Analysis of conidiogenesis and lifelong conidial production from single conidiophores of Podosphaera aphanis on strawberry leaves using digital microscopic and electrostatic techniques

In this study, strawberry plants (Fragaria × ananassa Duchesne ex Rozier) grown via elevated cultivation in a greenhouse were heavily infested with powdery mildew. We isolated the powdery mildew fungus from strawberry leaves and identified the isolate as Podosphaera aphanis (Wallroth) U. Braun & S. Takamatsu var. aphanis KSP-7 N, based on its morphological characteristics and sequencing of the ribosomal DNA internal transcribed spacer region. The host range of KSP-7 N was assessed by inoculating 17 plant families comprising 49 species (a total of 110 cultivars) with conidia. The fungus caused severe powdery mildew on the tested strawberry plants (commercial, wild, and false strawberries), produced scattered conidia on conidiophores, and formed completely catenated conidia within approximately 27 h from conidiophore erection to the first release of mature conidia. Six conidia were produced on each conidiophore; only those at the apex reached maturity. The cycles of conidial release were repeated 17 to 21 times at intervals of approximately 6 h, during gradual upward elongation of the conidiophores. At the final stage, conidia were released without growth or septation of the generative cells. Conidiophores produced an average of 38 conidia during a 96-h period. These results will help to clarify the modes of conidiogenesis, the lifetime of conidiophores, and the production of conidia on a conidiophore among powdery mildew pathogens isolated from strawberry leaves.


Introduction
Recently, severe powdery mildew disease has affected the leaves and fruits of strawberries (Fragaria × ananassa Duchesne ex Rozier) grown via elevated cultivation in our greenhouse. Powdery mildews occurring on the Rosaceae plants, such as that caused by Podosphaera aphanis (Wallroth) U. Braun & S. Takamatsu [formerly Sphaerotheca humuli (DC.) Burrill or S. macularis (Wallroth ex Fries) Jaczewski f. sp. fragariae (Peries)] on strawberry plants, are among the most important ground diseases; they occur on the surfaces of all epigeal organs including the leaves, petioles, stolons, fruits, receptacles, runners, and flowers (Peries 1962a, b;Jhooty and McKeen 1965;Spencer 1978;Maas 1998). In Japan, this disease occurs during the growing stage (May to June), after planting (September to November), and during the harvest period (March to April) (Tanigawa et al. 1993;Okayama et al. 1995;Nakazawa and Uchida 1 3 1998). Strawberry fruit production is significantly reduced by severe infection (Spencer 1978;Nelson et al. 1995;Maas 1998;Xiao et al. 2001;Carisse et al. 2013). Powdery mildew is mainly controlled by applying commercial protectants containing sulfur, or by systemic application of fungicides such as myclobutanil or triflumizole, beginning early in the flowering period (Legard and Chandler 2000;Hollomon and Wheeler 2002). Consequently, the fungi have become resistant to these fungicides (Bal and Gilles 1986;Hollomon and Wheeler 2002;Sombardier et al. 2010). Previous studies have tested the effects of various physical (Kanto et al. 2014;Janisiewicz et al. 2015), chemical (Freeman and Pepin 1967;Tanigawa et al. 1993;Okayama et al. 1995;Hukkanen et al. 2007), and biological (Miller et al. 2004;Davik and Honne 2005;De Cal et al. 2008;Pertot et al. 2008;Cockerton et al. 2018) methods to control powdery mildew on strawberry plants. However, effective eco-friendly methods against the fungi that cause powdery mildew remain to be identified.
In the present study, we combined these techniques to examine the morphological characteristics of a powdery mildew fungus isolated from strawberry plants, paying particular attention to the infection structures, including conidiophore morphology and the conidiogenesis. We first isolated a single conidium from a strawberry leaf causing powdery mildew symptoms, and then multiplied the fungal mycelia on powdery mildew-free strawberry leaves. Then, we identified the isolate on the basis of previously described morphological characteristics (Braun 1987;Nakazawa and Uchida 1998;Braun and Cook 2012) and sequence of the ribosomal DNA internal transcribed spacer (rDNA-ITS) region amplified via polymerase chain reaction (PCR) (Takamatsu et al. 2010). We clarified the host range of the fungus for the first time by testing various plant species, including rosaceous plants. To elucidate the mode of conidiation, including conidiophore longevity and productivity, we analyzed formation processes of conidiophores of the powdery mildew isolate in detail using the KH-2700 DM. To our knowledge, this is the first report of conidiogenesis in Podosphaera aphanis (=Sphaerotheca humuli, S. macularis). Deciphering the conidiogenesis will facilitate taxonomic classification of the fungi that cause powdery mildews.

Fungal materials
Mature conidia were collected from conidiophores on powdery mildew-infected strawberry leaves using an electrostatic spore collector (Nonomura et al. 2009). The conidia were transferred onto true leaves of 60-day-old healthy strawberry seedlings (cv. Sagahonoka), with the aid of a KH-2700 DM (Matsuda et al. 2005). New conidia were reisolated from the conidiophores. The isolation process was repeated three times. The strawberry powdery mildew isolate was subsequently designated KSP-7 N. Mature conidia of KSP-7 N were inoculated onto true leaves of 60-day-old healthy strawberry seedlings, as described above. The isolate was maintained through incubation for 14 days in a growth

Microscopic morphological observation of KSP-7 N
KSP-7 N conidia were transferred to glass slides and strawberry leaves to examine the morphological characteristics of the isolate. The conidia were observed under a light microscope (BX-60 LM; Olympus, Tokyo, Japan) and a KH-2700 DM. Experiments were conducted using 100 mature conidia. Leaves of potted strawberry plants were inoculated with KSP-7 N conidia in September 2019, and the inoculated plants were placed outside the greenhouse. Three months later, leaves forming chasmothecia were collected for microscopic observation. The chasmothecia were gently scraped from the leaves using a glass needle, transferred to a glass slide, and pressed with a cover glass to squeeze asci and ascospores out of the chasmothecia. Chasmothecia on strawberry leaves and broken chasmothecia were observed under a stereomicroscope (SZ-PT SM; Olympus) and a BX-60 LM, respectively.

rDNA-ITS sequencing
To sequence the rDNA-ITS region of KSP-7 N, conidia were collected from KSP-7 N-infected leaves. Chromosomal DNA was extracted from the mycelia using a MagExtractor Plant Genome system according to the manufacturer's instructions (Toyobo, Osaka, Japan). PCR was performed using a PCR Thermal Cycler Dice TP600 system (TaKaRa Bio Inc., Shiga, Japan), the ITS1 (5′-TCC GTA GGT GAA CCT GCG G-3′) and ITS4 (5′-TCC TCC GCT TAT TGA TAT GC-3′) fungal-specific primers (White et al. 1990), and GoTaq Green Master Mix (Promega, WI, USA). PCR was performed under the following conditions: an initial denaturation for 2 min at 94 °C, followed by 40 cycles of denaturation for 2 min at 94 °C, annealing for 30 s at 55 °C, and extension for 30 s at 72 °C, followed by a final extension cycle of 8 min at 72 °C. The nucleotide sequence of the amplified region was determined using a BigDye Terminator v3.1 Cycle Sequencing Ready Reaction kit (Thermo Fisher Scientific, CA, USA) on an ABI Prism 3130 Genetic Analyzer (Thermo Fisher Scientific) at the Nara Prefecture Agricultural Research and Development Center.

Host range testing work
We inoculated 17 plant families comprising 49 species (a total of 110 cultivars) with KSP-7 N conidia to determine the host range of the isolate ( Table 1). Seeds of all test plants were purchased from the originators or obtained from our university seed collection. Seeds were placed on wet filter paper in Petri dishes and germinated for 3-4 days in an LH-240 growth chamber under continuous illumination (22.2 µmoL m −2 s −1 ) at 22 ± 1 °C. Cotyledon seedlings were placed in sponge cubes (3 × 3 × 3 cm 3 ) and grown for 14-28 days in an LH-240 growth chamber under the conditions described above. Each inoculation involved three test plants. We obtained one or two leaves from large-leaved plants including Amaranthaceae, Apiaceae, Asteraceae, Brassicaceae, Convolvulaceae (Ipomoea nil), Cucurbitaceae, Lamiaceae, Malvaceae, and Solanaceae, four to six true leaves from small-leaved plants including Caryophyllaceae, Convolvulaceae (Ipomoea aquatica), Fabaceae (Leguminosae), Onagraceae, Pedaliaceae, Rosaceae, Scrophulariaceae, and Valerianaceae, and primary leaves from Gramineae (Poaceae) plants, with three strawberry plants (60-day-old seedlings of cv. Sagahonoka) serving as controls. The leaves were placed in a cubic frame covered with electrostatic spore precipitators . Conidia were sprayed onto the leaves (Nonomura et al. 2009). Successful inoculation was confirmed by observing pustule expansion on leaf surfaces of the control plants. Fungal development on the inoculated leaves was observed at 14 days after inoculation.
To observe cytological responses to the invasion of KSP-7 N, samples were prepared as follows. Leaf segments (approximately 1 × 1 cm 2 ) were cut from KSP-7 N-inoculated plants. The leaf segments were fixed and leaf color was removed by boiling in an alcoholic lactophenol solution (10 mL glycerol, 10 mL phenol, 10 mL lactic acid, 10 mL distilled water, and 40 mL 99.8% ethanol) for 1-2 min and staining with 0.1% aniline blue (Nacalai Tesque, Tokyo, Japan; Sameshima et al. 2004). The samples were observed under a BX-60 epifluorescence microscope (BX-60 EM; Olympus) with B excitation, a B absorption filter, and an O-515 barrier filter.

Conidiogenesis of strawberry powdery mildew isolate
Mature KSP-7 N conidia were collected from conidiophores that formed in the fungal colonies and transferred onto welldeveloped young leaves of 60-day-old strawberry seedlings using an electrostatic insulator probe (Nonomura et al. 2009). A cylindrical plastic tube containing a KSP-7 N-inoculated strawberry seedling was placed on the stage of the KH-2700 DM in a temperature-controlled room (at 22 ± 1 °C, 45-55% RH; Fig. 1A). An electric fan was placed 1 m from the inoculated seedling and air was blown continuously at 0.5 m s −1 onto the seedling. At 4-5 days after inoculation, immature conidiophores were selected and observed under the KH-2700 DM. Those selected were photographed at 30-min intervals using a ½-inch Interline transfer charge-coupled device camera on the KH-2700 DM. Photographs were treated using image-processing software (Adobe Photoshop ver. 5; Adobe Systems, CA, USA). More than 30 conidiophores were observed to trace their development. A fluorescent brightener, Calcofluor white (CFW; Sigma, MO, USA), was used to stain conidiophores on strawberry leaves (Oichi et al. 2004). A few drops of 10% potassium hydroxide containing 0.05% CFW were placed on KSP-7 N-inoculated strawberry leaves, which were then incubated at room temperature for 5 min. Fluorescencestained conidiophores were observed using a BX-60 EM (U excitation with a BP330-385 excitation filter and BA420 absorption filter).

Electrostatic collection of mature conidia on conidiophores
To collect mature conidia, we used a pencil-type electrostatic insulator probe consisting of an ebonite rod with a pointed tip (length, 7 cm; diameter, 4 mm; tip diameter, 5 µm; Moriura et al. 2006;Nonomura et al. 2009;Takikawa et al. 2015). The insulator probe was negatively charged using a directcurrent voltage generator (HVA 10K202PA; Max Electronics, Tokyo, Japan) and attached to the micromanipulator of a AF, A&F, Tottori, Japan; AT, Atariya Noen, Chiba, Japan; CR,Central Rose, Gifu, Japan; HG, Hagihara Farm, Nara, Japan; HK, Herbal K, Tokyo, Japan; JA, Jardin, Chiba, Japan; KA, Kaneko seeds, Tochigi, Japan; KO, Kokkaen, Osaka, Japan; KS, Kashihara shokubutuen, Nara, Japan; KU, Kindai University seed collection; MA, Marutane, Kyoto, Japan; NA, Nara Prefecture Agricultural Research and Development Center, Nara, Japan; NI, Nihon nousan shubyo, Nagano, Japan; SA, Sakata seeds, Yokohama, Japan; SU, Suntory Flowers, Tokyo, Japan; TA, Takii seeds, Kyoto, Japan; TO, Tohoku seeds, Tochigi, Japan; TS, Tsurusin, Aichi, Japan; UT, Utane seed, Tochigi, Japan b By 14 days after inoculation. HR, hypersensitive cell death; Pa, papilla formation; -, no necrosis c By 14 days after inoculation. R, non-infection; S, infection with sporulation and extensive spread of the colony the KH-2700 DM (Fig. 1A). Conidiophores located at the edge of the colony were selected as targets for consecutive conidium collection. The insulator probe charged with static electricity (5.2 × 10 -1 nC) was placed approximately 100-120 µm from the conidiophore apex to collect released conidia (Takikawa et al. 2015). When the septum of the apical conidial cell in a full-length conidial chain was fully constricted, the surface of the insulator probe was negatively polarized by the current provided by the generator Nonomura et al. 2009;Takikawa et al. 2015). A healthy strawberry leaf was inoculated with conidia trapped on the probe to test infectivity. The collection procedure was repeated and the intervals between consecutive conidium releases were timed until conidia were no longer released. Finally, all conidia from a given conidiophore were counted.

Morphological observation and molecular analysis
We isolated conidia from powdery mildew-infected strawberry leaves and then grew the fungus on true leaves of healthy strawberry plants (cv. Sagahonoka). True leaves of strawberry plants were inoculated with KSP-7 N conidia on glass slides to observe morphological characteristics, focusing on conidial shape and size, appressorial shape, conidiophores, and chasmothecia. Conidia 27-33 × 18-22 μm in size were hyaline, ellipsoid-ovoid to doliiform-limoniform in shape, and contained oil and fibrosin bodies (Fig. 1B).
All conidia germinated at 4-5 h after inoculation and then formed slightly swollen appressoria (width, ca. 4 μm) at the tips of unbranched germ tubes (Fig. 1C). The germ tubes arose from a side wall (lateral germination) (Fig. 1C). Conidiophores were catenated, with six conidia forming in straight chains, ca. 84-129 × 8-11 µm in size, at approximately 27 h after the formation of primordial cells from hyphae (Fig. 1D). Numerous chasmothecia formed within 3 months of conidial inoculation. Chasmothecia (100-125 × 65-80 μm in size) were round, dark brown, and scattered or densely gregarious on/in the mycelia. Each chasmothecium possessed a single ascus, and each ascus had eight ascospores and, simple, mycelioid, septate, unbranched, and rod-like appendages (Fig. 1E). Asci (60-94 × 55-76 μm in size) were broadly ellipsoid-ovoid and ascospores (16-25 × 14-22 μm in size) were narrowly ellipsoid to subglobose. Next, we conducted gene amplification of the rDNA-ITS region of KSP-7 N to determine the nucleotide sequences of the PCR products (490 bp). The rDNA-ITS sequence of KSP-7 N was compared with those of other rosaceous powdery mildew isolates registered in GenBank (data not shown). The nucleotide sequence of the KSP-7 N rDNA-ITS region was highly homologous to those of P. aphanis MUMH 1871 (AB525933) and Puyallup REC GH (MF919433). The rDNA-ITS sequence of KSP-7 N was registered under accession no. LC432335 in the DNA Data Bank of Japan (DDBJ). Thus, we identified the isolated fungus as Podosphaera aphanis (Wallroth) U. Braun & S. Takamatsu var. aphanis KSP-7 N, based on its morphological characteristics and molecular analysis of the rDNA-ITS region.

Host range testing work
The results of host range tests for the strawberry powdery mildew isolate KSP-7 N are shown in Table 1. KSP-7 N infected and then sporulated on inoculated leaves of strawberry plants (commercial, wild, and false strawberries) without causing hypersensitive cell death (HR) or necrosis or forming papillae in leaf epidermal cells. The HR reaction was induced in the epidermal cells of inoculated leaves of 38 tested cultivars. The HR reaction induced in epidermal cells by the invasion of KSP-7 N into tomato leaves (cv. Moneymaker) is shown in Fig. 1F. Additionally, papillae were induced in the epidermal cells of inoculated leaves of 39 tested cultivars. Papillae induced in epidermal cells by the invasion of KSP-7 N into barley leaves (cv. Kobinkatagi) are shown in Fig. 1G. Hyphal elongation of KSP-7 N was completely suppressed in resistant plant species. Among the remaining plant cultivars tested, KSP-7 N did not infect their leaves without causing resistance responses in epidermal cells.

Conidiophore development
KSP-7 N conidiophores formed on strawberry leaves (cv. Sagahonoka) were observed under a KH-2700 DM. Digital and epifluorescence micrographs of KSP-7 N conidiophores are shown in Figs. 2A and B, respectively. Septation and conidial development in the conidiophores were consecutively traced using the KH-2700 DM (Figs. 2A, 1 to 11). Septation was confirmed by staining conidiophores with CFW at designated stages (Fig. 2B). The septation sites of conidiophores observed with the KH-2700 DM coincided with those observed in CFW-stained conidiophores under a BX-60 EM.

Consecutive electrostatic spore collection of progeny conidia from conidiophores
We attempted to collect mature progeny conidia consecutively released from conidiophores, using an electrostatic spore collection technique (Fig. 1A). The non-polarized insulator probe did not attract conidia (Fig. 3A); however, conidiophores were attracted to the probe when it was negatively polarized (Fig. 3B), allowing collection of the mature progeny conidial cell (C1-1) at the top of the conidiophore (Fig. 3C). The probe was then depolarized and the conidiophore returned to its original position. This procedure was repeated, similarly attracting the next apical conidial cell (C1-2) to the probe 2 h later (Fig. 3D). We repeatedly collected progeny conidia on conidiophores using this electrostatic technique. Thus, apical cells C(n)-1 and C(n)-2 were attracted to the probe ca. 2 h apart. By collecting all of the released conidia, we determined the entire quantitative production of progeny conidia by individual conidiophores and the timing of conidial secession (Table 2). Individual conidiophores released an average of 38 progeny conidia within ca. 96 h during conidial secession.
At the final stage of conidial secession (Fig. 7), the gc ceased division and growth, although four to five progeny conidia were still released (7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22). Secession of the conidial cells (zero to two immature conidia) remaining on the conidiophores was not detected, even when the observation period was extended by 12 h. Consequently, the lifetime of the conidiophores was determined to be 125-150 h, from primordial formation until final conidial release on conidiophores (i.e., first to third phases). Hori (1932) first reported the infestation of strawberry plants in Japan with powdery mildew fungi, and proposed Sphaerotheca fragariae (Harz) Arthur as the pathogen without recording the sexual stages (teleomorph) of the fungus. Later, Homma (1937) classified strawberry powdery mildew fungus as Sphaerotheca humuli (DC.) Burrill, also without recording the sexual stages of the fungus, and then Nakazawa and Uchida (1998), proposed Sphaerotheca aphanis (Wallroth) Braun var. aphanis after observing the sexual stages. Based on the morphological characteristics of powdery mildew fungi and molecular phylogenetic analyses of isolates, the genus Sphaerotheca was recently merged with Podosphaera (Braun and Takamatsu 2000). The scientific name of strawberry powdery mildew fungus was then changed to Podosphaera aphanis (Wallroth) U. Braun & S. Takamatsu var. aphanis (Braun and Takamatsu 2000;Braun and Cook 2012). In our study, microscopic observations on glass slides revealed that the sizes and shapes of KSP-7 N conidia were very similar to those of previously described isolates (Salmon 1900;Braun 1987;Braun and Cook 2012;Karajeh et al. 2012;Calis et al. 2015). Fibrosin bodies were also present in KSP-7 N conidia, as described by Braun and Cook (2012) and Calis et al. (2015). Germ tubes were very similar to those of the orthotubus subtype within the Fibroidium type as described by Braun and Cook (2012). Moreover, the morphological characteristics of KSP-7 N chasmothecia were similar to those of previously described isolates (Jhooty and McKeen 1962;Howard and Albregts 1982;Nakazawa and Uchida 1998;Farooq et al. 2007). Thus, our microscopic observation data strongly imply that KSP-7 N is identical to Podosphaera aphanis (Wallroh) U. Braun & S. Takamatsu var. aphanis (syn. Sphaerotheca humuli, S. macularis, and S. aphanis), based on its morphological characteristics, including the shapes of chasmothecia formed in the sexual  (Braun 1987;Nakazawa and Uchida 1998;Braun and Takamatsu 2000;Braun and Cook 2012) and sequence analysis of the rDNA-ITS region (Takamatsu et al. 2010). As a result of experiments designed to understand the pathogenicity of KSP-7 N to various plant species, we confirmed its host range. The isolate KSP-7 N heavily infected all tested commercial strawberry cultivars (Peries 1962a, b;Abiko 1982;Nelson et al. 1996;Harvey and Xu 2010 Wild strawberry (F. vesca L.) was susceptible to the isolates described by Abiko (1982) but resistant to the isolate studied by Peries (1962a), perhaps reflecting the existence of different pathotypes among strawberry powdery mildew isolates. Consequently, the host range of KSP-7 N is very narrow, as previously described by Abiko (1982) and Harvey and Xu (2010). To our knowledge, no previous studies have assessed the pathogenicity of strawberry powdery mildew fungus based on broad testing of many families, genera, and species of flowering plants.

Discussion
Fungicides with modes of action targeting fungal respiration, nucleic acid synthesis, sterol biosynthesis, and signal transduction are commonly used to control strawberry powdery mildew. Previous studies have reported tolerance of the strawberry powdery mildew fungus to commercial fungicides, such as demethylation inhibitors (DMI), triflumizole, and thioquinox (Bal and Gilles 1986;Hollomon and Wheeler 2002;Sombardier et al. 2010). Fortunately, KSP-7N did not exhibit tolerance to the main commercial fungicides in the present study (data not shown). However, we must continuously screen for fungicide-resistant strawberry powdery mildew fungi on host plants in our greenhouses.
Powdery mildew widely expands the infection source by releasing progeny conidia from conidiophores (Willocquet et al. 2008). Therefore, the main objective of this study was to analyze the conidiogenesis of the strawberry powdery mildew isolate KSP-7N, because the morphological characteristics of strawberry powdery mildew conidiophores have not been known in detail. Conidiophores produce conidia in chains (i.e., catenescent conidia), as confirmed through  (Jhooty and McKeen 1965;Paulus 1990;de los Santos et al. 2002;Braun and Cook 2012;Calis et al. 2015). Salmon (1900) described the production of unicellular spores, or conidia, at the apex of the conidiophore, in long chains via basipetal succession. Thus, the reported characteristics of conidiophores have been variable. To our knowledge, the full developmental process of conidiophores has never been studied in living conidiophores of strawberry powdery mildew fungi. Based on the digital microscopic observations in this study, we assert that the number of conidia that accumulates through production on powdery mildew conidiophores should be naturally determined. In our previous studies, we demonstrated the conidiogenesis of tomato powdery mildew fungus (Pseudoidium neolycopersici L. Kiss) (Oichi et al. 2004, barley powdery mildew fungus (Blumeria graminis f. sp. hordei Marchal race 1) , and melon powdery mildew fungus (Podosphaera xanthii Pollacci) on host leaves (Takikawa et al. 2015). Japanese isolates of tomato powdery mildew fungi (KTP-01 and -02) formed non-catenated conidia with foot cells (conidia produced singly on conidiophores) under windy conditions (1.0 m s -1 ; Oichi et al. 2004Oichi et al. , 2006Seifi et al. 2012), whereas the barley powdery mildew isolate KBP-01 produced catenated conidia possessing eight immature conidia, with a globularly bulky gc , and a melon powdery mildew isolate KMP-6N produced catenated conidia possessing six immature conidia, with a straight and non-bulky gc (Takikawa et al. 2015). In the present study, the strawberry powdery mildew isolate KSP-7N also produced catenated conidia and accumulated six conidia on a conidiophore, with a straight and non-bulky gc, under windy conditions (1.0 m s -1 ). Among types of conidiophores, powdery mildew fungi possess conidia that form singly or are catenated in chains comprising cells composed of a gc and straight or sinuous-helicoid foot cells (Braun and Cook 2012). However, KSP-7N formed a straight gc on basal septa because the cell was divided via septation into two cells during conidiogenesis. This process clearly differed from that observed on the P. neolycopersici isolates KTP-01 and -02, as the Pseudoidium type possessed foot cells on basal septa (Oichi et al. 2004;Seifi et al. 2012). Boesewinkel (1980) reported that P. aphanis (=S. macularis) forms conidiophores with foot cells. We were unable to determine whether the straight hyphae on basal septa were foot cells or gcs until we observed the developmental process of conidiophores  (11) possessing full-length chains of six conidial cells (C1-1 to C3-2), an undivided conidial cell (C4), and a gc at the first phase, as shown in Fig. 2A. B and C) Process of conidial secession repeated with elongation of the conidial chain, and growth and septation of the gc at the second phase (12-14, 15-17, and 19-21) (B) and without elongation of the conid-ial chain or growth and septation of the gc at the third phase (22-25) (C). Note that the same conidiophore was consecutively observed from when it released the first progeny conidium (C1-1) of the conidiophore (12) to the release of the last [C(n + 1)-2] conidium (25). The gc length of the later conidiophore (25) was greater than that of an earlier conidiophore (11). Black left-pointing arrows indicate positions of septation in the gc. Bar = 60 µm directly under a KH-2700 DM. Thus, the results obtained in this study will contribute greatly to our understanding of morphological characteristics of conidiophores. Aspects of the division of the KSP-7N gc through septation ( Fig. 2A) were clearly similar to those of the barley powdery mildew isolate KBP-01 (Blumeria genus), except for initial conidiophore formation , and the melon powdery mildew isolate KMP-6N (Podosphaera genus) (Takikawa et al. 2015). These divisions may be the main mechanism by which powdery mildew conidia are catenated in chains. In the present study, KSP-7N produced conidiophores coincident with gradual upward elongation of the gc during conidiogenesis (second phase, Figs. 5B and 6), unlike conidiogenesis of the melon powdery mildew isolate KMP-6N. This observation reveals a newly discovered difference in the mode of conidiogenesis among Podosphaera species. Braun and Cook (2012) described a slight increase in the lengths of foot cells near the apex (30-160 μm), followed by 2-3 shorter cells in conidiophores (up to 300 μm long). These differences of foot cell length appear to reflect growth of gc accompanying the formation of conidiophores.
To confirm the longevity of conidiophores and lifelong production of progeny conidia by living conidiophores on strawberry leaves, we used an electrostatically activated insulator probe (Nonomura et al. 2009). This technique enabled us to analyze the release of mature conidia (progeny conidia) from conidiophores and count the total number of progeny conidia produced by individual conidiophores throughout their lifetime. Consequently, we clarified the precise cycle of conidial formation and release from the conidiophores of KSP-7N. The release of progeny conidia from the apex of conidiophores (during the second phase) was similar to those of barley (KBP-01) and melon powdery mildew fungi (KMP-6N), with a succession of two progeny conidia at the apex of the conidiophore repeatedly released at approximately 3-5 h Takikawa et al. 2015). The conidial release process of KSP-7N was identical to those of KBP-01 and KMP-6N during the third phase. Conidia at the apex of the conidiophore were continuously released, and the conidiophore became shorter, without elongation or division of the gc. However, release times varied among the fungi. KSP-7N released a conidium approximately every 1 h, as observed on KMP-6N (Takikawa et al. 2015), whereas KBP-01 released conidia approximately every 3 h . Thus, at the third phase, there were Diagrams of the successive secession of mature progeny conidia during conidiogenesis of KSP-7 N. Conidial secession was repeated with elongation of the conidial chain and growth and septation of the gc at the second phase (12-14, 15-17 and 19-21). The gc length of the later conidiophore (18) was greater than that of an ear-lier conidiophore (17) due to gradual upward elongation of the gc in the conidiophores (upward arrow). Time intervals (h) for each process are shown at the bottom of the figure. We observed 30 conidiophores; data are presented as means ± SDs. Bar = 20 µm clear differences in the timing of conidial release from conidiophores between Podosphaera and Blumeria. The number of progeny conidia released from conidiophores also varied among fungi; strawberry, barley, and melon powdery mildew isolates released four to six, three to five, and four conidia, respectively, during the third phase. However, the longevity of conidiophores and lifelong production of conidia were similar between KSP-7N, KBP-01, and KMP-6N Takikawa et al. 2015). Thus, we demonstrated conidiogenesis and the longevity of KSP-7N conidiophores in detail. The data obtained in the present study provide essential information describing the conidiophores and conidiogenesis of powdery mildew fungi, which will be useful for the accurate taxonomic classification of powdery mildew fungi. In future studies, we will compare and analyze the developmental process of conidiophores among powdery mildew fungi belonging to various genera and species, and among the fungi of the same genus. Fig. 7 Diagrams of the successive secession of mature progeny conidia during conidiogenesis of KSP-7 N. Conidial secession without elongation of the conidial chain or growth and septation of the gc at the third phase (22-25). C(n + 2)-1 and C(n + 2)-2 cells are immature conidial cells that remain in the conidiophore. C(n + 3) is an undivided cell without growth or septation. Time intervals (h) for each process are shown at the bottom of the figure. We observed 30 conidiophores; data are presented as means ± SDs. Bar = 20 µm