The Fusarium solani species complex (FSSC) has been studied intensively but its association with legumes, particularly under European agro-climatic conditions, is still poorly understood. In the present study, we investigated phylogenetic relationships and aggressiveness of 79 isolates of the FSSC collected from pea, subterranean clover, white clover and winter vetch grown under diverse agro-climatic and soil conditions within Temperate and Mediterranean Europe. The isolates were characterized by sequencing tef1 and rpb2 loci and by greenhouse aggressiveness assays. The majority of the isolates belonged to two lineages: the F. pisi comb. nov. lineage (formerly F. solani f. sp. pisi) mainly accommodating German and Swiss isolates, and the Fusisporium (Fusarium) solani lineage accommodating mainly Italian isolates. Based on the results of aggressiveness tests on pea, most of the isolates were classified as weakly to moderately aggressive. In addition, using one model strain, 62 accessions of 10 legume genera were evaluated for their potential to host F. pisi, the species known mainly as a pathogen of pea. A total of 58 accessions were colonized, with 25 of these being asymptomatic hosts. These results suggest a broad host range for F. pisi and challenge the forma specialis naming system in Fusarium.
Fusarium solani (sexual morph Nectria haematococca; syn. Haematonectria haematococca) is a filamentous fungus of significant agricultural importance that has been accommodated as a single species in the section Martiella within the genus Fusarium1. Re-evaluation of species taxonomy based on molecular phylogenetic analyses has revealed that F. solani is a species complex (FSSC) which includes at least 60 distinct phylogenetic species2,3. Members of the complex are globally distributed and of considerable ecological plasticity, causing infections in both plants and humans4,5.
Phytopathogenic species within the FSSC include some of the most economically important plant pathogens associated with vascular wilts and root rots in over 100 crops6. Despite the broad host range of the complex as a whole, individual species are often associated with only one or a few plant hosts. Consequently, populations of F. solani pathogenic on plants have been divided into 12 formae speciales and two races7,8,9,10.
Early studies on sexual compatibility of special forms and races has already shown that F. solani represents at least seven biological species classified as mating populations (MPs) I-VII, with Nectria haematococca as the most commonly referred sexual morph. Biological species correlated with a host range, as successful sexual crosses were found only among heterothallic isolates within each special form or race11. However, the designation forma specialis may lead to incorrect assumptions concerning the aggressiveness and host specificity of individual isolates. For example, studies on the host range of F. solani f. sp. pisi (MP VI), named by its specific pathogenicity on pea (Pisum sativum), revealed that the species was also pathogenic on chickpea (Cicer arietinum) as well as several other non-legume hosts, such as ginseng (Panax ginseng) and mulberry tree (Morus alba)12,13. Similar results have been reported for the host range and aggressiveness of F. virguliforme (formerly F. solani f. sp. glycines) and F. solani f. sp. eumartii14,15. Thus, the term forma specialis is often misleading and will most likely need to be reconsidered in the future.
Traditional taxonomic methods for identifying special forms and races rely on morphological criteria, aggressiveness tests and sexual compatibility. They are time consuming, labor intensive and often inconclusive. The use of morphological criteria for diagnostic purposes requires extensive knowledge of classical taxonomy and still remains difficult due to overlapping morphological characters among many closely related species. In the case of aggressiveness tests on a specific host plant, the environmental factors and genetic makeup of the host may have significant influence on the bioassay outcome. Similarly, the biological species concept on the basis of sexual crosses in F. solani has several problems including unequal frequencies of mating type alleles in different populations, failure of compatible isolates to reproduce due to male or female dominance, or simply environmental conditions suppressing sexual reproduction4.
In the past 10 to 15 years, molecular phylogenetic approaches have been extensively employed to facilitate accurate species identification in the genus Fusarium3,16,17. Polymorphisms in DNA sequences of the translation elongation factor 1 alpha (tef1) and the second largest subunit of RNA polymerase II (rpb2) have provided a robust and reliable means for phylogenetic species recognition within the FSSC and the genus Fusarium18,19. These protein coding gene regions show a high level of sequence polymorphism among closely related species, do not have non-orthologous copies and can be amplified from all species of the genus using single pairs of universal primers20.
The main objectives of this study were to investigate diversity, geographical patterns of host preference and phylogenetic relationships among the FSSC isolates collected from pea, subterranean clover, white clover and winter vetch grown under diverse agro-climatic and soil conditions within Temperate and Mediterranean Europe. To that end, we conducted gene sequence analysis, aggressiveness bioassays and sought to clarify the host range of F. solani f. sp. pisi by determining the potential of various legumes to symptomatically or asymptomatically host the fungus. Based on our results, we propose a taxonomic recombination by assigning isolates of Fusarium solani f. sp. pisi to Fusarium pisi comb. nov. Therefore, in this article the fungus is being referred to as Fusarium pisi.
Phylogenetic analyses inferred from the tef1 and the rpb2 sequences resolved the phylogenetic positions of the 83 isolates studied (including F. redolens) in relation to currently recognized monophyletic species in the FSSC complex. The tef1 data set included 122 strains and consisted of 720 characters including alignment gaps, of which 220 characters were parsimony informative; the rpb2 data set included 59 strains and consisted of 910 characters with alignment gaps, of which 168 were parsimony informative, and the concatenated tef1 and rpb2 gene sequence included 56 strains and comprised 1600 characters including alignment gaps, of which 262 were parsimony informative.
The isolates studied formed four different lineages, all nested within the FSSC clade 3 (Fig. 1). According to the single locus phylogenetic analysis, based on the tef1 tree, 51 isolates were placed in two closely related subclades in the F. pisi lineage: the first group of 27 isolates matched with F. pisi comb. nov. (NRRL 22820) with a 75% bootstrap value, and the second group of 24 isolates matched with Fusarium solani (FRC S485) with a 68% bootstrap value. Based on the tree generated from rpb2 sequencing with representative isolates from both sub-clades, the topological differences did not receive any additional significant support on either rpb2 or concatenated tef1 - rpb2 trees. The same strains that previously formed two sub-clades (tef1 tree, Fig. 1) were nested in one clade that matched F. pisi (NRRL 22820) with some strains showing low intraspecific variation (Figs 2 and 3). Thus, based on the phylogenetic network obtained from concatenated gene trees, all 51 strains were assigned to a single species i.e. F. pisi. With the exception of Fs66 recovered from the roots of subterranean clover grown in Italy and Fs76 recovered from white clover roots grown in Sweden, the strains nested in the F. pisi lineage originated from different localities in Germany and Switzerland. These strains were recovered from all four legume hosts and included 25 strains from subterranean clover, 7 strains from winter vetch, 16 strains from pea, one strain from faba bean roots and one strain recovered from yard waste compost.
A group of 25 isolates were placed in the Fusarium solani sensu stricto (FSSC 5) lineage, recently assigned an epitype specimen by Schroers et al.2 to define the species Fusarium solani s. str., which was recombined by the same authors in Fusisporium as Fusisporium solani. Similar to F. pisi, the relative distances using the single and the combined gene analyses (two locus tree) revealed intraspecific variation in Fusisporium (Fusarium) solani s. str. (Figs 1–3). This lineage primarily accommodated isolates recovered from subterranean clover and winter vetch roots collected in Italy. The few exceptions included six strains collected in Germany and Switzerland, namely two pea, one cherry and two white clover strains from Germany and one subterranean clover strain from Switzerland (Figs 1–3).
The results of the tef1 tree topology also revealed one strain (FK70) matching with F. keratoplasticum (FRC S2477, Fig. 1). Strains Fs29 and Fs30, recovered from compost and hibiscus, respectively, did not cluster with any of the currently recognized FSSC species (Figs 1–3). Based on distances from the nearest neighbor species and high bootstrap support values (≥96%) between sub-clusters using both the single and the two gene phylogenetic analyses (tef1 and the rpb2 sequences), our results suggest that these two isolates represent at least one new lineage within the FSSC clade 3.
All of the tested isolates were scattered within clade 3 without evidence of phylogenetic structure with respect to host. Four isolates of F. redolens, recovered from white clover grown in Sweden were placed in the F. fujikuroi species complex and used as an outgroup for species level resolution within the tef1 locus.
Fusarium pisi (Jones) A. Šišić, J. Baćanović-Šišić, S. A. Ahmed & A. M. S. Al-Hatmi, comb. nov.
=Fusarium martii Appel & Wollenw var. pisi Jones, J. Agric. Res., Washington 26: 459 (1923) (basionym).
=Fusarium solani (Mart.) Sacc. f. pisi (Jones) W.C. Snyder & H.N. Hansen, Amer. J. Bot. 28: 740, 1941. (presently considered as F. solani f. sp. pisi).
Following descriptions were based on CBS 142372 strain (Fs21 current study) growing in the darkness at 27 °C after 7 days on PDA, MEA and CLA. Colonies grew rapidly to a final diameter of 45 mm. Observed aerial mycelium cottony, white on MEA and white to milky on PDA (Fig. 4a,b). Reverse pigmentation yellowish to orange-brown. Sporodochia emerged after 7 days of incubation as white flowers on pieces of carnation leaf placed on CLA (Fig. 4c). Abundant production of erect, mononematous conidiophores from the agar surface and the aerial mycelium was also observed. Mononematous conidiophores were acremonium like and unbranched or occasionally branched once (Fig. 4d–f). Long conidiophores ranged from 2.1 to 162.2 µm with a mean length of 45.1 µm and 4.8 µm width. Short conidiophores ranged from 3.0 to 35.2 µm with mean length of 11.2 µm and 4.0 µm width at the base, terminating into a single sub-cylindrical phialide. Tip of the phialide with inconspicuous preclinical thickening, collarette not flared (4d-4f). Macroconidia was abundant, 4.3–46.1 × 5.4–6.2 μm, 2‒4 septate, slightly curved or arcuate with a rounded apical cell and wedge-shaped, weakly pedicellate basal cell (Fig. 4g–j). Microconidia ovoidal or with a rounded apex and truncate base, 2.5–12.7 × 1.5–2.1 μm (Fig. 4d–f). Chlamydospores absent.
Cardinal growth temperature test showed that the strain evaluated in this study had optimal development at 25–30 °C attaining diameter of 17 and 21 mm on PDA and SNA, respectively. Colonies showed no visible growth at 5 °C and at 35 and 40 °C on PDA. On SNA strain was still able to grow at 5 and 35 °C but not at 40 °C.
Aggressiveness of selected FSSC isolates to pea in greenhouse assay
The severity of pea root rot averaged over isolates varied significantly among different species within the FSSC. Fusarium pisi (mean Disease Index, DI = 60.91) together with F. redolens (DI = 63.41) caused the highest overall disease severity, followed by Fusisporium solani (DI = 50.23) and the F. solani group (DI = 39.65) (Tukey HSD, P < 0.05). There were no significant differences in mean fresh or dry plant biomass of inoculated treatments and un-inoculated controls (data not shown).
Significant variation in aggressiveness was also observed among individual isolates of F. pisi, Fusisporium solani, and those included in the F. solani group (Fig. 5). The majority of the isolates tested were pathogenic to pea. In general, weakly to moderately aggressive strains dominated the populations of the tested Fusarium species. Among the 48 F. pisi isolates, three isolates recovered from subterranean clover roots (Fs24, Fs36, Fs43) and one isolate collected from compost (Fs28) did not differ significantly in root rot severity from the un-inoculated control and were classified as non-aggressive (4/48, 8%). Among the pathogenic isolates, three (6%) were highly aggressive, 24 (50%) moderately aggressive, and 17 (35%) were weakly aggressive. Similar results were also observed for Fusisporium solani (Fig. 5a). Among the 24 isolates tested, three (Fs26, Fs60 and Fs71) recovered from the roots of white clover, subterranean clover and winter vetch, respectively, were classified as non-aggressive. The remaining isolates induced mild to modest symptoms on pea roots and were differentiated into weakly (12/24, 50%) and moderately aggressive (9/24, 38%). The two F. solani isolates collected from compost (Fs29) and hibiscus (Fs30), placed into one separate clade (lineage 1, Fig. 1) were rated as non- and weakly aggressive, respectively. The F. keratoplasticum isolate recovered from winter vetch in Italy (FK70), sorted in the F. solani group in Fig. 5, was weakly aggressive. Among the F. redolens isolates, two were weakly and two were moderately aggressive. Aggressiveness was not related to isolate phylogenetic position or the host plant from which it was recovered (Figs 1–5a).
In contrast to root rot severity, the effect of the tested isolates on pea biomass was much less pronounced. Significant reductions in fresh biomass compared to the corresponding control were observed only in plants inoculated with F. pisi isolates Fs1, Fs2, Fs17 and Fs18 recovered from pea roots, as well as for isolate Fs22 recovered from subterranean clover roots (Fig. 5b). In comparison to the un-inoculated control, dry plant biomass was significantly reduced only by isolate Fs1 (data not shown).
Host range of F. pisi in greenhouse assays
Fusarium pisi was re-isolated from the surface sterilized roots in 58 out of the 62 accessions tested. Among the 58 successfully colonized accessions, 33 (including the three Pisum cultivars) were symptomatic and 25 accessions were asymptomatic hosts of F. pisi (Table 1). The fungus was not re-isolated from one T. subterraneum (acc. 1001) and three of the T. repens accessions (acc. 1965, 1968 and 2010). As assessed by plating surface sterilized root segments, these accessions were considered as non-hosts for the tested F. pisi isolate. In addition, several potentially pathogenic fungal species were isolated from the un-inoculated roots in 13 of the tested accessions: one Pisum cultivar, seven Lathyrus, four T. subterraneum and one T. repens accessions. Contaminants were usually F. avenaceum and F. oxysporum, several other Fusarium spp. and in one case Didymella pinodella (Table 1).
Inoculation with F. pisi resulted in significantly different levels of root rot disease severity among the tested legumes (Table 1). The highest overall disease severity rating (DSR) was observed on Trigonella foenum-graecum (mean DSR = 7.8), followed by Scorpiurus muricatus (DSR = 5.4), Pisum cultivars (DSR = 5.1), Melilotus albus (DSR = 4.8) Crotalaria ochroleuca (DSR = 4.0) and Lathyrus accessions (DSR = 3.4). Inoculated Vicia, Medicago, Trifolium, and Galega accessions showed lower overall disease symptoms, with mean severity ratings of 2.1, 2.0, 1.6 and 1.4, respectively. Symptoms of fungal infection were not observed in Lotus pedunculatus (Table 1).
Responses of individual accessions within Lathyrus, Medicago, Trifolium and Vicia genera to inoculation with F. pisi varied greatly (Table 1). For Lathyrus, with the exception of L. aphaca (L045), all accessions developed disease severity rating >2. Mean root rot severity ranged between 2.2 for L. gorgoni acc. L1663 and 5.2 for L. sativus acc. L1668. Similar responses were observed for Medicago accessions. Inoculated M. arabica (acc. 1735 and 211) and M. polymorpha (acc. 365) did not develop significant disease symptoms (DSR < 2), whereas M. arabica acc. 624 and M. orbicularis acc. 44 and 46 had DSR of 3.4, 2.8 and 2.3, respectively.
Significant variation in response to F. pisi was also observed for Trifolium accessions. Out of the 25 accessions tested, only seven were found to be susceptible (DSR > 2), with five belonging to T. subterraneum (DSR between 3.2 and 4.3), one to T. diffusum (acc. 906, DSR = 3.2) and one to T. palestinum (acc. 910, DSR = 4.0). The remaining 18 accessions had no symptoms or developed low levels of disease (DSR < 2). These included all T. repens accessions (n = 12), two T. subterraneum and one of each T. angustifolium, T. arvense and T. campestre.
Among the 14 Vicia accessions, six were considered susceptible based on the disease severity symptoms whereas five did not develop disease symptoms higher than two. For susceptible accessions, the mean DSR ranged between 2.2 for V. articulata acc. 924 and 3.6 for V. sativa accessions 1576 and 1577. Vicia sativa acc. 1576 and V. villosa subsp. varia acc. 1644 were also classified as symptomatic hosts in the absence of visible disease symptoms due to a significant loss of biomass (Table 1).
Low variability in response to F. pisi was found among the tested Pisum cultivars. Winter pea cv. EFB 33 (DSR = 4.1) was generally less susceptible to root rot compared to spring pea cv. Santana and cv. IPR83 (both DSR = 5.6) (Table 1).
Compared to the corresponding controls, fresh or dry plant biomass of inoculated treatments was significantly reduced only in Pisum sativum cv. IPR83, Lathyrus incospicuus acc. L1672, Trifolium diffusum acc. 906, T. palestinum acc. 910, Vicia articulata acc. 924, V. sativa acc. 1576 and 1590, V. villosa subsp. varia acc. 1644, Scorpiurus muricatus acc. 69 and Trigonella foenum-graecum acc. 409. Additionally, in some accessions, inoculation with F. pisi even resulted in a significant biomass increase (Table 1).
In the current study, the single gene phylogeny inferred from tef1 gene sequences that included 79 isolates, as well as the single rpb2 and the concatenated tef1 – rpb2 phylogenetic tree topologies inferred for a selected subset of 28 isolates placed the examined strains in four lineages within the FSSC clade 317. Most isolates were associated with two major lineages, the F. pisi (F. solani f. sp. pisi) lineage accommodating mainly German and Swiss isolates, and the Fusisporium (Fusarium) solani s. str. lineage accommodating mainly Italian isolates. The aggressiveness tests on pea that included a subset of 75 isolates confirmed the pathogenicity of most of the FSSC isolates tested. Aggressiveness was not related to isolate phylogenetic position or the host plant from which it was recovered.
The predominant share of the identified strains examined here (n = 51) belonged to F. pisi, suggesting a significant pathogenic potential of this species and/or a common prevalence in different host plants. Previous studies have established F. pisi as a primary causal agent of pea root rot and one of the main reasons for the decline of pea production worldwide21,22. While the result of the aggressiveness tests on pea in our study showed that the most aggressive isolates belonged to F. pisi, it is important to note that the tested population was dominated by weakly and moderately aggressive strains. In addition, non-pathogenic strains were also present in the population of the species. More importantly, our study showed that both, pathogenic and non-pathogenic isolates of F. pisi can be found in a variety of habitats under diverse agro-ecological conditions, and that the fungus is, in addition to pea, able to colonize roots of various hosts such as subterranean clover, white clover, winter vetch and faba bean under field conditions.
The ability of F. pisi to occupy diverse ecological niches and the significant variation in the aggressiveness of individual isolates observed in this study is consistent with the previous work of VanEtten23, who found large variation in pea plant symptom severity following inoculation with 152 F. pisi isolates collected from diverse habitats and geographical locations. While some isolates were highly aggressive to pea, the authors also reported the presence of non-pathogenic strains. Additional studies on the aggressiveness factors of F. pisi revealed that a number of enzymes released from the fungi have a major influence on their ability to cause disease. The capacity of the pathogen to degrade the phytoalexin pisatin by the activity of pisatin demethylase is one of the main determinants of its aggressiveness in pea. All naturally occurring isolates without this ability were essentially non-pathogenic24.
Out of the 26 additionally identified isolates in our study, 25 belonged to Fusisporum solani (Fusarium solani sensu stricto ‘5’), the species mainly considered as a causal agent of dry rot in potatoes and known as an opportunistic human pathogen2. However, the fungus has not been reported as part of the root rot complex of pea. As for F. pisi, our results suggest a lack of host specificity and the ability of Fusisporium (Fusarium) solani s. str. isolates to colonize various hosts under field conditions, as well as their potential to cause significant damage to pea. In addition, with few notable exceptions, the population of F. pisi isolates in our study mainly originated from German and Swiss environments, whereas the Italian isolates mainly comprised the Fusisporium (Fusarium) solani s. str. lineage. These results indicate a biogeographic distribution pattern of the FSSC species distribution with respect to the host plants.
One additionally identified isolate belonged to F. keratoplasticum, the species mainly associated with human eye infections25. Our data suggest considerable ecological plasticity of the fungus which possesses pathogenic potential on pea, and points to soil and plant debris as potential environmental sources of human infections. Included in F. solani group, the isolates Fs29 and Fs30 which were recovered from compost and hibiscus, respectively, represent at least one new phylogenetic lineage in the complex. Additional studies are needed to fully understand their ecological importance.
The greenhouse data on the host range of F. pisi further supports our observations that the species is not adapted to a particular host. Data from the single isolate inoculation indicate that the host range of F. pisi should be expanded to include 33 symptomatic and 25 asymptomatic hosts. Only one accession of subterranean clover and three accessions of white clover tested could be classified as non-hosts for F. pisi isolate Fs21 pointing to potential sources of resistance. A broader host range has been demonstrated for other special forms of F. solani previously. Studies on F. solani f. sp. eumartii, named by its specific pathogenicity to potato, revealed that the species was also pathogenic to pepper, eggplant and tomato15. Besides the Solanaceae family, the pathogen also infected maple (Acer sp.) and citrus (Citrus sp.) trees26. Similarly, F. phaseoli comb. nov27 (formerly F. solani f. sp. phaseoli) generally considered as a root rot pathogen of bean (Phaseolus vulgaris), has been associated with at least four other legume host plants28. In an extensive survey of F. solani f. sp. glycines associated with sudden death syndrome of soybean, Aoki et al.27 found that the disease is in fact caused by two phylogenetically and morphologically distinct species, F. virguliforme (formerly F. solani f. sp. glycines) and F. tucumaniae in North and South America, respectively. More recent studies conducted by Kolander et al.14 showed that F. virguliforme can cause disease in a range of legumes and non-legumes including alfalfa, red clover, white clover, pea, bean, sugar beet and canola. In addition, the authors demonstrated the ability of F. virguliforme to asymptomatically infect wheat, maize, ryegrass and lambsquarters, which are commonly grown in rotations with legumes. More recently, two new special forms have been assigned to the FSSC, namely F. solani ff. sp. passiflorae7 and phalaenopsis8. However, the authors did not provide sufficient molecular data that would allow comparison of their strains with already existing species within the FSSC, and based their results solely on the pathogenicity to the specific host evaluated on a narrow range of closely related plant species. Thus, whether such problems exist in these and other special forms within the complex remains to be investigated.
In addition to causing economically important crop diseases, Fusarium spp. in general and the members of the FSSC in particular that were previously considered solely as plant pathogens are being increasingly reported as causal agents of superficial and systemic infections in humans and animals5,29. For example, these so called trans-kingdom fungal pathogens have been demonstrated for F. oxysporum f. sp. lycopersici, Fusisporium solani s. str., F. keratoplasticum and F. falciforme2,25,30,31. Similarly, F. pisi has recently come into focus as a species of clinical relevance associated with human eye infection in the Netherlands (Al-Hatmi, unpublished data). This confirms that the species is adapted to many different habitats and supports the idea that Fusarium spp. can serve as a model for studying trans-kingdom pathogenicity in fungi30.
Our results provide new insights into the diversity of the FSSC associated with the legume host plants in Europe and provide another example for a wide host range of a single lineage within the complex. Currently the concept of forma specialis represents an informal rank in taxonomic classification3,32, and may deserve revision and formal taxonomic treatment.
Material and Methods
A total of 79 FSSC isolates were collected for this study (Table 2). Among these, 18 isolates were recovered from pea (Pisum sativum), 39 from subterranean clover (Trifolium subterraneum), 3 from white clover (T. repens) and 14 from winter vetch (Vicia villosa). The isolates originated from Germany (n = 28), Switzerland (n = 24), Italy (n = 21) and Sweden (n = 1). Additionally, we included one isolate from faba bean (V. faba), and expanded the isolates associated with legumes with two isolates collected from compost and two (one of each) recovered from an infected hibiscus (Hibiscus sp.) and cherry tree (Prunus sp.), all from Germany (Table 2). All isolates were collected in the period between 2009 and 2016, morphologically identified as F. solani and maintained as single-spore cultures at the Internal Culture Collection of the Ecological Plant Protection Department at University of Kassel. Four isolates of F. redolens, recovered from white clover grown in Sweden, were additionally included in this study.
DNA extraction, PCR amplification and sequencing
Genomic DNA was extracted from cultures actively growing on half strength PDA agar plates (½ strength PDA; 19 g/l Difco PDA and 10 g/l agar) using the CTAB (cetyltrimethylammonium bromide) protocol described by Doyle and Doyle33. All DNA were diluted 20 times in milli-Q water and stored at −20 °C before use.
A portion of the translation-elongation factor 1 alpha (tef1) gene was amplified using primer pairs EF1 and EF234. Based on the tef1 tree topology, 28 strains were selected and the second-largest subunit of RNA polymerase II (rpb2) was amplified using primers RPB2-5F235 and fRPB2-7cR36. Each polymerase chain reaction (PCR) had a total volume of 50 μl and contained 1 μl of diluted genomic DNA, 10× TrueStart Hot Start Taq Buffer [200 mM Tris-HCl (pH 8.3 at 25 °C), 200 mM KCl, 50 mM (NH4)2SO4], 2.5 mM MgCl2, 0.2 mM of each of the dNTP, 0.4 µM of each primer, and 1 unit TrueStart Hot Start Taq DNA Polymerase (ThermoFisher Scientific, Darmstadt, Germany). The PCR reactions were performed in a Biometra TAdvanced Thermal Cycler (Applied Biosystems, Foster City, California, USA). Conditions for amplification for the tef1 gene region were an initial denaturation step of 3 min at 95 °C, followed by 30 cycles of denaturation (95 °C for 30 s), annealing (53 °C for 30 s) and elongation (72 °C for 45 s). The final elongation step was conducted at 72 °C for 7 min. For the rpb2 loci amplification consisted of 5 cycles of 45 s at 94 °C, 45 s at 60 °C and 2 min at 72 °C, then 5 cycles with a 58 °C annealing temperature and 30 cycles with a 54 °C annealing temperature37.
Amplicons were purified using the DNA Clean & Concentrator kit (Zymo Research, Freiburg, Germany) according to the manufacturer’s instructions and sequenced in both directions either by Eurofins Genomics (Ebersberg, Germany) or by MacroGen (Amsterdam, Netherlands) with the above-mentioned primers.
Obtained row sequences were assembled and errors identified and corrected in MEGA v638. Partial sequences of tef1 and rpb2 were used as queries for the Fusarium-ID v. 1.0 database20, and the Fusarium MLST databases (http://www.cbs.knaw.nl/fusarium ref.39) to confirm the taxonomic assignments of the isolates. The sequences were aligned with sequences of reference strains retrieved from the GenBank using MAFFT v.7 (http://mafft.cbrc.jp/alignment/server/index.html ref.40) and the alignments were adjusted manually with MEGA v6. The phylogenetic analyses, including the majority of known Fusarium species within the FSSC, was performed on tef1 and rpb2 sequences separately as well as on a combined data set (28 selected strains). The strains used in these analyses and GenBank accession numbers of the sequences are listed in Table 2 and Table 3. A bootstrapped Maximum-Likelihood (ML) analysis was performed using the RAxML-VI-HPC v. 7.0.3 with non-parametric bootstrapping and 1000 replicates implemented on the Cipres portal (http://www.phylo.org ref.41). For the outgroup purposes, the F. redolens and F. thapsinum (H05-557S-1 DCPA and CBS 130176) were used to generate the phylogenetic trees.
Growth rates and morphological characterization of Fusarium pisi (F. solani f. sp. pisi)
The isolate Fs21 (CBS 142372) collected from the roots of subterranean clover was chosen as the representative strain to study morphological characters. Cardinal growth rates were determined at eight temperatures (5–40 °C) at 5 °C intervals in darkness following methods adopted from Nalim et al.31. Briefly, a five mm diameter plugs were taken from the actively growing edge of a 15 days old colony cultured on potato dextrose agar (PDA; 39 g/l Difco), and placed mycelium side down one cm from the edge of a fresh PDA and synthetic nutrient-poor agar42 (SNA). Colony radius was measured after five days from the edge of the inoculum plug to the most distant part of the colony. Average growth rates were calculated from three replicate plates for each respective temperature and expressed as diametric growth per 24 h.
Plates for colony morphology and colors were prepared by placinga five mm agar plug in the center of PDA, malt extract agar (MEA, Oxoid, UK) and carnation leaf agar4 (CLA). Culture plates were grown at 27 °C in the dark and examined seven days after inoculation. For microscopic observations, a block of ca. one cm SNA agar was cut and placed on a microscopic slide, inoculated with the fungus and covered with a No. 1 cover glass. Slides were examined in a drop of lactic acid with cotton blue, and the pictures were taken with a Jenoptik ProgRes® digital camera (JENOPTIK, Germany) attached to a Zeiss Axiosskop2 plus microscope. A minimum of 20 measurements were made per structure using the CapturePro 2.8 (JENOPTIK, Germany) software.
Aggressiveness of selected FSSC isolates to pea
To compare aggressiveness (relative ability of the pathogen/isolate to colonize and cause damage to plants43) and to determine whether the FSSC strains from non-pea hosts are capable of causing disease on pea, a total of 75 isolates were tested in a greenhouse assay. The aggressiveness test included 48 isolates of F. pisi, 24 isolates of Fusisporium solani, 2 isolates of F. solani, and 1 isolate of F. keratoplasticum. In this study, the isolates that formed distinct groups based on the phylogenetic analysis and showed no strong phylogenetic relationship to any of the previously defined species within the FSSC were included in the F. solani group. Four isolates of F. redolens were also included in this experiment. The geographic origin and the host plants from which the isolates were collected are given in Table 2.
To prepare inoculum, each Fusarium isolate was cultured on ½ strength PDA at room temperature under alternating cycles of 12 h blacklight blue (BLB) fluorescent light (F40; range 315–400 nm with the peak at 365 nm) and 12 h darkness. After 15 days, spores were washed with sterile distilled water and enumerated in the suspension with a Fuchs Rosenthal hemocytometer.
Seeds of field pea cv. Santana were surface sterilized in 70% ethanol for 5 five minutes and rinsed with distilled water prior to planting. Four pea seeds (germination rate of 98%) were then planted into 500 ml pots filled with autoclaved sand, and 2 × 104 spores g−1 substrate of the respective isolate was applied to each pot. Un-inoculated controls were mock inoculated with sterile distilled water. Four replicate posts were sown per treatment and arranged in a completely randomized design. Experimental plants were kept in the greenhouse at 19 °C day and 16 °C night temperature. Natural day light was additionally supplemented with high-pressure sodium lamps (400 W) in order to provide a photoperiod of 16 h light day−1. Plants were watered daily with tap water and additionally fertilized with complex N:P:K fertilizer Wuxal Super (8:8:6 + microelements). A total of 120 mg of N l−1 of substrate was divided into four portions and given over the course of the experiment.
After 42 days of growing, plants were removed from pots, and the roots were separated from the shoots. Above ground plant parts of each pot were weighted and dried at 105 °C until constant weight was attained. Roots were washed under running tap water, and root rot severity was assessed using a visual 0–8 score scale based on external and internal root tissue discoloration levels adopted from Pflughöft44. The external disease severity was rated as follows: disease severity rating (DSR) 0 = no symptoms, 1 = streaks at the transition zone, epicotyl or hypocotyl, 2 = brown lesion cover up to 50% of root perimeter, 3 = brown-black lesion cover 51 to 99% of root perimeter, 4 = black lesion cover 100% of stem perimeter, 5 = black lesion spread up to 30–49% of the tap root, 6 = black lesions spread up to 50 to 70% of the tap root, 7 = black lesions spread >70% of the tap root, 8 = dead plant. The roots were then cut transversally across the lesions and internal disease severity was rated, where 0 = no visible symptoms, 1 = epidermis/rhizodermis is brown to black, 2 = brown discoloration of cortical tissues, 3 = cortical tissues is partially black, but the center and endodermis are still healthy, 4 = cortex tissue is completely black, 5 = cortex tissue begins to rot (bursting of epicotyl or rhizodermis on the root), 6 = cortex tissue is completely rotten, 7 = shedding of the cortex tissue and endodermis, and 8 = dead plant. Consequently, a disease severity index (DI) between 0 and 100 was calculated for each pot using equation (1):
where, SR = Mean external and internal disease severity rating (DSR), NR = Number of infected plants having that DSR, Nt = Total number of plants assessed, MR = Maximum rating scale number.
Four distinct aggressiveness classes were then assigned relative to the un-inoculated control and based on the gradual increase of severity of symptoms following inoculation, where: DI = 0–36 – non-aggressive; DI = 37–55 – weakly aggressive; DI = 56–85 – moderately aggressive; and DI = 86–100 – highly aggressive. A threshold disease index of 36 for classifying an isolate as aggressive was chosen because factors other than inoculation caused low levels of root discoloration in un-inoculated control plants. Up to this level the DI of inoculated treatments was in the same range as the DI of the un-inoculated control; e.g. mean DI of un-inoculated controls was 24 (±12) while the mean DI of inoculated treatment was ≤36. In addition, up to this level there was no statistically significant difference in DI of inoculated treatments and un-inoculated control.
Twenty one different inoculation treatments were selected at random and the fungi were re-isolated from the surface sterilized roots (1% NaOCl, 3 roots per treatment) and identified morphologically to confirm recovery of the isolate.
Evaluation of host range of Fusarium pisi
To determine the host range and evaluate plant response to inoculation with F. pisi, 60 accessions of 10 legume genera were tested in a greenhouse assay. This study was conducted over a set of four consecutive experiments. In each experiment two field pea cultivars, cv. Santana and cv. EFB 33, were included as additional controls (Table 1).
The F. pisi isolate (Fs21) classified as moderately aggressive to pea in experiment 1 was selected for the inoculation experiments. The inoculum was prepared by incubating the strain for 10 days in aerated malt extract broth (MEB, 17 g/l) at 20 °C under constant agitation/shaking at 100 rpm. After 10 days of incubation, conidia were collected by filtration and enumerated in suspension as described above.
Preliminary studies on seed germination using untreated seeds showed that the majority of accessions chosen for this experiment had a very low germination rate. Thus, to ensure adequate seedling emergence, seeds of all plant accessions, with the exception of pea, were treated with 97% sulfuric acid for 4 min, rinsed in distilled water and germinated for 48 h on wet filter paper in Petri dishes at room temperature. Pea seeds were treated with 70% ethanol prior to placing on wet filter paper. Single pre-germinated seeds were then transplanted into 200 ml pots filled with autoclaved sand. Each treatment consisted of five replicates with one germinated seed sown per pot. The experiment was arranged in a completely randomized design and the pots were inoculated 24 h after transplanting with 2 × 104 spores g−1 substrate. Plants were kept in the greenhouse for five weeks under the conditions described for experiment 1. After five weeks of growing, plants were harvested and the biomass and disease severity (external root tissue discoloration levels only) were assessed as described above.
Cultural methods in combination with disease severity data were used to determine the host range of F. pisi on tested plants. Three randomly selected roots from each treatment were surface-sterilized in 0.5% NaOCl for 10 s, thoroughly washed in distilled water and placed on filter paper under a laminar flow hood for 1 h to dry. Subsequently, the roots were cut into approximately 1 cm long fragments and placed in Petri dishes containing ½ strength PDA medium and incubated under alternating cycles of 12 h BLB fluorescent light and 12 h darkness. After 10 to 15 days of incubation, fungal colonies developing from the root pieces were sub-cultured separately in Petri dishes containing PDA and SNA agar, incubated as described previously and identified based on cultural characteristics and microscopic examination of conidiogenous cells4.
The response of the tested legume species to F. pisi was determined according to criteria adopted in a slightly modified form from Kolander et al.14. The accessions were considered symptomatic hosts if the inoculated isolate was re-isolated from surface sterilized roots, the average disease severity rating was higher than 2 and significantly greater than in un-inoculated control plants. Accessions were also considered symptomatic if DSR < 2 but there was a significant reduction in mean plant biomass of inoculated treatment compared to the corresponding control. Some of the control plants showed moderate symptoms on the roots (mean DSR > 2) caused by factors other than inoculation with F. pisi, and in this case the host was considered susceptible if the final disease severity level of inoculated treatments was significantly higher than that of the corresponding control plants. The accessions were considered asymptomatic hosts if disease rating was less than or equal to 2, there was no significant reduction in biomass, and the fungus was re-isolated from the root parts following surface sterilization.
All statistical analyses were done using R statistical software45 (version 3.3.1). For aggressiveness assay of selected FSSC isolates to pea, the normality of data distribution and homogeneity of variances were tested by Shapiro-Wilks-W-Test and Levene’s test, respectively. Prior to statistical analysis, disease severity index (DI) values were square root transformed. The data were first subjected to one way ANOVA to analyze differences in mean effects of different phylogenetic groups on root rot severity and plant biomass. Mean separations were made by Tukey HSD test. Differences among single isolates were tested separately by comparing means from inoculated treatments (each isolate) and un-inoculated control using Dunnett’s t test46. Treatments were considered significantly different if P ≤ 0.05.
Due unequal variances of many groups, the data on the host range of Fusarium pisi were subjected to nonparametric analyses. Differences in root rot severity ratings and biomass of inoculated treatments and corresponding un-inoculated controls were compared with 2 by 2 comparisons using the non-parametric ranking procedure of the Dunn’s test47.
The partial translation elongation factor 1-alpha (tef1) and the second-largest subunit of RNA polymerase II (rpb2) sequences of all strains used in this study were submitted to GenBank database. All fungal and plant materials are maintened at University of Kassel and are available for resesearch purposes upon request (representative fungal strains were deposited in the culture collection of the CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank Jason Wood for help with proofreading the manuscript. This research was supported by University of Kassel Zentralen Forschungsförderung (ZFF) grant number 1930 and the European Union FP7 Project n.289277: OSCAR (Optimizing Subsidiary Crop Applications in Rotations).