Characterization of Lasiodiplodia species associated with grapevines in Mexico

Copyright: © 2021 E.A. Rangel-Montoya, M. Paolinelli, P.E. Rolshausen, C. Valenzuela-Solano, R. Hernandez-Martinez. This is an open access, peerreviewed article published by Firenze University Press (http://www.fupress. com/pm) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Botryosphaeria dieback is a degenerative wood disease caused by Botryosphaeriaceae fungi, this disease has cosmopolitan distribution and predominates in warm climate regions (Úrbez-Torres, 2011;Gramaje et al., 2018). Fungi in this family are known as opportunistic or latent plant pathogens, as they can remain endophytic for long periods in host tissues without causing symptoms (Slippers et al., 2007).
More than 30 species in the Botryosphaeriaceae have been associated with Botryosphaeria grapevine dieback, and these are in Botryosphaeria, Diplodia, Dothiorella, Lasiodiplodia, Neoscytalidium, Neofusicoccum, Sphaeropsis, and Spencermartinsia (Úrbez-Torres, 2011;Rolshausen et al., 2013;Stempien et al., 2017;Gramaje et al., 2018). The main symptoms caused by these fungi are vascular discolouration and perennial cankers in host plant vascular bundles, by occlusion of xylem and phloem, which leads to the death of branches and eventually of entire plants. This disease is distinguished from Eutypa dieback because it is not known to cause particular foliar symptoms (Úrbez-Torres, 2011;Bertsch et al., 2013;Billones-Baaijens and Savocchia, 2019). Species in the Botryosphaeriaceae were commonly found in grapevines 7 to 10 years old and older, mainly in plants where large pruning wounds had been made in vines (Gubler et al., 2005). However, incidence of symptoms caused by this group of fungi has greatly increased in recent years, especially in young vineyards (Gramaje and Armengol, 2011;Gispert et al., 2020).
Among the Botryosphaeriaceae, the Lasiodiplodia has been reported as highly virulent on grapevines (Úrbez-Torres and Gubler, 2009), and has also been identified on more than 500 host species (Punithalingam, 1976). Some of the main morphological characteristics of Lasiodiplodia include hyaline and smooth conidiogenous cells, with cylindrical to conical shapes, which produce conidia with subovoid to ellipsoid-ovoid shapes and which are hyaline without septa, or dark-brown with single septae (Phillips et al., 2013). Lasiodiplodia are globally distributed, mainly in the tropics and subtropics, and are probably spread when plants are transported between regions due to the lack of restrictions on the movement of propagation material (Cruywagen et al., 2017;Mehl et al., 2017). Lasiodiplodia theobromae is the type species of the genus (Alves et al., 2008), and this species is comprised of many cryptic species because of their morphological similarity (Alves et al., 2008;Mehl et al., 2017). As a result, the taxonomy of Lasiodiplodia has undergone revisions, and new species have been introduced (Dissanayake et al., 2016;Tibpromma et al., 2018). Several Lasiodiplodia species have been reduced to synonymy, particularly those with morphology similar to Lasiodiplodia mahajangana, L. plurivora and L. theobromae. There are currently 34 accepted Lasiodiplodia species (Zhang et al., 2021).
The only Lasiodiplodia species causing perennial cankers and dieback that has been reported in Mexican vineyards is L. theobromae (Úrbez-Torres et al., 2008). However, given the recent taxonomical revision of Lasiodiplodia, we hypothesize that the species diversity within that group is broader than initially reported. Hence, the present study aimed to clarify and update the taxonomy of Lasiodiplodia present in vineyards from Baja California and Sonora, Mexico, and to evaluate the pathogenicity of these fungi to grapevine.

Fungal isolation and morphological characterization of Lasiodiplodia spp.
This study encompassed ten vineyards in the main grape-growing areas of the States of Baja California and Sonora, from which 35 samples from grapevines exhibiting Botryosphaeria dieback symptoms were taken from trunks and branches ( Figure 1). Small pieces of symptomatic plant tissue were obtained from each diseased plant, and these were immersed in 95% ethanol, quickly flamed, and then placed onto potato dextrose agar (PDA; Difco) supplemented with 25 mg mL -1 chloramphenicol in Petri plates. The plates were incubated at 30°C until fungal growth was observed. Smoke-gray fungal colonies with abundant aerial mycelium were sub-cultured onto PDA plates to obtain pure cultures, and were then preserved at 4°C in 20% glycerol.
Pure cultures were grown on PDA and incubated at 30°C for 7 d to determine morphological characteristics of fungal isolates, including their pigmentation and formation of aerial mycelium. Pycnidium production was induced using liquid Minimal Medium 9 (MM9) (10 g·L -1 glucose, 1.0 g·L -1 NH 4 Cl, 0.5 g·L -1 NaCl, 2.5 g·L -1 K 2 HPO 4 , 2.5 g·L -1 KH 2 PO 4 ) in flasks supplemented with sterile pine needles (5% w/v). The flasks were incubated at room temperature under an ultraviolet electromagnetic radiation lamp, using a 12 h light and 12 h darkness regime for 15 d. Formed pycnidia were suspended in 0.5% Tween 20 to obtain conidia, which were observed under a light microscope (Nikon Eclipse E200). Images of the conidia were captured with an Infinity 1 Lumenera camera, and analyzed using Infinity Analyze v 6.5.4 and ImageJ software. To compare conidium size across species, one-way ANOVA followed by a post hoc Fisher LSD analysis (α < 0.05) were carried on these data using STATISTICA 8.0.

Phylogenetic analyses
The sequences were analyzed using BioEdit v.7.0.5.3 (Hall, 1999) and a BLASTn analysis was carried out. Sequences with the greatest similarity were downloaded from the GenBank (Table 1) and aligned with ClustalW (pairwise alignment parameters: gap opening 10, gap extension 0.1, and multiple alignment parameters: gap opening 10, gap extension 0.2. Transition weight was set to 0.5, and delay divergent sequences to 25 %) (Thompson et al., 1994). The alignment was adjusted manually where necessary. Alignment of ITS and tef-1α were imported in BioEdit v.7.0.5.3 to obtain the concatenated matrix. Maximum Likelihood (ML) and Maximum Parsimony (MP) analyses were performed using MEGA-X (Kumar et al., 2018), based on the concatenated sequence alignment. The best model of nucleotide substitution was selected according to the Akaike Information Criterion (AIC). The T3+G+I model was used for the ML analysis (Tamura, 1992). Parameters for Maximum Likelihood were set to Bootstrap method using 1000 replicates. Initial tree(s) for the heuristic search were obtained automatically by applying the Maximum Parsimony method. Gaps were treated as missing data. The tree was visualized in MX: Tree Explorer. New sequences were deposited in the GenBank (https://www.ncbi.nlm.nih.gov/genbank/) ( Table 1).

Determination of optimum growth temperature of selected Lasidioplodia isolates
The optimum growth temperature of identified Lasiodiplodia species was determined. Selected isolates of identified species were grown on PDA plates by inoculating each plate with a 3-mm diam. plug of a 2-d-old colony at the edge of the plate. Three replicates of each isolate for each temperature were included, and plates were then incubated at 20, 23, 25, 28, 30, 37, or 40°C. This temperature range was chosen based on previous reports (Úrbez-Torres et al., 2006;Paolinelli-Alfonso et al., 2016), and considering the prevalent summer temperatures of the zone from which the isolates were obtained. The colony radius was measured every 24 h for 3 d. The optimum growth temperature was determined as the temperature that produced the maximum mycelial growth rate (mm d -1 ), which was calculated using the formula: when colony measured, and T i = Initial time (day 1).

Production of aerial mycelium in Lasiodiplodia spp.
To evaluate aerial mycelium production as a phenotypic characteristic to differentiate among species, 2 d-old cultures of selected isolates were each used to inoculate a 3 mm diam. plug of each culture into a glass tube containing 5 mL of PDA medium. Tubes were incubated at 28°C for 5 d and the elevations of mycelia were measured.

Pathogenicity tests of selected Lasiodiplodia isolates
Based on the analyses of the morphological and genetic results, the isolates MXL28BC, MXCS01BC, MX50BC, MXV5BC, MXVSM1b, MXVSM6, MXVS-M16a, MXVSM18, and MXVS21b were selected for pathogenicity tests. Grapevine plants of 'Cabernet Sauvignon' were used to evaluate the pathogenicity of these Lasiodiplodia isolates. Inoculation of each test plant was carried out through a mechanical wound in woody tissue made with a drill bit (2 mm diam.), and a mycelium plug of a selected isolate was placed inside the hole. An isolate of L. gilanensis UCD256Ma (formerly L. theobromae) (Úrbez-Torres et al., 2006; Obrador-Sánchez and Hernandez-Martinez, 2020) was used for comparisons. Plugs of sterile PDA were used in control plants, and all drill wounds were covered with Parafilm®. The grapevine plants were left in greenhouse conditions for 2 months. Samples were then taken to measure the length of the necrotic lesion caused by Lasiodiplodia isolates, and attempts were made to recover the inoculated fungus onto PDA. The experiments in plants were conducted twice. Statistical analyses were carried out using oneway ANOVA followed by post hoc Fisher LSD analyses, with α < 0.05 for determination of significant differences in virulence between isolates using STATISTICA 8.0.

Host symptoms, and morphological characteristics of fungal isolates
Botryosphaeria dieback symptoms observed on sampled grapevine plants were mainly dead spurs, cordons, and arms, and shorter shoot internodes. The collected wood exhibited wedge-shaped cankers and necrotic lesions in the vascular bundles.
From necrotic tissue placed in PDA, rapid fungus growth was observed after 2 d. From these colonies, 23 fungal isolates with a similar phenotype were recovered, seven from Baja California and sixteen from Sonora. According to their morphological characteristics, these isolates were identified as Lasiodiplodia. Morphological characteristics included initially white colonies with abundant aerial mycelium, which became smoke-gray and produced pycnidia in PDA as they aged (Figure 2). Pycnidium induction allowed observation of hyaline and pigmented conidia in all the isolates (Figure 3). Inside pycnidia, only hyaline aseptate conidia, with granular contents, were observed, while one-septate pigmented conidia with longitudinal striations were mainly found in cirri (Figure 3). The dimensions (length and width) of 30 conidia per isolate were measured, and minimum, maximum, mean, and standard deviations were calculated (Table 2). Statistically significant differences in conidium dimensions were observed among the four analyzed Lasiodiplodia species. Isolates characterized as L. gilanensis, MX50 (av. = 28.5 × 16.6 mm), and MXCS01 (av. = 30.2 × 15.6 mm), produced larger and wider conidia than L. brasiliensis, L. crassispora, or L. exigua. Lasiodiplodia brasiliensis and L. crassispora isolates had similar sized conidia (respective mean lengths = 24.0 and 25.6. mm. The L. exigua isolates had shorter conidia (av. = 21.2 × 12.2 mm).

Optimum growth temperature and aerial mycelium production of Lasiodiplodia spp.
The Lasiodiplodia isolates selected had optimum growth temperatures of 28°C. Most of the isolates grew at greater than 20 mm d -1 at 30°C (Table 3). Lasiodiplodia exigua grew at up to a mean of 24.6 mm d -1 at 37°C, and this was the only species that grew at 40°C. Lasiodiplodia gilanensis had the least mycelium growth rate, with a maximum mean growth rate of 19.8 mm d -1 at 28°C.
All the Lasiodiplodia isolates produced aerial mycelium, but in L. gilanensis this was less (mean = 0.8 ± 0.4 mm) than for the other species. The most abundant and longest aerial mycelium was observed in L. exigua isolate MXVS5a (16 ± 4.8 mm), followed by L. brasiliensis (9.0 ± 2.56 mm). The species Lasiodiplodia crassispora produced less abundant aerial mycelium (5.4 ± 2.3 mm) than the other species, and this species melanized more rapidly than the other species ( Figure 5).

Evaluation of the pathogenicity of selected isolates of Lasiodiplodia spp.
Pathogenicity assays on grapevine plants showed that two-months post inoculation L. brasiliensis MXB-CL28 and MXVS18, and L. gilanensis MXCS01 were the most virulent isolates (Figure 5, C, D, and F), in the woody shoots induced necrotic lesions up to 6 cm in length around the inoculation site, and were significantly different from the other inoculated isolates. L. exigua MXVS21b caused necrotic lesions in length, similar to L. gilanensis UCD256Ma (Figure 5 and 6). L. crassispora MXBCV5 and MXVS1b caused lesion below 1 cm in length (Figure 5 and 6) and showed a non-significant difference in comparison to control plants. All isolates were recovered from the inoculate site at three days after incubation at 30°C on PDA plates, which confirmed Koch's postulates. Non-necrotic lesions were observed in the control plants, only the wound effect; instead, green tissue was found, which indicated tissue regeneration of the caused wound.

DISCUSSION
In this study, four Lasiodiplodia species causing Botryosphaeria dieback symptoms were identified from Mexican vineyards. Lasiodiplodia theobromae, the type species of Lasiodiplodia, is one of the most common species associated with Botryosphaeria dieback in grapevine (Úrbez-Torres, 2011;Fontaine et al., 2016), and for several years, it was the only known species within the genus. Later, L. theobromae was shown to be a complex of cryptic species (Alves et al., 2008), which led to taxonomic revision of Lasiodiplodia. As a result, fungal isolates previously reported as L. theobromae have been re-  classified as new species (Dissanayake et al., 2016;Cruywagen et al., 2017;Mehl et al., 2017;Tibpromma et al., 2018). Some species were subsequently reduced to synonymy (Zhang et al., 2021). The fungal rDNA internal transcribed spacer region (ITS) is the primary barcode used to identify fungal species, but in Lasiodiplodia spp., this region has low interspecific variation. The translation elongation factor 1-α (tef-1α) is more variable than ITS, and has been recommended as a secondary barcode region to estimate species identity for Botryosphaeriaceae (Lawrence et al., 2017), and this locus allowed us to segregate L. brasiliensis from L. theobromae. Pathogens associated with wood dieback diseases are generally found in vineyards that are at least 10 years old (Gubler et al., 2005), but we have isolated these fungi in younger vineyards in Mexico. Lasiodiplodia exigua, L. brasiliensis, and L. crassispora were recovered from the two Mexican viticulture areas (Baja California and Sonora), whereas L. gilanensis was only found in Baja California. Lasiodiplodia exigua was the most prevalent species. Previously, only L. theobromae was reported in Mexico in grapevine (Úrbez-Torres et al., 2008), but our phylogenetic analyses indicated that those isolates clustered with L. brasiliensis, suggesting that L. brasiliensis has been in Mexico for a long time.
The pathogenicity tests showed that the L. brasiliensis isolates MXBCL28 and MXVS18, and L. gilanensis isolate MXCS01 were the most virulent to grapevine plants 'Cabernet Sauvignon'. These isolates caused necrotic lesions to the host vascular systems at 2 months post-inoculation. Lasiodiplodia brasiliensis was also reported for the first time on grapevine in Brazil, and this was the most virulent species on green shoots, followed by L. theobromae (Correia et al., 2016). Lasidiplo-dia gilanensis was described for the first time from Iran, from an unknown tree showing branch dieback, cankers, and fruit rot (Abdollahzadeh et al., 2010). Considering isolate UCD256Ma, formerly identified as L. theo-