PCR-based identification of pathogenic
Fusarium proliferatum from asparagus
Bachelorproject BiologyJoep Koomen
Abstract
Fusarium proliferatum isolates can be pathogenic on many crops including A. officinalis (asparagus), and this infection can cause a serious loss of crop yield. To reduce such yield loss, one of the solutions is developing an early detection method for pathogenic F. proliferatum isolates from asparagus seeds. In this way, after detection, the seeds can still be treated. F. proliferatum produces several mycotoxins, which are also toxic to humans. Fumonisin is the most predominant of such mycotoxins. This compound is also thought to have an effect on pathogenicity. In this study, at Bejo Zaden B.V., we searched for variation in pathogenicity using a novel bioassay. These differences in pathogenicity were then linked to genetic differences in the candidate gene FUM1 which is a part of the gene cluster for the biosynthesis of fumonisin. Genetic differences were determined by sequencing and analysis is done by creating a phylogenetic tree. The differences found could be used for the development of a pathogen specific Taqman PCR so that pathogenic F. proliferatum isolates can be early detected.
Introduction
The genus Fusarium consists of fungi that can infect a wide range of plant hosts (Ma et al., 2013). Some of these fungi can be phytopathogenic. Some examples of pathogenic fusaria are F. oxysporum, F. redolens and F. proliferatum. When Fusarium is pathogenic on plants, it can enter through wounds in the roots. Fusaria therefore, often cause plant diseases like root rot (Figure 1) and wilting through obstruction of the vasculature (Arias et al., 2013). Pathogenic fungi are a widely distributed and a serious problem in agriculture resulting in loss of yield (McMullen et al. 1997). This crop loss is believed to be as high as one third of the total food crops annually (Fisher et al., 2012). To prevent this loss, a seed breeding company like Bejo Zaden B.V. has a responsibility to sell seeds that are disease free. To guarantee selling disease free seeds, a seed breeder has a few options, including breeding for pathogen resistance when this is possible. Also, additives like fungicides can be added to the seeds (i.e. in a seed coating) for additional defense. However, the first step in treatment is identification. Here, we focus on pathogen detection, because when pathogens can be detected prior to selling the seeds, the seeds can still be treated preventing disease in the field.
Apart from the loss of yield, it is a problem that Fusarium species can produce and excrete mycotoxins (Munkvold, 2016). When crops are infected by these species, the mycotoxins can be released into the human food chain. This is a problem because mycotoxins are toxic to both plant and animal (Marasas, 2001). The most common mycotoxin produced by Fusarium species is fumonisin B1 (Thiel et al., 1991). This compound disrupts the production of sphingolipids (He et al., 2006) . Sphingolipids are structural lipids with an important role in signal transduction present in the lipid bilayer of membranes. It is also suspected that sphingolipids are involved in the plants defense system, as they are is possibly involved in the
Figure 1. Picture of a young
asparagus with root rot, caused by a F. proliferatum infection.
Salycic Acid (SA) pathway which induces Programmed Cell Death (PCD) (Berkey et al., 2012). Fumonisin B1 is produced by a biosynthetic gene cluster under key influence of the Fum1 protein, a polyketide synthase. This protein is transcribed from the gene FUM1. Fumonisin B1 inhibits ceramide synthase by occupying a specific binding site of sphinganine, a precursor of sphingolipids. Inhibition is caused by the similarity in molecule structure between Fumonisin B1 and sphinganine (Figure 1). Ceramide synthase is an important regulatory enzyme for sphingolipid metabolism (Riley et al., 2001) and consequently, Fumonisin B1 alters the sphingolipid metabolism, modifying the membrane properties and is linked to cell apoptosis. Moreover, this compound is also thought to be carcinogenic (IARC, 2002). Therefore, in addition to producing pathogen-free seeds, early detection of the fungus is important for Bejo to ensure a high level of seed quality and prevention of infected crops entering the food chain.
Figure 2. A: Molecule structure of fumonisin B1. source: Wikipedia. B: Molecule structure of sphinganine. source: Brogden et al. in 2019 To be able to prevent infected crops entering the food chain through early detection of pathogens on seeds, a standardized detection method was needed for Fusarium. A common method for pathogen detection is a bioassay (Kathe et al., 2019), which comes down to inoculating the soil, seed or root with a spore suspension and see if the seedling emerges and if the emerged seedling dies off. This test is not very specific and is time consuming because it takes time for a seedling to emerge and when it does not, the exact cause is often unknown. In this study several pathogenicity assays were tested and compared. The University of Amsterdam and Naktuinbouw developed a pathogen-specific test for Fusarium oxysporum f.sp. cepae based on the screening for presence of two fungal effector genes, using a PCR with Taqman probes. Taqman probes are used for indication of the presence of a specific sequence (Holland et al., 1991). A Taqman probe consists of a probe sequence with a fluorophore and a quencher covalently attached to it. When the fluorophore is in proximity of the quencher, the emitted fluorescence is absorbed. When the probe is incorporate in double strand DNA, the fluorophore and the quencher are released, and the fluorescent signal can be registered by the qPCR machine. Because this assay is carried out using a PCR reaction it is very rapid, compared to a bioassay. It is also specific because it only reacts to the presence of specific gene sequences, in this case genes that are specific for F. proliferatum. It is useful to be able to detect F. proliferatum. However, not all F. proliferatum isolates are harmful. Pathogenicity varies between isolates (Stankovic et al., 2007). A detection method specific for pathogenic F. proliferatum isolates was not yet available. Due to the frequent occurrence of F. proliferatum on asparagus, this study is focused developing an assay on isolates coming from asparagus. The aim of this research was to design this detection method with pathogen specificity and the objectives were to determine if pathogenicity varies per isolate. Upon finding this, the genes or specific differences in DNA
sequence in these isolates that are associated with pathogenicity were searched for. Such an approach was already shown feasible by Abd Murad et al. in 2017 with a variety of Fusarium species on banana. Here, we focus only on F. proliferatum on asparagus. These possible genetic differences could then be used as targets for Taqman probes that can be designed for the PCR assay. To this end we tried to design a test that is pathogen specific.
Figure 3. Schematic of the working of a Taqman probe source: Wikipedia
The origin of pathogenicity was researched by checking the presence of a candidate gene. Furthermore, we looked for polymorphisms in this candidate gene that corresponds with pathogenicity. These polymorphisms were searched for by sequencing the amplicons and aligning the sequences. To sort the isolates in groups that have similarity in sequence, phylogenetic trees were constructed. The candidate gene is FUM1, which is involved in the production of fumonisin, described above. This gene was selected because Dissanayake et al. in 2009 found that some F. proliferatum strains do produce fumonisin and others do not. It has not yet been researched whether the ability to produce fumonisin is linked to pathogenicity of the isolate. Another gene that is analyzed in this study is the tef-1α gene, which is useful for identification of fungi (Kristensen et al., 2005). Furthermore, household genes are used for further identification and confirmation of F. proliferatum isolates. If pathogen specificity could be detected using the genes described above, they could possibly be used as a diagnostics tool as target for Taqman probes in the PCR assay.
Materials and Methods
Biological material
A. officinalis seeds were used to test pathogenicity of the F. proliferatum isolates. The seeds originate from the seed storage of Bejo Zaden and were all from the same Cumulus variety. Sixty different F. proliferatum isolates were used. The isolates were partly from collections stored at Bejo Zaden B.V. in Warmenhuizen and the other part was isolated from contaminated asparagus seed batches that were harvested at the Limburg location of Bejo zaden B.V. The fungi were isolated from the seeds by placing the contaminated seeds on Potato Dextrose Agar (PDA) and letting the fungi grow out. Next, the fungi were transferred to a clean PDA plate to grow further. F. proliferatum isolates were grown on PDA at 20 °C under a Near UV (NUV) light for 16 hours and 8 hours under normal light. NUV is near visible ultraviolet light with a wavelength between 300 and 400 nm, it helps increase spore production (Cooke & Gareth Jones, 1970).
DNA isolation
DNA isolation of the Fusarium isolates was performed with the Roche MagNA Pure 96 System. Using the DNA and Viral NA Small Volume Kit and the Pathogen Universal 2000 protocol. Beforehand, conidia were collected from Fusarium cultures growing on PDA by scraping off with sterile water and 0.05% Tween-20. Cells were lysed and proteins were degraded by adding 200 μl Roche Bacterial Lysis Buffer and 4 μl Protein Kinase K.
qPCR identification
Fusarium isolates were identified by NCBI nucleotide BLAST of the sequence of the Tef1 gene. Other information on the identity of isolates was retrieved by qPCR with primers that target the mating type genes, Mat1 and Mat2. Primers pairs that targeted sequences specific for F. proliferatum were Fpa1 (Bejo database), Fpa2 (Bejo database) , Fp3 (Jurado et al., 2006) and Gib2 (Jurado et al., 2006). A more general primer pair was used for the identification of Fusarium species, Fo (Bejo database). This primer pair was initially used for the identification of F. oxysporum but appeared to react with a variety of Fusarium species. A primer pair that targets the 28S rDNA of fungi (Pan fungal) was used as an internal control for qPCR reaction. For the detection and amplification of FUM1 gene sequences, two primer pairs were used. One of them, FUM1-5, is from literature (Gonzalez-Jean et al., 2004; Jurado et al., 2010). The other one, FUM1-5_2, was derived and optimized from the sequence of the amplicon of the primer pair FUM1-5. Optimization was done following an optimization protocol designed by the Seed Pathology Research group at Bejo. Optimal amplicon length, melting temperature, primer concentration and input volumes were determined for most efficient qPCR reactions. Criteria for efficient reactions were a low Cq value and one single melting peak at the right temperature. qPCR reactions were performed with the Roche Lightcycler 96. A primer pair from Stępień et al. from 2011 was used to try to amplify a partial sequence of the FUM1 gene but the sequence that was amplified could not be identified as FUM1. Instead, when BLAST was used to identify, it appeared to be part of a signal recognition particle receptor (SRPR). Because we presumed these primers amplified a partial reaction of FUM1 another primer pair was derived and optimized based on this sequence. These primer pairs were named SRPR and SRPR_2 respectively.
Sequencing and analysis
Sequencing was done at the company BaseClear B.V. located in Leiden. Before amplicons were sent for sequencing, amplicon length was confirmed in agarose gel. When the length was confirmed, amplicons and primers were sent to BaseClear for sequencing. Analysis of the DNA sequences and phylogenetic tree construction was done with CLC genomics workbench 12.
Bioassay
The selected F. proliferatum isolates were tested for pathogenicity against seeds using three methods. The first, seed in soil; a method were seeds were grown on inoculated filter paper and a method were the soil was inoculated. Isolates were inoculated onto one variety of asparagus seeds from Bejo. Seed inoculation was done following an adjusted method of Taylor et al. from 2013. The filter inoculation and the soil inoculation methods were designed by the Seed Pathology Research group at Bejo Zaden B.V.. Soil inoculation
Conidia of F. proliferatum were harvested from 5-day old cultures growing in liquid CBD-Y medium. Cultures were filtered through autoclaved cheesecloth to filter out the mycelium. The concentration of conidia was adjusted to 105 spores per gram soil, 1500 gram of soil was used. Spores were suspended in one liter of water and then added to the ground. Spores were distributed by mixing with gloved hands. Asparagus seeds were sowed in two rows of 25 seeds per container, two containers per isolate were prepared. In total 100 for every isolate. Two control containers were treated with one liter of water without spores.
Inoculated and control plants were maintained at 25°C by light and 18°C by dark under a 16-hour light, 8 hours dark photoperiod and 80 percent relative humidity. After 14 and 21 days, the emergence of the seedlings was monitored.
Seed in soil
Seeds were placed in plastic 96-well trays that were previously filled with germination soil. Conidia of F. proliferatum were harvested from actively growing 14 day-old cultures on PDA with sterile distilled water. The concentration of conidia was adjusted to 105 conidia ml-1using replicate haemocytometer counts. 200 μL of conidial suspensions were added to the seeds. Following watering, seeds were covered with a 1 cm deep layer of vermiculite. When covered in vermiculite the trays were watered again. Control seeds were treated with 200 μL of water instead of conidial suspension. One isolate was used to inoculate one row of 12 wells with one asparagus seed per well, this was repeated 11 times.
Inoculated and control plants were maintained at 25°C by light and 18°C by dark under a 16-hour light, 8 hours dark photoperiod and 80 percent relative humidity. After 14 and 21 days, the emergence of the seedling was monitored.
Filter inoculation
Conidia of F. proliferatum were harvested from actively growing 14 day-old cultures on PDA with sterile distilled water. The concentration of conidia was adjusted to 105 conidia ml-1. 50 mL of spore suspension was poured onto the filter paper. 50 asparagus seeds were placed on top of the filter paper. This was done twice per isolate. Two control trays were treated with 50 mL of sterile distilled water without spores. Inoculated and control seeds were placed in a dark climate cabinet, maintained at 20°C for 16 hours and at 30 °C for 8 hours.
Table 1. Overview of all primers that amplify specific targets that were used for identification of the F. proliferatum isolates.
Target (NAME)
Primer Sequence Reference 28S rDNA
(Pan fungal)
F: ZUP2889 TAA AGC TAA ATA YYG GCC RGA GA Bejo database
R: ZUP2890 CTT TYC AAA GTG CTT TTC ATC
SIX3 (Fo) F: Fo alg F CGC TGA GCT CGG TAA GGG Bejo database
R: Fo alg R CCA GAG AGC AAT ATC GAT GGT GA Unknown
(Fpa1)
F: Fpa F1 ATT AGC ACT TGG GTT GCC AGA A Bejo database
R: Fpa R1 ATT TGC CGT CAC TCT TGG AT Unknown
(Fpa2)
F: Fpa F2 GCG CTC AAG TCC AGA GTG ACT Bejo database
R: Fpa R2 CAG CGG GAC CGT TGG A
tef-1α (Tef) F: Ef728M CAT CGA GAA GTT CGA GAA GG Stępień et al., 2011
R: Tef1R GCC ATC CTT GGA GAT ACC AGC
Signal recognition partical receptor (SRPR)
F: Fum1F1 ACA TCY GTG GGC GAT CC Stępień et al., 2011
R: Fum1R2 ATA TGG CCC CAG CTG CRT A
FUM1 (FUM1-5)
F: Fum5-5F GAA ATG GAT CTM TTC GAG GC Gonzalez-Jean et al., 2004
R: FUM1P2-R TGG GTC CGA TAG TGA TTT GTC A Jurado et al., 2010
Signal recognition partical receptor (SRPR_2)
F: Fum1 212F TGA AGG TGA ACT TGG AGG AAC Designed in this study
FUM1 (FUM1-5_2)
F: Fum5 467F TCT TCT CCC GGT CGG CTG Designed in this study
R: Fum5 467R GTT TTG CGC CAG GTA GCC AA
IGS (Gib2) F: Gib2-F GAG GCG CGG TGT CGG TGT GCT TG Jurado et al., 2006
R: Fgc-R CTC TCA TAT ACC CTC CG
IGS (Fp3) F: Fp3-F CGG CCA CCA GAG GAT GTG Jurado et al., 2006
R: Fp4-R CAA CAC GAA TCG CTT CCT GAC
MAT1 (Mat1) F: Mat1F GAC CAA CTC AAA CCT CGT GGC G Kerenyi et al., 2002
R: Mat1R TCA TCA AAG GCA AGC GAT ACC
MAT2 (Mat2) F: Mat2F ACC GTA AGG AGC GTC ACC ATT Kerenyi et al., 2002
Results
Genetic identification and differentiation of F. proliferatum isolates
To identify the isolates, different PCR-based reactions were used. From the sequences of the TEF1 gene and from the PCR with the Fp3 primer pair, the identification of the isolate could be determined by BLAST of the PCR product. The MAT1 and MAT2 genes appeared to be both present in different isolates (Table 2). This shows that both mating types were found. The presence of the sequences amplified by primer pairs Fpa1 and Fpa2 varied between isolates (Table 2). From the original 60 isolates, a selection of 20 was made for the bioassay. Selection was done by choosing for variation in the sequences from primer pairs SRPR_2 and FUM1-5_2 (Table 2).
Table 2. Cq values for qPCR reactions with primer pairs SRPR_2, FUM1-5_2, Fp3, Mat1, Mat2, Fpa1 and Fpa2 on isolates used in the bioassay. –
means no detectable PCR signal. Identity was determined by nucleotide BLAST of the partial sequence of the Tef-1α gene.
Isolate SRPR_2 FUM1-5_2 Fp3 Mat1 Mat2 Fpa1 Fpa2 Identity
2 20.66 22.2 24.08 - 25.7 23.12 23.54 F. proliferatum 5 - - 22.64 - 26.22 28.52 - F. proliferatum 7 19.62 20.04 21.29 - 23.88 21.65 22.15 F. proliferatum 9 19.77 20.45 21.19 - 26.55 23.54 23.61 F. proliferatum 11 23.35 23.9 24.44 - 27.15 24.38 24.34 F. proliferatum 13 19.4 19.85 24.2 - 25.4 23.23 23.18 F. proliferatum 14 33.76 - 23.56 - 25.92 24.04 24.21 F. proliferatum 19 19.48 19.91 22.28 - 24.08 21.91 22.01 F. proliferatum 23 21.46 22.15 20.86 - 25.14 23.15 23.24 F. proliferatum 26 40.67 33.01 24.55 - 25.23 23.22 23.15 F. proliferatum 29 19.96 25.15 21.76 - 23.32 20.91 21.14 F. proliferatum 36 21 21.42 - - 25.47 - - F. proliferatum 38 20.75 23.08 20.47 20.37 - 22.39 22.49 F. proliferatum 40 29.14 - 21.62 - 23.86 21.64 21.71 F. proliferatum 44 21.49 22.22 - - 22.94 20.76 20.94 F. proliferatum 45 22.1 22.44 - - 24.42 22.35 22.74 F. proliferatum F1 23.23 23.45 22.74 - 28.225 - 29 F. proliferatum F38 23.15 - 20.73 - 26.36 - 27.22 F. proliferatum P6491 21.29 21.73 - 24.59 - - - F. proliferatum P6903 22.86 23.29 22.805 24.085 - - 29.28 F. proliferatum
To determine the exact genetic differences in the FUM1 gene and the SRPR gene between isolates, a partial DNA sequence of both the FUM1 and the SRPR gene were amplified in a PCR with primer pairs SRPR_2 (212 nt) and FUM1-5_2 (467 nt). These amplicons were next sequenced by Sanger sequencing. To analyze the differences, a phylogenetic tree was constructed on the basis of the sequences (Figure 5). Some isolates are not present in the tree because they did not possess the sequence that was used. Varying genetic differences between the isolates are visible in Figure 5. The tree of the partial sequence from the SRPR gene (Figure 5A) showed that the isolates were divided into four groups, this is caused by the multiple nucleotide differences between the isolates. The tree for the partial sequence from the FUM1 gene (Figure 5B) showed that the isolates are divided in two groups, caused by a single nucleotide difference.
Figure 4. A. Consensus phylogenetic tree for 19 F. proliferatum isolates used in the bioassay. The tree was constructed on the basis of a DNA
sequence amplified with primer pair SRPR_2. A Fusarium oxysporum f.sp. cubense, GenBank accession number: XM_031203921, isolate was used as an outgroup. B. Consensus phylogenetic tree for 16 F. proliferatum isolates used in the bioassay. The tree was constructed on the basis of a partial sequence of the FUM1 gene, amplified with primer pair FUM1-5_2. A Fusarium aywerte, GenBank accession number: KU179902, isolate was used as an outgroup. The trees were constructed using the UPGMA approach and tested by bootstrapping (250 replicates). Only branches with 50% and above support from bootstrapping were shown. Numbers indicate groups.
Bioassay comparison
To determine whether the genetic differences identified above also cause a difference in pathogenicity, a bioassay was attempted. Three different methods of bioassay were tested and compared for reliability and whether differences in pathogenicity were clearly visible. Eventually, the soil inoculation method did not provide reliable results (data not shown) probably due to irregular watering. The soil inoculation method is therefore omitted from the comparison.
Next, we attempted a germination assay on filter paper. Three weeks after placing the inoculated A. officinalis seeds between filter paper, the emergence of the seedlings was recorded. Relative emergence was calculated by dividing the emergence of the test treatments by the emergence of the control treatment. Clear differences in relative emergence were visible between treatments (Fig. 3). The positive control, P6475, shows an emergence of 0, as expected Some isolates, such as P6461 score as high as the negative control, with a relative emergence of near 1, meaning no effect of the fungi is visible. Isolates that stand out are JF2, JF5, JF14 and JF45. They all show a relative emergence lower than 0.1 pointing out that they are highly pathogenic. An overview of the isolates classified by pathogenicity are presented in Table 3.
Figure 5. Bar graph of the relative emergence of A. officinalis seedlings (n=50, repeated twice) inoculated by F. proliferatum isolates on the
x-axis. Emergence values are relative to the emergence of the negative control treatment. Isolate P6475 was used as a positive control.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 JF 2 JF 5 JF 7 JF 9 JF 11 JF 13 JF 14 JF 19 JF 23 JF 26 JF 29 JF 36 JF 38 JF 40 JF 44 JF 45 F38 P6491 P6903 P6475+ F1 JF 10 JF 39 C-Re lat iv e em erge n ce
Table 3. A categorization of pathogenicity based on the relative emergence of the filtermethod. – is given when the average of both values is
above 0.6. + is given when the average is between 0.6 and 0.1. ++ is given when the average is below 0.1.
ISOLATE PATHOGENICITY JF2 ++ JF5 ++ JF7 + JF9 + JF10 + JF11 - JF13 - JF14 ++ JF19 + JF23 + JF26 - JF29 + JF36 + JF38 + JF39 + JF40 + JF44 + JF45 ++ F38 - P6491 - P6903 - P6475+ ++ F1 -
The third assay performed entailed seed germination in soil, rather than filter paper. Three weeks after sowing the inoculated A. officinalis seeds in soil, the emergence of the seedlings was counted. The relative emergence was calculated by dividing the emergence of the test treatments by the emergence of the control treatment. The positive control shows a relative emergence of 0.48. Differences in relative emergence is less clearly visible using this method (Fig. 4). However, isolates that
show the highest pathogenicity are JF2, JF5, JF7, JF14 and JF45. They all show a relative emergence of lower than 0.5. An overview was made in which isolates were classified by pathogenicity (Table 3). All combined, the assay best fitting with the criteria that were used was the filter method. The
differences in pathogenicity were the most visible and it was the most reliable method because watering was done the most regular.
Figure 6. Bar graph of the relative emergence of A. officinalis seedlings (n=12, repeated 11 times) inoculated by F. proliferatum isolates on the
x-axis. The percentages are relative to the emergence of the control treatment. Isolate P6475 was used as a positive control.
Table 4. A categorization of pathogenicity based on the relative emergence of the seed in soil method. – is given when the average of both
values is above 0.8. + is given when the average is between 0.8 and 0.5. ++ is given when the average is below 0.5.
ISOLATE PATHOGENICITY JF2 ++ JF5 ++ JF7 ++ JF9 + JF11 - JF13 + JF14 ++ JF19 + JF23 + JF26 + JF29 + JF36 + JF38 + JF40 + JF44 + JF45 ++ F38 - P6491 - P6903 + P6475+ ++ F1 - 0 0.2 0.4 0.6 0.8 1 1.2 1.4 2 5 7 9 11 13 14 19 23 26 29 36 38 40 44 45 F38 P6491 P6903 P6475+ F1 C-Re lat iv e em erge n ce (% ) Isolates
Discussion
A detection method for pathogenic F. proliferatum isolates is needed to reduce the loss of crop yield caused by this pathogen. A detection method that is currently used is a bioassay (Kathe et al., 2019). This method however, is not very specific. That is why a PCR based detection is needed. This method would be both specific and fast. To this end, genetic differences between pathogenic and
non-pathogenic isolates were searched for using a partial sequence of candidate genes. Pathogenicity was determined with a novel bioassay.
To obtain these genetic differences, partial sequences of the FUM1 and SRPR genes were amplified and sequenced. When these sequences were analyzed, differences in sequence between isolates were present. This is in line with findings from study by Kristensen et al. in 2005, were genetic differences were also found in the Tef-1α gene between different F. proliferatum isolates. Moreover, PCR reactions with F. proliferatum specific primer pairs were used to confirm that the isolates were in fact F. proliferatum. When a tree was constructed from the partial sequences of the FUM1 and SRPR genes, the genetic differences were made visible. FUM1 showed a division into two groups, based on a single nucleotide difference. SRPR showed a division into four groups, based on multiple nucleotide differences. This confirms that genetic differences between the F. proliferatum isolates are present.
In addition, the pathogenicity of these isolates was determined with a bioassay. Out of three methods, the filtermethod was chosen to be the most reliable. This was based upon the fact that differences in pathogenicity were the most clearly visible and it was most reliable due to consistent watering. To search for a possible marker for pathogenicity, the relative emergence values from the bioassay and the sequence analysis were compared. Considerable differences in pathogenicity can be observed in the results from the bioassay. Pathogenicity in F. proliferatum appears not to be binary distributed (Fig. 5&6). Some isolates seem to be highly pathogenic and others to a lesser extent. However degrees of pathogenicity have been found to be different between Fusarium species (Abd Murad et al., 2017), this study also found that pathogenicity varies between F. proliferatum isolates. This has already been found by Stankovic et al. in 2007, but this study showed that F. proliferatum isolates from a single host also vary in pathogenicity.
The genetic differences in the partial sequence of the FUM1 gene did not seem to lead to a pathogenic or a non-pathogenic isolate. With two genetic variants, a binary distribution is expected, while the pathogenicity distribution in this study was not. Some isolates also did not possess the partial sequence of FUM1 but were still found to be slightly pathogenic. Despite the fact that the genetic differences in the partial sequence of FUM1 do not explain the pathogenicity, other genes possibly can. Isolates JF2, JF5, JF14 and JF45 showed the highest pathogenicity. When pathogenicity and genetic differences were compared, it stands out that group 4 (JF2, JF14 and JF45) from the tree of SRPR, contains three of the five isolates with the highest pathogenicity. This could indicate that there is a genetic difference between highly pathogenic and less pathogenic isolates. This is in line with expectations from a study by Taylor et al. in 2012. They also expected that genetic loci exist that correlate with pathogenicity.
Our expectation was that a genetic difference in the partial sequence of FUM1 would cause a difference in pathogenicity, the fact that this study did not show this result can be explained. Only a small part from one of many genes in the fumonisin biosynthetic cluster was sequenced (Fig. 7), so there was a small change to capture genetic differences. Also, many other genes apart from fumonisin genes could be possible candidates for pathogenicity markers.
Figure 7. An overview of the fumonisin biosynthetic gene cluster. Partial sequence of FUM1 is indicated with arrows. Credits: Proctor et al. in
2003
The information found in this study is useful for further research into the identification of pathogenic F. proliferatum isolates from asparagus because when pathogenic isolates can be early detected with the fast and easy Taqman PCR method, seeds can be treated so healthy seeds can be delivered to the grower. When this is possible, crop loss can be reduced and the yield can be increased. Also, the release of mycotoxins into the human food chain can be reduced. To be able to develop such a method, further research is needed. A more extensive genetic study could be done into the fumonisin biosynthetic gene cluster, the entire cluster could be sequenced and analysed. This could in turn be compared to the results of a bioassay such as in this study. Another option is to explore the entire genome with a Genome Wide Association Study. With this method SNPs can be detected that possibly cause pathogenicity.
In conclusion, this study found that F. proliferatum isolates from asparagus have varying pathogenic abilities. This variety in pathogenicity can not yet be explained by genetic differences in the candidate gene FUM1. However, only a partial sequence from a gene in the biosynthetic gene cluster of fumonisin was used for analysis, so further analysis is needed.
Acknowledgments
I would like to express my gratitude towards Bejo Zaden for the opportunity to work on my bachelor project at their facilities. I would also like to thank my supervisors Wilfried Jonkers and Petra Bleeker for their advice and guidance.
Literature
Abd Murad, N. B., Mohamed Nor, N. M. I., Shohaimi, S., & Mohd Zainudin, N. A. I. (2017b). Genetic diversity and pathogenicity of Fusarium species associated with fruit rot disease in banana across Peninsular Malaysia. Journal of Applied Microbiology, 123(6), 1533–1546.
https://doi.org/10.1111/jam.13582
Armitage, A. D., Taylor, A., Hulin, M. T., Jackson, A. C., Harrison, R. J., & Clarkson, J. P. (2019). Draft Genome Sequence of an Onion Basal Rot Isolate of Fusarium proliferatum. Microbiology Resource Announcements, 8(1). https://doi.org/10.1128/mra.01385-18
Berkey, R., Bendigeri, D., & Xiao, S. (2012). Sphingolipids and Plant Defense/Disease: The “Death” Connection and Beyond. Frontiers in Plant Science, 3. https://doi.org/10.3389/fpls.2012.00068
Brogden, G., Husein, D. M., Steinberg, P., & Naim, H. Y. (2019). Isolation and Quantification of Sphingosine and Sphinganine from Rat Serum Revealed Gender Differences. Biomolecules, 9(9), 459.
https://doi.org/10.3390/biom9090459
Dissanayake, M. L. M. C., Tanaka, S., & Ito, S. (2009). Fumonisin B1 production by Fusarium proliferatum strains isolated from Allium fistulosum plants and seeds in Japan. Letters in Applied Microbiology, 48(5), 598–604. https://doi.org/10.1111/j.1472-765x.2009.02576.x
Fisher, M. C., Henk, Daniel. A., Briggs, C. J., Brownstein, J. S., Madoff, L. C., McCraw, S. L., & Gurr, S. J. (2012). Emerging fungal threats to animal, plant and ecosystem health. Nature, 484(7393), 186–194.
https://doi.org/10.1038/nature10947
González-Jaén, M. T., Mirete, S., Patiño, B., López-Errasquín, E., & Vázquez, C. (2004). Genetic Markers for the Analysis of Variability and for Production of Specific Diagnostic Sequences in Fumonisin-Producing Strains of Fusarium Verticillioides. European Journal of Plant Pathology, 110(5/6), 525–532.
https://doi.org/10.1023/b:ejpp.0000032392.20106.81
He, Q., Suzuki, H., Sharma, N., & Sharma, R. P. (2006). Ceramide Synthase Inhibition by Fumonisin B1 Treatment Activates Sphingolipid-Metabolizing Systems in Mouse Liver. Toxicological Sciences, 94(2), 388–397. https://doi.org/10.1093/toxsci/kfl102
Holland, P. M., Abramson, R. D., Watson, R., & Gelfand, D. H. (1991). Detection of specific polymerase chain reaction product by utilizing the 5’----3’ exonuclease activity of Thermus aquaticus DNA
polymerase. Proceedings of the National Academy of Sciences, 88(16), 7276–7280.
https://doi.org/10.1073/pnas.88.16.7276
Jurado, M., Marín, P., Callejas, C., Moretti, A., Vázquez, C., & González-Jaén, M. T. (2010). Genetic variability and Fumonisin production by Fusarium proliferatum. Food Microbiology, 27(1), 50–57.
https://doi.org/10.1016/j.fm.2009.08.001
Jurado, M., Vázquez, C., Marín, S., Sanchis, V., & Teresa González-Jaén, M. (2006). PCR-based strategy to detect contamination with mycotoxigenic Fusarium species in maize. Systematic and Applied
Kenényi, Z., Mulé, G., Moretti, A., Waalwijk, C., & Hornok, L. (2002). Fertility and mating type assessment within Fusarium proliferatum isolates from different host plants. In P. Chelkowski, P. Golínski, M.
Kozlowska, H. Kwasna, & J. Perkowski (Eds.), Fusarium - Mycotoxins, Taxonomy and Pathogenicity. Proceedings of the Seventh International Symposium, September 4 - 7, 2002, Poznán, Poland (Vol. 43A, pp. 55-68). Poznán, Poland: Warsaw Agricultural University.
Kristensen, R., Torp, M., Kosiak, B., & Holst-Jensen, A. (2005). Phylogeny and toxigenic potential is correlated in Fusarium species as revealed by partial translation elongation factor 1 alpha gene sequences. Mycological Research, 109(2), 173–186. https://doi.org/10.1017/s0953756204002114
Ma, L.-J., Geiser, D. M., Proctor, R. H., Rooney, A. P., O’Donnell, K., Trail, F., … Kazan, K. (2013). Fusarium Pathogenomics. Annual Review of Microbiology, 67(1), 399–416. https://doi.org/10.1146/annurev-micro-092412-155650
Marasas, W. F. (2001). Discovery and occurrence of the fumonisins: a historical perspective.
Environmental Health Perspectives, 109(suppl 2), 239–243. https://doi.org/10.1289/ehp.01109s2239
McMullen, M., Jones, R., & Gallenberg, D. (1997). Scab of Wheat and Barley: A Re-emerging Disease of Devastating Impact. Plant Disease, 81(12), 1340–1348. https://doi.org/10.1094/pdis.1997.81.12.1340
Munkvold, G. P. (2016). Fusarium Species and Their Associated Mycotoxins. Methods in Molecular Biology, 51–106. https://doi.org/10.1007/978-1-4939-6707-0_4
Riley, R. T., Enongene, E., Voss, K. A., Norred, W. P., Meredith, F. I., Sharma, R. P., Merrill, A. H., Jr. (2001). Sphingolipid perturbations as mechanisms for fumonisin carcinogenesis. Environmental Health Perspectives, 109(suppl 2), 301–308. https://doi.org/10.1289/ehp.01109s2301
Stępień, Ł., Koczyk, G., & Waśkiewicz, A. (2011). Genetic and phenotypic variation of Fusarium proliferatum isolates from different host species. Journal of Applied Genetics, 52(4), 487–496.
https://doi.org/10.1007/s13353-011-0059-8
Taylor, A., Vagany, V., Barbara, D. J., Thomas, B., Pink, D. A. C., Jones, J. E., & Clarkson, J. P. (2012). Identification of differential resistance to six Fusarium oxysporum f. sp. cepae isolates in commercial onion cultivars through the development of a rapid seedling assay. Plant Pathology, 62(1), 103–111.
https://doi.org/10.1111/j.1365-3059.2012.02624.x
The International Agency for Research on Cancer. (2002). Some Traditional Herbal Medicines, Some Mycotoxins, Naphthalene and Styrene. Retrieved from https://publications.iarc.fr/100
Thiel, P. G., Shephard, G. S., Sydenham, E. W., Marasas, W. F. O., Nelson, P. E., & Wilson, T. M. (1991). Levels of fumonisins B1 and B2 in feeds associated with confirmed cases of equine
leukoencephalomalacia. Journal of Agricultural and Food Chemistry, 39(1), 109–111.
