Multigene Phylogenetics and Morphology Reveal Five Novel Lasiodiplodia Species Associated with Blueberries

Botryosphaeriaceous fungi cause stem blight, canker and dieback in woody plants. During a survey on the fungal pathogens associated with blueberries in China, 135 blighted, cankered or dead blueberry branches were collected from Fujian and Shandong Provinces. Based on the morphological characterization and phylogenetic analyses of a concatenated ITS rDNA, tef1-α, TUB, and RPB2 loci, five new species of Lasiodiplodia, viz., L. clavispora, L. fujianensis, L. henanica, L. nanpingensis and L. paraphysoides were recognized. Detailed descriptions and illustrations, as well as multigene phylogenies, are provided in this paper. The diversity of plant pathogens on agriculturally and economically important plants is higher than anticipated.


Introduction
Blueberries (Vaccinium spp.) are perennial shrub fruit trees. The fruits are well known and are widely consumed for their protective properties against heart diseases and cancer, they can help to maintain bone strength and mental health and can regulate blood pressure [1]. Blueberries are widely distributed in temperate regions, such as North America, Europe, Canada and Northern China [2][3][4][5][6][7][8][9][10][11][12][13]. Due to their health benefits and economic value, blueberries have been commercially cultivated worldwide, particularly in the USA, Canada and European countries [14,15]. Blueberry cultivation started in 1981 in China, and productivity has reached 43,244 tons per year [16].
Botryosphaeriaceous fungi are a group of economically important plant pathogens [17][18][19][20]. They cause stem blight, canker or dieback in a wide range of hosts, including blueberries [8,9,11,15,[21][22][23][24][25]. In the USA, blueberry stem blight, caused by Botryosphaeria ribis, has been a major disease in commercial plantations in North Carolina [26,27]. Pathogenicity studies conducted show that stem dieback is caused by B. dothidea and canker by Lasiodiplodia corticis in blueberries in North Carolina. Neofusicoccum parvum was identified as the causal agent for blueberry stem blight and dieback in California and Mexico [2,4]. In Florida, the blueberry stem blight and dieback caused by Neofusicoccum ribis and Lasiodiplodia theobromae led to huge economic losses and were one of the most severe diseases in the local blueberry planting industry [8,9,28]. The incidence of blueberry stem blight and canker caused by Neofusicoccum parvum has been a limiting factor for blueberry production in Chile [29]. The incidence of blueberry blight and crown rot caused by N. ribis and L. theobromae was so severe in New Zealand that it resulted in an annual loss of about USD 500,000 due to yield losses and replanting [6,8,9,28]. Neofusicoccum parvum and N. austral caused blueberry stem dieback and canker in Spain [3,30]. Many more

Sample Collections and Fungal Isolation
One hundred and thirty-five blighted, cankered or dead blueberry branches were collected from Fujian (69 samples) and Shandong (66 samples) Provinces in China from April to November 2018. Diseased or dead twigs of blueberries (ca. 30 cm) were cut for sampling, from which the fungal strains were isolated. Wood segments (0.5 × 0.5 × 0.2 cm) were cut from the diseased lesion boundary or dead tissues and were subsequently surface sterilized and incubated in malt extract agar (MEA) at 28 • C for fungal strain isolation [13,[37][38][39]. The isolates were kept at ambient temperatures (about 26-28 • C) and grown in the dark.

Morphological Characterization
Fungal colonies were initially identified based on morphological characteristics. Fungal isolates were transferred to synthetic nutrient-poor agar (SNA) with sterilized pine needles for 3 weeks in order to induce sporulation. The pycnidia produced on the pine needles were morphologically described following the work by Dou et al. [35,40]. Microscopic observations were made from material mounted in water. Measurements of paraphyses, conidiogenous cells and conidia were made in water. For each new species, the measurements of 20 paraphyses, 20 conidiogenous cells and 50 conidia were taken under a Nikon Eclipse E600 microscope. Fungal isolates and herbarium specimens were deposited at the China General Microbiological Culture Collection Center (CGMCC) and the Mycological Herbarium of the Institute of Microbiology, Chinese Academy of Sciences (HMAS). The new species were established based on the guidelines outlined by Jeewon and Hyde [41].

DNA Extraction, PCR Amplification
DNA was extracted with the CTAB plant genome DNA fast extraction kit (Aidlab Biotechnologies Co, Ltd., Beijing, China) from the mycelium grown on MEA. PCR amplifications were performed using the Easy Taq PCR Super Mix kit (Beijing Transgen Biotech Co., Ltd., Beijing, China). The internal transcribed spacers of rDNA (ITS) were amplified and sequenced with the primers ITS-1 and ITS-4 [42]. The translation elongation factor-1α (tef1-α) was amplified and sequenced with primers EF1-688F and EF1-1251R [43]. The TUB gene was amplified and sequenced with primers Bt2a and Bt2b [44]. The RPB2 were amplified and sequenced using primers RPB2-LasF and RPB2-LasR [45]. PCR amplification and sequencing followed the protocol outlined by Zhang et al. [46]. PCR amplifications were performed using the Easy Taq PCR Super Mix kit (Beijing Transgen Biotech Co., Ltd., Beijing, China). For the ITS regions, the following PCR profile was used: 95 • C for 3 min, followed by 34 cycles of denaturation at 95 • C for 1 min, annealing at 52 • C for 30 s and elongation at 72 • C for 1 min, with a final extension step of 72 • C for 10 min. The PCR profiles for the tef1-α, TUB and RPB2 genes were same, except that 35 cycles of denaturation were used and the annealing temperature was 55 • C.

Sequence Alignment and Phylogenetic Analysis
The concatenated loci of ITS, tef1-α, TUB and RPB2 were used to infer the phylogenetic relationships of taxa within Lasiodiplodia. Alignments were conducted in MEGA v. 6, and phylogenetic analyses performed in PAUP v. 4.0b10 and MrBayes v. 3.1.2 [47][48][49]. Prior to phylogenetic analysis, ambiguous sequences at the start and end were deleted and gaps manually adjusted in order to optimize the alignments. Maximum Parsimony (MP) was used to conduct heuristic searches, as implemented in PAUP with the default options method [50]. Analyses were conducted under different parameters of maximum parsimony criteria [50]. Clade stability was assessed in a bootstrap analysis with 1000 replicates, random sequence additions with maxtrees set to 1000 and other default parameters, as implemented in PAUP. For the MrBayes analysis, the best-fit model of nucleotide evolution (GTR+I+G) was selected by the Akaike information criterion [51] in MrModeltest v. 2.3. The metropolis-coupled Markov Chain Monte Carlo (MCMCMC) approach was used to calculate posterior probabilities [47]. Bayesian inference (BI) analysis with MrBayes revealed that the Markov chain Monte Carlo (MCMC) steady state was reached after fewer than 19,820,000 generations (the average standard deviation of split frequencies was constantly below 0.01). A conservative burn-in of 198,200 trees was chosen, and a full analysis of 20,000,000 generations was carried out with sampling every 100 generations. Trees were viewed in TREEVIEW [52]. The nucleotide sequences generated in this paper were deposited in GenBank (Table 1). Trees and alignments were deposited in TreeBase (http://purl.org/phylo/treebase/phylows/study/TB2: S24322?x-access-code=1443788eea51ad240fcd94b3927ffb1a&format=html, accessed on 15 June 2021).
Culture characteristics: Colonies on MEA were initially white with moderately dense aerial mycelia reaching the lid of the plate and became olive grey on the surface after 5 d, with the reverse side of the colonies being pale grey to grey. Colonies reached 18 mm on MEA after 24 h in the dark at 28 • C, and were more than 55 mm after 48 h.
Lasiodiplodia fujianensis Y. Zhang ter., Y. Wang, sp. nov ( Figure 3). MycoBank: MB 830996. The etymology is in reference to the location, Fujian province, where the species was first reported.
The sexual stage was not observed. Conidiomata were stromatic, produced on both sterilized pine needles and SNA within 10 days, semi-immersed, uniloculate, black, covered by greyish brown mycelium, and were up to 1.3 mm in diameter. Paraphyses were filiform, arising from the conidiogenous layer that extended above the level of developing conidia and were up to 95 µm long and 3 µm wide, aseptate, hyaline, tip rounded, and unbranched. Conidiophores reduced to conidiogenous cells. Conidiogenous cells were holoblastic, hyaline, discrete, smooth, and thin walled, Culture characteristics: Colonies on MEA were initially white with moderately dense aerial mycelia reaching the lid of the plate and became ash-grey on the surface after 5 d, with the reverse side of the colonies being pale grey to grey. Colonies reached 45 mm on MEA after 24 h in the dark at 28 • C, and more than 90 mm after 48 h.
Culture characteristics: Colonies on MEA were initially white with moderately dense aerial mycelia reaching the lid of the plate and becoming ash-grey on the surface after 5 d, with the reverse side of the colonies being pale grey to grey. Colonies reached 17 mm on MEA after 24 h in the dark at 28 • C, and more than 60 mm after 48 h.
Lasiodiplodia paraphysoides Z. P. Dou, Y. Wang, Y. Zhang ter sp. nov. (Figure 6). Mycobank: MB 817655. The etymology is in reference to the long and multiseptate paraphyses. The sexual stage was not observed. Conidiomata were stromati, produced on both sterilized pine needles on SNA within 14 days, and were solitary, globose, semi-immersed or superficial, uniloculate, dark brown to black, covered with brown mycelium, up to 1.8 mm diam, and often had a long papilla, which was up to 383 µm long and 113 µm wide. Paraphyses were filiform, arising from the conidiogenous layer, extending above the level of developing conidia, up to 125 µm long and 7 µm wide, and were cylindrical, thin-walled, hyaline, tip rounded, initially aseptate, becoming up to 1-2-septate when mature, branched, occasionally basal, and were middle or apical swollen cells. Conidiophores were reduced to conidiogenous cells. Conidiogenous cells were holoblastic, hyaline, discrete, smooth, thin-walled, and were cylindrical to ampulliform, (8-  Notes: Phylogenetically, L. paraphysoides was closely related to L. citricola and an unidentified taxon, Lasiodiplodia sp. Lasiodiplodia paraphysoides (CGMCC 3. 19174) and differred from its closest phylogenetic neighbor L. citricola (IRAN1522C) (Figure 1) by unique fixed alleles in two loci based on alignments of the separate loci deposited in TreeBASE (S25538), by 4 bp in tef1-α (0.72 %, gaps included) ( Table 2). Morphologically, the long papilla of the conidiomata of L. paraphysoides delineated itself from the non-papillate conidiomata of L. citricola [54]. Furthermore, the conidiogenous cells of L. citricola had 1-2 annellations, which also differed from the holoblastic conidiogenous cells of L. paraphysoides [54].

Discussion
In this study, we recovered five new species of Lasiodiplodia associated with stem blight and/or canker of blueberries, namely, L. clavispora, L. fujianensis, L. henanica, L. nanpingensis and L. paraphysoides, and they were characterized in terms of their morphology and their phylogenetic relationships to other species of Lasiodiplodia. Phylogenetically, each of these five newly described species formed a well-supported subclade close to other species (Figure 1). Species of Lasiodiplodia were mostly differentiated based on the morphology of the conidia (especially dimensions) and paraphyses [17,35]. In this study, we attempted to use other morphological characters, such as the dimensions and papillate nature of conidiomata, as well as annelations of conidiogenous cells, but to what extent these are phylogenetically significant warrants further investigation.
Geographically, Lasiodiplodia tends to be distributed in tropical or subtropical areas or in warm temperate areas associated with various stem diseases of woody substrates [8,9,11,22,33,35]. For instance, L. mediterranea and L. pseudotheobromae have been reported as canker-causing agents of grapevine and other woody hosts in Italy, Algeria and Tunisia [58]. The stem blight and crown rot of blueberry caused by L. theobromae have been reported in Florida in the USA, as well as in Zhejiang Province and Shanghai in China [8][9][10][11]. The cane dieback of blueberry caused by L. mediterranea has been reported in Washington in the USA [33]. In China, blueberry stem blight and dieback caused by L. chinensis have been reported in Shandong Province [40,56]. The stem blight of blueberry caused by L. vaccinii was reported in a greenhouse plantation in Beijing, where it was warm with high humidity [13]. All the five species of Lasiodiplodia newly described in this study were from Fujian and Shandong Province, which belong to subtropical or warm temperate areas in China. The distribution of Lasiodiplodia spp seems largely influenced by environmental conditions, such as temperature, humidity, elevation, as well as the prevalence of alternative hosts instead of their host associations [28,59].
We also compared our species with newly described species recently published by de Silva et al. [60]. From a phylogenetic perspective, our new species are quite different, except for one, L. fujianensisis. The latter is basal to L. thailandica and L. iraniensis, which are known species. de Silva et al. [60] also reported that their new species, L. endophytica from Magnolia plant, are phylogenetically closely related to L. thailandica and L. iraniensis albeit in a distinct independent lineage with weak support. To avoid any ambiguous taxonomic interpretation in connection with the identification of L. fujianensisis, we compared DNA base pair differences with L. endophytica. DNA sequences from the TEF protein coding region for L. endophytica is quite short (271 bp) and we still found two major differences, which supports that our species is different. With respect to the DNA sequences from the Beta tubulin gene region, L. fujianensisis was 100% similar to L. endophytica. Could this be pointing to the fact that these two taxa are conspecific? This might be true, but we compared existing DNA sequences of the Beta tubulin from other published species, such as L. pseudotheobromae, L. jatrophicola, L. vitis and L. iraniensis and found that they are identical to L. fujianensisis and L. endophytica. The taxonomy of Lasioplodia has been rather controversial [35]. While some are proponents of a taxonomy based on morphological characteristics, others argue that more protein genes should be included in the taxonomy, especially at the species level. However, the protein genes might not be useful, at least in some fungal groups, because they have reached saturation and are possibly less informative than has been anticipated. Mycologists also encounter difficulties when analyzing DNA sequence data for many bitunicate fungi. In this case, even with L. endophytica, de Silva et al. [60] demonstrated that single gene phylogenies could reveal extensive incongruence (Figures 1-3), which can be found in the supplementary information provided by de Silva et al. [60]. We could not compare the morphs of L. fujianensisis to L. endophytica as the latter was isolated as an endophyte and did not produce any fruiting bodies in culture. There is also a need to update the name of the GB accession numbers of MK501838, MK584572, and MK550606 as these are labelled as "Lasiodiplodia sp. NIS-2019a isolate", but we presume that it should be Lasiodiplodia endophytica.