Downy mildews (Peronosporaceae; Oomycota) are an obligate biotrophic group that infects a wide range of mono-and dicotyledonous plants, including many economically relevant crops [1]. Pseudoperonospora cubensis is a notorious species infesting many cucurbitaceous crops, such as cucumber, gourd, pumpkin, and watermelon [2, 3], and Ps. humuli is one of the most important threats to the cultivation of hops (Cannabaceae) [4, 5]. Given their association with high economic losses, many recent studies have focused on the biology, host specificity, population structure, detection, and control of Pseudoperonospora species [3, 6-11], as well as their taxonomy and phylogeny [12, 13].
To date, four species of Pseudoperonospora have been reported in Korea [14, 15], Ps. cannabina, Ps. celtidis, Ps. cubensis, and Ps. humuli. In September 2009, symptoms typical of downy mildew were found on the leaves of Urtica angustifolia Fisch. ex Hornem. (Urticaceae) growing near Jangjeon valley in Pyeongchang, Korea (N37°29'36"; E128°32'33"). Urtica angustifolia is distributed in the wastelands, grasslands, valleys, wet places, and ridges of East Asian countries, including China, Korea, and Russia [16], and it is used as a traditional medicinal plant due to its high hypoglycemic activity [17]. The downy mildew infection resulted in slight discoloration of the leaf tissues, with yellow or pale green spots on the upper leaf surfaces that developed dark grey fungal growth on the lower surfaces. The lesions were poly-angular, and were delimited by the leaf veins (Fig. 1A, 1B). As the disease progressed, the spots turned blackish and often merged to cover larger areas. A representative sample was deposited in the National Institute of Biological Resources (KZITFG0000000017) and the Korea University Herbarium (KUS-F24488).
A detailed microscopic examination was performed using an Olympus BX53 microscope (Olympus, Tokyo, Japan), and DIC micrographs were captured with a DigiRetina 16M camera (Tucsen, Fuzhou, China). The following morphological characteristics were observed at 100~200 × for sporangiophores and at 400 × for sporangia and ultimate branchlets. The sporangiophores emerging through the stomata were tree-like, hyaline, straight, slender, and 120~400 µm in height (Fig. 1C, 1D). Trunks were straight, 4~6 µm wide below the first branch, of nearly uniform width, with no callose plug, and slightly swollen at the bases. Branching was monopodial, ramified 3~5 orders, but sometimes appeared subdichotomous. Ultimate branchlets were mostly in pairs but rarely single, straight to substraight, 7~12 µm long, and 1.5~2.5 µm wide at the base, with obtuse or subtruncate tips (Fig. 1E, 1F). Sporangia were pale brown to violet, ellipsoidal, 22~32(~42) µm long, and 14~20 µm wide, with a round or gradually narrowing tip and base and a somewhat protruding pedicel. The length to width ratio of the sporangia was 1.4~1.8 (n = 69), with the greatest width mostly at the median, but rarely supra-median. In the dehiscence apparatus, the inner layer of the wall was discontinuous, with a pore of 3~5 µm diameter and a papilla of 1.5~2.3 µm thick (Fig. 1G, 1H). Resting organs were not seen. The morphological observations revealed that this fungus unequivocally belongs to the genus Pseudoperonospora, and were well consistent with the known characteristics of Ps. urticae (Lib.) E.S. Salmon and Ware [18-20], except for the slight differences in sporangial size (Table 1). Waterhouse and Brothers [19] noted that the sporangia of Ps. urticae (maximum 40 µm and average 30 µm) were larger than those of other Pseudoperonospora species; however, others reported smaller sporangia: 19~32 × 13~21 (ave. 25.5 × 16.6) µm by Ito [21], 18~25(~30) × 12~16(~20) µm by Kochman and Majewski [22]; 19~33.5 × 12.5~23 µm by Vanev et al. [23]; 16~28 × 14~21 (mostly 23~25 × 18~20) µm by Ul'yanishchev et al. [24]; 22~32 × 14~22 µm by Mazelaitis and Staneviciené [25]; and 25~30 × 17~19 µm by Constantinescu [20]. In the Korean sample, such large sporangia were quite rare, but unambiguously present among mature sporangia with darker color, sympathizing with the opinion of Waterhouse and Brothers [19], along with a study of Yu [26] (maximum 40 µm).
Genomic DNA was extracted from the infected plant tissue of the herbarium specimen using the DNeasy Plant Mini Kit (Qiagen, Germantown, MD, USA). PCR amplification was performed with the primers ITS1-O & LR-0 for internal transcribed spacer (ITS) rDNA [27] and cox2-F & cox2-RC4 for cox2 mtDNA [28]. Amplicons were sequenced by a DNA sequencing service (Macrogen, Seoul, Korea) with the primers used for amplification. The resulting sequences ware deposited in GenBank (accession numbers: KY986684 for ITS rDNA and cox2 mtDNA for KY986683). The sequences were edited using the DNASTAR Lasergene software package (DNASTAR, Madison, WI, USA), version 5.05. Alignments of each locus were generated using MAFFT 7 [29] with the Q-INS-1 algorithm. Minimum evolution (ME) and maximum likelihood (ML) methods were used to construct two different trees. ME analysis was done using MEGA 7.0 [30], with the default settings of the program, except for replacement with the Tamura-Nei model. For ML analysis, 1,000 rounds of random addition of sequences as well as 500 fast bootstrap replicates were performed with RAxML 7.0.3 [31] using the GTRCAT model. In the ITS rDNA and cox2 mtDNA regions, the barcoding loci of oomycetes [28], the Korean isolate from Urtica angustifolia exhibited a high similarity of 99.5% (4 out of 750 characters are different) with Ps. urticae sensu stricto from Urtica dioica (AY198307, HM636048, HM636049) for the ITS sequencse, but 98% (12 out of 550) with three sequences (DG3657644, HM635952, HM635953) for the cox2 sequences. The phylogenetic trees for a combined alignment of ITS rDNA and cox2 mtDNA were inferred using the ME and ML methods. As the two trees were congruent, only a ME tree is shown in Fig. 2. The Korean sample was a close sister-lineage to Ps. urticae s. s., with a maximum support in both ME and ML trees. The phylogenetic divergence between the Korean sample and Ps. urticae s. s. may be due to either the different host species (Urtica angustifolia vs. Urtica dioica) or the distant geographic origins (Korea vs Europe), as in this study, no morphological differences were found between them (data not shown). Further study with additional collections is needed to investigate the precise relationship of the two lineages.
|
Table 1. Morphological comparison of the Korean specimen and previously reported Pseudoperonospora urticae |
|
|
|
*Constantinescu [20] stated that U. urens is not a host plant of Ps. urticae. |
Based on the morphological and phylogenetic analyses, the pathogen was identified as Pseudoperonospora urticae. This fungus has now been reported on six species of Urtica; U. angustifolia, U. dioica, U. fissa, U. gracilis (often regarded as a subspecies of U. dioica), U. kioviensis, and U. urens [32]. After a detailed review of downy mildews parasitic to Urtica spp., Constantinescu [20] suggested that both U. gracilis and U. urens are not the host plant of Ps. urticae, but instead they are infected by Peronospora debaryi, another downy mildew species occurring on Urtica spp. The present study confirmed that U. angustifolia is a rare host plant of Ps. urticae, and infection by Pseudoperonospora has been reported only once in Far Eastern Russia [33]. To our knowledge, this is the first report of Ps. urticae occurring on U. angustifolia in Korea.
