First report of Robbsia andropogonis causing bacterial leaf spot of bougainvilleas in Taiwan.

Bougainvilleas (Bougainvillea spp.) are popular ornamentals commonly grown as bushes, vines, or trees worldwide (Kobayashi et al. 2007). Leaf spot symptoms were observed on a bougainvillea hedge located in North District, Taichung, Taiwan during August of 2022. The lesions were brown, necrotic and had yellow halos (Fig. S1). All the plants at the location showed similar symptoms. Leaf samples were collected from five plants and symptomatic tissues were minced in 10 mM MgCl2. The samples were streaked onto nutrient agar (NA) and after culturing at 28°C for 2 days, small, round, creamy white colonies were consistently isolated from all the samples. A total of five strains (BA1 to BA5) were obtained; each of them was isolated from a different plant. All five strains induced hypersensitive response in tobacco leaves. Amplification and sequencing of the isolated strains' 16S rDNA using primers 27F and 1492R (Lane 1991) revealed that all five strains shared identical sequences (GenBank accession no. OQ053015) with Robbsia andropogonis LMG 2129T (formerly Burkholderia andropogonis and Pseudomonas andropogonis; GenBank accession no. NR104960; 1,393/1,393 bp). Further testing of BA1 to BA5's DNA samples using the pathogen's species-specific primers Pf (5'-AAGTCGAACGGTAACAGGGA-3') and Pr (5'-AAAGGATATTAGCCCTCGCC-3'; Bagsic et al. 1995) successfully amplified the expected 410-bp amplicon in all five samples; the sequences of the PCR products completely matched to those of BA1 to BA5's 16S rDNA. Strains BA1 to BA5 also tested negative for arginine dihydrolase and oxidase activity, and failed to grow at 40°C, all of which are consistent with descriptions of R. andropogonis (Schaad et al. 2001). Pathogenicity of the isolated bacteria was confirmed by spray inoculation. Three representative strains, BA1 to BA3, were used for the assay. Bacterial colonies were scraped from NA plates and suspended in 10 mM MgCl2 added with 0.02% Silwet L-77. The concentrations of the suspensions were adjusted to 4.4-5.8 x 108 cfu/ml. The suspensions were sprayed onto three-month-old, cutting-propagated bougainvillea plants (to runoff). Controls were treated with bacteria-free solutions. Three plants were used for each treatment group (and the controls). The plants were placed in a growth chamber (27/25°C, day/night; 14-hour photoperiod) and bagged for three days. Within 20 days post inoculation, brown, necrotic lesions resembling those observed in the sampling site were observed on all inoculated plants, but not on the controls. One strain was re-isolated for each treatment group and the re-isolated strains all shared the same colony morphology and 16S rDNA sequence with BA1 to BA5. Additional PCR testing of these re-isolated strains using Pf and Pr also produced the expected amplicon. This is the first formal report of R. andropogonis affecting bougainvilleas in Taiwan. The pathogen has been reported causing diseases of betel palm (Areca catechu), corn and sorghum in Taiwan (Hseu et al. 2007; Hsu et al. 1991), some of which are economically important (Lisowicz 2000; Navi et al. 2002). As such, infected bougainvilleas could potentially serve as an inoculum source for these diseases.

First report of bacterial soft rot on Epipremnum aureum caused by Pectobacterium aroidearum in Taiwan.

Pothos (Epipremnum aureum) is an Araceae foliage plant with great ornamental values, which has long been enjoyed by consumers (Chen et al. 2010). In September 2021, pothos showing soft rot symptoms were found in 2 nurseries in Taichung, Taiwan. The petioles of the infected plants were macerated; some lesions extended to the leaves (Figure S1). The disease incidence was 50% in one nursery and 37.5% in the other; two and three plants were respectively collected from the two sites. Macerated tissues were homogenized in 10 mM MgCl2 and the samples were observed microscopically without dyeing. Motile, rod-shaped bacteria were observed in the samples, and the bacteria were isolated onto nutrient agar (NA) and grown at 28°C for 2 days. Fast-growing, round, creamy colonies were isolated from all 5 plants. One strain was isolated from each plant and the strains were named Ea1 to Ea5. The bacteria could ferment glucose and induce maceration on potato tuber slices (Schaad et al. 2001), but did not produce indigoidine on NGM medium (Lee and Yu 2006) and were tested negative for phosphatase activity (Schaad et al. 2001). The bacteria's DNA samples were tested using primers specific to Pectobacterium (Y1/Y2; Darrasse et al. 1994). The expected 434-bp amplicon was amplified in all five strains. Multilocus sequence analysis was conducted as previously described (Portier et al. 2019). A concatenated sequence (1,592 bp) comprising partial dnaX (492 bp), leuS (452 bp) and recA (648 bp) sequences was obtained for each strain. Two genotypes were detected among the strains; Ea1 and Ea2 belonged to one genotype (i.e., they had identical sequences), while Ea3, Ea4 and Ea5 belonged to the other (GenBank accession nos. OK416015-OK416020). Phylogenetic analysis was conducted using these data and those of representative strains of known Pectobacterium species (Klair et al. 2022). A maximum-likelihood tree showed that Ea1 to Ea5 clustered with P. aroidearum CFBP8168T (Figure S2). Sequence comparison (Table S1) showed that the similarity between the two genotypes' concatenated sequences was 99.1% (Ea1 vs. Ea3; 1,578/1,592 bp); Ea1 and Ea3 shared 99.2% and 99.3% sequence similarity with P. aroidearum CFBP8168T, respectively. The sequences obtained in this work were searched against GenBank and all of their top hits were those of strains belonging to P. aroidearum (supplementary information). Koch's Postulates were fulfilled by stab inoculating cutting-propagated pothos (8-cm tall) using toothpicks carrying bacteria grown on NA. The pathogen loads used were estimated by suspending cells (attached to individual toothpicks) in 10 mM MgCl2 and spread-plating them onto NA (after dilution); the loads were 5.5 x 106 - 2.2 x 107 CFU. Three plants were inoculated for each strain (3 petioles per plant). Control plants were stabbed with sterile toothpicks. Each plant was then bagged and placed in a growth chamber (28°C; 14 h light). After 24 h, all inoculated plants produced symptoms resembling those found in the nurseries, and the controls did not. For every treatment group, a strain was re-isolated onto NA; each of them shared the same recA sequence with the original strain inoculated. This is first report of P. aroidearum causing pothos soft rot in Taiwan. Local nurseries often grow pothos and other Araceae plants together in humid areas. Since other Araceae species are also known to be susceptible to P. aroidearum (Xu et al. 2020), growers should be cautious of the pathogen's spread across hosts.

First report of a 'Candidatus Phytoplasma aurantifolia'-related strain (16SrII-V) associated with phyllody, virescence, and shoot proliferation of sweet William (Dianthus barbatus) in Taiwan.

The sweet William (Dianthus barbatus) is an ornamental belonging to the Caryophyllaceae family; the species produces clusters of flowers that comes in various colors and is grown commonly as garden plants (Lim 2014). In February 2021, sweet Williams showing symptoms typical of phytoplasma diseases were found in a garden located in Wufeng District, Taichung, Taiwan (24°04'37.6"N 120°43'20.4"E). Infected plants exhibited virescence and phyllody symptoms and produced an abnormal number of new shoots from the base of the flowers/flower-like structures (Figure S1) as well as the base of the plants. Among the fifteen plants grown in the area, two exhibited such symptoms. The two symptomatic plants, along with five symptomless plants were sampled. Two flower-like structures were collected from each of the symptomatic plants, and two flower samples were collected for each symptomless plant (Figure S2). Total DNA were extracted from each sample using the Synergy 2.0 Plant DNA Extraction Kit (OPS Diagnostics) and subjected to diagnostic PCR using primers P1/P7 (Schneider et al. 1995). All four symptomatic samples produced a 1.8-kb amplicon and the ten symptomless samples did not. The amplification products were diluted fifty-fold and used in a second round of PCR using primers R16F2n/R16R2 (Gundersen and Lee 1996). Again, only the symptomatic samples produced an expected 1.25-kb amplicon. A sample was selected for each plant and the PCR products from the first round of PCR were cloned using the pGEM-T Easy Vector System (Promega Inc.) and sequenced (three clones per sample). Fragments of the 16S rRNA gene (1,248 bp; GenBank accession: MW788688) were analyzed using iPhyClassifier (https://plantpathology.ba.ars.usda.gov/cgi-bin/resource/iphyclassifier.cgi). Sequences obtained from the two infected plants were identical, and were classified to the 16SrII-V subgroup with similarity coefficients of 1.0; they also shared 98.6% similarity with the sequence of a 'Candidatus Phytoplasma aurantifolia' reference strain (accession: U15442). BLASTn results indicated that the 16S rRNA gene sequences detected were identical to those of 16SrII-V phytoplasmas affecting mungbean (accession: MW319764), lilac tasselflower (accession: MT420682), peanut (accession: JX403944) and green manure soybean (accession: MW393690) found in Taiwan. To corroborate the above results, 16SrII group-specific primers were used to conduct nested and semi-nested PCR targeting the pathogen's 16S rRNA gene (outer primers: rpF1C/rp(I)R1A; inner primers: rp(II)F1/rp(II)R1; Martini et al. 2007) and immunodominant membrane protein gene (imp; outer primers: IMP-II-F1/IMP-II-R1; inner primers: IMP-II-F2/IMP-II-R1; Al-Subhi et al. 2017). In both assays, the symptomatic samples produced the expected amplicons and the symptomless samples did not. The coding sequence of the imp gene (519 bp; accession: MW755353) was the same among all symptomatic samples, and shared 100% identity with that of the peanut witches'-broom phytoplasma (16SrII; accession: GU214176). To our knowledge, this is the first report of a 16SrII-V phytoplasma infecting sweet Williams in Taiwan. Since 16SrII-V phytoplasmas have also been found infecting mungbeans and peanuts in Taiwan (Liu et al. 2015), the findings here suggest that by serving as a natural host in the field, the sweet William may potentially contribute to the spread of 16SrII-V phytoplasmas to food crops.

First report of Pectobacterium aroidearum causing bacterial soft rot of carrot in Taiwan.

Carrot (Daucus carota) is an important root vegetable planted and consumed worldwide (Stein and Nothnagel 1995). In June 2020, carrots (cv. New Kuroda) showing soft rot symptoms were observed in a 600 sqft plot located in Pitou, Changhua, Taiwan (23°54'00.9"N, 120°28'37.3"E; with around 400 plants). About 10% of the plants on site had similar symptoms; infected taproot tissues were macerated (Figure S1) and emitted a foul odor. In most cases, the peels above the rotten tissues remain intact. Two infected plants were brought to the lab. Macerated tissues were suspended in water and examined under a microscope at 600X (without staining). Rod, motile bacteria were observed in all of the samples and the bacteria were isolated onto nutrient agar. Three bacterial strains were obtained from two taproots; strain Car1 was isolated from one plant, and strains Car2 and Car3 were isolated from the other. Their colonies were translucent, round and convex. All isolates could ferment glucose and induce soft rot symptoms on potato tuber slices (Schaad et al. 2001). They were not able to produce indigoidine on yeast dextrose calcium carbonate agar and were tested negative for phosphatase activity (Schaad et al. 2001). The 16S rDNA of Car1 to Car3 were amplified using primers 27F/1492R (Lane 1991). Cloning and sequencing of their 16S rDNA (GenBank accession no. MT889640) revealed that their sequences shared 99.9% identity (1,463/1,464 bp) with that of Pectobacterium aroidearum CFBP 8168T (SCRI 109T; GenBank accession no. NR_159926.1). Multilocus sequence analyses targeting the three isolates' dnaX, leuS and recA genes were conducted. The concatenated sequences (1,596 bp) of Car1 to Car3 and those included in a previous work (Portier et al. 2019) were subjected to phylogenetic analysis. The sequences of Car1 to Car3 were identical (GenBank accession nos. MT892671-MT892673). A maximum-likelihood tree showed that the three isolates belonged to the same clade as P. aroidearum CFBP 8168T (GenBank accession nos. MK516971, MK517115 and MK517259; Figure S2). For the concatenated sequences analyzed, the identity between P. aroidearum CFBP 8168T and our three isolates was 99.4% (1,587/1,596 bp). The pathogenicity of these isolates was determined by inoculating the bacteria into carrot (cv. Xiangyang No.2) taproots. Strains Car1 to Car3 were grown on NA for 48 h (28 °C) and cell suspensions with OD600 values of 0.3 (2.4 x 108 CFU/ml; in water) were prepared. The suspensions of each strain (100 µl) were loaded into 200 µl pipette tips. The tips were then pierced into intact carrot taproots (2.4 cm deep), ejected and left on the plants (one tip per plant). Three taproots were tested for each strain. Tips loaded with 100 µl of water were used for the controls (three replicates). The plants were incubated in a sealed plastic container kept in a growth chamber set at 28°C. After 48 h, all of the inoculated taproots produced soft rot symptoms resembling those observed in the field and plants in the control group did not. Bacteria were re-isolated from macerated tissues of the artificially infected plants and found to share the same leuS sequence with Car1 to Car3. Occurrences of carrot soft rot in Taiwan have only been attributed to Dickeya spp. (Erwinia chrysanthemi) in previous studies (Hsu and Tzeng 1981). The present study is the first report of P. aroidearum infecting carrots in Taiwan. The findings may add to our understanding of the diversity of soft rot pathogens affecting carrot production in Taiwan.