Introduction

The plant pathogenic fungus Bipolaris sorokiniana (Sacc.) Shoem. causes various diseases like seedling blight, spot blotch, root rot, and black point of wheat, barley, and other cereal and grasses (Kumar et al. 2002; Duveiller et al. 2005). Among them, spot blotch affects about 25 million hectares of wheat in the warmer regions of Bangladesh, Nepal, Bolivia, eastern India, Brazil, southeast China, southeast Australia, northeast Argentina, Paraguay, Zambia, northern Kazakhstan, and the Great Plains of the USA and Canada (Singh et al. 2016). Although resistant varieties and fungicide treatments have been used to control the disease, severe yield losses due to spot blotch have been documented by several reports in warm and humid areas of South Asia and South America (Acharya et al. 2011; Singh et al. 2016; Sultana et al. 2018). B. sorokiniana may produce pathogenicity factors including phytotoxins and hydrolytic enzymes to induce necrosis in plant tissues (Han et al. 2010). The phytotoxin sorokinianin produced by B. sorokiniana inhibited the germination of barley seeds, but its pathogenicity mechanism is still unknown (Nakajima et al. 1998).

The necrotrophic effector ToxA, firstly identified from Pyrenophora tritici-repentis, is able to confer fungal pathogenicity in wheat genotypes with the corresponding susceptible factor Tsn1 (Ciuffetti et al. 1998; Faris et al. 2010). Isolated or purified ToxA treatment in susceptible wheat leaves results in thylakoid disruption and chlorophyll loss, leading to decrease in photosystem II (PS II) activity (Manning et al. 2007). Friesen et al. (2006) showed evidence that ToxA was obtained by P. tritici-repentis from Parastagonospora nodorum via horizontal gene transfer, leading to the emergence of a highly virulent pathogen that causes tan spot worldwide. Later, ToxA was found in a third fungal pathogen Phaeosphaeria avenaria f. sp. tritici, and it was evidenced that interspecific hybridization between this pathogen and its sister species P. nodorum played the key role for the transfer of ToxA (McDonald et al. 2012). Recently, ToxA was identified in a fourth fungal pathogen, B. sorokiniana, from its populations in Australia (McDonald et al. 2018), the USA (Friesen et al. 2018), and India (Navathe et al. 2020). Comparative analysis of a conserved ~ 14 kb regions surrounding ToxA between three wheat pathogens (P. nodorum, P. tritici-repentis, and B. sorokiniana) provides the evidence that terminal inverted repeats flank the ToxA region, and this complete transposon features may facilitate horizontal gene transfer between fungal genomes (McDonald et al. 2019).

A resistance-like gene Tsn1 on wheat chromosome arm 5BL is required for ToxA sensitivity, conferring disease susceptibility to fungal pathogens harboring ToxA (Adhikari et al. 2009; Faris et al. 2010). Tsn1 does not interact directly with ToxA, suggesting that Tsn1 may mediate ToxA recognition or transmembrane transport (Manning and Ciuffetti 2005). Once internalized into cells, ToxA localizes to cytoplasmic compartments and chloroplasts only in susceptible wheat genotypes and inhibit a chloroplast protein ToxABP1 (ToxA binding protein 1), which may be important in assembly or disassembly of functional PSII complexes (Manning et al. 2007).

Germplasm screening for spot blotch resistance has been performed at the International Maize and Wheat Improvement Center (CIMMYT) since 1990s (Singh et al. 2015), and hence, isolates of B. sorokiniana have been collected from different regions in Mexico. The present study reports, for the first time, the occurrence of ToxA in the Mexican population of B. sorokiniana.

Materials and methods

Isolate collection

Wheat, durum, and barley fields in different locations in Mexico were scouted for spot blotch–infected leaves that had distinct necrosis symptoms for the isolation of B. sorokiniana isolates, following the methods described in Gilchrist-Saavedra et al. (2006). The leaf samplers were cut into pieces of 1 to 2 cm and surface sterilized with 10% NaClO, which were then incubated on a petri dish with moistened Whatman filter paper for 2–3 days at 23 °C. A single spore was transferred onto 30% V8 medium (300 mL V8 juice, 700 mL distilled water, 3 g calcium carbonate, and 20 g agar) to induce mycelial and subsequent conidia production. Five to 7 days later, the medium with conidia and mycelium were cut into pieces and stored at − 20 °C after drying in room temperature.

Fungal DNA extraction and ToxA gene amplification

The liquid 2YEG medium (2 g yeast extract, 10 g glucose, 1 L distilled water) cultures for mycelium production of B. sorokiniana were prepared according to Miki et al. (2009). The fungal mycelia were collected by vacuum filtration through two layers of Whatman filter papers and were then ground to a fine powder in Eppendorf tubes for extracting DNA with CTAB method (Nicholson et al. 1998). About 600 μL extraction buffer (1.4 M NaCl, 2% (wt/vol) CTAB, 0.1 M Tris-Base pH 8.0, 0.02 M EDTA pH 8.0, 1% (wt/vol) PVP 40000) was added to the tubes with fungal samples and incubated for 10 min at 65 °C. After centrifuging for 10 min at 7200g, the supernatant was transferred into a new tube and 50 μL 10% CTAB solution (10% (wt/vol) CTAB, 0.7 M NaCl) was added. An equal amount of chloroform: isoamyl alcohol (24:1) was added and gently stirred for 20 times. After centrifuging for 10 min at 7200g, supernatant was collected to a new tube and three volumes of precipitation buffer were added. Genome DNA should be visible at the bottom of the tube after centrifuging for 15 min at 7000g. The precipitate was rinsed with 70% (vol/vol) ethanol twice and dried at room temperature until the ethanol completely evaporated. DNA samples were then dissolved in 100 μL TE buffer and quantified using a NanoDrop spectrophotometer (ND-1000, Thermo Scientific Inc.). Primers ToxA1 and ToxA2 designed by Friesen et al. (2006) were used to amplify ToxA coding regions. Each PCR reaction contained 1 × REDTaq® ReadyMix™ PCR reaction mix (Sigma-Aldrich, USA), 1 μM forward/reverse primer, and 50 ng genome DNA. The amplification parameters were as follows: 3 min at 94 °C followed by 30 cycles of 30 s at 94 °C, 60 s at 60 °C, 40 s at 72 °C, with a final extension of 5 min at 72 °C. The PCR product was detected by electrophoresis on a 2% agarose gel, stained with EnviroSafe® DNA/RNA stain and photographed under UV light.

Fungal culture filtrate assay

The culture filtrates of B. sorokiniana isolates were produced in liquid Fries medium following the method used by Liu et al. (2004). For filtrates of each isolate, the mycelium was removed by vacuum filtration through two layers of Whatman filter papers and resultant culture filtrates were then passed through a Millipore Durapore 0.45-μm filter. Culture filtrates were adjusted to pH 6.8 and stored at 4 °C overnight. Salts and other precipitates were removed from filtrates by centrifugation at 12,000g for 5 min. Culture filtration assay followed the protocol described by Liu et al. (2004). Twenty BsToxA+ and 10 BsToxA− isolates identified in this study, along with a P. tritici-repentis isolate MexPtr1 and a P. nodorum isolate MexSn4 (Hu et al. 2019), both being ToxA producers, were evaluated in this assay. Pure ToxA solution provided by Dr. Friesen, USDA-ARS, Fargo, USA, and 2YEG medium were used as positive and negative controls, respectively. Wheat varieties Glenlea with Tsn1 (Tsn+) and Erik with tsn1 (Tsn−) were grown in plastic pots with two replications, each having three to four plants, for culture filtrate assays against each isolate. Infiltration took place at 2 weeks after sowing, when two leaves of each seedling were infiltrated with about 20 μL of culture filtrate. Reaction types were scored at 5 days after infiltration as sensitive (with necrosis) or insensitive (without necrosis).

Results

Collection of B. sorokiniana isolates

A total of 196 isolates were collected from experimental stations and farmers’ fields in nine states of Mexico. Nearly half (92) of the isolates were collected from the state of Mexico, followed by Oaxaca (33), Veracruz (29), and Tlaxcala (21), whereas only a small number of isolates were collected from the remaining five states (Fig. 1). Collection of the isolates spanned from 1992 to 2017, with 34 isolates collected in 1990s, 72 in 2000s, and 90 in 2010s; but a big gap was identified between 1994 and 2007, during which no isolate was collected (Table S1). Majority (168) of the isolates were collected from wheat, with only 20 from durum and 8 from barley (Table S1).

Fig. 1
figure 1

Geographical distribution of the 196 Mexican Bipolaris sorokiniana isolates used in this study. Numbers in parentheses indicate the numbers of isolates collected from the state

Full size image

Identification of ToxA in the B. sorokiniana collection

Based on the PCR results of the ToxA-specific marker, 20 of the 196 B. sorokiniana isolates turned out to be ToxA carriers (Table 1, Fig. 2), exhibiting an occurrence frequency of 10.2%. All the ToxA-positive (ToxA+) isolates were collected from wheat, and none from durum or barley. Geographically, these isolates were distributed in six out of the nine states surveyed. Among the states with more than 20 isolates collected, Veracruz showed the highest frequency (10 out of 29, 34.5%), followed by the state of Mexico (6.5%) and Oaxaca (3.0%), whereas no ToxA+ isolate was found in Tlaxcala (Table 1). Temporally, ToxA+ isolates were identified in six out of the 12 years surveyed, with a higher frequency of 17.4% in isolates collected in 1990s and lower frequency of 6.4% in the ones collected after 2007. The earliest ToxA+ samples were CIMFU472 and CIMFU475, which were collected from Poza Rica, Veracruz, in 1993. In terms of individual survey events, the highest frequency of ToxA+ isolate was found in Cotaxtla, 2017 (50%); but considering the fact that three of the positive isolates were from a same sample CIMFU491 (Table 1), the actual frequency must be lower. Generally, around 30% of the collected isolates in those survey events were ToxA+ (Table 2).

Table 1 Information of the 20 Mexican ToxA-positive Bipolaris sorokiniana isolates identified from the CIMMYT collection
Full size table
Fig. 2
figure 2

PCR amplification of ToxA in Bipolaris sorokiniana isolates with (lanes 1–10) and without (11–20) the gene. M stands for DNA ladder, P for a Pyrenophora tritici-repentis isolate used as the positive control, and N for a Fusarium graminearum isolate used as negative control

Full size image
Table 2 Frequencies of ToxA-positive (ToxA+) Bipolaris sorokiniana isolates among their sister isolates collected in the same survey event
Full size table

Culture filtrate assay

All filtrates of the 20 ToxA+ isolates produced clear necrosis symptom (Fig. 3) on the Tsn1 carrier Glenlea, whereas those of the 10 ToxA- isolates did not. As expected, infiltration with ToxA and filtrates of MexPtr1 and MexSn4 induced necrosis reaction on Glenlea, whereas the 2YEG medium did not cause the necrosis reaction. None of the treatments caused necrosis on the tsn1 carrier Erik.

Fig. 3
figure 3

Symptoms of Glenlea (left) and Eric (right) after filtration with the ToxA-positive Bipolaris sorokiniana isolate CIMFU1357

Full size image

Discussion

In this study, we investigated the occurrence of ToxA in a panel of the Mexican population of B. sorokiniana collected from nine states of Mexico, spanning a period of 1992 to 2017. However, the sampling was imbalanced geographically, which could be ascribed to the limited overlap of regions where the three cereal crops are grown and where climate conditions are favorable for spot blotch infection. For example, no isolate was collected from Sonora and Baja California, the two main wheat producers of Mexico with around 50% of wheat production of the country, due to the dry climate unfavorable for spot blotch. On the other hand, the state of Veracruz has a hot and humid climate conducive for spot blotch, but its wheat acreage is almost negligible (SAGARPA 2016), making isolate sampling a very difficult task there. Nevertheless, 196 isolates of B. sorokiniana were collected from wide geographical locations, representing well the spot blotch epidemic regions in Mexico. The 20 ToxA+ isolates identified in this study also showed a wide distribution, indicating that such isolates present in most, if not all, Mexican B. sorokiniana populations.

Previous studies indicated high frequencies of ToxA+ isolates of B. sorokiniana in the USA (86.7%) and India (70%) (Friesen et al. 2018; Navathe et al. 2020). In Australia, however, the frequency was 34.3% that was dramatically lower compared those in the USA and India, as well as to the nearly 100% occurrence of ToxA in Australian P. nodroum and P. tritici-repentis populations (Antoni et al. 2010; McDonald et al. 2013; McDonald et al. 2018). In our study, however, the overall frequency was much lower at 10.2%. Only in Veracruz with conducive environmental conditions for spot blotch, the frequency (34.5%) was similar to that in Australia. The low frequency of ToxA+ isolates in Mexico was further evidenced in individual survey events, where the corresponding percentages were around 30% (Table 2). Therefore, it might not be risky to conclude that majority of the Mexican B. sorokiniana isolates do not carry ToxA. This imply a fact that the Mexican B. sorokiniana population might have not been subjected to strong selection against the susceptible gene Tsn1 in wheat. This might have been caused by the wide cultivation of CIMMYT wheat germplasm, which has high frequency of tsn1 (S. Dreisigacker, personal communication, 2020), making the ToxA+ isolates do not have additional competitive advantage over the ToxA− isolates.

Given the effects of ToxA on the fungal pathogenicity of B. sorokiniana, P. tritici-repentis, and P. nodorum, it is recommended to investigate its prevalence in global populations of the three pathogens, especially in the wheat production regions of central Asia, where less information is available. If the frequency of ToxA+ isolates is high in a region, breeding efforts should be paid to reduce the frequency of Tsn1 in local varieties, to reduce the risk brought by this susceptible factor against the three diseases. Nevertheless, we admit that resistance to spot blotch involves a range of genes/QTL beyond Tsn1 (Singh et al. 2018), and extensive breeding work is required to achieve satisfactory varietal resistance.