RESEARCHFree Access icon

Two New Bacterial Pathogens of Peanut, Causing Early Seedling Decline Disease, Identified in the Texas Panhandle

    Authors and Affiliations
    • Ken Obasa1
    • Leonard Haynes2
    1. 1Department of Plant Pathology and Microbiology, Texas A&M University, College Station, TX 77843
    2. 2Texas A&M AgriLife Extension Services, Texas A&M University System, College Station, TX 77843


    Peanut (Arachis hypogaea L.) is cultivated in tropical and subtropical regions of the world as an important source of oil and protein. Until now, bacterial wilt, caused by Ralstonia solanacearum, was the only known bacterial disease of peanut. In 2020, widespread incidence of poor stand establishment was observed in multiple production fields planted to the Spanish-type peanut varieties in the Texas Panhandle. The observed symptoms included seed rot, pre- and postemergence damping-off, poor seedling vigor, poorly developed root systems with little or no nodule formation, and death. Subsequent diagnosis of symptomatic seedlings recovered two bacterial species identified by BLAST using 676- and 661-bp 16S rRNA fragments as a Ralstonia sp. and a Pantoea sp., respectively. To investigate a possible causative role of these bacteria in the observed peanut disease, the pathogenicity of the two isolates was evaluated under greenhouse conditions relying on Koch’s postulates. Cell suspensions of the two bacteria, separately and in combination, were used to inoculate seeds of a Valencia-type peanut variety with no history of the disease and found to be pathogenic on the resultant seedling plants. Symptoms that developed on the inoculated plants were similar to the symptoms initially observed in the field, including seed rot, pre- and postemergence damping-off, poor seedling vigor, and root establishment. The two bacteria were also successfully recovered from inoculated and symptomatic plants, thus satisfying Koch’s postulates. Given the early onset of symptom development on affected seeds and seedlings, a seedborne origin of the disease, described here as early-decline bacterial disease of peanut, was investigated in the same batches of peanut seeds that were planted, as well as seeds later harvested in some of the affected fields. Identical bacterial species, on the basis of 16S rRNA identity, were recovered from all of the seeds evaluated indicating that the bacteria are both seedborne and seed-transmissible. Multilocus sequence analysis involving six genes (dnaK, fumC, gyrB, murG, trpB, and tuf) showed that these new strains are most closely related to R. pickettii and Pantoea dispersa, but also phylogenetically distinct. The two bacteria were designated Ralstonia sp. strain B265 and Pantoea sp. strain B270. Losses from the disease in affected fields in 2020 averaged 50% (US$1.12 million) from a total of nine production fields. Findings from this study provide evidence for two new bacterial pathogens of peanuts capable of infecting Spanish and Valencia peanut varieties.

    Bacterial wilt, caused by Ralstonia solanacearum, is a species complex comprising several strains with phenotypic and genotypic variations (Fegan et al. 1998; Gillings and Fahy 1994), and is an important disease affecting >250 plant species (Buddenhagen 1986); it causes up to 30% yield loss, with total crop failure in extreme cases (Yu et al. 2011). The R. solanacearum species complex is subdivided into four phylotypes, a taxonomic equivalent of subspecies; and at least 23 sequevars, the equivalent of infra-subspecific groups (Fegan and Prior 2005). R. solanacearum isolates that cause wilt disease in peanut belong to phylotype I. At the infra-subspecific level, they are represented in at least six sequevars (Jiang et al. 2017) Bacterial wilt disease of peanut caused by R. solanacearum was first reported in the coastal belt of Natal, in South Africa, during 1924 to 1925, where it has since occurred every year thereafter (McClean 1930). The disease is now known to be distributed across peanut-growing regions in tropical and subtropical humid countries (Wicker et al. 2007). The bacterial wilt disease of peanut is considered soilborne and favored by soils with high-moisture-holding capacities, such as heavy loam soils. In the soil, infection proceeds with the penetration of root cortical cells; this is followed by an increase in bacteria cell density, which results in sudden wilting of affected plants. Management of bacterial wilt in peanut is challenging, given the broad host range of the pathogen and the soilborne nature of the disease. Conventional management strategies, such as soil treatment, crop rotation, and other cultural methods, have not proven effective against the disease (Cao et al. 2009). Consequently, use of resistant varieties remains the most cost-effective strategy for the management of bacterial wilt in peanut worldwide, despite the known inverse relationships between the available genetic resistances and peanut yield and quality (Lu et al. 2010).

    During the 2020 production season, widespread incidences of poor stand establishments were reported in production fields in Donley County in the Texas Panhandle planted with Spanish-type peanut varieties. Symptoms observed in the affected fields included seed rot, pre- and postemergence damping-off, poor seedling vigor, poorly developed roots with little to no nodule formation, and death (Fig. 1). The observed above-ground symptoms looked identical to those of peanut bacterial wilt caused by R. solanacearum, a disease that is not known to be present in the United States. Unlike peanut wilt caused by R. solanacearum, the observed symptoms and disease severity were greater in fields with sandier soils (sandy-loam) compared with those with heavier loam soils.

    Fig. 1.

    Fig. 1. Field-grown peanuts in Donley County in the Texas Panhandle in 2020 showing symptoms including A, seed rot; B, poor seedling vigor/senescence; C, normal root development; and D, poorly developed roots lacking nodules.

    Download as PowerPoint

    The observed constant association of two bacterial species with symptomatic peanut plants from spatially isolated production fields suggested a possible association of at least one of the two bacterial isolates with the observed diseased symptoms. This study therefore investigated a possible causative role for the observed disease symptoms by the two bacterial isolates through a possible association of either or both bacteria with peanut seeds to determine if they are seed-transmissible and/or seedborne; and the identity of the two bacterial isolates by sequence comparison with representative members of closely related species.

    Materials and Methods

    Bacterial isolation, purification, and identification.

    Symptomatic peanut seedling stems, cotyledons, and seeds were rinsed in sterile-distilled water and surface-sterilized by soaking in 10% hypochlorite solution for 1 min and rinsing in five changes of sterile distilled water. The surface-sterilized seedling tissues were blotted dry on Kimwipes (catalog #34155; Kimberly-Clark, Roswell, GA) and cut into 0.5- to 1-cm-long sections with sterile scalpel blades. The cut tissue sections were placed in Petri dishes containing water agar (WA) using sterile forceps and the plates were incubated overnight in the dark at 25°C. Surface-sterilized seeds were split in half along the divide between the two cotyledons of each seed, and each half was placed face-down on WA and incubated using the same parameters described above. After the incubation period, bacterial outgrowth from the sectioned tissue samples or seeds were purified for single colonies by streaking onto Luria Bertani (LB) agar plates and incubated overnight in the dark at 28°C. Representative colonies from the resulting single colonies were selected and further purified through a second round of streaking on LB agar plates and similarly incubated overnight. Single colonies from the second purification round were transferred separately into LB broth and incubated overnight at 28°C and 240 rpm. Genomic DNA was extracted, as described by Maniatis et al. (1982), from the overnight cultures and used in downstream analysis. The 16S rRNA PCR amplification was carried out using the primer pair 27F (AGAGTTTGATCCTGGCTCAG) and 1492R (GGTTACCTTGTTACGACTT; Galkiewicz and Kellogg 2008) at a final concentration of 0.5 μM and 1× final concentration of 2× HF Master-Mix (Bio-Rad, Hercules, CA) in a 25-μl total reaction volume. The PCR amplification cycle consisted of an initial denaturing step at 98°C for 3 min, 35 cycles of three steps consisting of 98°C for 10 s, 57°C for 30 s, 72°C for 30 s, and a final step of 72°C for 10 min. The PCR products, ∼1,500-bp amplicons, were sequenced, and the resulting sequences used to identify the respective bacteria through Basic Local Alignment Search Tool (BLAST; searches.

    Phylogenetic analysis and whole-genome alignments.

    DNA library preparation and sequencing.

    For Illumina short-read sequencing, the genomic DNA of each sample was randomly sheared into short fragments of ∼350 bp, which was used for library construction using the NEBNext DNA Library Prep Kit (New England Biolabs, Ipswich, MA) according to the manufacturer’s instructions. This was followed by end repairing, dA-tailing, and ligation with the NEBNext adapter. The required fragments (300- to 500-bp size) were PCR-enriched by P5 and indexed P7 oligos. The prepared DNA library was then purified and verified for concentration and quality using a Qubit 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA), and subsequently assessed for insert size using the 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Finally, the constructed libraries were pooled and paired-end sequenced on the Illumina HiSeq (Illumina, San Diego, CA) sequencing platform. For long-read sequencing with Oxford Nanopore MinION (Oxford Nanopore Technologies, Cambridge, MA), library preparation was carried out with the Rapid Barcoding Sequencing kit (SQK-RBK004; Oxford Nanopore Technologies) according to the manufacturer’s instructions. Sequencing was done using the R9 flow cells according to the manufacturer’s instructions.

    Assembly and annotation of sequence reads.

    The generated raw reads from all sequencing outputs were processed to remove adapters and primers using the post-run processing softwares for the respective sequencing technologies. The processed reads were assembled using the default settings of the Geneious SPAdes DNA seq de novo assembler (Biomatters, Inc., San Diego, CA; Bankevich et al. 2012; Gurevich et al. 2013). The assembly was analyzed to identify prokaryotic genes using the tool GLIMMER (Delcher et al. 1999) and the identified prokaryotic genes were subjected to BLAST search and the gene ontology terms associated with hits obtained from the BLAST search were retrieved by mapping (Gotz et al. 2008). Next, Blast2GO annotation (Gotz et al. 2008) and InterProScan annotation (Jones et al. 2014) were independently run on the mapped assembly and the resulting annotation results merged. A final round of annotation was carried out using the tool EggNOG mapper v.1.0.3 in the program EggNOG v.5.0.0 (Huerta-Cepas et al. 2019) for the functional annotation of novel sequences using precomputed EggNOG-based orthology assignments, and the resulting annotations were merged similarly with those from the earlier annotations.

    Phylogenetic analysis and whole-genome alignments.

    The sequences for dnaK, gyrB, murG, trpB, tuf, and fumC (Table 1) for the peanut bacterial isolates were concatenated and aligned to similarly concatenated genes from closely-related species (Table 1) identified by BLASTn (Altschul et al. 1990). A phylogenetic tree was constructed using the Tamura–Nei genetic distance model (Tamura and Nei 1993) and the default settings of the Geneious Tree Builder application in the program Geneious Prime (Kearse et al. 2012). Furthermore, alignments of the respective assembled genomes of the peanut bacterial isolates with those of the type strains of their respective closest relatives identified by BLASTn was carried out using the JSpeciesWS online taxonomic threshold service (Richter et al. 2015; to determine the average nucleotide identity (ANI) between each pair.

    Table 1. List of 37 bacterial species and genes used for multilocus sequence comparisons and phylogenetic analysis of Ralstonia sp. strain B265 and Pantoea sp. strain B270

    Bacterial inoculum preparation and seed inoculation.

    Bacterial cultures were grown overnight in LB broth at 28°C and 240 rpm. From the overnight cultures, a subculture was transferred into fresh LB broth and similarly incubated at 28°C and 240 rpm to an absorbance value (OD600) of 0.5 as measured with a DS-11 FX+ Spectrophotometer (DeNovix, Wilmington, DE). To estimate the cell density in CFUs for each bacterial culture, a subsample of the final culture was serially diluted in sterilized nuclease-free water (Ambion-Life Technologies Corp., Austin, TX), then spread on LB agar plates in triplicate and incubated overnight in the dark at 28°C. Healthy peanut seeds were inoculated with each bacterium separately or with a mixture consisting of equal volumes by soaking them for 30 min in the bacterial suspensions, with a final concentration of 8.0 × 106 CFU. Peanut seeds of the Valencia type with no history of the disease that had been soaked in sterile water, and seeds of the Spanish type from a batch planted to one of the affected fields, served as the noninoculated healthy and unhealthy controls, respectively. Lastly, the treated seeds were transferred onto Kimwipes and air-dried for ∼2 h in a biosafety hood.

    Pathogenicity assay.

    Inoculated and noninoculated seeds were seeded at a rate of six seeds per pot in sterile potting media (Berger BM1 Nutrient Retention General Purpose Media; Hummert International, Topeka, KS) contained in 1.87-liter plastic pots. Three replicate pots per treatment were used. The pots were lightly watered daily and maintained at 28°C, and a 14-h photoperiod was achieved with supplemental lighting under greenhouse conditions for 4 weeks. Pots were arranged in a complete randomized design. After 4 weeks, the replicated pots of each treatment were assessed for seed germination percentage, then qualitatively assessed for postemergence damping-off, level of root development, stunting, and overall plant vigor relative to the noninoculated controls. To verify the causal pathogen, stem and root tissues from symptomatic plants were collected, surface-sterilized, placed on WA in Petri dishes, and diagnosed as described above for the presence of the bacteria used to inoculate the seeds for the particular treatment. Recovered bacteria were purified by streaking to obtain a single colony. Their 16S rRNA sequence fragments were aligned and compared with those of the bacterial isolates originally used to inoculate the seeds. The study was conducted twice.

    Data analysis.

    Statistical analysis of plant count data was performed with the statistical software Minitab (v.19.2020.1; Minitab Inc., State College, PA). Count data were subjected to analysis of variance, and treatment means were compared using Fisher’s individual error rate at P ≤ 0.05.


    Bacteria associated with symptomatic tissues.

    Bacteria were isolated from stems and cotyledons of symptomatic peanut seedlings collected from affected fields. Two bacteria were consistently recovered from the symptomatic tissues, and the BLAST results for the 16S rRNA fragments identified these as Ralstonia sp. and Pantoea sp. Multilocus sequence analysis using six genes (dnaK, fumC, gyrB, murG, trpB, and tuf) obtained from the respective draft genome sequences indicated that the two bacterial isolates were most closely related to R. pickettii and Pantoea dispersa (Figs. 2 and 3). The branching of the two bacteria also indicated that they likely represent unique strains of their respective genera (Figs. 2 and 3). To evaluate the relatedness of the two bacterial isolates with their closest relative identified in the phylogenetic analysis, alignments of whole-genome sequences of the two bacteria and a type strain of their respectively most closely related species, i.e., R. pickettii ATCC 27511 (GenBank accession: GCA_000743455.1; Table 1), and P. dispersa CCUG 25232 (GenBank accession: GCA_008692915.1; Table 1), was performed. Additionally, whole-genome sequence alignment between the Ralstonia isolate and a R. solanacearum K60-type strain, a phylotype-II strain (GenBank accession: NCTK01000000; Table 1), was also performed for comparison.

    Fig. 2.

    Fig. 2. Phylogeny of Ralstonia sp. B265 with 14 closely related species and an outgroup using concatenated sequences of six genes (dnaK, fumC, gyrB, murG, trpB, and tuf) showing their relatedness. Scale bar indicates branch length, estimated using the Tamura–Nei genetic distance model.

    Download as PowerPoint
    Fig. 3.

    Fig. 3. Phylogeny of Pantoea sp. B270 with 22 closely related species and an outgroup using concatenated sequences of six genes (dnaK, fumC, gyrB, murG, trpB, and tuf) showing their relatedness. Scale bar indicates branch length, estimated using the Tamura–Nei genetic distance model.

    Download as PowerPoint

    For the comparison of the genomes of the Ralstonia isolate and R. pickettii ATCC 27511-type strain, the calculated ANI values and aligned nucleotide percentages were 99.90 and 94.89%. In contrast, for comparison of the genomes of the Ralstonia isolate and R. solanacearum K60-type strain, the calculated ANI values and aligned nucleotide percentages were 82.08 and 62.44%. For the Pantoea genomes, the calculated ANI values and aligned nucleotide percentages were 97.65 and 88.71%. Therefore, the two bacteria were designated Ralstonia sp. strain B265 (GenBank accession: JAHCSY000000000) and Pantoea sp. strain B270 (GenBank accession: JAHCSX000000000).

    Bacterial isolates pathogenic on peanut.

    Peanut seeds (Valencia-type) inoculated with either or both bacterial inoculum resulted in seedlings with disease symptoms similar to those observed on the symptomatic plants in affected fields. Inoculations with Ralstonia sp. strain B265 or Pantoea sp. strain B270 also resulted in identical disease symptoms that included significantly (P = 0.05) reduced seed germination, postemergence damping-off, stunting, poor seedling vigor, poor and abnormal development of the root systems, and death (Figs. 4 and 5). Mixed infections involving both bacteria resulted in increased symptom severity relative to the noninoculated healthy control, especially in poor and abnormal development of the shoots and roots (Fig. 5D and I). Seeds from the noninoculated healthy control had a significantly (P = 0.05) higher germination percentage and nonsymptomatic seedlings with normal root developments (Fig. 5A and F). In contrast, the noninoculated unhealthy control seeds had reduced germination, postemergence damping-off, stunting, poor seedling vigor, abnormal development of the root systems similar to those of the inoculated seed treatments, and death (Figs. 4 and 5E and J). Bacterial isolates recovered from symptomatic seedlings, including the noninoculated unhealthy controls, had 16S rRNA sequence fragment identities that were 100% identical (data not shown) to those of the original bacterial isolates used to inoculate the seeds, thus satisfying Koch’s postulate. Neither type of bacteria was recovered from the noninoculated healthy control plants.

    Fig. 4.

    Fig. 4. Comparison of the effect of inoculation of healthy peanut seeds with Ralstonia sp. B265, Pantoea sp. B270, and a combination of both on peanut stand count showing a significant reduction relative to the non-inoculated healthy control. Each scale bar represents means from two independent experiments with three replicates each. Interval bars represent the standard error of the means. Bars with the same letter are not statistically different (P ≤ 0.05).

    Download as PowerPoint
    Fig. 5.

    Fig. 5. Comparison of non-treated healthy peanut seeds (A and F), and non-treated unhealthy peanut seeds (E and J) grown in sterile potting media under greenhouse conditions 4 weeks after planting with the effect of inoculation of healthy peanut seeds with Ralstonia sp. B265 (B and G), Pantoea sp. B270 (C and H), and a combination of Ralstonia sp. B265 and Pantoea sp. B270 (D and I), respectively.

    Download as PowerPoint

    Bacterial pathogens are seedborne and seed-transmissible.

    At the end of the cropping season in 2020, evaluation of seeds from the same batches that were planted in the affected fields, and seeds harvested from the affected fields, resulted in the successful recovery of bacteria of the Ralstonia and Pantoea genera. These bacteria all had 16S rRNA sequence fragments with a 100% match to those of Ralstonia sp. strain B265 and Pantoea sp. strain B270, respectively (data not shown).


    The widespread incidence of early-season poor stand establishments and failure of peanut crops in several fields in the Texas Panhandle in 2020 led to the discovery of two new bacterial pathogens of peanut that were initially identified based on their respective 16S rRNA fragments and BLAST (Altschul et al. 1990) searches ( Using 676- and 661-bp 16S rRNA sequence fragments, the first bacteria and the second bacteria returned a 99.7 and 100% identity match with R. pickettii and P. dispersa, respectively. Although neither R. picketti (syn. Pseudomonas pickettii and Burkholderia pickettii) nor any Pantoea species had previously been reported as a direct cause of disease in peanut, a pathogen of R. pickettii had been reported, in Italy in 2008, to be involved with leaf spot and blight symptoms disease on bird of paradise tree (Strelitzia reginae; Polizzi et al. 2008). In another case, a strain of R. pickettii was isolated from the rhizosphere of tomato plants as a congeneric strain of R. solanacearum, and then utilized as an effective biological control of R. solanacearum (Zhong et al. 2013). Other reports of R. pickettii-associated diseases have also included clinical cases in humans (Fujita et al. 1981; Ryan et al. 2006; Verschraegen et al. 1985). On the other hand, the only known reported association of a Pantoea species with peanut involves the symbiosis of P. ananatis (syn. Erwinia uredovora) with the peanut rust fungus (Arias et al. 2013). P. dispersa has been described as both a plant pathogen and a beneficial organism. In a 2020 publication, P. dispersa was implicated for the first time as the principal causative agent of inner boll rot in cotton (Nagrale et al. 2020). P. dispersa was also reported as an effective biocontrol agent against the pathogenic fungus Ceratocystis fimbriata, the causative agent of black rot, which is an important postharvest disease of sweet potato (Jiang et al. 2019). Also, a strain of P. dispersa was identified in the rhizosphere of sugarcane and shown to induce plant growth promotion, nitrogenase activity, and stress- and defense-related genes (Singh et al. 2021). Lastly, P. disersa has also been associated with clinical cases in humans including bacteremia (Asai et al. 2019) and neonatal sepsis (Mehar et al. 2013).

    Overall, symptoms from infections by the two bacteria were identical and included seed rot, poor germination, pre- and postemergence damping-off, poor seedling vigor, poor and abnormal root and root nodule development, and death. The similarity in symptoms suggests that the mechanism of infection by the two bacteria might be similar. Under field conditions, the poor root development in affected plants contributed to increased susceptibility to wilting during hot periods, decreased plant vigor, and, consequently, to colonization and infection by secondary invaders and opportunistic fungal pathogens like Alternaria. Furthermore, the combination of instances of none to poorly developed root nodules likely negatively impacted the nutritional nitrogen status and vigor of affected peanut plants, contributing to overall reduction in yield from affected fields. Early-season (2020) projected loss estimates based on incidence levels and observable disease severity from the disease in affected fields were put at 10 to 30%. However, actual losses at the end of the production season averaged 50% across the affected fields in the Texas Panhandle. In Donley County, a total loss of $1.12 million is from the disease; ∼2,800 acres (∼$400/acre) was reported. Unlike the bacterial wilt disease of peanut caused by R. solanacearum that is favored by soils with higher moisture-holding capacities such as heavy-loam soils, the affected peanut fields in the Texas Panhandle had sandy-loam soils, and the symptoms and disease severity were greater in fields with sandier soils.

    Whole-genome sequence analysis of the two peanut bacterial pathogens confirmed their closest relatives as belonging to the genera Ralstonia and Pantoea, respectively. Furthermore, the calculated ANI value for Ralstonia sp. B265 indicated it is more closely related to R. pickettii than it is to R. solanacearum (ANI < 90%). And, because the calculated ANI from the comparison of Ralstonia sp. B265 and the R. solanacearum K60-type strain was less than the threshold for both bacteria to be considered as belonging to the same species (Goris et al. 2007; Kim et al. 2014; Richter and Rossello-Mora 2009; Scortichini et al. 2013), additional analysis to determine the phylotype and sequevar of Ralstonia sp. B265, within the R. solanacearum species complex, was not undertaken. However, the percentages of aligned whole-genome nucleotides associated with their respective calculated ANI also suggests Ralstonia sp. B265 and Pantoea sp. B270 could represent distinct species within their respective genus. The uniqueness of both isolates is also supported by their respective phylogenetic clustering based on multilocus sequence analysis of six genes.

    The successful recovery of the two bacterial pathogens, Ralstonia sp. strain B265 and Pantoea sp. strain B270, from the same batches of peanut seeds planted and those harvested from the affected fields indicated that both bacteria are seedborne and seed-transmissible. This finding is supported by the fact that all of the affected fields, though spatially removed from each other, were planted with seeds purchased from the same source. The seedborne attribute of this new disease contrast with that of peanut bacterial wilt, which is soilborne. Investigation is underway to determine if, additionally, either or both bacteria can become established and soilborne in infested soils. The successful infection of peanut varieties of the Spanish- and Valencia-types, however, suggests that other peanut varieties and types could potentially also be susceptible to this new bacterial disease designated “bacterial early-decline disease of peanut.”

    The seed-transmissible attribute of this disease pathogens, unlike the peanut bacterial wilt pathogen, provides disease management opportunities for limiting the spread of the pathogens. For instance, seed-testing could be used to detect the presence of the bacterial pathogens in peanut seeds, thus allowing for the exclusion of infected seeds from the supply chain, trade, and planting in farmer’s fields. Implementation of such management strategy can help to mitigate the risk to farmers of the potential economic losses associated with bacterial early-decline disease of peanut, as well as safeguard peanut trade at all levels, including international trade.

    The author(s) declare no conflict of interest.

    Literature Cited

    Funding: This research was supported in part with funding from the Texas Peanut Producers Board with award no. M2102710.

    The author(s) declare no conflict of interest.