Molecular and physiological characterization of Fusarium strains associated with different diseases in date palm

Several species of Fusarium cause serious diseases in date palm worldwide. In the present work, 14 SSR markers were used to assess the genetic variation of Fusarium strains isolated from diseased trees in Saudi Arabia. We also studied the effect of different temperatures on mycelial growth of these strains. The pathogenicity of four strains of F. proliferatum was also evaluated on local date palm cultivars. Eleven SSR markers amplified a total of 57 scorable alleles from Fusarium strains. Phylogenetic analysis showed that F. proliferatum strains grouped in one clade with 95% bootstrap value. Within F. proliferatum clade, 14 SSR genotypes were identified, 9 of them were singleton. Four out of the five multi-individual SSR genotypes contained strains isolated from more than one location. Most F. solani strains grouped in one clade with 95% bootstrap value. Overall, the SSR markers previously developed for F. verticillioides and F. oxysporum were very useful in assessing the genetic diversity and confirming the identity of Saudi Fusarium strains. The results from the temperature study showed significant differences in mycelial growth of Fusarium strains at different temperatures tested. The highest average radial growth for Fusarium strains was observed at 25°C, irrespective of species. The four F. proliferatum strains showed significant differences in their pathogenicity on date palm cultivars. It is anticipated that the assessment of genetic diversity, effect of temperature on hyphal growth and pathogenicity of potent pathogenic Fusarium strains recovered from date palm-growing locations in Saudi Arabia can help in effectively controlling these pathogens.


Introduction
Date palm (Phoenix dactylifera L.) is one of the most important fruit crops in the arid climates including the Arabian Peninsula, North Africa and the Middle East regions [1]. The total world date palm production is around 9.08 million tons harvested from a total area of 1.38 million ha [2]. The kingdom of Saudi Arabia produces ca 17% of the total world production, was extracted using CTAB method according to Murray and Thompson (1980) and modified by Saleh et al. (2017) [7,18]. DNA solutions were quantified on 1% agarose gels stained with 10 μg/100 mL Acridine Orange in 0.5× TBE buffer and using Nanodrop 2000 Spectrophotometer (Thermo Fisher Scientific, USA). DNA solutions were diluted with sterilized distilled water (sdH 2 O) to a final concentration of 20 ng/μL. Developing of SSR markers. Fourteen SSR primer-pairs, ten developed for F. verticillioides [8] and four developed for F. oxysporum [15,19], were used to genotype 59 Fusarium strains belonging to the following species: F. proliferatum (47/59), F. solani (9/59), F. oxysporum (2/59) and F. verticillioides (1/59). Forward primers were labeled with Applied Biosystems standard dyes, NED (yellow), VIC (green), PET (Red) and FAM (blue). The optimum annealing temperature for each SSR primer-pair was initially determined by gradient PCR using DNA of seven Fusarium strains: three F. proliferatum, two F. solani, one F. oxysporum and one F. verticillioides. PCR reactions contained 5 μL of 2× Go Taq Green Master Mix (Promega, WI, USA), 0.5 μL of each 10 μM primer, 0.5 μL of 20 ng/μL DNA and 3.5 μL of sdH 2 O. The PCR conditions were as follows: an initial denaturation step at 94˚C for 4 min, followed by 35 cycles of a denaturation step at 94˚C for 40 s, an annealing step at 50-60˚C for 30 s and an extension step at 72˚C for 30 s, then a final extension step at 72˚C for 5 min in the Mastercycler1 nexus gradient PCR machine (Eppendorf AG, Hamburg, Germany). After determining the optimum annealing temperature for each SSR primer-pair (Table 1), PCR reactions were set up as previously described but in a final volume of 25 μL. PCR amplification program was also as above but with the suitable annealing temperature for each SSR primer-pair. The PCR products generated from SSR primer-pairs were primarily run on 2% NuSieveTM agarose (Cambrex Bio Science, Rockland, ME, USA) gels stained with Acridine orange in 0.5× TBE buffer and visualized under UV light using the G-Box gel documentation system (Syngene, Cambridge, UK). Then, 0.5 μL of each SSR-PCR product was mixed with 9.25 μL of HiDi formamide and 0.25 μL of GeneS-can1 LIZ 500 size standard. The resultant mixtures were denatured at 95˚C for 5 min and chilled on ice for another 5 min. Two microliters of the previous mixture were loaded into ABI Prism1 3730 Genetic Analyzer (Applied Biosystems, Foster City, USA) and raw data were collected using genetic analyzer data collection software version 3.0 (Applied Biosystems). SSR data were analyzed with the Peak Scanner software 1.0 (Applied Biosystems).
Analysis of SSR data. SSR alleles were scored as "1" for the presence of a peak (allele) and "0" for its absence, assuming that peaks (alleles) with the same molecular size in different individuals were homologous. Unweighted pair grouping by mathematical averages (UPGMA) subroutine of PAUP 4.8b was used to construct dendrograms and determine the number of haplotypes among fungal strains. To quantitatively evaluate the informativeness of SSR markers, the polymorphism information content (PIC) was calculated as follows: PICj = 1-S i n P i 2 , where i is i-th allele of the j-th SSR marker, n is the number of alleles of the j-th SSR marker and is allele frequency [20].

Effect of different temperatures on mycelial growth of Fusarium species
The effects of temperature on mycelial growth of Fusarium species were evaluated at different temperatures, ranging from 15 to 40˚C. Fungal strains were grown on PDA Petri plates (9 cmdiameter) by transferring 6 mm diameter disks of pure fungal cultures to the center of the plates. The plates were incubated at appropriate temperatures (15, 20, 25, 30, 35 and 40˚C). After 6 days of incubation, diameter of each colony was then measured in two perpendicular directions and an average of the two measurements was calculated after subtracting the 6 mm diameter of the colonized plug. Three PDA plates were used as replicates for each Fusarium strain.
Data analysis of temperature effect on mycelial growth of Fusarium species. Data obtained from growing Fusarium strains at different temperatures were analyzed by one-way ANOVA using SAS software version 9.2 (SAS Institute, Cary, NC, USA). Duncan's multiple range test was performed for comparing means of different treatments of all experiments at P < 0.05 using SAS software. To get the optimum growth temperature for each Fusarium species, mycelial growth was plotted against different temperatures and each curve was fitted by the polynomial regression model "y = a+bx+cx2" using Microsoft excel 2016.

Pathogenicity of Fusarium strains on Saudi date palm cultivars
Three local date palm cultivars, namely Sheeshee, Khalas and Ruziz were used for pathogenicity experiments under greenhouse conditions. The seeds were surface disinfected in 1% of sodium hypochlorite for 10 min and then immersed in sdH 2 O overnight. The soaked seeds were planted in 9×9 cm pots containing sterilized peat:soil mixture (2:3 v/v). Each pot had only one seed. Seeded pots were irrigated and fertilized as needed under greenhouse conditions. As the F. proliferatum was the dominant Fusarium sp. recovered from Saudi date palms, four potent pathogenic strains of F. proliferatum and one strain of F. oxysporum were further selected to infect six-month-old Sheeshee, Khalas and Ruziz seedlings under greenhouse conditions [7,21]. PDA disks (5 mm in diameter) were cut from the edges of 5-day old cultures of Fusarium and transferred to 500-mL flasks containing 150 mL of PDB. Inoculated flasks were incubated and agitated on a horizontal shaker (100 rpm) at room temperature. After two weeks, the spores were collected using four layers of cheesecloth and spore filtrates were diluted with sdH 2 O and adjusted to 1×10 7 conidia/mL by using a hemocytometer.
Six-month-old Sheeshee, Khalas and Ruziz seedlings were gently removed from the pots and immersed up to the crown region in 1×10 7 conidia/ml solutions for 10 min. Then, the inoculated seedlings were transferred to 12×15 cm pots, one seedling per pot. The control seedlings were immersed in sdH 2 O for 10 min before transferring into the 12×15 cm pots. Four seedlings (replicas) were inoculated with each Fusarium strain. All pots were maintained in greenhouse until the end of the experiments. The experimental design of the greenhouse work was in a randomized complete block design (RCBD). Four weeks after seedlings' inoculation, the disease severity was estimated using the following disease rating scale: 0 = healthy (no observable disease symptoms), 1 = shrinking of one leaf, 2 = stunting and shrinking of one or two leaves, 3 = stunting and shrinking of two leaves with one leaf died, 4 = completely dead seedling ( Fig 1A).
Isolation of Fusarium strains from above and underground parts of infected seedlings of four date palm cultivars. To isolate Fusarium strains from above and underground parts of infected seedlings, plants were gently removed from the pots and washed thoroughly under tap water. From each seedling, four plant pieces (ca 1 cm in length each) were cut, almost 1 cm apart from the crown region, from both above and underground parts. Plant pieces were sterilized for 5 min in 1% Sodium hypochlorite, dried on paper tissues and transferred to PDA plates. Four tissue pieces were arranged sequentially from 1 (close to crown region) to 4 (far from the crown region) onto the surface of PDA plates ( Fig 1B). PDA plates containing plant pieces were incubated for 5 days at 25˚C. The average recovery of a fungal strain was calculated by summing plant pieces that showed the fungal growth and divided by the total number of pieces used for fungal isolation ( Fig 1B).
Data analysis of Fusarium pathogenicity. The data obtained from the pathogenicity experiments of the five Fusarium strains on Sheeshee, Khalas and Ruziz seedlings were analyzed, using SAS software, with a two-way (Fusarium strains and date palm seedlings cultivars) ANOVA in RCBD with four replicates. Duncan's multiple range test was performed for comparing means of different treatments of all experiments at P < 0.05.

Genotyping of Fusarium strains with SSR markers
A total of 14 SSR primer-pairs were used to fingerprint Fusarium strains recovered from date palm in Saudi Arabia. The primer-pairs Fv-114 and Fv-269 did not amplify PCR products from all Fusarium strains. Although Fv-403 amplified PCR products from Fusarium strains, bands were not clearly scorable. Collectively, 11 out of 14 SSR primer-pairs amplified a total of 57 scorable alleles from F. proliferatum, F. solani, F. oxysporum and F. verticillioides strains ( Table 1). The primer-pair Fv-120 amplified only one allele from F. verticillioides. The number of alleles for individual SSR primer-pair ranged from 3 (Fv-312) to 9 alleles (FoAB11), with a mean number of alleles 5 per SSR marker. In addition, PIC values of SSR markers ranged from 0.078 to 0.511 (Table 1). According to the number of nucleotides in SSR motifs, hexa-and penta-nucleotide repeats showed higher PIC values compared with and tetra-, tri-and dinucleotide ones (Table 1). Among 47 F. Proliferatum strains, 6 out of 10 SSR markers were found to be polymorphic. The number of alleles per SSR marker varied, ranging from 1 (FoDC5, Fv-284, FO18 and Fv-312) to 5 (FoDD7), with a total of 24 alleles. The size of alleles ranged from 111 to 374 bp detected by the ABI genetic analyzer. The PIC values were low, ranging from 0.04 (Fv-338) to 0.297 (Fv-47).
SSR fingerprints of Fusarium strains were used to construct a phylogenetic tree using UPGMA algorithm (Fig 2). Strains of F. proliferatum grouped together in one clade that received 95% bootstrap value. Within F. proliferatum clade, 14 SSR genotypes were identified, of which nine were singleton, i.e. represented only with one strain, and five genotypes had more than one strain. Moreover, four out five genotypes contained strains isolated from more than one location. The most frequent genotype contained 21 strains isolated from root and shoot tissues collected from all the surveyed locations (Fig 2). Fusarium verticillioides E52 was closely related to the F. proliferatum clade, as both species belonging to Fusarium fujikuroi species complex. Most F. solani strains (8 out of 9) grouped in one clade that received 95% bootstrap value (Fig 2). The F. solani H13 strain was slightly distant from F. solani clade. The clade that contained the two F. oxysporum strains received 69% bootstrap value.

Temperature effects on mycelial growth of different Fusarium species
According to ANOVA, the average effects of different temperatures on fungal radial growth of Fusarium species were highly significant at P < 0.0001. In general, among different Fusarium species, there were differences in their radial growth. For example, at low temperatures 15 and 20˚C, F. proliferatum strains had the highest radial growth and F. solani had the lowest growth (Table 2). At moderate temperatures 25 and 30˚C, F. verticillioides showed the highest radial mycelial growth (Table 2). At high temperatures 35 and 40˚C, the mycelial growth of F. solani had the highest significant growth. Fusarium proliferatum, F. verticillioides and F. oxysporum didn't grow at 40˚C ( Table 2). The optimum growth temperature for each Fusarium species was estimated using the polynomial regression model (Fig 3). Fusarium solani had the highest optimum temperature (27.2˚C), followed by F. proliferatum, F. verticillioides and F. oxysporum (25, 24.9 and 24.9˚C, respectively) (Fig 3).
Within F. proliferatum strains, there were significant differences in their mycelial growth on PDA (P < 0.0001) ( Table 2). Generally, different temperatures promoted variable mycelial growth of F. proliferatum strains ( Table 2). The highest average mycelial growth was recorded at 25˚C, followed by 20˚C and 30˚C, respectively. Both 35˚C and 15˚C temperature regimes showed lower average mycelial growth for F. proliferatum strains. Apparently, temperatures between 20˚C to 25˚C would be better for the growth of F. proliferatum strains. For F. solani strains, there were significant differences in mycelial growth on PDA at P < 0.0001 ( Table 2). The highest average mycelial growth was recorded at 25˚C, followed by 30˚C. Strains of F. solani showed slight mycelial growth at 40˚C. It is expected that temperatures between 25˚C to 30˚C would be better for the growth of F. solani strains.

Pathogenicity of Fusarium strains on Saudi date palm cultivars
The F. proliferatum H78, E128, WF3D and RF93 strains, along with F. oxysporum KH20 had different degrees of disease severity on the date palm seedlings of Sheeshee, Khalas and Ruziz cultivars. Disease symptoms started on the first (old) leaf that showed chlorosis and/or were shrunken, then proceeded to the second leaf. As the disease progressed, either one or two leaves died, according to strain virulence. Overall, the five Fusarium strains demonstrated significant differences (P < 0.0025) in their pathogenesis on date palm cultivars (Table 3). Two F. proliferatum strains RF93 and H78 had the highest disease severity values, followed by WF3D, E128 and KH20 (Table 3). Based on average effects of disease severity, there was no significant differences among date palm cultivars. However, Ruziz was slightly susceptible than Khalas and Sheeshee ( Table 3).

Isolation of F. proliferatum and F. oxysporum strains from above and underground parts of infected date palm seedlings
To investigate the fungal progress inside Fusarium-infected seedlings of Sheeshee, Khalas and Ruziz cultivars, direct isolation of Fusarium from above and underground parts were conducted. All root tissues of the three cultivars showed complete colonization by the five Fusarium strains mycelial. However, there were significant differences (P < 0.0001) in strain recovery from aboveground tissues ( Table 4). The strain recovery from Ruziz was higher compared with Khalas and Sheeshee. The high recovery value could explain the slight susceptibility of Ruziz to Fusarium strains compared with Khalas and Sheeshee seedlings (Table 4).

Discussion
A better understanding of genetic variation within and between phytopathogen strains would be useful in the improvement of disease management strategies of date palm trees against pathogenic Fusarium spp. Among DNA-based methods used to assess genetic variations of organisms, SSR loci become the markers of choice for studying genetic diversity of organisms. Many studies have also evaluated the transferability of SSR across species and genera of different kingdoms including fungi and plants [22]. In this study, SSR markers previously developed for F. verticillioides and F. oxysporum have been used to determine genetic diversity of Fusarium strains recovered from date palm trees in Saudi Arabia. The four SSR primer-pairs of F. oxysporum amplified fingerprints from Saudi Fusarium strains. However, eight out of the ten F. verticillioides SSR primer-pairs amplified PCR amplicons from Fusarium strains. Although, In general, SSR markers are less abundant in fungal genomes compared with other organisms' genomes [24]. Fungal genomes also have unique distribution and occurrence of different oxysporum. �� Disease severity on date palm seedlings was estimated 4 weeks after inoculation using a 0-4 scale (Fig 1A). Values  oxysporum. �� Average recovery of a fungal strain was calculated by summing plant pieces that showed the fungus divided by the total number of pieces (Fig 1B). Values within a column followed by the same small letters are not statistically significant from each other. Main average effect values for the seedlings and strains followed by the same capital letters are not statistically significant from each other.
https://doi.org/10.1371/journal.pone.0254170.t004 SSR motifs [24]. For example, di-and tetra-nucleotide SSR motifs are less frequent in fungi [8,24]. In Fusarium, di-and tri-nucleotides SSR are mainly used to assess genetic diversity in its populations [13,16]. However, Leyva-Madrigal et al. (2014) found that penta-and hexanucleotides are the most abundant microsatellites in the F. verticillioides genome. In this study, penta-and hexa-nucleotide repeats showed higher PIC values compared with and tetra-, triand di-nucleotide ones. The number of alleles for individual SSR primer-pair ranged from 3 (Fv-312) to 9 alleles (FoAB11), with a mean of 5 alleles per locus. The number of detected alleles in this study were lower than the number of alleles obtained previously for F. verticillioides and F. oxysporum strains [8,15,19]. The explanation of getting lower number of alleles could be due to the power of SSR transferability. Based on the PIC values, Botstein et al. (1980) categorized molecular markers into three groups: (1) highly informative markers with PIC values > 0.50, (2) reasonably informative markers with PIC values between 0.50 and 0.25 and (3) slightly informative markers with PIC values < 0.25 [25]. Five of the SSR markers used in this study were reasonably informative, among Fusarium species, with PIC values between 0.43 and 0.29. FoDD7 marker was highly informative among Fusarium species. However, within F. proliferatum strains, two SSR markers were reasonably informative (Fv-47 and FoDD7) and two were slightly informative (Fv-140 and FoAB11). Other than the transferability of the SSR markers, the low PIC values obtained in this study could be due to the sample size of Fusarium species. Kalinowski (2004) reported that there is a proportional relationship between the number of alleles and population sample size [26].
When the SSR fingerprints used to construct a phylogenetic tree using UPGMA algorithm, Fusarium strains of each species clustered in separate clades supported with bootstrap values � 69%. Although F. proliferatum and F. verticillioides are closely related, SSR markers distinguished strains of the two species. SSR markers have been successfully used to distinguish many closely related species of Fusarium, including F. graminearum and F. pseudograminearum [27]; F. asiaticum and F. graminearum [28] and F. culmorum and F. crookwellense [16]. Moreover, many formae speciales of F. oxysporum were distinguishable by SSR markers [14,15].
Among various environmental factors that impact disease development in the host plants, the temperature has significant influences on the pathogen-host interaction. Many studies have shown that the temperature can affect the growth and multiplication of pathogens as well as the development of diseases they cause [29,30]. In this study, different temperatures affected the radial growth of Fusarium strains (P < 0.0001) on PDA plates, with the best growth obtained at 25˚C. Indeed, many studies showed that 25˚C was most suitable temperature for Fusarium mycelial growth [31,32]. At both low (15-20˚C) and high (30-35˚C) temperature regimes, Fusarium strains showed low average mycelial growth. At the 40˚C, no mycelial growth was observed for F. proliferatum, F. verticillioides and F. oxysporum strains. Similarly, Mogensen et al. (2009) reported that strains of five Fusarium species, including F. proliferatum, F. verticillioides and F. oxysporum, were not able to grow at temperatures � 40˚C [33]. The only Fusarium strains that showed mycelial growth at 40˚C belonged to F. solani. It is well known that strains of F. solani are the most common plant pathogens in tropical and temperate regions [17,34]. Saremi and Burgess (2000) studied the population dynamics of five Fusarium species representing different climatic conditions at three levels of temperatures and found that F. solani populations had the highest mycelial growth at high temperatures. In contrast, Ramteke [17,35]. Samapundo et al. (2005) reported that F. verticillioides has optimal temperature higher than F. proliferatum [36]. The present study showed the same trend where F. verticillioides showed faster mycelial growth at temperatures 25 and 30˚C compared with other tested Fusarium strains. Moreover, Reid et al. (1999) demonstrated that F. verticillioides can grow well above 28˚C [37]. The optimum temperature for mycelial growth of Saudi F. proliferatum recovered from date palms was 24.6˚C. Interestingly, the optimum growth temperature for F. proliferatum strains, isolated from garlic bulb rot in Egypt and from wheat grains in Argentina, was 25˚C [31,38]. However, Samapundo et al. (2005) and Marín et al. (1999) showed that the optimum growth temperature for F. proliferatum strains isolated from maize grains was 30˚C [36,39]. High temperatures are associated with higher levels of Fusarium wilt on several crops. For instance, Landa et al. (2006) noted that early seed sowing that coincided with low temperatures suppressed Fusarium wilt of chickpea, whereas late sowing that coincided with warm temperatures increased disease severity. Also, F. solani f. sp. pisi, the causal agent of chickpea root rot, was very severe on plants grown at 30˚C, whereas the plants grew at lower temperatures (10, 15, 20 and 25˚C) showed less symptoms [40]. Likewise, in a field study, F. solani colonized perennial ryegrass roots and produced more propagules under high temperature regimes [35].
Recently, F. proliferatum has become the most important pathogenic Fusarium species on date palms not only in Saudi Arabia but also worldwide [5,6,41,42]. In the present work, pathogenicity of four potent pathogenic strains of F. proliferatum and one F. oxysporum strain were conducted on date palm seedlings generated from seeds belonging to three local cultivars (Khalas, Ruziz and Sheeshee). These strains were previously isolated from diseased trees and they also showed high colonization ability on detached leaf experiments [7]. Moreover, Sharafaddin et al. (2019) showed that these four strains of F. proliferatum produced different cell wall degrading enzymes enabling them to colonize leaflet cuttings of date palm tissues. Indeed, we found that these strains caused yellowing, shrinking and drying of leaves, and eventually the death of inoculated seedlings. The symptoms resembled those produced by F. proliferatum strains on seedlings of P. canariensis [43]. In some cases, different symptoms can develop on F. proliferatum-inoculated date palm seedlings, such as leaf blight [6]. The symptoms produced by F. proliferatum on majesty palm seedlings included browning areas on unopened leaves along with reddish leaf spots [44]. Fusarium proliferatum can cause other disease symptoms on date palm trees, e.g. bunch fading, inflorescence rot and fruit spots [41,42]. The ability of F. proliferatum strains to produce a wide spectrum of disease symptoms is referred to its ability to attack a wide range of host plants [17]. Moreover, F. proliferatum strains are able to secrete a wide spectrum of mycotoxins that play essential roles in their pathogenicity and ecology [6]. Testing four potent pathogenic strains of F. proliferatum and one strain of F. oxysporum on local cultivars confirmed their virulence ability. The tested fungal strains represented different locations and were isolated from different date palm cultivars [7].

Conclusions
In conclusion, the SSR markers developed for F. verticillioides and F. oxysporum were very useful in assessing the genetic diversity of Fusarium strains collected from Saudi date palm trees. Multi-individual SSR genotypes of F. proliferatum contained strains isolated from more than one region, suggesting that infected date palm tissues and/or Fusarium-contaminated tools are moving among different regions in Saudi Arabia. The results of the temperature study showed that the Fusarium strains can grow under a wide range of temperatures. This ability enables them to survive in the extreme temperatures over different seasons. This study also provides an evidence that strains of F. proliferatum can consistently cause disease to date palm seedlings under controlled conditions.