RSSC-Lineage Multiplex PCR Assay Detects and Differentiates Ralstonia solanacearum, R. pseudosolanacearum, R. syzygii, and the R3bv2 Subgroup

The Ralstonia solanacearum species complex (RSSC) (Smith 1896; Yabuuchi et al. 1995) (syn: Pseudomonas solanacearum) is a group of related bacteria in the Burkholderiaceae that causes vascular wilt of plants. The RSSC infects a wide range of solanaceous and non-solanaceous hosts in the lowland tropics, subtropics, and temperate regions (Hayward 1994; Janse et al. 2004; Lowe-Power et al. 2021; Paudel et al. 2020; Pradhanang et al. 2000). The pathogens survive in soil, water, and infected plant tissues and are transmitted through contaminated soil, plants, and irrigation water. Most plant-pathogenic Ralstonia spp. enter host plants via wounds or natural openings at sites of lateral root emergence. However, a few lineages are vectored by pollinating insects or Hemipteran tube-building cercopoids (Buddenhagen 1960; Buddenhagen and Kelman 1964; Ray et al. 2022; Safni et al. 2018). Infected tubers, rhizomes, and stem cuttings are common means of disseminating these pathogens to new geographical locations (Champoiseau et al. 2009; Ray et al. 2022; Remenant et al. 2011). Because of their ability to survive in soil and alternate hosts, remarkable diversity, worldwide distribution, and an ever-expanding host range, RSSC pathogens have an extraordinary economic impact on global agricultural production systems and continue to present new research challenges to the scientific community (Paudel et al. 2020).

A near-clonal pandemic lineage of R. solanacearum strains in phylotype IIB sequevar 1 has been disseminated to most continents (Lowe-Power et al. 2021; Paudel et al. 2020), but it has not yet become established in North America. Known historically and for regulatory purposes as race 3 biovar 2 (R3bv2), strains in this global pandemic lineage are cold-tolerant, aggressive on solanaceous plants (potato and tomato), and form both latent and symptomatic infections on geranium (Buddenhagen and Kelman 1964; Swanson et al. 2005). Thought to have originated in cooler, highland regions of the Andes, R3bv2 is now endemic in Africa, Europe, the Middle East, Asia, Central America, and many island nations. R3bv2 has caused repeated outbreaks of potato brown rot in European seed potato production areas and has also damaged tomato, eggplant, and ornamental industries (Champoiseau et al. 2009; Janse 1996). R3bv2 can asymptomatically infect the semi-aquatic weed Solanum dulcamara, which sheds the bacteria from roots into water that is used for irrigation. This helped the pathogen become established and disseminated in European waterways (Wenneker et al. 1999). Global trade and international movement of vegetative propagative materials and cuttings pose a threat with the latest introduction of R3bv2 to the United States in 2020 (Roman-Reyna et al. 2021; Dickstein et al. 2024). If R3bv2 becomes established in North America and can overwinter and spread in potato-growing regions, it could severely impact the $3.65 billion/year U.S. potato industry (Lawrence and Delheimer 2022). Because this lineage is considered a serious threat to potato production, R3bv2 is a quarantine pest in Europe and Canada and is a strictly regulated Select Agent pathogen in the United States. Thus, availability of a rapid and robust detection method for R3bv2 is critical for early surveillance, quarantine, and eradication of the pathogen.

Since the first identification in 1896 by E. F. Smith, the taxonomy of the bacterial wilt pathogen has been revised multiple times (Paudel et al. 2020). Early efforts classified RSSC strains into five races based on phenotypic characters and host range (Buddenhagen and Kelman 1964; Buddenhagen et al. 1962). However, accumulating data made it abundantly clear that the host range in this group is too elastic and complex to be used for classification (Albuquerque et al. 2014; He et al. 2021; Safni et al. 2018; Wicker et al. 2007). When DNA-based analyses revealed close relationships between some R. solanacearum strains, the Sumatra disease of clove pathogen (currently Ralstonia syzygii subsp. syzygii), and the banana blood disease pathogen (Ralstonia syzygii subsp. celebesensis), the field adopted the term “Ralstonia solanacearum species complex (RSSC).” The RSSC was sub-classified into four phylotypes based on multiplex amplification of a region of the 16S-23S rRNA (Prior and Fegan 2005). The phylotype scheme, which is phylogenetically robust and widely used, correlates with the inferred geographical origin of strains. Diversity of phylotype I strains is widest in continental Asia, phylotype II strains in the Americas, phylotype III strains in Africa, and phylotype IV strains in the Indonesian archipelago, Japan, and Australia. In 2010, Remenant et al. proposed that the RSSC was composed of three genomospecies (species 1: phylotypes I and III; species 2: phylotype II; and species 3: phylotype IV) (Remenant et al. 2010; Safni et al. 2014). Strains from phylotypes I and III are now designated as R. pseudosolanacearum (Rpseu). Phylotype II strains, which include R3bv2, retained the original name, R. solanacearum (Rsol). Phylotype IV strains are reclassified as R. syzygii (Rsyz) with three distinct subspecies: R. syzygii subsp. celebesensis causing banana blood disease, R. syzygii subsp. syzygii causing Sumatra disease of clove, and R. syzygii subsp. indonesiensis strains causing typical soilborne bacterial wilt diseases on a wider range of hosts (Safni et al. 2014). This reclassification was corroborated with supporting data from phenotypic, genomic, and proteomic analyses (Prior et al. 2016). Strains are often further classified into sequevars based on partial sequences of the egl endoglucanase gene. The sequevar system robustly reflects whole-genome phylogenies in Rsol, but sequevars in phylotype I Rpseu do not correspond to whole-genome phylogeny (Sharma et al. 2022).

Effective tools to detect and differentiate the RSSC are urgently needed to exclude the pandemic lineage (R3bv2) from North America. Almost all locations with mild winters and year-round precipitation have one or more endemic or exotic RSSC lineages established in the region. Multiple detection-based tools can identify the pathogen at the RSSC level without indicating whether it is the highly regulated R3bv2 lineage. Agdia Immunostrips that rely on antibodies specific to RSSC extracellular polysaccharides are commonly used for detection of the RSSC (Agdia, Elkhart, IN). The robust RSSC Universal primers 759/760 are used for DNA-based detection of all plant-pathogenic Ralstonia spp. (Opina et al. 1997). A few PCR-based approaches provide more insight into the sub-RSSC identity of a strain. A multiplex PCR assay differentiates the phylotypes of the species complex (Fegan and Prior 2005). A multiplex real-time TaqMan PCR has been developed for the detection of R. solanacearum, R. pseudosolanacearum, and Clavibacter sepedonicus from the asymptomatic potato tubers (Vreeburg et al. 2018). A 16srRNA gene-based PCR assay was developed for the detection of the RSSC in potato tubers (Pastrik and Maiss 2000). Similarly, a duplex PCR to detect phylotype II banana strains and a tube-wise diagnostic microarray for the detection of R. solanacearum have also been developed (Cellier et al. 2015, 2017). There are also R3bv2-specific assays using 630/631 primer pairs and the B2 probe (Fegan et al. 1998). Stulberg et al. (2015) developed a multiplex assay that identifies strains in phylotype IIB sequevars 1 and 2, intended to detect and differentiate Select Agent R3bv2 strains from others in Rsol. Because other lineages of Rsol, Rpseu, and Rsyz can also infect the most common hosts of R3bv2 (solanaceous plants and geraniums) and also infests soil and surface water, we sought to develop a rapid detection tool that can be broadly useful for identifying and tracking RSSC strains in epidemiological studies.

We developed and validated a fiveplex endpoint PCR assay that simultaneously detects and differentiates all three genomospecies, the Select Agent R3bv2, and plant-pathogenic Ralstonia in general. This tool will facilitate early disease detection for Plant Quarantine services, certification of planting stocks, and application of timely intervention practices. The use of the economical GoTaq Green Master Mix improves its use as an economic, convenient tool for routine day-to-day diagnostics and culture characterization studies.

Materials and Methods

Source of strains, growth conditions, and DNA extraction

A total of 113 Ralstonia strains including strains from the three genomospecies, Rpseu, Rsol (including R3bv2), and Rsyz, and 24 other strains encompassing closely related species of Ralstonia and other plant-pathogenic bacterial genera were included in the inclusivity (Supplementary Table S1) and exclusivity panels (Supplementary Table S2), respectively, to validate the fiveplex assay. Strains starting with “A” and “PL” were selected from the Pacific Bacterial Collection at the University of Hawaii at Manoa, and those starting with “LMG” were obtained from the Belgian Coordinated Collections of Microorganisms (BCCM). Cultural conditions and media for the growth of bacteria and bacterial genomic DNA extraction procedures were performed as described by Arizala et al. (2022). Bacterial DNA concentration was analyzed using a NanoDrop 2000/c Spectrophotometer (Thermo Fisher Scientific).

Identity confirmation of the strains

Confirmation of strain identity for the exclusivity panel was performed as previously described (Ahmed et al. 2018; Boluk et al. 2020; Dobhal et al. 2019; Marrero et al. 2013; Ocenar et al. 2019) or was reconfirmed using Sanger sequencing of the partial dnaA region or 16S rRNA partial gene region (Boluk et al. 2020; Dobhal et al. 2019). Ralstonia-specific dnaA primers (dnaA-F: CTGAACCCCATCCTCACCTT; dnaA-R: TCTGCTCCAGCACGTGCA) were used to amplify the partial dnaA chromosomal replication initiator gene for the Ralstonia strains in the inclusivity panel. The PCR conditions were as follows: initial denaturation at 94°C for 5 min followed by 35 cycles of denaturation at 94°C for 20 s, annealing at 58°C for 60 s, extension at 72°C for 60 s, and a final extension at 72°C for 120 s. All PCR reactions were performed using a Bio-Rad T100 Thermocycler (Bio-Rad Laboratories, Hercules, CA). Enzymatic purification of the PCR product and sequencing procedures were followed as described by Arizala et al. (2022). Briefly, The PCR product was purified by adding 2 µl of EXOSAP-IT and incubated at 37°C at 15 min followed by 80°C for 15 min. Purified products were sequenced using both sense and antisense strands at the GENEWIZ facility (Genewiz, La Jolla, CA). After sequencing, the sense and antisense strands were aligned, trimmed, and edited, and an error-free consensus sequence was generated using GENEIOUS Prime v 2020.0.4. The consensus sequences were used for strains’ identity confirmation using the NCBI GenBank BLASTn tool.

Target selection, species complex, and genomospecies-specific primer design

Whole genomes of the representative strains of Rpseu GMI1000 (NC_003295.1), FJAT1463.50 (NZ_CP052114.1), 202 (CP049789.1), EP1(NZ_CP015115.1), CMR15 (NC_017559.1), UW386 (NZ_CP039339.1), CRMRs218 (CP021764.1); Rsol IBSBF 2571 (CP026307.1), SFC (CP026092.1), CIAT_078 (NZ_CP051296.1), R3bv2 UY031 (NZ_CP012687.1), CFBP 8695 (CP047138.1), UW163 (NZ_CP012939.1), Po82 (NC_017574.1); Rsyz PSI07 (NC_014311.1), T82 (NZ_CP022763.1), T51 (NZ_CP022770.1), A2-HRMardi (CP019911.1), T101 (NZ_CP022757.1), SL2312 (NZ_CP022796.1), SL2064 (NZ_CP022798.1); R. mannitolilytica SN82F48 (NZ_CP010799.1), R. pickettii FDAARGOS_410 (NZ_CP023537.1), Dickeya dianthicola RNS04.9 (CP017638.1), Pseudomonas aeruginosa PAO1 (NC_002516.2), Xanthomonas campestris pv. campestris ATCC 33913 (NC_003902.1), and Clavibacter nebraskensis NCPPB 2581 (NC_020891.1), were retrieved from the NCBI GenBank genome database. One representative genome from each species was aligned using the progressive MAUVE (Darling et al. 2010) plugin in GENEIOUS Prime v 2020.0.4. To ensure the robustness and reliability of the target selection, protein sequences from the representative strains of each genomospecies were also downloaded from the NCBI database and used as input for OrthoMCL (Li et al. 2003) to identify the core genes among the strains. Orthologous gene pairs were grouped to identify unique core genes associated with respective genomospecies and R3bv2. The unique genes identified manually from progressive MAUVE and core genes from OrthoMCL were screened for specificity using BOWTIE short and accurate mapper (Langmead et al. 2009) plugin GENEIOUS Prime v 2020.0.4 and BLASTn/BLASTp suite of the NCBI GenBank database. The unique target gene region specific for each of the genomospecies, RSSC, and R3bv2 was individually screened for specificity in the aforementioned genomes. The genomic regions found to be best for the respective group were extracted from the representative strains and aligned against each other using the pairwise alignment tool in GENEIOUS. Multiplex PCR primers were designed for each group using the conserved region of consensus sequences using Web-interface application Primer3 (Rozen and Skaletsky 2000). The designed primers were checked for specificity in silico using BLASTn and GENEIOUS Prime v 2020.0.4. Primers were validated in vitro using multiple strains from different genomospecies and Select Agent groups. The primers found to be specific in vitro were made thermodynamically competent by adding AT-rich flap sequences to the 5′ end to minimize the internal structure, self-dimer formations and to optimize the annealing temperature (Arif and Ochoa-Corona 2013; Larrea-Sarmiento et al. 2019). After adding 5′ flap sequences, primers were validated for in vitro specificity with representative strains from each group. Details of the primer sequences plus flaps, the target gene region, and product size are presented in Table 1. A circular plot was created in the CIRCOS software (Krzywinski et al. 2009) to highlight the presence or absence of unique genomic target regions selected for the five different groups in the genomes of the representative strains mentioned above (Fig. 1).

TABLE 1 Primers and their corresponding target regions used to develop multiplex PCR assay to differentiate three genomospecies and the Select Agent R3bv2 group and Ralstonia solanacearum species complex (RSSC)

Target pathogen	Primer name	Primer sequence 5′–3′a	Amplicon	Target gene
Ralstonia solanacearum species complex	RssC-wF3	TATATATCCTCGACTTTTCCATGAAGCTGTG	141 bp	Glycosyl transferase CDS
	RssCwR3	CTATATATATACCCCACTTGTTGAGGAACTG	(162 bp)b
Ralstonia pseudosolanacearum	Rpseu-wF5 Rpseu-wR5	TTTTATTTTTTTGGTGTCCGGGCCAAGATAG TTATATTACTCGAACGTGCTGCAAAACCACT	228 bp (251 bp)	Intergenic region between Flp family type IV b pilin and TPM domain containing protein
R. solanacearum	RsolP2wF2	ATTCTATTTATATCATGAGCGTTCCCCGACT	454 bp	DUF692 domain-containing protein CDS
	RsolP2wR2	TTTTTTTTTTTGAGGTAGCTGCTGGGGTTC	(478 bp)
R. syzygii	RsyzP4wF4	TTTCTTTTATTATAGTGTCGCGTCCGAACAG	642 bp	Ankyrin repeat domain-containing protein CDS
	RsyzP4wR4	TTTTTCTTTTTCGGTCTCTCCGTCTATCGTT	(664 bp)
Race 3 biovar 2	RsR3B2wF	TTTCATTACTCATGACTGCAGAAACGCTTGA	932 bp	Thymidylate synthase CDS
	RsR3B2wR	TATACATAACTATAGTTCGCCGTGCTCATCT	(954 bp)

^a Sequences in bold are the AT-rich flap sequences incorporated at the 5ʹ end of the primer sequence to improve the primer thermodynamics. Flap sequences were not included in the BLAST analysis.

^b Amplicon length with the flap sequences.

View as image HTML

Download as PowerPoint

Single and multiplex endpoint PCR assays

All the single and multiplex PCR assays were performed in a T100 Thermal Cycler (Bio-Rad). The PCR conditions used to test the individual single primer set, and the primers with a flap were 94°C for 3 min followed by 35 cycles of denaturation at 94°C for 20 s, annealing at 56°C for 30 s, extension at 72°C for 45 s, and final extension at 72°C for 3 min. The multiplex PCR assay was performed with a mixture of five primer sets in a single reaction. Conditions for multiplex PCR were optimized using multiple parameters, including ranges of annealing temperature from 54 to 60°C, annealing durations, multiple concentrations of MgCl₂, dNTPs, and two cycling conditions (35 and 40 cycles). The assay was performed in 50-μl reactions with 25 μl of GoTaq Green Master Mix (Promega, WI) (2× GoTaq Green reaction buffer, 2 mM of each of four deoxynucleotide triphosphates and 3 mM of MgCl₂), 3 μl of MgCl₂ (25 mM), 5 μl primer mix (10 μl from 100 μM normalized stock concentration of each primer sets in 400 μl of nuclease-free water/2 μM final concentration of each primer), 1 μl of dNTPs (2.5 mM), 1 μl of DNA (genomic DNA or infected plant material DNA), and 15 μl of nuclease-free water.

The developed assay was also validated with the Qiagen Multiplex PCR Kit (Qiagen, Germantown, MD). The assay was performed in 25-μl reactions with 12.5 μl of Qiagen multiplex PCR Master Mix, 2.5 μl of Q-solution, 5 μl of primer solution (10 μl from 100 μM normalized stock concentration of each primer sets in 400 μl of nuclease-free water), 1 μl of genomic DNA, and 4 μl of nuclease-free water. A total of 16 μl of the PCR product was mixed with 4 μl of 6× Gel loading dye and electrophoresed using a 2% agarose gel in 1× TAE buffer at 100V for 120 min. Ethidium bromide (Invitrogen, Carlsbad, CA) was used to visualize the agarose gels under UV light using the gel documentation system (Bio-Rad Gel Doc-XR+).

Specificity validation with members of inclusivity and exclusivity panels

Specificity validation of RSSC-specific primer sets, representative strains from all genomospecies of the RSSC were included in the inclusivity panel (Supplementary Table S1), whereas other species in the Ralstonia genus and representative strains from plant-pathogenic bacterial genera were kept in the exclusivity panel (Supplementary Table S2). To validate the specificity of the primer set designed for a specific species or group (R3bv2), strains from the target species/group of Ralstonia were included in the inclusivity panel, whereas strains from other genomospecies were incorporated into the exclusivity panel. All specificity assays were conducted using individual primer sets and by multiplexing all primer sets together. Selected strains from all the genomospecies, including the Select Agent R3bv2, and other closely related plant-pathogenic bacterial genera were also tested for specificity using the Qiagen Multiplex PCR kit.

Limit of detection determination of the single and fiveplex assays

Sensitivity assays were performed using genomic DNA from each target group to determine the limit of detection for both single and multiplex assays. Pure genomic DNA for each target pathogen was quantified using a Qubit 4 fluorometer (Thermo Fisher Scientific) and the starting concentration was adjusted to 10 ng/µl. Tenfold serial dilutions from 10 ng (∼1.59 × 10⁶ CFU) to 1 fg (a total of eight 10-fold dilution series) were prepared using nuclease free-sterile water. For the spiked test, 100 mg of healthy potato stem was crushed, DNA was isolated, and 1 µl of it was added to the reaction containing 10-fold serially diluted DNA. The multiplex assays were performed by challenging all target template groups against all primer sets in a single reaction and single DNA against all five primer sets in a single reaction, whereas for the singleplex assays, sensitivity was determined by using single target DNA against a single primer set in a reaction. Sensitivity tests with a single primer set using the Qiagen Multiple Kit were not performed, as this kit is designed for multiplex reactions.

Validation of the multiplex assay with naturally and artificially infected samples

Two naturally RSSC-infected tomato plants from the Poamoho Research Station, Oahu, Hawaii, were processed for bacterial isolation, and their identity was confirmed by PCR and Sanger sequencing. A healthy-appearing area of vascular tissue near the site of vascular browning was selected for bacterial isolation. Samples were washed with 10% sodium hypochlorite (NaOCl) for 30 s followed by two rinses with sterile distilled water for 30 s each. Washed samples were crushed with sterile plastic pestles, and plant tissue was streaked onto modified SMSA media (Aley and Elphinstone 1995; Elphinstone et al. 1996; Norman and Alvarez 1989) and incubated at 28°C for 48 h. Colonies from each plate were picked, mixed in 50 μl of sterile nuclease-free water, denatured at 95°C for 10 min, and centrifuged for 2 min. The pelleted colonies were used as a template to run endpoint PCR with dnaA primers for the RSSC. Both forward and reverse strands of the purified PCR products were sent for Sanger sequencing. The obtained sequences were edited manually, consensus sequences were generated and blasted against the NCBI GenBank database to confirm the identity of both strains. After the identity confirmation of the strains (PL208 and PL209), DNA was extracted from the infected plant material using the DNeasy Plant Mini Kit (Qiagen). DNA concentration was determined using a NanoDrop Spectrophotometer. Extracted DNA was used as a template to validate the specificity of the multiplex assay. A total of 10 naturally infected ironwood plant samples DNA were also used as a template to test the multiplex PCR assay.

For artificially infected samples, 6-week-old tomato seedlings, grown under greenhouse conditions, were inoculated with eight Rpseu strains (GMI1000, 19-127-H1/PL237, 19156/PL214, 19170/PL225, A5664, A5711, A5712, A5715), five Rsol strains (A5345, A3908, A3913, A4126, A5685), and six Rsyz strains (A3533, A4609, A4611, A5515, A5516, A5719). The plants were wounded at the lower part of the stem with sterile scalpels, and a loopful of live bacterial cells from a 48 h grown culture was applied and sealed with water-soaked cotton and parafilm to create a favorable environment for the disease development; control plants were inoculated with sterile water. In addition, Rpseu strains 19-200/PL221 and 19-228/PL220 isolated from ironwood were inoculated into six 6-week-old ironwood (Casuarina equisetifolia) plants grown in the greenhouse following the procedure described above. DNA was extracted from infected and healthy (control) plants using Qiagen DNeasy Plant Mini Kit following the manufacturer's instructions. This DNA was used as a template to analyze the specificity of the multiplex PCR assay using the PCR conditions and components described above. Representative DNA from each target group and healthy control were used as positive and negative controls, respectively, along with a water only non-template control (NTC) in each run.

Results

Target selection, primer design, and in silico validation

We used multiple in silico comparative genomic approaches to select a highly conserved unique target region for each group. The whole-genome-based comparative genomic analysis of 20 complete RSSC representative whole genomes and the related species (R. pickettii, R. mannitolilytica) identified a unique conserved glycosyl transferase gene region for the RSSC strains tested. The gene was independently identified as a core orthologous RSSC gene by the OrthoMCL software. The selected region was present in an additional 100 RSSC genomes and absent from closely related bacterial genera retrieved from the NCBI GenBank database. Using the same strategies mentioned above, we found a conserved Rpseu target in the intergenic region between Flp family type IV b pilin and a TPM domain-containing protein. A DUF692 domain-containing protein was specifically conserved in the Rsol clade. A partial ankyrin repeat domain-containing protein shared high homology among the Rsyz strains. For the R3bv2 strains, we identified a specifically conserved thymidylate synthase. The relative position of the target gene region in the whole genomes of representative RSSC strains from all genomospecies is presented in Figure 1. When the primer sequences were BLASTn-searched against the NCBI GenBank database, they showed 100% nucleotide identity and 100% query coverage for the strains in the respective target groups. The primer pairs did not match the strains from other target groups and other plant-pathogenic bacterial genera as per the in silico analysis using BLASTn and GENEIOUS Prime v 2020.0.4.

Identity confirmation of the strains using partial dnaA gene-based sequencing

The partial dnaA gene region is commonly used to confirm the identity of plant pathogenic bacteria. Sanger sequencing using both sense and antisense strands of 16s rRNA and dnaA gene regions confirmed the identity of the RSSC and other closely related strains used for the validation of the assay (Table 1). The manually edited ∼875- and 843-bp sequenced region of the 16s rRNA and dnaA genes, respectively, were used as an input for the NCBI GenBank database BLASTn analysis. All the amplicons of the strains used in the study gave 98 to 100% identity and 100% query coverage with the expected reference strains. A partial dnaA gene-based maximum likelihood phylogenetic tree was generated with our strains and the dnaA sequence of the representative RSSC strains: (Rsol: UY031, UW551, UA1609, UA1591, UW163, IBSBF 1900, Grenada 9-1, CFBP 3858), (Rsyz: PSI07, SL2312, SL3175, T98, SL3022, SL2064), and (Rpseu GMI1000, UW386, CMR15, FQY_4, EP1, YC45). The dnaA gene tree differentiated the RSSC strains into four phylotypes and three genomospecies, consistent with the egl gene trees that are typical in RSSC diversity studies. The NCBI accession number of the deposited sequences and the other strain characteristics are presented in Supplementary Table S1 and Table 1.

Multiplex PCR assay

The primer sets for five different target groups were designed with enough difference in amplicon size to visualize clear separation of amplicons via gel electrophoresis. The annealing temperature, number of cycles, and concentrations of reagents were all optimized for the multiplex PCR assay (data not shown). A gradient PCR for annealing temperature from 54 to 60°C gave the best amplification results at 59°C. Likewise, the best results were obtained with 3 µl (25 mM) of MgCl₂ per reaction, 1 µl of dNTPs (2.5 mM) per reaction, and 35 PCR cycles. The optimum conditions for the multiplex PCR with best amplification results were as follows: initial denaturation at 94°C for 4 min followed by 35 cycles at 94°C for 30 s, 59°C for 60 s, 72°C for 45 s, and a final extension of 72°C for 3 min. The best results with the Qiagen Multiplex PCR kit were obtained under the following conditions: 95°C for 15 min, 94°C for 30 s, 59°C for 60 s, 72°C for 45 s, and a final extension at 72°C for 3 min.

In vitro specificity validation

To validate the specificity and broad-range detection ability of all primer sets used in the assay, genomic DNA from 113 RSSC strains (56 Rpseu strains; 47 Rsol strains, including 21 R3bv2 strains; and 10 Rsyz strains, including 2 from R. syzygii subsp. syzygii, 1 from R. syzygii subsp. indonesiensis, and 7 from R. syzygii subsp. celebesensis) were tested with both single and multiplex assays. All except two RSSC strains had positive amplification results from their respective primer sets with no false positives or false negatives (Fig. 2; Supplementary Table S1). The expected amplicon sizes of 162, 251, 478, 664, and 954 bp (with flap sequences) were obtained with the RSSC (plant-pathogenic Ralstonia), Rpseu, Rsol, Rsyz, and Select Agent R3bv2, respectively (Fig. 2).

Download as PowerPoint

The exclusivity panel also included 24 strains from closely related bacterial species or plant-pathogenic bacteria infecting diverse hosts. No amplification was observed for any of the exclusivity panel strains (Fig. 3; Supplementary Table S1). When the primer sets were challenged against all the target DNA in the single reaction, no cross-amplification was observed. Likewise, when a single target DNA was challenged against all the primers in the same reaction, there was no nonspecific amplification.

Download as PowerPoint

To validate that the assay worked with PCR reaction mixes other than Promega GoTaq Green, the primers were tested in the Qiagen Multiplex PCR kit against DNA from representative strains of each of the four target groups. All tested strains had positive amplification with their respective primer sets, and no cross-amplification was observed when a single DNA sample was challenged against all primer sets in a single reaction. No amplification was observed with the DNA of six representative strains from closely related bacterial species in the exclusivity panel. The primer pairs showed no nonspecific results with the Qiagen Multiplex Kit.

Determination of the limit of detection

The detection limit of the multiplex PCR assay was determined using 10-fold serial dilutions of genomic DNA from 10 ng to 1 fg. Genomic DNA from Rpseu A5715, Rsol A3903, Rsyz A5720, and R3bv2 A3447 were used as templates for the corresponding primer sets. When all DNA templates were added with all primer sets in a single reaction, the limits of detection were 100 pg (Rsol, Rsyz, and R3bv2 DNA) or 10 pg (Rpseu) using the GoTaq Green-based multiplex assay (Fig. 4A). No change in the limit of detection was observed when a single DNA sample was challenged against all primer sets in a single reaction (Fig. 5A to D). Similarly, no change in the limit of detection was observed when healthy potato tuber DNA was added to the reaction containing 10-fold serially diluted DNA (Fig. 4B).

Download as PowerPoint

Download as PowerPoint

The limit of detection was also determined using the Qiagen Multiplex Kit by adding a single target DNA along with all primer sets in a single reaction. The Qiagen Multiplex Kit was more sensitive and had a detection limit of 1 pg for each of Rpseu, Rsol, Rsyz, and R3bv2 DNA (Fig. 5E to H).

Validation of the assay with naturally and artificially infected plant materials

Specificity of the multiplex assay was also assessed using DNA isolated from two naturally infected tomato plants with visible wilting and ooze symptoms. The two naturally infected tomato plants were confirmed as R. pseudosolanacearum using partial dnaA gene-based phylogenetic analysis and NCBI BLASTn web interface platform. The assay was also validated with 10 naturally infected ironwood plants that were confirmed as R. pseudosolanacearum using partial dnaA gene-based phylogenetic analysis and NCBI BLASTn analysis. All the infected plant DNA samples yielded bands corresponding to the RSSC- and Rpseu-specific primers sets. Bacteria isolated from the infected tomato were confirmed to be Rpseu using partial dnaA gene sequence. The assay performed equally well with DNA extracted from tomato, ironwood, and potato plants artificially inoculated with 10 representative RSSC strains from different genomospecies. Artificially infected symptomatic plants inoculated with a total of 21 strains were positively and accurately amplified by the primer set specific for the respective genomospecies group (Fig. 6; Supplementary Fig. S1). No false-positive or false-negative results were observed. No amplification was observed from healthy plant DNA control or NTC for the multiplex PCR reaction. Thus, the fiveplex assay demonstrated high accuracy and specificity for the detection and differentiation of RSSC strains from both naturally and artificially infected plant DNA.

Download as PowerPoint

Discussion

The RSSC is a well-studied group of phytopathogenic bacteria because of its potential economic consequences in high-value crops planted worldwide. The existence of a high-consequence select agent group, worldwide distribution, and high degree of genetic diversity necessitates accurate detection and differentiation of genomospecies within the RSSC complex. Because related species in the RSSC can infect the same hosts, a robust user-friendly diagnostic tool that can identify and differentiate the major groups is urgently needed for quarantine decisions and further characterization.

Currently available serological, multiplex PCR, and endpoint PCR-based detection tools for the RSSC did not differentiate the genomospecies groups within the species complex (Stulberg et al. 2015; Fegan et al. 1998, Ji et al. 2007). To our knowledge, no detection tools were available that differentiated the three genomospecies and the strictly regulated R3bv2 Select Agent strains in a single reaction. In this study, we developed a cost-effective endpoint multiplex PCR assay based on GoTaq Green Master Mix for convenient and simultaneous detection of three genomospecies and the Select Agent strains belonging to R3bv2. The assay was validated with 113 strains from different hosts and geographic origins. None of the outgroups yielded bands, and the amplification of target bands was specific across two PCR kits. The developed PCR assay also accurately detected the pathogen from naturally and artificially infected plant material with no cross-amplification among the groups.

Robust primer design is a crucial consideration in diagnostic assays (Arif et al. 2021). With the decreased cost of whole-genome sequencing and the availability of more sequencing resources in the public database, it is important to encompass the diversity of strains while selecting the region. We analyzed whole-genome sequences of Ralstonia strains using multiple comparative genomics software and web interface platforms to ensure a unique target region was selected for the five different groups. Multiple in silico analyses suggested that the five selected target regions are unique and highly conserved among the available genome sequences. Several published RSSC diagnostic assays target genes from a potential mobile element or phage related sequence in the R3bv2 lineage (Guidot et al. 2009; Kubota et al. 2011) and the Banana Blood Disease lineage (Rincón-Flórez et al. 2022). The instability and mobility of these elements potentially reduce the stability and specificity of the diagnostic assay. Thus, the selection of a unique and conserved target region is a critical step in the development of a reliable, robust, and specific diagnostic tool (Arif et al. 2021; Arizala et al. 2022; Dobhal et al. 2020).

Primer optimization is another important step to develop an efficient diagnostic assay (Arif and Ochoa-Corona 2013). The use of AT-rich flap sequences at the 5′ end has improved the primer thermodynamics and reduced secondary structure formation (Arif and Ochoa-Corona 2013). Also, addition of these flaps increases the sensitivity and reaction efficiency (Larrea-Sarmiento et al. 2019). The uniformity in the GC content and melting temperature by adding about 10 to 12 AT-rich bases to the primer increases the PCR amplicon intensity and equilibrate competition among primer sets in a multiplex PCR reaction (Arif and Ochoa-Corona 2013; Arizala et al. 2022; Dobhal et al. 2020; Larrea-Sarmiento et al. 2019; Ramachandran et al. 2021).

The primer set RsolP2WF2/R2 amplified all 47 Rsol strains. Surprisingly, one strain of Cupriavidus necator (a sister genus of Ralstonia) yielded a band with the Rsol primer set. However, the strain did not amplify with the RSSC primer set. Despite being related to the RSSC group, no species of C. necator (formerly Ralstonia eutropha) is known as a plant pathogen. The Rsol primer set targets a gene encoding a DUF692 domain-containing protein. DUF692-containing proteins are within the AP2Ec superfamily that contains enzymes that target phosphodiester bonds in DNA (endonucleases with a role in DNA repair) or phosphorylated sugars (isomerases) (Lu et al. 2020). It is possible that both the C. necator strain and Rsol lineage horizontally acquired a similar DUF692-containing gene.

Of the 56 Rpseu strains tested, 54 yielded the expected bands with the RpseuWF5/R5 primer set. Two Rpseu strains, A5713 (Reunion Island) and A5714 (Zimbabwe), did not amplify with the primer set RpseuWF5/R5. When the new primers were designed to amplify a larger fragment of the same region, A5713 and A5714 DNA still did not yield a band, suggesting that these strains lack genes encoding the Flp family type IV b pilin, the TPM domain-containing protein, or both. When a new primer set was designed for a different Rpseu-specific target, the product amplified from A5713 and A5714 DNA but failed to amplify a band in A5711 and A5715 from Burkina Faso and Cameroon, respectively (data not shown). Nevertheless, all four African strains (A5711, A5713, A5714, A5715) yielded the expected band with the RSSC primer set. The results may be due to the high diversity among the phylotype III African strains of the RSSC. Only a few whole-genome sequences are available for the African strains, but preliminary analysis indicates that phylotype III is significantly more diverse than the phylotype I Rpseu strains (Paudel et al. 2020; Sharma et al. 2022). Although the specific primers designed for the Rpseu target group in the study may not encompass the remarkable diversity among the phylotype III strains, the RSSC-specific primers along with partial dnaA- or egl-based phylogenetic analysis will be helpful in species level identification of the strains. In the future, it would be interesting to analyze the complete genomes of more phylotype III African strains.

An important goal of the RSSC-Lineage Multiplex PCR is to facilitate routine diagnostic work and colony identification. Diagnostic assays are more likely to be adopted by research and diagnostic laboratories if they are cost-effective, use accessible reagents and techniques, and work effectively on a diversity of sample types (Arif et al. 2014). Well-designed routine endpoint PCR assays meet these requirements. The RSSC-Lineage Multiplex PCR assay identified strains in infected plant samples from a farmer's field and correctly identified RSSC strains in artificially inoculated plants, including seedlings of woody host ironwood (Casuarina equisetifolia) grown in the greenhouse. We tested the RSSC-Lineage Multiplex PCR against multiple PCR reaction mixes and showed that although it works better in the economical GoTaq Green PCR Mix, the assay also worked in the Qiagen Multiplex Kit. Unlike the GoTaq Green-based PCR, the Qiagen Multiplex Kit did not amplify all five targets simultaneously in the same reaction. This should not present a problem for detection of the RSSC from the infected plant parts, as to our knowledge, no two genomospecies of Ralstonia have ever been isolated from the same host plant.

The limit of detection (lowest concentration of the template that can be accurately detected) is an important characteristic of a diagnostic tool (Arif et al. 2021; Armbruster and Pry 2008). In this assay, no differences in detection limits were observed when either pure genomic DNA or DNA spiked with the host was used. No noticeable differences were observed in the detection limits when comparing the GoTaq Green-based single and multiplex PCR. Likewise, we found no change in the limit of detection when either a single DNA template or all five DNA targets were used in the same multiplex reaction with GoTaq green-based mix (100 pg each for Rsol, Rsyz, and R3bv2; 10 pg for Rpseu). However, using the Qiagen Multiplex PCR kit increased the detection limit by 10 to 100 fold when a single DNA target was tested against all primers, which may be due to the use of a more effective polymerase, PCR buffers, or the Q-solution in the kit. This kit is specifically designed for a multiplex assay, but it will increase the cost of detection per sample.

The developed assay has potential uses in characterizing bacterial cultures and in routine diagnostics and screening of plant materials. Our data demonstrate that it can accurately identify members of the RSCC in a timely and cost-effective way.