First report of Leptosphaeria maculans and Leptosphaeria biglobosa causing blackleg disease of oilseed rape in Tunisia

Summary. Blackleg has been observed in oilseed rape in Tunisia since 2017. Morphological observations, pathogenicity tests, and sequencing of the internal transcribed spacer regions for four fungal isolates from affected plants confirmed the presence of Leptosphaeria maculans and Leptosphaeria biglobosa. These results provide the first record of L. maculans and L. biglobosa as causes of blackleg of oilseed rape in Tunisia.


INTRODUCTION
Brassica napus L. is one of the most common domesticated Brassica crops for human and animal nutrition (Friedt et al., 2018).Oilseed rape is estimated to occupy more than 35.6 million hectares (ha) of world agricultural area with average production of 71 million tons (t) in the 2021-2022 growing season (World Agricultural Production, 2022).The oilseed rape crops have been reintroduced into the Tunisian national cultivation systems since 2014 (Medimagh et al., 2018), and average yields have increased from 1.3 t ha -1 in 2014-2015 to 1.8 t ha -1 in 2018-2019(Maghreb Oléagineux, 2022).With more than 15,000 ha of current rapeseed crop area, these crops are important in Tunisia, especially in the northern regions of the country (Maghreb Oléagineux, 2022).
Five hybrid European spring varieties are currently subscribed in the Tunisian catalogue of varieties and dominate Brassica napus cultivation in Tunisia (Maghreb Oléagineux, 2022).Reductions in oilseed rape yields have been noticed due to biotic stress (Wang et al., 2020;Zheng et al., 2020).Blackleg (Phoma stem canker) is an important fungal disease threat to international oilseed rape production (Howlett, 2004;Fitt et al., 2006).Leptosphaeria maculans (Desm.)Ces. and de Not.(anamorph Phoma lingam) is the principal cause of this disease together with L. biglobosa Shoemaker and H. Brun (L.biglobosa) (Rouxel et al., 1994;Shoemaker and Brun, 2001).Leptosphaeria biglobosa is considered to be less aggressive than L. maculans, and often attacks the upper parts of host plants (Williams, 1999;Shoemaker and Brun, 2001;Mendes-Pereira et al., 2003).In 2017, symptoms similar to blackleg were observed in Beja, Bizerte, Nabeul and Manouba, the four main oilseed rape production areas of Tunisia.
The objective of the present study was to identify the causal agents responsible for the blackleg on oilseed rape in Tunisia.Cultural and morphological features, molecular sequencing of the internal transcribed spacer (ITS) region, phylogenetic analysis, and pathogenicity tests were performed for isolates of fungi obtained from oilseed rape crops.

Isolation and morphological identification of causal agents
To isolate the causal agent, diseased oilseed rape plants were sampled from four northern regions of Tunisia (nine fields) during April and May 2018.The samples were conveyed to the Pests and Integrated Protection in Agriculture research laboratory in Tunisia.For each sample, five infected stem sections (5 cm length) were surface-sterilized in 1% sodium hypochlorite solution for 30 s, followed by 70% ethanol for 20 s, and three rinses in sterile water.The stem sections were then transferred into 90 mm diam.Petri dishes con-taining V8 juice agar supplemented with 25 mg mL -1 of chloramphenicol.After 15 d incubation under 12 h photoperiod at 20°C, serial dilutions were performed to obtain single conidium isolates that were then maintained on V8 juice agar at 20°C.From the initial isolate collection (Table 1), four isolates (obtained from four fields) were randomly selected for morphological and genetic identifications.Isolates L31 and L36 were from two fields in Manouba and isolates L48 and L50 were from two fields in Nabeul.
For macroscopic identification, 5 mm mycelium agar discs of each isolate were inoculated onto 90 mm diam.Petri dishes containing malt agar, V8 juice agar or potato dextrose agar (PDA).Colonies were photographed at 7 and 14 d after incubation at 20°C in complete darkness.For each isolate, colony colour, and conidium size and shape were analyzed under light microscope and then measured using ImageJ software (Schneider et al., 2012).

Molecular and phylogenetic analyses of the causal agents
Genomic DNA was extracted from the four selected isolates using the CTAB protocol (Doyle and Doyle, 1987).PCR was performed to amplify the ITS region using the forward ITS1 (5'-TCCGTAGGTGAACCT-GCGG-3') and the reverse ITS4 (5'-TCCTCCGCT-TATTGATATGC-3') primer, as described by White et al. (1990).The amplifications were carried out using the Thermo Cycler 2720 (Applied Biosystems) in 25 µL reaction mixtures.PCR conditions were as follows: 4 min of initial denaturation at 94°C, followed by 30 cycles of denaturation each at 94°C for 1.5 min, 2 min of annealing at 55°C and 3 min of extension at 72°C, with a final elongation at 72°C for 10 min.A Sanger sequencing using both directions was carried out by CarthaGenomics Advanced Technologies (Tunis, Tunisia).The consensus ITS sequence of each  et al., 2003, Voigt et al., 2005, Liu et al., 2006, Vincenot et al., 2008, De Gruyter et al., 2012, Amirdehi et al., 2017, Zou et al., 2019, Zhao et al., 2021, Luo et al., 2021, Zamanmirabadi et al., 2022) (Table 2).A phylogenetic tree was obtained using the Maximum Likelihood method and the Tamura-Nei model (Tamura and Nei, 1993) with 1000 bootstrap replications, under the Mega XI software (Tamura et al., 2021).

Pathogenicity tests
Seedlings of oilseed rape cv.Topas (which has no major resistance genes (Larkan et al., 2016)) were grown under controlled conditions of 20°C, 90% relative humidity and 16 h light, 8 h dark cycles.Using the cotyledon assay for pathogenicity (Bonman et al., 1980), nine plants were slightly wounded in each cotyledon lobe with a sterile needle, and were then each inoculated with a 10 µL spore suspension at 1 × 10 7 conidia mL -1 (Winter and Koopmann, 2016;Alnajar et al., 2022).Mock inoculations with only sterile distilled water were also carried out in a similar manner.Each isolate and water controls were inoculated onto nine plants.Symptom evaluations were carried out 14 d post-inoculation using the IMASCORE rating scale (Volke, 1999;Balesdent et al., 2001).Leptosphaeria maculans that gave sporulating grey-green tissue collapse in inoculated seedlings was re-isolated and morphologically identifi ed, to assess Koch's postulates.

RESULTS AND DISCUSSION
In the northern visited oilseed rape fi elds of Tunisia, typical symptoms of blackleg were observed on the crop plants, that included large green to grey-coloured leaf spots and basal stem lesions, which were cream to pale brown thick dark brown borders.Lesions on living plants and on 2-month-old crop residues contained multiple pycnidia oft en releasing pink mucilage.Disease incidence was variable from one sampled fi eld to another, but ranged between 48 and 100%.No severe attack leading to plant lodging was observed in the northern oilseed rape growing regions of Tunisia in 2018.
Aft er 14 d of incubation, the fungal isolates L31, L36, and L50 had regular white-dark green colonies on PDA and white-brown mycelium on V8 agar, and irregular small brown to black colonies on malt agar.Colonies of isolate L48 varied from distinct brownishyellow with yellow pigmentation on PDA and malt agar   to white and dark-brown with notable aerial mycelium on V8 agar (Figure 2).Similar morphological characteristics have been previously reported, and supported identifi cation of L. maculans for isolates L31, L36, and L50; and L. biglobosa for isolate L48 (Somda et al., 1996;Howlett et al., 2001;Chen et al., 2010;Vakili Zarj et al., 2017).Th e pycnidia in cultures were black and globose, and 200-500 µm in diameter.Th e pycnidia each had an ostiole, from which a conidial cirrhus protruded (Figure 3a).Each cirrhus had a mucilaginous texture and was light pink.No cirrhus colour variations toward bright red or "oxblood", as described by Rouxel et al. (1994), were observed for the isolates.All conidia were singlecelled, hyaline, ovoid to cylindrical, and of dimensions 2-4 × 1-2 µm (Figure 3b).Th e shape and size of the observed conidia and pycnidia for all four isolates corresponded to the descriptions for Leptosphaeria species (Somda et al., 1996;Howlett et al., 2001).
Sequenced ITS fragments for the four isolates were registered in GenBank under the accession numbers MZ542280 to MZ542283.Molecular analyses confi rmed the identifi cation of L. maculans and L. biglobosa from oilseed rape fi elds in Tunisia.Blast results of the ITS sequences against the NCBI database showed that isolate L48 had 91% similarity with L. biglobosa subgroup 'canadensis' (GenBank number KJ574217).Th e remaining three isolates showed 86% (for L31), 93% (for L36), and 97% (for L50) similarity with L. maculans subgroup 'brassicae' (GenBank number KT225526).
Th e phylogenetic analyses using the ITS sequences of the four isolates from Tunisia and L. maculans reference isolates revealed that L. maculans isolates L31, L36 and L50 were most related to the reference isolate IBCN80 (GenBank number AJ550883) belonging to the L. maculans 'brassicae' subgroup (Figure 4).Th e L. biglobosa isolate L48 was closely related to all L. biglobosa 'canadensis' reference isolates, including IBCN63 (Gen-Bank number AJ550868) and IBCN82 (GenBank number AJ550872).
Th e pathogenicity tests showed that for the three L. maculans isolates, typical host symptoms of grey-green tissue collapse with numerous pycnidia (IMASCORE = 6) were visible at 14 d post-inoculation (Figure 5a).For the L. biglobosa isolate L48, an hypersensitive reaction was developed (IMASCORE = 1) (Volke, 1999;Balesdent et al., 2001) for all replicates (Figure 5b).Th ese First report of Leptosphaeria maculans and Leptosphaeria biglobosa on oilseed rape in Tunisia phenotypic results were similar to those found by Zou et al. (2019), and confirmed the grouping of L48 to the subclade 'canadensis'.To fulfil Koch's postulates, L. maculans was re-isolated from the artificially inoculated plants.All control plants had no disease (Figure 5c).This report is the first of Leptosphaeria maculans and Leptosphaeria biglobosa on oilseed rape in Tunisia, causing blackleg.L. maculans has been previously identified on wild radish (Raphanus raphanistrum L.) in Tunisia (Djebali et al., 2009).The information from the present study will be useful for future blackleg diagnosis and disease management in oilseed rape in Tunisia.

Figure 1 .
Figure 1.Blackleg (Phoma stem canker) symptoms on oilseed rape plants observed during fi eld sampling in Tunisia in spring 2018.(a) Lesions observed on an upper leaf of the cv.PR45H73, in the Manouba region.(b) Severe symptoms on a stem of the cv.Trapper, in the Bizerte region.

Figure 2 .
Figure 2. Front and reverse sides of cultures of Leptosphaeria maculans isolate L50 (a) and Lepotosphaeria biglobosa isolate L48 (b), grown for 7 or 14 d at 20°C in the dark on Potato Dextrose agar (PDA), V8 agar or malt agar.

Figure 4 .
Figure 4. Phylogenetic tree of Leptosphaeria maculans and Leptosphaeria biglobosa isolates performed on the ITS sequenced region based on the Maximum Likelihood method and Tamura-Nei model.Th is analysis was carried out using Mega XI soft ware (Tamura et al., 2021), with 1000 bootstrap replications, and involved 28 nucleotide sequences of 327 fi nal nucleotide positions [four sampled isolates from Tunisia, 21 other representative isolates of each subclade available in GenBank, and three diff erent Leptosphaeria species (Câmara et al., 2002): L. typharum (AF439465), L. dryadis (AF439461) and L. weimeri (AF439466)].

Table 1 .
Details of Tunisian regions from which Leptosphaeria spp.isolates were sampled in Brassica napus crops in 2018.First report of Leptosphaeria maculans and Leptosphaeria biglobosa on oilseed rape in Tunisia isolate was then analyzed using the Basic Local Alignment Search Tool (BLAST) soft ware on NCBI available from: https://www.ncbi.nlm.nih.gov/.The best match for each sequence with the lowest E-value and the highest query cover and percentage of identity was recorded.In addition, a phylogenetic analysis was carried out by aligning, using the ClustalW algorithm, the ITS sequences of the four isolates from Tunisia with available reference sequences of L. maculans and L. biglobosa in the GenBank database.Isolates from diff erent countries of origin, representative of previously published Leptosphaeria subclades, were used in this assessment (Mendes-Pereira

Table 2 .
Isolates of the Leptosphaeria maculans and Leptosphaeria biglobosa species complex used as references in the phylogenetic analyses.