409283.7ABySS71
I. mors-panacis type 2IMP.RD3U14-5 cScotland, ON65.248.99%18,451283.0ABySS75
I. mors-panacis type 1IMP.DAOM226727 eDelhi, ON64.848.98%18,153280.1SOAPdenovo85
I. mors-panacis type 1IMP.DAOM226729 eDelhi, ON65.048.94%18,150279.2SOAPdenovo85
a Jesse Macdonald, Agriculture and Agri-Food Canada; b S.J. Hughes, Agriculture and Agri-Food Canada; c Amy Fang Shi, Ontario Ginseng Growers Association; d Zamir Punja, Simon Fraser University, and e Richard D. Reeleder, Agriculture and Agri-Food Canada.
Table 2. Number of total non-secreted and secreted predicted proteins and select categories of predicted secreted proteins of I. mors-panacis (IMP) and I. robusta (IR) isolates (Table 1). Non-secreted and secreted proteins are mutually exclusive, and the secreted proteins includes both classically secreted (signal peptide detected) and non-classical secreted proteins predicted with a Neural Network score ≥ 0.5.
Table 2. Number of total non-secreted and secreted predicted proteins and select categories of predicted secreted proteins of I. mors-panacis (IMP) and I. robusta (IR) isolates (Table 1). Non-secreted and secreted proteins are mutually exclusive, and the secreted proteins includes both classically secreted (signal peptide detected) and non-classical secreted proteins predicted with a Neural Network score ≥ 0.5.
IsolateTotal Non-Secreted ProteinsTotal Secreted ProteinsSecreted SSNPsSecreted SSCPsSecreted CAZymesSecreted ProteasesSecreted
Lipases		Other Secreted Proteins
IR.NR1BC16-18716854112646746635734123809
IR.NR1BC16-29440963113247026976284284453
IR.NR2BC16-48838855412636666715764213778
IR.DAOM1393988481821712236666595623952314
IMP.ND3P14-1A8973927613716976705954094359
IMP.ND3P14-39012927713726956696014114339
IMP.ND3A16-18933921713716996675974084306
IMP.ND4Z159060938113817146726054084382
IMP.ND3A16-28922923913646986705954084229
IMP.ND3P14-18977927513777006715994094352
IMP.RD3U14-88948922613676956695894082536
IMP.RR3A14-19969844113727026195503752952
IMP.RD3U14-58842961013827066996224443355
IMP.K1129020924413557066695994114306
IMP.DAOM2267278916923813687026675974132551
IMP.DAOM2267298943920813736956685964082515
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Goodwin, P.H.; Valliani, M.; Hsiang, T. Variation in the Number of Genes in the Secretomes of Isolates of Ilyonectria robusta and Ilyonectria mors-panacis Pathogenic to American Ginseng (Panax quinquefolius). Horticulturae 2026, 12, 135. https://doi.org/10.3390/horticulturae12020135

AMA Style

Goodwin PH, Valliani M, Hsiang T. Variation in the Number of Genes in the Secretomes of Isolates of Ilyonectria robusta and Ilyonectria mors-panacis Pathogenic to American Ginseng (Panax quinquefolius). Horticulturae. 2026; 12(2):135. https://doi.org/10.3390/horticulturae12020135

Chicago/Turabian Style

Goodwin, Paul H., Moez Valliani, and Tom Hsiang. 2026. "Variation in the Number of Genes in the Secretomes of Isolates of Ilyonectria robusta and Ilyonectria mors-panacis Pathogenic to American Ginseng (Panax quinquefolius)" Horticulturae 12, no. 2: 135. https://doi.org/10.3390/horticulturae12020135

APA Style

Goodwin, P. H., Valliani, M., & Hsiang, T. (2026). Variation in the Number of Genes in the Secretomes of Isolates of Ilyonectria robusta and Ilyonectria mors-panacis Pathogenic to American Ginseng (Panax quinquefolius). Horticulturae, 12(2), 135. https://doi.org/10.3390/horticulturae12020135

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop