Leptosphaeria biglobosa inhibits the production of sirodesmin PL by L. maculans

BACKGROUND: Phoma stem canker is caused by two coexisting pathogens, Leptosphaeria maculans and L. biglobosa . They coexist because of their temporal and spatial separations, which are associated with the differences in timing of their ascospore release. L. maculans produces sirodesmin PL, while L. biglobosa does not. However, their interaction/coexistence in terms of sec- ondary metabolite production is not understood. RESULTS: Secondary metabolites were extracted from liquid cultures, L. maculans only (Lm only), L. biglobosa only (Lb only), L. maculans and L. biglobosa simultaneously (Lm&Lb) or sequentially 7 days later (Lm+Lb). Sirodesmin PL or its precursors were identi ﬁ ed in extracts from ‘ Lm only ’ and ‘ Lm+Lb ’ , but not from ‘ Lm&Lb ’ . Metabolites from ‘ Lb only ’ , ‘ Lm&Lb ’ or ‘ Lm+Lb ’ caused signi ﬁ cant reductions in L. maculans colony area. However, only the metabolites containing sirodesmin PL caused a signi ﬁ cant reduction to L. biglobosa colony area. When oilseed rape cotyledons were inoculated with conidia of ‘ Lm only ’ , ‘ Lb only ’ or ‘ Lm&Lb ’ , ‘ Lm only ’ produced large gray lesions, while ‘ Lm&Lb ’ produced small dark lesions similar to lesions caused by ‘ Lb only ’ . Sirodesmin PL was found only in the plant extracts from ‘ Lm only ’ . These results suggest that L. biglobosa prevents the production of sirodesmin PL and its precursors by L. maculans when they grow simultaneously in vitro or in planta . CONCLUSION: For the ﬁ rst time, L. biglobosa has been shown to inhibit the production of sirodesmin PL by L. maculans when interacting simultaneously with L. maculans either in vitro or in planta . This antagonistic effect of interspeci ﬁ c interaction may affect their coexistence and subsequent disease progression and management.


INTRODUCTION
Offspring of most living organisms are provided with enough resources for initial development, but they must obtain resources for the fundamental energy required for further development, growth, survival and reproduction. 1 Resource conservation rules state that resources acquired by one organism are not immediately available to another, so if there is a limited supply of resources, competition will ariseresource competition. [1][2][3] Interactions between pathogens and host plants do not occur in isolation; a host is likely to interact with many different species. 4 There will be resource competition between different species, called interspecific competition. 3,5 Plant pathogens that share the same host face resource competition for successful infection of the host.
Phoma stem canker is an economically important disease of oilseed rape caused by two closely related hemi-biotrophic ascomycete fungal pathogens: Leptosphaeria maculans (Plenodomus lingam) and L. biglobosa (P. biglobosus). [6][7][8] This disease caused annual yield losses of >£74M in the UK, despite the use of fungicides. 9,10 The distribution of L. maculans and L. biglobosa varies globally. Whereas in parts of eastern Europe and China only L. biglobosa is found, both pathogens can be found to coexist in northern and western Europe, Australia, and Canada. 11 These two pathogens can coexist because they occupy different ecological niches due to slight differences in their biological and epidemiological characteristics. 11 One example is the difference in their temperature optima for pseudothecial (fruiting body) maturation, which causes a temporal separation of ascospore release in Western Europe. [7][8][9][10][11][12] This enables the resource to be partitioned 2 MATERIALS AND METHODS 2.1 Pathogen and media preparation 2.1.1 Pathogen preparation L. maculans isolate ME24, obtained from stem canker of oilseed rape cultivar Apex in 2002, 28 and L. biglobosa isolate WH17-Why-1, obtained from phoma leaf lesion of oilseed rape cultivar Whisky in 2017, were used in this study. The L. maculans and L. biglobosa isolates were grown on 20% V8 agar plates for 14 days, and agar plugs were cut at the growth front of the plate for inoculation of liquid culture or agar plates.
2.1.2 Identifying the optimal media for culturing of the pathogens Agar plugs (10 mm) of L. maculans isolate ME24 and L. biglobosa isolate WH17-Why-1 were inoculated in the middle of plates (9 cm) containing five different growth media: malt extract agar (MEA), water agar (WA), potato dextrose agar (PDA), 20% V8 juice agar (V8) or 20% clarified V8 agar (V8 Clar). V8 juice was clarified by adding 15 g L −1 of calcium carbonate to the V8 juice agitated using a magnetic stirrer for 15 min before being centrifuged at 1400 g for 5 min. The pH of the clarified V8 media was 5.86, whereas the pH of the V8 only media was 6.15. The pellet was discarded. The supernatant was stored at −20°C until required. The supernatant was classed as clarified V8 juice. The inoculated plates were incubated for 7 days at 21°C in continuous darkness before the colony diameter was measured. Each treatment was replicated three times. The colony diameter was converted to colony area (cm 2 ).
2.2 Concentration of sirodesmin PL production by L. maculans Three plugs of L. maculans (8 mm diameter) were cultured in 250-mL conical flasks containing 75 mL of clarified 20% V8 juice broth in a rotary shaker at 80 rpm and 18°C for 14 days in continual darkness. Secondary metabolite extractions were taken at six time points. There were three replicates for each time point. The experiment was replicated two times (experiments 1 and 2). The time points for experiment 1 were 1, 3, 7, 8, 10, and 14 days post inoculation (dpi) and for experiment 2 were 1, 3, 6, 8, 10, and 14 dpi. The secondary metabolites were extracted from each sample and analyzed using high-performance liquid chromatography (HPLC).
The secondary metabolites from liquid culture were extracted using ethyl acetate. The culture filtrate from each replicate was split equally into two 50-mL Falcon tubes (∼37.5 mL in each). Each Falcon tube was then topped up to 50 mL using ethyl acetate (EtOAc) (∼12.5 mL), then gently inverted 30 times to increase secondary metabolite uptake, then left for 1 h so that the phases could settle before being centrifuged at 3200 g for 5 min. For each replicate, 20 mL of the organic phase was pipetted into a clean 50-mL Falcon tube and the EtOAc was evaporated under a constant stream of nitrogen using a sample concentrator. The dried metabolites were resuspended in 0.5 mL of EtOAc. Falcon tubes containing the same treatment were combined into sterile plastic sample bottles. Using a syringe, the resuspended metabolites were passed through a 0.45-μm syringe filter into a clean HPLC sample vial. A different HPLC vial was prepared with a glass insert and 200 μL of the resuspended metabolites was pipetted into the insert. The samples were stored at 20°C until required for HPLC analysis.
To measure the concentration of sirodesmin PL, a gliotoxin standard curve (R 2 = 0.999, y = 2 6 x -56 900) was created using maximal areas of gliotoxin samples visualized at 254 nm at concentrations of 0, 10, 50, 100, 250, 500, and 1000 mg L −1 using HPLC analysis. This gliotoxin standard curve was used to calculate the concentration of sirodesmin PL at each time point. The level of detection (LOD) and level of quantification (LOQ) were 70 and 200 mg L −1 , respectively. The mean concentration at each time point was calculated. Gliotoxin was used for the standard curve because it belongs to the same class of fungal secondary metabolites as sirodesmin PL, ETP, has been used in previous work as a standard curve, 23 and is commercially available.
2.3 Effects of L. biglobosa on secondary metabolite production by L. maculans in vitro To investigate the effects of L. biglobosa on secondary metabolite production by L. maculans, L. biglobosa was simultaneously or sequentially inoculated with L. maculans. Clarified 20% V8 juice broth was inoculated with four different treatments: L. maculans only (Lm only), L. biglobosa only (Lb only), L. maculans and L. biglobosa simultaneously (Lm&Lb), and L. maculans and L. biglobosa sequentially (Lm+Lb). For the 'Lm+Lb' treatment, the liquid media was inoculated with L. maculans followed by L. biglobosa inoculation after 7 days. Three plugs of each pathogen were used per corresponding treatment, therefore cultures containing both the pathogens had six fungal plugs. Each inoculation treatment was done in triplicate. The experiment was repeated three times. To identify the secondary metabolite composition of the different treatments, secondary metabolites were extracted from liquid media using ethyl acetate after 14 dpi. The samples were prepared and analyzed using HPLC. To identify the compound responsible for each unique HPLC maximum, individual 1-mL fractions were taken for liquid chromatography-mass spectrometry (LC-MS) analysis.
To identify the secondary metabolite composition, the samples were analyzed using a Shimadzu Prominence HPLC machine (Kyoto, Japan) with a diode array detector (SPD-M20A) using 20-μL injections. Separations were performed on a C 18 column (Varian pursuit 5, 150 × 4.6 mm). A linear gradient from 85% water and 15% acetonitrile going to 100% acetonitrile in 40 min, then maintaining 100% acetonitrile for 3 min before starting a linear gradient back to 85% water over 5 min was used. It was then held at 85% water before starting the next sample. Each sample had a total run time of 53 min. A flow rate of 1 mL min −1 was used. Results were visualized at 254 nm using Lab solution version 5.92 (Shimadzu Corporation) but wavelengths were detected from 190 to 400 nm. The LOD and LOQ were calculated to be 70 and 200 mg L −1 , respectively. All samples and chemicals used were passed through a 0.45-μm nylon microfilter before being analyzed, although all chemicals used were of HPLC analytical grade and water was double distilled. The expected pattern of maxima for deacetylsirodesmin PL, phomamide, and sirodesmin PL were identified using information from Pedras et al. 27 2.4 Identification of secondary metabolites using LC-MS A Waters I-Class UPLC system coupled to a Xevo Micro TQ-S mass spectrometer was used for analysis. Chromatographic separation was performed on a Waters BEH C18 column (2.1 × 50 mm, 1.8 μm) held at 40°C with mobile phase A = 0.2% formic acid in water, and mobile phase B = acetonitrile. The flow rate was set at 0.4 mL min −1 and a gradient separation performed by ramping initial starting conditions of 85% A to 0% A over 9 min, holding 100% B for 1 min, returning to starting conditions of 85% A over 0.1 min, and holding 85% for 4.9 min to re-equilibrate the column. The injection volume was 10 μL. A full mass scan was performed in positive ion mode over the range 50-650 m/z using a scan time of 0.2 s. The probe capillary voltage was 3 kV, the cone voltage was fixed at 20 V, and the collision energy was fixed at 3 V. Desolvation gas flow was set at 1000 L h −1 , with a temperature of 500°C, and a cone gas flow of 150 L h −1 was used, with source temperature set at 150°C.

2.5
Effects of secondary metabolite extracts on growth of L. maculans and L. biglobosa on agar plates Fungal plugs of L. maculans or L. biglobosa (8 mm in diameter) were inoculated onto the middle of clarified V8 juice (V8 Clar) agar plates. Each fungal plug was then inoculated with 20 μL of the corresponding secondary metabolite extract from each treatment ('Lm only', 'Lb only', 'Lm&Lb', 'Lm+Lb') or ethyl acetate (control). Each treatment had five replicate plates. Plates were sealed using Parafilm and incubated at 18°C in continual darkness. Colony diameters for L. maculans and L. biglobosa were recorded at 7 dpi. The colony diameter was converted to colony area (cm 2 ). The experiment was repeated three times.
2.6 Effects of L. biglobosa on disease development and sirodesmin PL production by L. maculans in planta 2.6.1 Preparation of L. maculans and L. biglobosa inoculum and plant material L. maculans and L. biglobosa conidial inoculum was obtained from 2-week-old cultures grown on 20% V8 juice agar as described by Huang et al. 28 The suspensions were measured using a hemocytometer slide under a light microscope and suspensions were adjusted to 1 × 10 7 conidia mL −1 . The 'Lm&Lb' inoculum was prepared by combining a sample of each conidial spore suspension in a 1:1 ratio. Cultivar Charger, which has a moderate disease resistance rating score of 4 according to the AHDB Recommended List resistance ratings (0-9 scale, with 9 being good resistance), was used for this study (https://ahdb.org.uk/knowledge-library/ recommended-lists-archive). Cotyledons were prepared by sowing seeds of winter oilseed rape cv. Charger into 40-well plug trays placed in a controlled environment cabinet set at 20/18°C and 12 h/12 h day/night. Plug trays were filled with John Innes (No. 3) soil and Miracle Grow multipurpose compost mixed in a 1:1 ratio. Plants were regularly watered to ensure soil remained moist. True leaves were removed to prolong the lifespan of cotyledons.

Experimental design and plant inoculation
There were four treatments, 'Lm only', 'Lb only', 'Lm&Lb' and a sterilized distilled water control, 'SDW'. There were 30 plants per treatment. The experiment was designed using a randomized block design with three replicates. Cotyledons of 14-day-old seedlings were inoculated. The inoculation sites were prepared by wounding the cotyledon using a sterilized needle. Each cotyledon was wounded in two locations, therefore there were four inoculation sites per plant. For inoculation, 10 μL of conidial suspension was pipetted onto each inoculation site according to the treatment and experimental design. Inoculated plants were covered with tray covers to maintain high humidity for 48 h. The inoculated plants were kept in a controlled environment cabinet at 18/16°C and 12 h/12 h day/night.

Assessment of disease development
To measure the size of lesions, cotyledons were excised from each plant at 17 dpi, placed inside a 9 × 9 cm square, and a standardized photograph was taken. The lesion area was calculated by processing the standardized photographs using ImageJ software. 29 The total number of pixels were converted into cm 2 . The experiment was repeated three times.

Assessment of sirodesmin PL production in planta
Due to the in planta secondary metabolite detection requiring very severely diseased plant tissue, a separate experiment with the same design as those used for disease development experiments was done and the cotyledons were excised from each plant at 26 dpi, rather than at 17 dpi. The standardized photograph was also taken for assessing the lesion area using ImageJ software. After lesion phenotype photographs were taken, lesions from 30 plants of each treatment were excised. The lesions were excised according to their size. For small (e.g. plants inoculated with Lb only) or no lesions (e.g. control), an 8 mm diameter corer centered over each inoculation site was used. For larger lesions, a scalpel was used to excise the lesion. Lesions from five plants were grouped together and placed into 2-mL screw cap tubes. There was a total of six 2-mL screw cap tubes per treatment. After 48 h of freeze-drying, three small metal balls (3 mm) and 600 μL of ethyl acetate were added to each tube. Each sample was lysed using a FastPrep machine for 40 s at 6.0 m s −1 ; this step was repeated three times, samples were placed on ice in between runs of the FastPrep. Lysed samples were centrifuged at 15 700 g for 5 min and 500 μL of ethyl acetate supernatant was pipetted into a sterile 50 mL Falcon tube. Supernatants for each treatment were combined and evaporated to residue. Each treatment sample was resuspended in 500 μL of ethyl acetate and passed through a 0.45 μm Polytetrafluoroethylene (PTFE) syringe filter into a 2 mL HPLC vial for HPLC analysis. For identification of the compound responsible for each unique HPLC maximum, individual 1.5 mL fractions were taken for between 11-12.5 min and 16-17.5 min for each sample. Fractions were evaporated to residue before being resuspended in 110 μL of ethyl acetate before being analyzed by LC-MS.

Statistical analysis
The statistical analysis of the data was done using GenStat 22nd edition. 29 To analyse the effects of different media types on colony area of different pathogens, the effect of co-inoculation of L. maculans and L. biglobosa on the concentration of sirodesmin PL and its precursors, a factorial analysis of variance (ANOVA) was done to determine significant differences between treatments on each pathogen. For the effect of secondary metabolite extracts on colony area, unbalanced ANOVA was used. For data at 17 dpi, a factorial ANOVA was done using 'treatment' and 'experiment number' as factors. For data at 26 dpi, an ANOVA was done using only 'treatment' as the factor. Variance ratio values (i.e., F-test) that were significant at P < 0.05 were used to provide evidence for significant effects of factors. Post hoc Fisher least significant difference tests were applied to separate means between different media and different secondary metabolite extracts for each pathogen.

Optimal media for colony growth
There were significant differences in the colony area on different media for L. maculans (F 4.13 = 169.37, P < 0.001, LSD = 1.51) and L. biglobosa (F 4.13 = 21.31, P < 0.001, LSD = 3.55). Both pathogens had significantly larger colony areas when cultured on 20% clarified V8 juice agar than on all other media tested (Fig. 1). For L. maculans, the colony area was not significantly different between water agar (5.6 cm 2 ) and MEA (6.4 cm 2 ), but was significantly smaller than that on PDA (10.4 cm 2 ). The largest colony area was on clarified V8 agar, which was significantly larger than that on V8 agar (Fig.1). For L. biglobosa, there were no significant differences between MEA (13.6 cm 2 ), WA (15.4 cm 2 ) and PDA (15.7 cm 2 ) or between MEA and V8 (11.6 cm 2 ). However, the colony area was significantly larger on clarified V8 agar (23.3 cm 2 ) than those on other media (Fig.1). The clarified V8 was the optimum medium for both L. maculans and L. biglobosa, therefore clarified V8 was used for liquid culture and agar plates in this study.

Concentration of sirodesmin PL production by L. maculans over 14 days
When the concentration of sirodesmin PL was measured at regular time intervals over 14 days in culture filtrates of L. maculans, sirodesmin PL was not detected until 3 dpi. The concentration of sirodesmin PL increased from 3 dpi until 10 dpi and then decreased from 10 dpi (Fig. 2).
3.3 Effects of L. biglobosa on sirodesmin PL production by L. maculans in vitro 3.3.1 Comparison of secondary metabolite chromatograms For the secondary metabolites extracted from liquid culture, there were three unique maxima that were present only in the 'Lm only' and 'Lm+Lb' samples (Maximum 1 -Rt 11.2; Maximum 2 -Rt 16.2 and Maximum 3 -Rt 19.25), so these maxima were unique to these samples. When L. maculans was cultured simultaneously with L. biglobosa ('Lm&Lb') the three maxima unique to 'Lm only' and 'Lm+Lb' were not present and its chromatogram was like that of 'Lb only'.
3.3.2 Identification of maxima found on the chromatograms from extracted secondary metabolites using LC-MS Maximum 1 contained ions in its LC-MS positive ion spectra (Fig. 3  (a)) at m/z 445.20, which corresponds to a monocharged molecule with a molecular weight 444.5 Da; deacetylsirodesmin is a known secondary metabolite produced by L. maculans and has a molecular weight of 444.5 Da. 30,31 Additionally, ions were observed at m/z 381.24 [M -S 2 + H] + , corresponding to deacetylsirodesmin lacking two sulfur atoms; m/z 403.23 [M -S 2 + Na] + , corresponding to the sodium adduct of sirodesmin lacking two sulfur atoms; and m/z 467. 19 [M + Na] + , corresponding to the sodium adduct of deacetylsirodesmin ( Fig. 3(a)). In addition to these, ions were observed at m/z 319.26, which corresponds to monocharged molecule with a molecular weight of 318.4 Da. Phomamide is a known secondary metabolite produced by L. maculans and has a molecular weight of 318.4 Da. 30 Maximum 2 from the chromatogram contained ions in its LC-MS positive ion spectra ( Fig. 3(b)) at m/z 487.19 (97%), which corresponds to a monocharged molecule with a molecular weight 486.6 Da. Sirodesmin PL is a known secondary metabolite produced by L. maculans with a molecular weight of 486.6 Da. 30 Maximum 3 from the HPLC chromatogram could not be identified using LC-MS. The fraction was further concentrated by resuspending in only 0.3 mL ethyl acetate, but a unique maximum could not be identified from the LC-MS spectra, therefore the compound responsible for maximum 3 on the HPLC chromatogram remains unknown.

Production of sirodesmin PL in planta
There was a difference in the presence or absence of sirodesmin PL and its precursors between treatments in planta. Sirodesmin PL and its precursors were present only in the 'Lm only' treatment, and were absent from the 'SDW', 'Lb only' and 'Lm&Lb' treatments (Fig. 8). The LC-MS chromatograms showed that there were three unique maxima at retention times 4.95, 5.03, and 5.11 min in the 'Lm only' sample. At retention time 4.95 min, the positive ion spectra contained ions with m/z 242.29 (30%) and 243.35 (5%). These ions remain unidentified.
At retention time 5.03 min (Fig. 8(a)), the positive ion spectra (Fig. 8(b)) contained ions at m/z 319.26 (100%) and 320.29 (15%) with a mean m/z of 319.4. This corresponds to a monocharged molecule with a molecular weight of 318.4 Da. Phomamide is a known secondary metabolite produced by L. maculans and has a molecular weight of 318.4 Da. 30,31 At retention time 5.11 min (Fig. 8(c)), the positive ion spectra (Fig. 8(d)) contained an ion at m/z 487.18, which corresponds to a monocharged molecule with a molecular weight of 486.57 Da. Sirodesmin PL has a molecular weight of 486.6 Da. 30,31 The positive ion spectra (Fig. 8(d)) also contained an ion at m/z 445.24, which corresponds to a monocharged molecule with a molecular weight of 444.24 Da. Deacetylsirodesmin PL is a known secondary metabolite produced by L. maculans and has a molecular weight of 444.5 Da. 30,31 The following ions were also found in the positive ion spectra (Fig. 8(d)

DISCUSSION
The results of this study for the first time provide evidence that L. biglobosa can inhibit L. maculans production of sirodesmin PL and its precursors to increase its competitiveness over L. maculans both in vitro and in planta. No sirodesmin PL nor its precursors were produced when L. maculans and L. biglobosa were grown together simultaneously in vitro (liquid culture) or in planta (cotyledons of oilseed rape), suggesting that the production of sirodesmin PL and its precursors by L. maculans had been inhibited by L. biglobosa. The inhibition of sirodesmin PL production has advantages for L. biglobosa growth because sirodesmin PL has antibacterial and antifungal properties. 19 There was an inhibition of L. biglobosa in vitro growth when secondary metabolite extracts containing sirodesmin PL were applied; this suggests that L. maculans uses sirodesmin PL as a form of interference interspecific competition strategy to outcompete L. biglobosa. This supports previous findings that L. maculans

SDW
Lm only The only extracts that contained sirodesmin PL and its precursors were 'Lm only' and 'Lm+Lb' (L. maculans followed by L. biglobosa inoculation 7 days later), not the 'Lm&Lb' (L. maculans and L. biglobosa inoculation simultaneously), suggesting that L. biglobosa inhibits sirodesmin PL production by L. maculans when L. maculans and L. biglobosa grow together simultaneously. Unique detectable maxima were not identified in the 'Lm&Lb' extracts, suggesting that the production of precursors may have been inhibited. This could be due to inhibition of gene expression of 18 co-regulated genes that have been reported to be involved in the biosynthesis pathway of sirodesmin PL. 23 It has been shown that O-prenyl-L-tyrosine is the first committed precursor for sirodesmin PL biosynthesis, a step catalyzed by prenylase (SirP). 30 The sirP mutants have been shown to inhibit the production of sirodesmin PL 23 so denaturation of sirP or another gene involved early in the biosynthesis pathway would probably inhibit the production of the precursors and sirodesmin PL production. 23,30,35 Also, it has been shown that the production of sirodesmin PL is regulated by the cross-pathway control gene cpaA using the transcription factor sirZ, 36 therefore if L. biglobosa disrupts the expression or products of the genes responsible for the biosynthesis of the sirodesmin PL precursors early in the pathway then downstream products and compounds will not be produced. Once the biosynthesis of sirodesmin PL precursors has started, L. biglobosa could not disrupt or inhibit sirodesmin PL production. This hypothesis is supported by the results showing that sirodesmin PL and its precursors were detected in extracts from 'Lm+Lb' but not 'Lm&Lb'. There is a need to investigate the expression of genes related to biosynthesis of sirodesmin PL in the presence of L. biglobosa to identify the potential gene inhibitors.

Lb only Lm&Lb
The results of this study suggest that the timing when L. maculans and L. biglobosa meet and interact is important for the outcome of the interaction between them. In liquid culture, sirodesmin PL and its precursors were identified in the secondary metabolite extracts from 'Lm+Lb' but not from 'Lm&Lb'. This effect of timing is mainly due to the timing of L. maculans production of sirodesmin PL since L. maculans did not produce sirodesmin PL until 3 dpi, which agrees with the study by Gardiner et al. 23 Therefore, if L. biglobosa can exploit the resource before L. maculans can produce sirodesmin PL, L. biglobosa can exploit its exploitation interspecific competitive strategy to use the resource and inhibit the production of sirodesmin PL and its precursors to outcompete L. maculans. This is supported by the results from in planta experiments. Cotyledons inoculated with 'Lm&Lb' produced lesions like those of L. biglobosa but different from those of L. maculans and no sirodesmin PL was found. This finding may help to explain why the effectiveness of L. biglobosa in inducing resistance to L. maculans infection was compromised if L. biglobosa was added at 64 h or later after L. maculans 24 because 64 h is roughly 2.5 days which is required for sirodesmin PL to be produced (Fig. 2). The production of sirodesmin PL was a maximum at 10 dpi before reducing at day 14; this may be due to conservation of energy because the resource is a relatively nutrient poor by 14 days. 37 Only one cultivar of winter oilseed rape was used in this study. The antagonistic effects of interspecific competition between L. maculans and L. biglobosa may differ, with cultivars having different resistance ratings or between isolates having different pathogenicity, therefore further investigation is required to fully understand these phenomena.
The results of this study have important practical and agricultural relevance. If the Leptosphaeria spp. ascospores are released and infect the host at the same time, then a simultaneous infection will occur that could result in small dark lesions (L. biglobosa type of lesions) and affect the timing of fungicide application because the timing of fungicide application is determined based on the L. maculans type of lesions (large gray lesions). In the UK, as part of an integrated pest management strategy, growers are advised to accurately apply the first fungicide application when 10-20% of the plants in a crop have L. maculans phoma leaf spots in autumn, followed by another application if/when reinfection occurs (https://ahdb.org.uk/knowledge-library/how-to-managephoma-in-oilseed-rape). This study showed that when cotyledons were co-infected with L. maculans and L. biglobosa at the same time, the lesions were small and had an appearance like L. biglobosa lesions. Therefore, in seasons where there is simultaneous infection of plants by both Leptosphaeria spp., the grower may apply the fungicide later or not at all due to the advised threshold of L. maculans lesions not being met. In addition, L. biglobosa is less sensitive to azole fungicides than L. maculans. 14,38 In the UK, there is substantial reliance on the use of azole fungicides, particularly prothioconazole and tebuconazole. 39 Furthermore, recent studies have shown that L. biglobosa can also cause stem basal cankers as well as upper stem lesions 15,18,40,41 and cause yield losses. 16 Therefore, results of this study suggest that the current advice on fungicide application threshold needs to be revised to integrate the appearance of L. biglobosa leaf spot lesions to protect the crops from both Leptosphaeria spp. because their interspecific interactions and inhibition of sirodesmin PL production may cause unforeseen challenges in the future.