All diseased samples were collected from the Medical Plants Herb Garden, in Chongqing City, Nanchuan County, China. This garden is located in a region of subtropical humid monsoon climate and has conserved more than 3000 kinds of medicinal plants. In this study, all fungal strains were isolated by the single-spore technique in order to obtain pure cultures following the method of Chomnunti et al. (2014). Single spores were transferred to potato-dextrose agar (PDA) and incubated at room temperature (28 °C). After several weeks of incubation, the morphological characters were recorded following the methods of Manamgoda et al. (2011, 2012). Conidia and conidiophores were observed using a compound microscope (Nikon Eclipse E600 DIC microscope and a Nikon DS-U2 camera or a Nikon 80i compound microscope fitted with a Canon 450D digital camera). The holotype specimen was deposited in the Herbarium of the Department of Plant Pathology, Agricultural College, Guizhou University (HGUP). Ex-type cultures were also deposited in the culture collection at the Department of Plant Pathology, Agriculture College, Guizhou University, P.R. China (GUCC).

Fungal cultures were grown on PDA until nearly covering the whole Petri-dish (90 mm) at 28 °C. Fresh fungal mycelia were scraped with sterilised scalpels. A BIOMIGA Fungus Genomic DNA Extraction Kit (GD2416) was used to extract fungal genome DNA. DNA Amplification was performed in a 25 μL reaction volume which contained 2.5 μL 10 × PCR buffer, 1 μL of each primer (10 μM), 1 μL template DNA and 0.25 μL Taq DNA polymerase (Promega, Madison, WI, USA). Primers ITS4 and ITS5 (White et al. 1990) were used to amplify the ITS region. The thermal cycling programme was: 3 min initial denaturation at 95 °C, followed by 30 cycles of 30 s denaturation at 94 °C, 30 s primers annealing at 52 °C, 1 min extension at 72 °C and a total 10 min extension at 72 °C. To amplify the GPD1 gene, the primers gpd1 and gpd2 were used (Berbee et al. 1999). The amplification programme included an initial denaturation step at 96 °C for 2 min, followed by 35 PCR cycles with 1 min at 96 °C, 1 min at 52 °C and 45 s at 72 °C with a final 10 min extension at 72 °C. The TEF1 and 28S regions were amplified using EF-526F/1567R and LR5/LROR primers respectively (Schoch et al. 2009). The 28S amplification programme included an initial denaturation step at 95 °C for 3 min followed by 30 cycles of 40 s denaturation at 94 °C, 50 s primer annealing at 52 °C, 1 min extension at 72 °C. The same PCR reaction was used to amplify TEF1 with the only change being the annealing temperature at 54 °C.

DNA sequences from these isolates and reference sequences were downloaded from GenBank and analysed by maximum parsimony (MP) and maximum likelihood (ML) (Table 1). Sequences were optimised manually to allow maximum alignment and maximum sequence similarity, as detailed in Manamgoda et al. (2012). The alignment document of four phylogenetic markers has been submitted to TreeBase (https://treebase.org/; Accession number: 21970). A partition homogeneity test (PHT) was performed with 1000 replicates via PAUP v. 4.0b10 (Swofford 2003) to evaluate statistical congruence amongst sequence data of 28S, ITS, GPD1 and TEF1 gene regions. MP analyses were performed in PAUP v. 4.0b10 (Swofford 2003), using the heuristic search option with 1,000 random taxa addition and tree bisection and reconnection (TBR) as the branch swapping algorithm. Maxtrees were set to 10,000. The characters in the alignment document were ordered accordingly: 28S+ITS+GPD1+TEF1, with equal weight and gaps were treated as missing data. The Tree Length (TL), Consistency Indices (CI), Retention Indices (RI), Rescaled Consistency Indices (RC) and Homoplasy Index (HI) were calculated for each tree generated. Maximum likelihood (ML) trees of DNA sequences were obtained by a heuristic search using the TrN + I + G model, which was deduced as the best fit for the data by the likelihood ratio test using the MODELTEST wer3.7 and MrMTgui version 1.01 (Posada and Crandall 1998).

Pathogenicity of this species was determined by inoculating healthy leaves of Hippeastrum striatum and Canna indica L. with 5 mm diameter mycelial plugs, cut from the margins of 10-day-old actively growing cultures; the control was treated with sterile agar plugs. Both inoculated and control plants were kept in a moist chamber at 25 °C for 7 days and observed for disease symptom development. Infected leaves were collected and the fungus was re-isolated in PDA medium and compared against the original strains. Control plants were sprayed with sterilised distilled water.

Nine isolates of Curvularia were sequenced from two plants in Chongqing Municipality, China (seven from Hippeastrum striatum and two from Canna indica). PCR products of approximately 900 bp (28S), 540 bp (ITS), 530 bp (GPD1) and 1200 bp (TEF1) were obtained. In the molecular phylogenetic analyses, the partition homogeneity test (P = 0.06) indicated that the individual partitions were not highly incongruent (Cunningham 1997) and thus 28S, ITS, GPD1 and TEF1 sequences were combined for sequence analyses. By alignment with a single gene region and then combination according to the order of 28S, ITS, GPD1 and TEF1, only 2689 characters were obtained, viz. 28S: 1–848, ITS: 849–1330, GPD1: 1331–1771 TEF1: 1772–2689 with 104 parsimony-informative characters and 157 parsimony-uninformative characters. The analysis produced three equally parsimonious trees, one of which (TL = 366, CI = 0.81, RI = 0.82, RC = 0.66 and HI = 0.19) is shown in Figure 1 and the topologies of MP and ML analysis were congruent, thus only MP topology was shown. Phylogenetic analysis confirmed nine strains (GUCC 6272, GUCC 6273, GUCC 6274, GUCC 6275, GUCC 6276, GUCC 6277, GUCC 6278, GUCC 6279 and GUCC 6280) with the same DNA sequences in four phylogenetic markers grouped into an independent clade supported by high bootstrap values (MP: 100%; ML: 99%). These strains were placed in trifolii-clade with strong bootstrap support (MP: 95%; ML: 95%) and had a close relationship with Curvularia gaudauskasii, C. gladioli, C. trifolii, C. borreriae and C. pallescens with a high MP support (MP: 87%), but its ML bootstrap value was lower than 50%.

Figure 1. 

The only one parsimonious tree obtained from combined analyses set of ITS, LSU,β-tubulin and tef1 sequence data. MP values (>50 %) resulting from 1000 bootstrap replicates. The tree is rooted with Alternaria alternata (EGS 34-0160). The branch of our new Curvularia is shown in blue.

Test plants (Hippeastrum striatum) were inoculated with 5 mm diam mycelial plugs of Curvularia microspora with two replicates of each plants and the inoculation experiment was repeated two times (with different sporulation generations). Hippeastrum striatum leaves both exhibited brown to dark brown necrotic spots (Figure 3a, b) after 7 days, which were very similar to those of natural infection (Figure 2a, b). The DNA sequencing result (ITS region), after re-isolation, identified this as C. microspora. The successful re-isolation of C. microspora from the inoculated leaves of H, striatum established a credible proof of pathogenicity. All test plants were covered with polyethylene bags for 7 days. However, on Canna indica, disease symptoms did not appear again.

Figure 2. 

Curvularia microspora (HGUP 6272). a–c Leaf diseases symptoms on Hippeastrum rutilum and Canna indica. d–f Conidiophores, conidiogenous loci and conidia g–j Immature and mature conidia k–l Upper (k) and lower (l) surface of colony. Scar bars: d, i (10 μm), e–f = 20μm, g–h, j = (5 μm).

Figure 3. 

Curvularia microspora inoculated to Hippeastrum rutilum (7 days). a the first time for inoculation b the second time for inoculation.