Introduction

The purple spot disease or Stemphylium leaf spot of asparagus (Asparagus officinalis) has become a significant problem in all asparagus growing regions worldwide (Suzui 1973; Menzies 1980; Lacy 1982; Blancard et al. 1984; Falloon et al. 1984; Gindrat et al. 1984), and become a serious problem in Germany in the late 80s (Menzinger and Weber 1990). Due to the premature defoliation of fern in autumn and the resulting reduced photosynthetic potential, up to 52 % yield loss can be caused (Bansal et al. 1992).

Stemphylium Wallr. is a genus of filamentous ascomycets with S. botryosum Wallr. (Teleomorph: Pleospora tarda E.G. Simmons) as the type species. The most important morphological characteristic to distinguish Stemphylium from the closely related genera Ulocladium and Alternaria is the pre-currently proliferating conidiophore (Simmons 1969). There are more than 30 recognizable species known for the genus Stemphylium (Câmara et al. 2002).

S. vesicarium (Wallr.) E.G. Simmons (Teleomorph: Pleospora allii), which is also known to cause the brown spot disease of pears, was identified also as causal agent of the purple spot disease of asparagus in the USA (Michigan (Hausbeck et al. 1997; Meyer et al. 2000), California (Falloon et al. 1987) and Washington (Johnson and Lunden 1984)), Australia (Cunnington and Irvine 2005), South Africa (Thompson and Uys 1992) and New Zealand (Bansal et al. 1992). The closely related species S. botryosum was detected in asparagus samples from Germany (Leuprecht 1988; Neubauer 1998), Japan (Suzui 1973) and Greece (Elena 1996). For many years, the identification of Stemphylium species was mainly based on morphological characteristics, such as conidial shape, size, length/width ratio, colour, ornamentation and septation (Simmons 1967, 1969, 1985). Many of these characters overlap among related species, making determinations on species level difficult and in some cases incorrect (Cunnington and Irvine 2005; Shenoy et al. 2007; Wang et al. 2010). However, S. botryosum seems to be highly variable on plant material and in pure culture. Stemphylium species are also known to show high conidial variability at different temperatures and on different substrates (Leach and Aragaki 1970; Leuprecht 1990). With durations of 3 months to form fertile ascospores for S. vesicarium and 8 months for S. botryosum (Simmons 1985; Chaisrisook et al. 1995), this morphological characteristic is not suitable for rapid species identification. As the two species may differ in etiology, the identification of the causal agent is crucial for the development of a suitable pest management.

Sequences of the internal transcribed spacers (ITS) of the nuclear ribosomal DNA and protein coding genes, for example gapdh, are important molecular markers in phylogenetic analyses of fungi (White et al. 1990; Begerow et al. 2010; Schoch et al. 2012). Besides the monophyly of the genus Stemphylium, taxonomic relations could be verified using ITS regions and gapdh gene sequences (Câmara et al. 2002; Inderbitzin et al. 2009; Köhl et al. 2009; Wang et al. 2010).

Furthermore, considerably large amounts of sequence data are available of the mitochondrial coded cytochrome b gene (cyt b), which was frequently analysed in regard to QoI resistance (Grasso et al. 2006; Sierotzki et al. 2007; Stammler et al. 2013). Due to this fact, the resolving power of this region as taxonomic marker could be verified for many agronomic important fungi. Besides the relatedness within the order Uredinales (Grasso et al. 2006), species identifications in the genera Monilinia and Phyllosticta (Miessner and Stammler 2010; Stammler et al. 2013) were done using cyt b sequences.

The aim of this study was: (i) to develop a molecular and genetic based differentiation method for S. vesicarium from S. botryosum, and (ii) to analyze asparagus samples from Germany with this method to identify the prevalent causal agent of the purple spot disease in Germany. The sequences of three molecular markers (ITS, gapdh and cyt b) were used to distinguish the two described causal agents (S. vesicarium and S. botryosum) of the purple spot disease of asparagus. In addition, we developed a method based on sequence aberrations in the cyt b gene, suitable for quantification of both pathogens in field samples.

Materials and methods

Origin of isolates

Four S. vesicarium and five S. botryosum isolates were obtained from international culture collections. Strain 224 is the type of S. botryosum, which was also used in the analyses of Köhl et al. (2009). Another eight Stemphylium isolates provided by Zapf et al. (2011) and two Stemphylium isolates provided by A. Wichura / R. Weber (Agricultural Chamber of Lower Saxony), which were all isolated from asparagus, were also analysed. Details of the used isolates are given in Table 1. Isolates of related Stemphylium species, like S. herbarum, S. alfalfa, S. sedicola and S. tomatonis, were not included as these species are from different hosts and assumed to be identical (synonymous) to S. vesicarium (Câmara et al. 2002; Inderbitzin et al. 2009; Köhl et al. 2009).

Table 1 Stemphylium Isolates investigated in this study
Full size table

Analyses of the DNA marker regions

Fungal mycelium (~25 mg) from 14 days old cultures grown on potato dextrose agar (PDA) was used for DNA and RNA extraction using the Nucleospin® Plant II Kit and Nucleospin® RNA Kit (Machery and Nagel, Düren, Germany), respectively. According to the manufacturer’s intructions, reverse transcription of the RNA for the cDNA synthesis was made with the Verso cDNA kit (Thermo, Ulm, Germany). PCR reactions were carried out for sequence analyses. Primers KES 1968 (5′-GCACCGACCACAAAAATC-3′) and KES 1969 (5′-GGGCCGTCAACGACCTTC-3′) were used for amplification of gapdh according to Câmara et al. (2002). For the ITS region primer sequences of White et al. (1990) ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and modified ITS4 (5′-TCCTCCGCTTATTGATATGCTT AA-3′), were used. The following reaction conditions were used for both the gapdh and ITS regions. Phusion MasterMix 2 × (Thermo, Ulm, Germany) with an initial heating step for 30 s at 98 °C followed by 35 cycles of 5 s at 98 °C, 5 s at 64 °C, 10 s at 72 °C and 1 min at 72 °C for final elongation. With the help of the Sanger method, the PCR products were sequenced. Alignments of these DNA sequences and database sequences were done to identify the different Stemphylium spp. The partial cyt b gene was amplified using the primers KES 183 (5′-CGATAGCTGCAGGAGTTTGC-3′) and KES 184 (5′-GCTTCAGCATTTTTCTTCATAGTT-3′). PCR was performed using 2 × HotStart-IT FideliTaq Mastermix (USB, Staufen, Germany) with an initial heating step of 1 min at 95 °C, followed by 35 cycles of 15 s at 95 °C, 30 s at 52 °C, 10 min at 68 °C and the final elongation for 5 min at 68 °C. PCR products were sequenced by primer walking sequencing. Alignments of DNA and corresponding cDNA sequences were used for the identification of the exon/intron structure of the amplified cyt b gene.

Molecular genetic species identification

The pre-assigned species assignment were checked using sequences of the ITS region and the partial gapdh gene. Sequences were aligned by using the Lasergene Programms (DNASTAR, Madison, USA). ITS and gapdh sequences of strains identified as S. vesicarium (ITS: AF442803; AF229484; gapdh: AF443902, AY278821.1) and S. botryosum (ITS: AF442782, KC584238; gapdh: AF443879; AF443881), obtained from Genbank, were used in this analysis in addition to the 19 Stemphylium isolates sequenced in this study.

Species-specific identification of Stemphylium spp.

With the sequence information from the study above, a primer pair for the qualitative differentiation of S. vesicarium and S. botryosum KES 1999 (5′-GACCGTCGGCCATATAAAGGGTCG-3′) and 2000 (5′-AACCGTCTCCGTCTATCAATCCT GCT-3′) was selected, which amplify the cyt b gene of the two species but generated products of different, species-specific length. Reaction conditions with an initial heating step for 30 s at 98 °C followed by 35 cycles of 10 s at 98 °C, 5 s at 72 °C and 1 min at 72 °C with 2 × Phusion Mastermix were used.

qPCR based differentiation of S. vesicarium and S. botryosum

Quantitative detection of S. vesicarium and S. botryosum was achieved by real-time PCR by coupling allele-specific primers with a 5′Nuclease Assay. A 214 bp fragment specific for S. vesicarium was amplified with the primers KES 1995 (5′-AGGGTCGCTACAGA CTGGGTCACT-3′), KES 1997 (5′-GCACTCATAAGGTTAGTAA TAACTGTAGC-3′) and the double-dye probe 5′-FAM-CTGCTTAATGTACAGGCGAAAC-BHQ1–3′. For the amplification of S. botryosum (215 bp), the primers KES 1995 and KES 1998 (5′-CAGCTATTACTTCGCCTTTTTAACTGTAGCA-3′) and the previous mentioned double-dye probe were used. The reactions were performed on a Rotor-Gene Q (Qiagen, Hilden, Germany) with Takyon qPCR MasterMix Plus dTTP No Rox reagents (Eurogentec, Köln, Germany) under the following conditions: 3 min at 95 °C and 40 cycles at 95 °C for 10 s and 60 °C for 45 s. The received Ct-values of the S. vesicarium reaction and the S. botryosum reaction were used to calculate the relative allele frequencies based on the method described by Germer et al. (2000).

Validation of the developed differentiation methods

To validate the developed differentiation methods (qualitative PCR and quantitative qPCR) six S. vesicarium strains isolated from pears and six isolates provided by J. Bohlen-Janßen (Agricultural Chamber of Lower Saxony) were used. Details of these isolates are given in Table 1.

Monitoring of Asparagus fields

Samples of infested asparagus fields were taken all over Germany in the years 2010, 2011, 2013 and 2014 at different stages of epidemiological development.

Results

Species identification

PCR with primer pair KES 1968 and KES 1969 for the gapdh gene resulted in an amplification product of 850 bp. Amplification of the ITS region with primer pair KES 1806 and KES 1816 resulted in a product of 558 bp. For the identification at species level, the gapdh gene and the ITS region of the reference strains, ordered from different international culture collections or received from Zapf et al. (2011), were sequenced and compared to database sequences of S. vesicarium (ITS: AF442803 (T); AF229484; gapdh: AF443902 (T), AY278821.1) and S. botryosum (ITS: AF442782 (T), KC584238; gapdh: AF443879; AF443881 (T)). The sequences were compared with sequences from type material (T). All strains except strain 224 were found to be identical with the S. vesicarium sequences in both regions, even though four of them (Table 1) were preliminary classified as S. botryosum. Strain 224 and the published S. botryosum sequences were identical in their gapdh and ITS sequence. Comparing S. vesicarium and S. botryosum, six single substitutions in the ITS region and 24 in the amplified gapdh region were found. Identical sequences were found in strains of the same species originating from different hosts and different countries.

Cyt b of S. vesicarium and S. botryosum

With the primer pair KES 183 and KES 184, the cyt b region could be amplified with DNA and cDNA, respectively. The cDNA fragments showed identical sizes of 550 bp for both Stemphylium species. While the DNA fragments showed a length of 3 kb for S. vesicarium (KJ934233) and 6 kb for S. botryosum (KJ934234). The discrepancy between the DNA and cDNA fragments indicated intron regions which are removed during the splicing process. Alignments of the DNA and cDNA sequences revealed two introns (1323 bp and 1252 bp) for S. vesicarium. The same introns plus an additional intron of 2931 bp were found in S. botryosum (Fig. 1).

Fig. 1
figure 1

Intron-exon organization of the cyt b gene (partial) of S. vesicarium and S. botryosum

Full size image

Differentiation of S. vesicarium from S. botryosum

Due to the large size differences in the cyt b gene it was possible to differentiate S. vesicarium and S. botryosum via fragment length comparisons.

A simple PCR-reaction with the primer pair KES 1999 and KES 2000 showed fragments of 420 bp for S. vesicarium and 3350 bp for S. botryosum (Fig. 2). When using a mixture (1:1) of the DNA of S. vesicarium and S. botryosum only the short fragment of S. vesicarium was amplified. Therefore only pure cultures can be analysed with this method. Cross reactions could be excluded using DNA extracts from other fungal pathogens such as Alternaria alternata and Botrytis cinerea (for A. solani, A. brassicae data not shown). Additionally, DNA of Puccinia triticina was used as a related species of P. asparagi which is causing the rust disease on asparagus.

Fig. 2
figure 2

Principle of the qualitative differentiation of the Stemphylium spp.. 1: S. vesicarium, 2: S. botryosum, 3: 1:1 Mixture of S. vesicarium and S. botryosum, 4: A. alternata, 5: B. cinerea, 6: P. triticina, 7: NTC, M: 1 kb DNA Ladder

Full size image

To quantify the two Stemphylium species directly from asparagus samples, a specific qPCR assay was developed using the cyt b gene as a suitable basis. The identical forward primer KES 1995 and probe in combination with the species specific reverse primers KES 1997 and KES 1998 were used to determine the abundances of S. vesicarium (KES 1997) and S. botryosum (KES 1998). The calculation of the frequency was done with the formula based on Germer et al. (2000). Each DNA was analysed in two separate PCR reactions, each with a primer pair specific to one or the other Stemphylium species. Theoretically, only the fragment of the matching species would be amplified. In practice the mismatching fragment was amplified as well, but in a much less quantity. A delay of eight cycles could be observed between mismatched and correct amplification. To test the sensitivity of this assay, the following parameters were determined. With a serial dilution of the DNA of both organisms, the linearity (regression line which gives information about the proportionality of the Ct-value to the amount of DNA) and the PCR efficiency could be investigated. The efficiency of the PCR reactions was calculated and found to be very similar (S. vesicarium: 99 %, S. botryosum: 97 %), which makes the reactions comparable and suitable for the quantification of S. vesicarium and S. botryosum. The detection and the quantification limit (0.001 ng/μg) was also defined by determining the lowest amount of target DNA that the tested assay can detect and quantify, respectively. The latter was accomplished by using a serial dilution of a 1:1 mixture of S. vesicarium and S. botryosum DNA. To exclude cross reactions with other fungal pathogens, DNA of A. alternata, A. solani, A. brassicae, B. cinerea and P. triticina were analysed. None of these organisms could be detected with the developed qPCR assay.

Analysed asparagus samples

With the qPCR method above, 15 samples from 2014, 87 samples from 2013, 90 isolates from 2011 and 40 samples from 2010 were analysed. All tested samples and isolates were from asparagus growing regions in Germany (Fig. 3).

Fig. 3
figure 3

Distribution of analysed asparagus samples and Stemphylium-isolates

Full size image

S. vesicarium could be identified with a frequency of 98–100 %. There was no evidence for S. botryosum to be the causal agent of the purple spot disease in Germany in the years of sampling.

Discussion

In this study a cost-efficient alternative, with which large quantities of asparagus samples can be analysed, was used and the causal agent of the purple spot disease was developed based on differences in the intron-exon structure of the cyt b gene. Characteristic substitutions in the ITS region and the gapdh gene, which were preliminary defined by Câmara et al. 2002, could also be verified. Based on these two loci S. vesicarium and S. botryosum could be placed into two distinct clusters, in which S. vesicarium showed identical sequences with S. herbarum and S. alfalfa. Even a four locus phylogeny made by Inderbitzin et al. (2009), could not differentiate between these three Stemphylium species as well as S. tomatonis and S. sedicola. Therefore, the separating into single species could not be supported by phylogenetic analyses and should be regarded as synonymous (Câmara et al. 2002; Inderbitzin et al. 2009; Köhl et al. 2009). So far, cyt b was not sequenced for S. herbarum, S. alfalfa, S. tomatonis and S. sedicola and therefore it is not known if there are differences in the sequence. Further analyses regarding cyt b sequences of these five closely related or even synonymous Stemphylium species could provide a further step on the clarification of the taxonomic relationship.

All used reference strains, except strain 224 were identified as S. vesicarium even though some of them were initially classified as S. botryosum.

The location of this gene on mitochondrial DNA (mtDNA) and the resulting high copy number gives it an advantage and reduces the detection limit for the development of reliable PCR assays (Stammler et al. 2013). Based on the development of resistances against QoI fungicides, the cyt b sequences of many different plant pathogenic fungi have been analysed. The cyt b region is intraspecific very conserved but shows interspecific differences. These differences relate to the base sequence as well as the location of non-coding regions (Grasso et al. 2006; Sierotzki et al. 2007; Stammler et al. 2013). Referring to this, the species-specific presence or absence of intron-regions could be shown for other fungal pathogens such as Monillinia spp. and Phyllosticta spp. (Miessner and Stammler 2010; Stammler et al. 2013).

The Stemphylium species could be qualitatively differentiated in a single PCR reaction with one primer pair. In addition to this qualitative differentiation it was possible to create a quantitative method with the help of qPCR to estimate the frequencies of the two Stemphylium species directly from asparagus material. With this Taqman based assay asparagus samples and Stemphylium isolates from Germany were analysed and the frequencies were determined. Only S. vesicarium could be identified in every sample. There was no hint for S. botryosum as causal agent of the purple spot disease.

So far, only S. botryosum had been described as the causal agent in Germany, a description based on morphological characteristics (Leuprecht 1988, 1990; Neubauer 1997, 1998; Zapf et al. 2011). Although a misidentification of German isolates was supposed by Falloon et al. (1987), there was no clear proof for a high abundance of S. vesicarium in Germany until now. Due to the results of this study the prevalent pathogen of asparagus leaf spot in Germany could be identified as S. vesicarium.