In February 2020, during a disease survey conducted in three commercial plantations of Eucalyptus on six-month-old to one-year-old trees, diseased leaves with typical symptoms of CLB (small, circular or elongated pale grey to pale brown to dark brown spots, that extend throughout the leaf blade), were observed and collected for fungal isolation and species characterization. On average, 50 diseased leaves were sampled from each Eucalyptus genotype, one leaf per tree, depending on the planted areas. The sampled Eucalyptus genotypes corresponded to E. urophylla, localized in the municipalities of Cidelândia (5°09'24"S, 47°46'26"W) and Itinga do Maranhão (4°34'43"S, 47°29'48"W), in the state of Maranhão, and to the E. grandis × E. brassiana hybrid genotype, in the microregion of Paragominas (3°10'51"S, 47°18'49"W), in the state of Pará, Brazil.

Samples were stored in paper bags and transported to the Laboratory of Forest Pathology at the Universidade Federal de Lavras. From each leaf, small segments of 1 cm² from the transition section between healthy and diseased tissue were cut and the surface was disinfected by washing with 1% sodium hypochlorite for 1 min, with 70% ethanol for 30 s and with sterilized water three times before culture on 2% malt extract agar (MEA; malt extract 20 g·L^-1, agar 20 g·L^-1, yeast extract 2 g·L^-1, sucrose 5 g·L^-1) plates at 25 °C. After 48 h of incubation, Calonectria-like mycelial plugs, 5 mm in diameter, were transferred to a fresh MEA plate and incubated at 25 °C until the fungus covered the plate completely. Induction of sporulation on MEA plates and single spore cultures was obtained following the procedures described by Alfenas et al. (2013). Each single spore culture was stored and maintained in a metabolically inactive state in dry culture and sterile water following Castellani’s method (Castellani 1939). Holotypes were deposited as herbaria in the Coleção Micológica do Herbário da Universidade de Brasília (UB). Ex-types were deposited as pure cultures in the Coleção de Culturas de Microrganismos do Departamento de Ciência dos Alimentos/UFLA (CCDCA) at Universidade Federal de Lavras (UFLA), Minas Gerais, Brazil. Ex-paratypes were deposited as pure cultures in the Laboratory of Forest Pathology (PFC) at UFLA.

Total genomic DNA was extracted from fresh mycelia of single spore cultures grown on malt extract broth (MEB; malt extract 20 g·L^-1, yeast extract 2 g·L^-1, sucrose 5 g·L^-1) for ten days at 25 °C in the dark. The protocol described by Lee and Taylor (1990) was followed with slight modifications; by adding 1.5 M NaCl and 2% polyvinylpyrrolidone (MW: 40000) to the lysis buffer; the DNA was precipitated directly with isopropanol without the use of 3 M NaOAc, and the DNA pellet was dried at room temperature overnight. A NanoDrop 1,000 spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) was used to quantify its concentration.

Based on a previous study (Liu et al. 2020), actine (act), calmodulin (cmdA), histone H3 (his3), RNA polymerase II (rpb2), translation elongation factor 1-alpha (tef1), and β-tubulin (tub2) genes were used as DNA barcodes due to provide a stable and reliable resolution to distinguish all Calonectria species. The primers ACT-512F and ACT-783R (Carbone and Kohn 1999) were used to amplify the act gene region; CAL-228F and CAL-2Rd (Carbone and Kohn 1999; Quaedvlieg et al. 2011) for the cmdA gene region; CYLH3F and CYLH3R (Crous et al. 2004) for the his3 gene region; fRpb2-5F and fRpb2-7cR (Liu et al. 1999; Reeb et al. 2004) for the rpb2 gene region; EF1-728F (Carbone and Kohn 1999) and EF2 (O’Donnell et al. 1998) for the tef1 gene region and the primer pairs T1 (O’Donnell and Cigelnik 1997) and CYLTUB1R (Crous et al. 2004) for the tub2 gene region.

The PCRs were carried out in a 25 μL final volume containing molecular biology-grade water (Sigma–Aldrich, St. Louis, MO, USA) 1X PCR buffer (Promega, Madison, WI, USA), 2.5 mM MgCl₂, 0.2 mM deoxyribonucleotide triphosphate (dNTP) mix (Promega, Madison, WI, USA), 1 U GoTaq Flexi DNA Polymerase (Promega, Madison, WI, USA), 0.2 mM each primer, and 30 ng DNA template. DNA amplifications were conducted in a thermal cycler (5 PRIME G gradient Thermal Cycler, Techne, Staffordshire, UK). The PCR conditions for the act, cmdA, his3, tef1, and tub2 gene regions were as follows: an initial denaturation step at 95 °C for 5 min; then 35 amplification cycles at [94 °C for 30 s; 52 °C for 1 min; 72 °C for 2 min], and a final extension step at 72 °C for 5 min. For the rpb2 gene region, a touchdown PCR protocol was used: an initial denaturation step at 95 °C for 5 min, then (95 °C for 30 s, 57 °C for 30 s, 72 °C for 90 s) × 10 cycles, (95 °C for 30 s, 57 °C for 45 s, 72 °C for 90 s + 5 s/cycle increase) × 30 cycles, and a final extension step at 72 °C for 10 min.

PCR products were separated by electrophoresis at 120 V for 1 h in a 1.2% agarose gel, stained with Diamond Nucleic Acid Dye (Promega, Madison, WI, USA), and visualized using an ultraviolet light transilluminator. Successful PCR products were purified and sequenced in both directions using the same primer pairs used for amplification by Macrogen Inc. (Macrogen, Seoul, Korea). Raw sequences from each gene region were edited, consensus sequences were generated using SeqAssem software ver. 07/2008 (Hepperle 2004), and the sequences generated in this study were deposited in the NCBI/GenBank database (http://www.ncbi.nlm.nih.gov).

The generated sequences were aligned with other sequences of closely related Calonectria spp. obtained from GenBank (Table 1), using the online interface of MAFFT v. 7.0 (Katoh et al. 2019, http://mafft.cbrc.jp/alignment/server) with the alignment strategy FFT-NS-i (Slow; interactive refinement method). Alignments were manually corrected using MEGA7 (Kumar et al. 2016).

The partition homogeneity test (PHT) described by Farris et al. (1995) was conducted to determine if data for six genes could be combined using PAUP 4.0b10 (Swofford 2003). To determine the phylogenetic relationships among species, phylogenetic analyses based on maximum parsimony (MP), maximum likelihood (ML), and bayesian inference (BI) were conducted on the individual gene regions and their concatenated dataset, depending on the sequence availability.

Maximum parsimony analysis was performed using PAUP 4.0b10 (Swofford 2003), with phylogenetic relationships estimated by heuristic searches, random stepwise addition sequences, and tree bisection and reconnection (TBR) branch swapping. Gaps were treated as missing data, and all characters were unordered and weighted equally. The measures calculated for parsimony included the tree length (TL), consistency index (CI), homoplasy index (HI), retention index (RI), and rescaled consistency index (RC). Statistical support for branch nodes was assessed with 1,000 bootstrap replicates.

The best evolutionary model of nucleotide substitution for each gene region was selected according to the Akaike Information Criterion (AIC) using MODELTEST v. 3.4 (Posada and Crandall 1998) for ML analyses and MRMODELTEST v. 2 (Nylander 2004) for BI analyses.

ML analyses for individual gene regions were performed using PAUP 4.0b10 (Swofford 2003). The ML models used were K80 + G (act), TVM + G (cmdA), TrN + G (his3), SYM + G (rpb2), TVM + I + G (tef1) and HKY + I (tub2). Statistical support for branch nodes was assessed with 1,000 bootstrap replicates. A partitioned ML analysis was performed using IQ-TREE (Nguyen et al. 2015) as implemented in the IQ-TREE web server (http://iqtree.cibiv.univie.ac.at, Trifinopoulos et al. 2016) by using partition models (Chernomor et al. 2016). Branch support values were evaluated based on 10,000 replicates for ultrafast bootstrapping (UFBoot2) (Hoang et al. 2018).

Individual and partitioned BI analyses were performed using MRBAYES v.3.2.7a (Ronquist et al. 2012) on XSEDE at the CIPRES Science Gateway v.3.3 (http://www.phylo.org/). The BI models used were K80 + G (act), GTR + G (cmdA and his3), SYM + G (rpb2), GTR + I + G (tef1) and HKY + I (tub2). A Markov Chain Monte Carlo (MCMC) algorithm was employed, and two independent runs of four MCMC chains (three hot and one cold) were run in parallel simultaneously starting from random trees for 10⁷ generations (individual gene regions) and 30⁷ generations (concatenated dataset), sampling trees every 1,000 generations. The distribution of log-likelihood scores was examined with TRACER v.1.5 (Rambaut and Drummond 2007) to determine the whether the stationary phase of each search was reached and whether chains had achieved convergence. The convergence of the chains was also assessed by the convergent diagnostics of the effective sampling site (ESS), the potential scale reduction factor (PSRF), and the average standard deviation of split frequencies (ASDSF) (Ronquist et al. 2019). The first 25% of saved trees were discarded as the “burn-in” phase, and posterior probabilities (PP) were computed using the remaining trees. Trees were visualized in FIGTREE v. 1.4.4 (Rambaut 2009) and edited in INKSCAPE v. 1.0 (https://inkscape.org).

Phylogenetically closely related species were analyzed using the Genealogical Concordance Phylogenetic Species Recognition (GCPSR) model (as described by Taylor et al. 2000) by performing a pairwise homoplasy index (Φw) test (PHI) (Bruen et al. 2006). The PHI test was performed in SPLIT TREE4 v.4.16.1. (https://uni-tuebingen.de) (Huson and Bryant 2006) to determine the recombination level within phylogenetically closely related species. Only the gene regions that were available for all compared individuals were used. Gaps’ sites were excluded. Significant recombination was considered at a PHI index below 0.05 (Φw < 0.05). The relationships between closely related taxa were visualized by constructing a phylogenetic network from the concatenated datasets using the LogDet transformation and the NeighborNet method; the resultant networks were displayed with the EqualAngle algorithm (Dress and Huson 2004). Bootstrap analysis was then conducted with 1,000 replicates.

The mating-type idiomorph of each Calonectria species isolate was determined through PCR by using the primer pairs Cal_MAT111_F/Cal_MAT111_R and Cal_MAT121_F/Cal_MAT121_R, which amplify the MAT1-1-1 and MAT1-2-1 genes using the protocol described by Li et al. (2020). Additionally, sexual compatibility tests were performed for all the single-spore isolates of both species on minimal salt agar (MSA; Guerber and Correll 2001) by crossing them in all possible combinations, following the procedure described by Lombard et al. (2010a). The plates were stacked in plastic bags and incubated at 25 °C for 12 weeks.

Morphological characterization of representative isolates of each Calonectria species identified by phylogenetic analyses was performed as described by Liu and Chen (2017). Optimal growth temperatures were determined by incubating the representative isolate at temperatures ranging from 5 °C to 30 °C at 5 °C intervals in the dark on MEA plates (three replicates per isolate were used). Colonial characteristics (diameter, color, and texture of colonies) were determined by inoculating the isolates on MEA plates at 25 °C in the dark after seven days of incubation.

One representative isolate of each Calonectria species was selected for inoculation. Healthy leaves of three short cut branches from an approximately eleven-month-old Eucalyptus plants were inoculated with suspensions of 1 × 10⁴ conidia·mL^-1 obtained from single spore cultures. The conidia suspensions for each isolate were prepared using the method described by Graça et al. (2009). Calonectria paragominensis was inoculated on E. grandis × E. brassiana hybrid genotype and C. imperata on E. urophylla genotype. The inoculation consisted of spraying the conidia suspension until the suspension run off the leaves. Sterile water was sprayed onto healthy leaves as the negative control. The branches with inoculated leaves were covered with plastic bags to maintain high humidity and kept at 25 °C under a photoperiod of 12 h for 72 h. After that time, the plastic bags were removed, and necrotic symptoms were observed.

A total of 34 isolates with the typical morphology of Calonectria species were obtained from infected leaves of the Eucalyptus genotypes sampled. Based on preliminary phylogenetic analyses of the tef1 and tub2 gene regions (data not shown), nine isolates were selected for further studies (Table 1).

Sequences from 50 isolates corresponding to 25 Calonectria species closely related to the isolates obtained in this study were downloaded from GenBank (Table 1). For the nine isolates selected in this study, five resided in the Calonectria spathiphylli species complex (CSSC), and four resided in the Calonectria candelabrum species complex (CCSC). Both Calonectria complexes belong to the Prolate Group, whose species are characterized by their clavate to pyriform to ellipsoidal vesicles (Liu et al. 2020). Therefore, both complexes were combined into a single sequence dataset for phylogenetic analyses, including two strains of Calonectria gracilipes as the outgroup taxa.

Alignments for each gene region and the concatenated dataset were as follows: act (36 isolates, 267 characters), cmdA (58 isolates, 485 characters), his3 (59 isolates, 439 characters), rpb2 (28 isolates, 863 characters), tef1 (58 isolates, 496 characters), tub2 (59 isolates, 511 characters) and concatenated (59 isolates, 3061 characters). The PHT generated a p value of 0.01 for the concatenated dataset, suggesting some incongruence in the datasets for the six regions and the accuracy of the combined data could have suffered relative to the individual partitions (Cunningham 1997). Although the p value was low, the different gene regions were combined because the significance threshold of 0.05 may be too conservative and it has been shown that combining incongruent datasets improves phylogenetic accuracy (Sullivan 1996; Cunningham 1997); moreover, this approach was followed by several previous studies (Lombard et al. 2016; Pham et al. 2019; Liu et al. 2020, 2021).

Tree topologies derived from the MP, ML, and BI analyses of the individual gene regions were similar overall, but the relative positions of some Calonectria species slightly differed. Moreover, the concatenated dataset formed well-supported lineages in the MP, ML, and BI analyses. Only the ML trees are presented in this study (Fig. 1, Suppl. material 1: Figs S1–S6). The concatenated dataset had 466 parsimony-informative characters, 67 parsimony-uninformative characters, and 2,528 constant characters. Analysis of the 466 parsimony-informative characters yielded 2 equally parsimonious trees, with TL = 862, CI = 0.7042, HI = 0.2958, RI = 0.9192, RC = 0.6472. For the partitioned BI analysis, the convergence of the chains was confirmed by an ESS > 200, a PSRF approaching 1, and an ASDSF equal to 0.000793. The aligned sequences were deposited in TreeBASE (http://treebase.org; No. 29573).

Figure 1. 

Phylogenetic tree based on maximum likelihood analysis of concatenated act, cmdA, his3, rpb2, tef1 and tub2 gene regions. Bootstrap support values ≥ 80% for maximum parsimony (MP), Ultrafast bootstrap support values ≥ 95% for maximum likelihood (ML), and posterior probability (PP) values ≥ 0.95 from BI analyses are presented at the nodes (MP/ML/PP). Bootstrap values below 80% (MP), 95% (ML) and posterior probabilities below 0.80 are marked with “-”. Ex-type isolates are indicated by “▲”, isolates highlighted in bold were sequenced in this study, and novel species are in blue and orange. C. gracilipes was used as outgroup. The scale bar indicates the number of nucleotide substitutions per site.

Phylogenetic analyses of the six individual gene regions showed that the five isolates from the CSSC were clustered in an independent clade (Suppl. material 1: Figs S1–S6). Based on the concatenated dataset of the six genes, the five isolates formed a new, strongly defined phylogenetic clade that was distinct from the other Calonectria species of the CSSC and was supported by high bootstrap values (MP = 100%, ML = 100%) and high values of posterior probability (1.0) (Fig. 1). A total of 41 fixed unique single nucleotide polymorphisms (SNPs) were identified in the new phylogenetic clade of the five isolates in comparison with their phylogenetically closely related Calonectria species in the six-gene concatenated dataset (Table 2). The results of these phylogenetic and SNP analyses indicate that the five isolates in the CSSC represent a distinct, undescribed species, which we named C. paragominesis.

Phylogenetic analyses of the individual gene regions of act, cmdA, his3, rpb2, and tub2 showed that the four isolates that resided in the CCSC were clustered in an independent clade (Suppl. material 1: Figs S1–S4, S6). However, the phylogenetic tree based on tef1 showed that three of those isolates formed an independent clade, while one isolate was closely related to C. metrosideri, C. pseudometrosideri, and C. candelabrum (Suppl. material 1: Fig. S5). Based on the concatenated dataset of the six genes, the four isolates formed a new, strongly defined phylogenetic clade that was distinct from other Calonectria species in the CSSC and was supported by high bootstrap values (MP = 91%, ML = 99%) and high values of posterior probability (1.0) (Fig. 1). The four isolates of the new phylogenetic clade were distinguished from their phylogenetically closely related Calonectria species using SNP analyses for the six-gene concatenated dataset, by presenting eight unique SNPs from a total of 78 SNPs (Table 3). The results of these phylogenetic and SNP analyses indicate that the four isolates in the CCSC represent a distinct, undescribed species, which we named C. imperata.

A PHI test using a five-locus concatenated dataset (act, cmdA, his3, tef1, tub2) was performed to determine the recombination level among C. paragominensis and its phylogenetically closely related species, C. densa, C. humicola, C. spathiphylli and C. pseudospathiphylli. A value of Φw = 0.2879 revealed no significant genetic recombination events, and this relationship was supported with a high bootstrap value (100%) in the phylogenetic network analysis, indicating that they are different species (Fig. 2A).

Figure 2. 

Results of the pairwise homoplasy index (PHI) test for C. paragominensis and C. imperata. Phylogenetic networks constructed using the LogDet transformation and the NeighborNet method and displayed with the EqualAngle algorithm. Bootstrap support values > 80% are shown. Φw < 0.05 indicate significant recombination. New species described in this study are highlighted in bold, with blue (A) and orange (B) lines.

A PHI test using a four-locus concatenated dataset (cmdA, his3, tef1, tub2) was performed to determine the recombination level among C. imperata and its phylogenetically closely related species, C. brassiana, C. glabeicola, C. piauiensis, and C. venezuelana. A value of Φw = 0.1587 revealed no significant genetic recombination events, and this relationship was supported with a high bootstrap value (94%) in the phylogenetic network analysis, indicating that they are different species (Fig. 2B).

MAT1-1-1 and MAT1-2-1 genes were amplified in all isolates of each identified species, indicating that they are putatively homothallic. However, after a twelve-week mating test on MSA, all isolates failed to yield sexual structures, indicating that they have lost the ability to be self-fertile or have retained the ability to favor outcrossing rather than selfing.

Based on phylogenetic analyses, GCPSR, and network analyses, the nine isolates presented two strongly defined phylogenetic clades in both the Calonectria spathiphylli species complex and the Calonectria candelabrum species complex. Morphological differences, especially in the macroconidia and stipe dimensions, were observed between each phylogenetic clade and its phylogenetically closely related species (Table 4). Thus, the fungi isolated in this study represent two new species of Calonectria and are described as follows:

 Calonectria paragominensis E.I.Sanchez, T.P.F.Soares & M.A.Ferreira, sp. nov.

MycoBank No: 843460

Fig. 3

Etymology

The term “paragominensis” refers to the microregion of Paragominas, Brazil, which is the place where the fungus was collected.

Diagnosis

Calonectria paragominensis differs from the phylogenetically closely related species C. densa, C. humicola, C. spathiphylli and C. pseudospathiphylli with respect to its macroconidia dimensions.

Type

Brazil,• Pará state, Paragominas microregion; 3°10'51"S, 47°18'49"W; From infected leaves of E. grandis × E. brassiana; 20 Feb. 2020; M.A. Ferreira; holotype: UB24349, ex-type: CCDCA 11648 = PFC1. GenBank: act = ON009346; cmdA = OM974325; his3 = OM974334; rpb2 = OM974343; tef1 = OM974352; tub2 = OM974361.

Description

Sexual morph unknown. Macroconidiophores consisted of a stipe, a suite of penicillate arrangements of fertile branches, a stipe extension, and a terminal vesicle; stipe septate, hyaline, smooth, (112–)135–207(–281) × (2–)2.6–3.5(–4) μm; stipe extension septate, straight to flexuous, (123–)147–220(–295) μm long, (1.5–)1.9–2.4(–3) μm wide at the apical septum, terminating in a globose to sphaeropedunculate vesicle, (8–)8.5–10.5(–12) μm diam; lateral stipe extensions (90° to the axis) also present. Conidiogenous apparatus was (40–)56–88(–113) μm long, (45–)67–107(–129) μm wide; primary branches aseptate or 1-septate, (15.7–)18.4–25.9(–30.6) × (3.3–)4–6(–6.5) μm; secondary branches aseptate, (12.7–)14.3–19.6(–22.1) × (3–)3.5–5(–6) μm; tertiary branches aseptate, (9.9–)11.6–15.3(–17.9) × (2.8–)3.6–5.3(–6.4) μm; additional branches (–4), aseptate, (10.3–)11–13.2(–14) × (3–)3.2–4.4(–5) μm; each terminal branch produced 2–4 phialides; phialides doliiform to reniform, hyaline, aseptate, (8–)9.1–11.8(–14) × (2–)2.7–4.1(–6) μm, apex with minute periclinal thickening and inconspicuous collarette. Macroconidia were cylindrical, rounded at both ends, straight, (47–)56–66(–71) × (4–)4.8–5.9(–7) μm (av. = 61 × 5 μm), (1–3) septate, lacking a visible abscission scar, held in parallel cylindrical clusters by colorless slime. Megaconidia and microconidia were not observed.

Figure 3. 

Calonectria paragominensis A, B macroconidiophore C lateral stipe extensions D, E conidiogenous apparatus with conidiophore branches and doliiform to reniform phialides F, G globose to sphaeropedunculate vesicles H, I one, two, and three-septate macroconidia. Scale bars: 20 μm.

Culture characteristics

Colonies formed abundant white aerial mycelium on MEA at 25 °C after seven days, with irregular margins and moderate sporulation. The surface had white to buff outer margins, and sienna to amber in reverse with abundant chlamydospores throughout the medium, forming microsclerotia. The optimal growth temperature was 23.8 °C, with no growth at 5 °C; after seven days, colonies at 10 °C, 15 °C, 20 °C, 25 °C, and 30 °C reached 7 mm, 23 mm, 38.3 mm, 36.1 mm, and 31.8 mm, respectively.

Substratum

Leaves of E. grandis × E. brassiana.

Distribution

Northeast Brazil.

Other specimens examined

Brazil,• Pará state, Paragominas microregion; From infected leaves of E. grandis × E. brassiana; 20 Feb. 2020; M.A. Ferreira; cultures PFC2, PFC3, PFC4, PFC5.

Notes

C. paragominensis is a new species in the C. spathiphylli species complex (Liu et al., 2020). Morphologically, C. paragominensis is very similar to C. densa, since both form lateral stipe extensions, which have not been reported for the other three species in the complex. However, the macroconidia of C. paragominensis (av. 61 × 5 μm) are longer than those of C. densa (av. 54 × 6 μm), C. humicola (av. 51 × 5 μm) and C. pseudospathiphylli (av. 52 × 4 μm) but smaller than those of C. spathiphylli (av. 70 × 6 μm).

 Calonectria imperata E.I.Sanchez, T.P.F.Soares & M.A.Ferreira, sp. nov.

MycoBank No: 843461

Fig. 4

Etymology

The term “imperata” is in honor of the city of Imperatriz, Brazil, which was close to the place where the fungus was collected.

Diagnosis

Calonectria imperata differs from the phylogenetically closely related species C. brassiana, C. glaebicola, C. piauiensis and C. venezuelana with respect to the number of unique alleles and stipe dimensions.

Type

Brazil,• Maranhão state, Cidelândia municipality; 5°09'24"S, 47°46'26"W; From infected leaves of E. urophylla; 20 Feb. 2020; M.A. Ferreira; holotype: UB24350, ex-type: CCDCA 11649 = PFC6. GenBank: act = ON009351; cmdA = OM974330; his3 = OM974339; rpb2 = OM974348; tef1 = OM974357; tub2 = OM974366.

Description

Sexual morph unknown. Macroconidiophores consisted of a stipe, a suite of penicillate arrangements of fertile branches, a stipe extension, and a terminal vesicle; stipe septate, hyaline, smooth, (135–)151–198(–227) × (2–)2.6–3.4(–4) μm; stipe extension septate, straight to flexuous, (151–)169–220(–254) μm long, (1.5–)1.9–2.7(–3) μm wide at the apical septum, terminating in an ellipsoidal to narrowly obpyriform vesicle (3–)3.1–4.6(–6) μm diam. Conidiogenous apparatus was (50–)66–100(–127) μm long, (41–)62–89(–110) μm wide; primary branches aseptate, (14.6–)19–24.8(–28.5) × (2.5–)3.2–4(–4.5) μm; secondary branches aseptate, (12.1–)13.5–18.2(–24.2) × (2.3–)2.8–3.7(–4) μm; tertiary branches aseptate, (10.1–)11–15(–18.1) × (1.9–)2.3–3.2(–4.1) μm; each terminal branch producing 2–4 phialides; phialides doliiform to reniform, hyaline, aseptate, (8–)9.1–13(–15) × (2–)2.7–3.3(–4) μm, apex with minute periclinal thickening and inconspicuous collarette. Macroconidia were cylindrical, rounded at both ends, straight, (38–)43–49(–52) × (2–)2.7–3.2(–4) μm (av. = 46 × 3 μm), (–1) septate, lacking a visible abscission scar, held in parallel cylindrical clusters by colorless slime. Megaconidia and microconidia were not observed.

Figure 4. 

Calonectria imperata A–C macroconidiophore D–G ellipsoidal to narrowly obpyriform vesicles H–J conidiogenous apparatus with conidiophore branches and doliiform to reniform phialides K, L macroconidia. Scale bars: 20 μm.

Culture characteristics

Colonies formed moderate aerial mycelium on MEA at 25 °C after seven days, with moderate sporulation. The surface had white to buff outer margins, and sepia to umber in reverse with abundant chlamydospores throughout the medium, forming microsclerotia. The optimal growth temperature was 25 °C, with no growth at 5 °C; after seven days, colonies at 10 °C, 15 °C, 20 °C, 25 °C, and 30 °C reached 10.1 mm, 25.5 mm, 29.1 mm, 44.5 mm, and 40.6 mm, respectively.

Substratum

Leaves of E. urophylla.

Distribution

Northeast Brazil.

Other specimens examined

Brazil,• Maranhão state, Cidelândia municipality; 5°09'24"S, 47°46'26"W; From infected leaves of E. urophylla; 20 Feb. 2020; M.A. Ferreira; cultures PFC7, PFC8, PFC9. Brazil• Maranhão state, Itinga do Maranhão; 4°34'43"S, 47°29'48"W; from infected leaves of E. urophylla; 20 Feb. 2020; M.A. Ferreira; culture PFC9.

Notes

C. imperata is a new species in the C. candelabrum species complex (Liu et al., 2020). Morphologically, C. imperata is very similar to its closest relatives, from which it can be distinguished based on stipe dimensions and phylogenetic inference. Stipe of C. imperata (135–227 × 2–4 μm) is larger than those of C. piauiensis (50–110 × 4–6 μm), C. glaebicola (50–130 × 5–7 μm), and C. venezuelana (35–100 × 4–8 μm) but narrower than those of C. brassiana (55–155 × 5–8 μm). Additionally, C. imperata lacks lateral stipe extensions, which are present in C. piauiensis.

The conidia suspensions of the representative isolates of C. paragominensis and C. imperata produced lesion symptoms on leaves (Fig. 5E, F, I, J), but no lesions were observed on the negative control inoculations (Fig. 5G, H, K). The pathogens were reisolated from inoculated leaves but not from the negative controls and identified by the same morphological characteristics as the originally inoculated species, thus, fulfilling the requirements of Koch’s postulates.

Figure 5. 

Pathogenicity tests on leaves of Eucalyptus genotypes A, B surface and reverse of C. paragominensis on MEA plates after 14 days grown at 25 °C C, D surface and reverse of C. imperata on MEA plates after 14 days grown at 25 °C E, I lesions on leaves of E. grandis × E. brassiana induced by C. paragominensis 72 h after inoculation F, J lesions on leaves of E. urophylla induced by C. imperata 72 h after inoculation G, H, K no disease symptoms on leaves inoculated with sterile water (negative controls). Scale bars: 5 cm (E–K).