In Japan, pea sprouts (Pisum sativum L.) are grown in plant factories and harvested as young, tender stems for consumption as a vegetable (Japanese, tomyo). In 2019, root rot symptoms on hydroponically grown pea sprouts caused up to 50% yield losses in a plant factory in Japan. The symptoms included root browning and darkening of the seed coat; severe symptoms included root rot, a shorter main root, and fewer lateral roots. Aboveground growth was also reduced in affected plants, although leaves and stems showed neither rot nor necrosis (Fig. 1a, b). Bacterial ooze was observed in symptomatic roots under light microscopy, which suggests that bacteria were the causal agents. However, bacterial root disease had not previously been reported in pea plants in Japan (Phytopathological Society of Japan 2022). Therefore, a series of experiments were performed to determine the causal agent of the observed root rot disease.

Fig. 1
figure 1

Bacterial brown root rot in pea sprouts and isolated bacterial colonies. (a, b) Original symptoms in pea plants. Plants are arranged by disease symptom severity, from healthy plants (left) to severe symptoms (right). Severe symptoms included brownish roots and stunted growth compared to healthy plants. (c–i) Inoculation assay results at 4 days after inoculation with UTLPPB19801 (c), UTLPPB19802 (d), UTLPPB19803 (e), UTLPPB19804 (f), sterile water (g), P. alliivorans NL7703 (h), or P. asplenii 6801 (i). Each bacterial isolate was suspended in sterile water. Only the pea sprout isolates induced symptoms such as root browning with lateral root suppression and darkening of the seed coat. Scale bar = 1 cm. (j) UTLPPB19802 isolate colonies on yeast tryptone medium after 2 days at 28 °C

Full size image

To isolate the pathogen, three roots and two seed coats for each of the three disease stages (mild, moderate, and severe) pea sprouts were washed with running tap water, surface sterilized with 70% (v/v) ethanol for a few seconds and with 0.5% sodium hypochlorite solution for 1 min, and rinsed three times with sterile water. Each tissue type was ground in 400 µL saline solution, and then 1 µL of the resulting suspension was spread on nutrient agar medium consisting of 0.5% peptone, 0.3% Lab-Lemco powder, 0.5% sodium chloride, and 1.5% agar. After 3 days of incubation at 28 °C, 16 bacterial colonies were moved to yeast tryptone medium consisting of 1% tryptone, 0.5% yeast extract, and 1.5% agar. For taxonomic analysis, the 16S rDNA sequence of each isolate was amplified through polymerase chain reaction (PCR) and sequenced directly as described previously (Kitazawa et al. 2014). The 16S rDNA sequence analysis revealed that seven out of 16 isolates appeared to belong to the genus Pseudomonas, and were especially dominant in the isolates from roots (Supplemental Table 1). Since these seven isolates could be divided into two groups based on the 16S rDNA sequence, representative five isolates (#7, 9, 11, 12, and 14) including both groups were inoculated on pea seedlings as follows. Pea seeds for sprouts (Noguchi Seed, Hanno, Japan; cultivar unknown) were washed under running tap water, surface sterilized as described above, and allowed to absorb sterile water at 28 °C for 4 h. Then the seeds were sown on cotton pads moistened with ca. 35 mL sterile water in sterile plastic boxes (L × W × H = 75 × 75 × 100 mm) and incubated at 28 °C for 3 days. The roots and seed coats of the germinated pea seedlings were inoculated by dropping 1 mL of bacterial suspension (ca. 107 cfu/mL) per individual without wounding, and then incubated at 28 °C for 7 days. Four isolates (#7, 9, 12, and 14) with identical 16S rDNA sequences caused brown root rot symptoms that were similar to the original symptoms (Fig. 1c–f), whereas no symptoms were observed on mock-treated sprouts (Fig. 1g) nor on those treated by the other isolate (#11) with a 16S rDNA sequence different from that of the four isolates (data not shown). Re-isolation and partial 16S rDNA sequencing of bacteria from the inoculated roots with brown rot symptoms revealed that they were identical to the inoculum, indicating that these isolates are causal agents of brown root rot of pea sprouts. Out of the four pathogenic isolates, three isolates obtained from symptomatic roots were named as UTLPPB19801, –19803, and –19804; and one obtained from seed coats was named as UTLPPB19802 (Supplemental Table 1). The partial 16S rDNA sequence (1360 nt) was deposited in the DNA Data Bank of Japan (DDBJ) (accession no. LC752793 for UTLPPB19802).

To elucidate the taxonomic position of the pathogen, the 16S rDNA sequence described above was further analyzed by a BLAST search against the National Center for Biotechnology Information (NCBI) RefSeq genome database. The sequence shared high sequence identity with those of Pseudomonas species belonging to distinct subgroups (SGs): P. vanderleydeniana (99.34%, accession no. NZ_CP077093.1) in the P. asplenii SG, P. xanthosomae (99.27%, NZ_CP077075.1) and P. fakonensis (99.27%, NZ_CP077076.1) in the P. xamthosomae SG, and P. xantholysinigenes (99.27%, NZ_CP077095.1) in the P. mosselii SG (Girard et al. 2021). To further clarify the phylogenetic relationships of the pathogen within the genus Pseudomonas, a multilocus sequence analysis was performed as described previously (Lalucat et al. 2020). Briefly, partial sequences of the gyrase B subunit (gyrB), β subunit of RNA polymerase (rpoB), and σ subunit of RNA polymerase (rpoD) of the isolates were determined. The primer pairs for rpoB and rpoD were LAPS5/LAPS27 (Tayeb et al. 2005) and PsEG30F/PsEG790R (Mulet et al. 2009), respectively. Because gyrB was not amplified with the primers used in Lalucat et al. (2020), a primer pair targeting the same region, gyrB_F and gyrB_R, was newly designed based on the gyrB sequences of several Pseudomonas species (Supplemental Table 2). PCR conditions for each primer set were shown in Supplemental Table 2. Each fragment was analyzed by direct sequencing as described above. Since the partial sequences of these three genes were also identical among the four isolates, those of the isolate UTLPPB19802 were deposited in the DDBJ database (gyrB [863 nt]: LC752796, rpoB [1097 nt]: LC752794, and rpoD [706 nt]: LC752795). The sequences were combined in the order of 16S rDNA–gyrB–rpoB–rpoD and used for phylogenetic analysis with the corresponding sequences of several other Pseudomonas species (Supplemental Table 3). A phylogenetic tree was constructed using the neighbor-joining method and the Jukes–Cantor algorithm as described previously (Lalucat et al. 2020). UTLPPB19802 formed a monophyletic clade with some Pseudomonas species belonging to the P. asplenii SG, i.e., P. vanderleydeniana and P. asplenii, including a strain formerly classified as P. fuscovaginae (Tohya et al. 2020), with high support (bootstrap values of 100%) (Fig. 2). In particular, UTLPPB19802 and P. vanderleydeniana formed a monophyletic clade (bootstrap values of 83%). Therefore, the causal agent of brown root rot of pea sprouts was a Pseudomonas species that is closely related to P. vanderleydeniana and belongs to the P. asplenii SG.

Fig. 2
figure 2

Neighbor-joining tree based on the concatenated gene sequences 16S rDNA–gyrB–rpoB–rpoD of isolate UTLPPB19802 and Pseudomonas species. Escherichia coli was used as an outgroup (data not shown). The tree was constructed using the Jukes–Cantor algorithm, as described by Lalucat et al. (2020). Numbers at nodes indicate bootstrap values (≥ 70%) based on 1000 replicates. Scale bar indicates nucleotide substitutions per site. Bacterial strain names and accession numbers are listed in Supplemental Table 3

a Formerly classified as P. fuscovaginae 6801

Full size image

We characterized the four isolates in detail following methods described previously (Buelow 1964; Goto and Takikawa 1984a, b, c, d), with appropriate negative and positive control strains for each test (Supplemental Table 4). The isolates formed round, yellowish-white colonies on yeast tryptone medium (Fig. 1j). Further characteristics were shown in Table 1 with the previously reported characteristics of the P. asplenii SG members. The four isolates had differences from P. asplenii in production of fluorescent pigment and arabinose assimilation; from P. agarici in production of fluorescent pigment, arginine dihydrolase, gelatin hydrolysis, and assimilation of arabinose, D-mannose and trehalose; from P. batumici in the production of levan, gelatin hydrolysis, and assimilation of arabinose, sucrose, myo-inositol, and L-malate. P. vanderleydeniana had similar assimilation profiles except for D-mannitol, whereas the other characteristics were unknown (Table 1). These results indicate that the four isolates share distinct characteristics in the P. asplenii SG, although P. vanderleydeniana may have partially similar bacteriological characteristics to the isolates.

Table 1 Bacteriological characteristics of the isolates from symptomatic pea sprouts and species in Pseudomonas asplenii subgroup
Full size table

To compare the pathogenicity of the isolates with known phytopathogenic Pseudomonas, two Pseudomonas species were inoculated to pea seedlings under the same conditions: P. alliivorans NL7703, which have been shown to be pathogenic on pea leaves (Tsuchiya et al. 1980) and P. asplenii 6801, a rice-pathogenic member of the P. asplenii SG (Miyajima et al. 1983). Neither of these species induced clear symptoms (Fig. 1h,i), suggesting that the four isolates exhibited higher virulence to pea roots compared to the previously reported phytopathogenic Pseudomonas.

In this study, we showed that bacteria belonging to the P. asplenii SG is the causal pathogen of brown root rot of pea sprouts. In inoculation tests, the bacteria induced clear symptoms, while P. alliivorans, which is pathogenic on aboveground peas, and rice-pathogenic P. asplenii did not. This suggests that the bacteria behave as a genuine pathogen rather than an opportunistic pathogen under hydroponic conditions. Since this disease was observed in a particular plant factory, it remains uncertain if it poses a risk to peas under field conditions, and further study, including leaf inoculation tests, is necessary.

Based on our results, the isolates demonstrate a close phylogenetic relationship and similar physiological properties to P. vanderleydeniana. However, our understanding of P. vanderleydeniana is limited, as only one strain’s genes have been reported and many of its bacterial characteristics remain unstudied. Moreover, P. vanderleydeniana was isolated as a rhizobacterium of non-diseased rice and has not been reported as a plant pathogen (Girard et al. 2021). Due to the lack of knowledge about P. vanderleydeniana, it was difficult to definitively identify the isolates as P. vanderleydeniana in this study. Since the draft genome of P. vanderleydeniana has been revealed (Girard et al. 2021), whole-genome sequence analysis of the isolates will be critical for assigning the species of the pea sprout pathogen.

To the best of our knowledge, this is the first report of a bacterial root disease in pea plants and of a P. asplenii SG member being pathogenic to legumes. Therefore, we propose the name “bacterial brown root rot” (Japanese, kassyoku negusare saikin byo) for this disease.