Highly specific and super-sensitive Dot-ELISA and colloidal gold immunochromatographic strips for the detection of Burkholderia glumae and Burkholderia plantarii of Rice bacterial panicle blight

Abstract

Burkholderia glumae (B. glumae) and Burkholderia plantarii (B. plantarii) are primary causal agents of rice bacterial panicle blight (RBPB) and cause substantial yield losses in rice worldwide. Given their seed-borne transmission characteristics, quarantine status and destructive hazards, rapid, super-sensitive, highly specific on-site detection technologies are urgently needed. Here, using B. glumae Os48 and B. plantarii ZJ171 as immunogens, we prepared two highly specific and ultra-sensitive monoclonal antibodies (mAbs) against B. glumae (4A7 and 8C5) and two highly specific and ultra-sensitive mAbs against B. plantarii (12B5 and 14B3). We then developed dot-enzyme-linked immunosorbent assays (Dot-ELISA) and colloidal gold immunochromatographic strip (CGICS) assays for detecting B. glumae and B. plantarii with the prepared mAbs as the detection antibodies. These developed mAb-based serological techniques enabled the rapid, broad-spectrum, and specific detection of B. glumae and B. plantarii, respectively, and showed no cross-reaction with other control plant bacteria included in the analysis. Moreover, the detection limits of the Dot-ELISAs for B. glumae or B. plantarii were up to 1.96 × 10⁴ colony-forming units (CFU)/mL, and CGICSs could detect B. glumae and B. plantarii at concentrations as low as 9.78 × 10³ CFU/mL. Surprisingly, these two serological techniques are 2 ~ 8 times more sensitive than conventional polymerase chain reaction (PCR). Collectively, the two newly developed serological techniques in this work provide two simple, rapid, broad-spectrum, highly specific, ultra-sensitive and on-site means to detect B. glumae and B. plantarii in rice grains and leaves, thereby offering reliable and practical tools for the detection and quarantine of these two RBPB pathogens.

Supplementary Information

The online version contains supplementary material available at 10.1007/s44297-026-00067-6.

Introduction

At present, rice bacterial panicle blight (RBPB) is one of the most destructive rice bacterial diseases caused by the Gram-negative bacterium Burkholderia glumae (B. glumae), Burkholderia plantarii (B. plantarii) or Burkholderia gladioli (B. gladioli) and seriously threatens global rice production [1, 2]. ‌RBPB was first documented in Japan in the 1950 s [3]. In recent years, this disease has become prevalent in rice-growing regions across Africa, Asia, North America, and South America, emerging as one of the most serious rice bacterial diseases [4]. RBPB often causes yield losses of approximately 15% [5]. However, a severe outbreak in Vietnam in 1992 demonstrated its devastating potential, with recorded losses reaching as high as 75% [6]. The pathogen of this disease is phylogenetically classified as a plant-pathogenic member of the Burkholderia genus. Under the culture conditions of a semi-specific S-PG medium, Yuan isolated and identified 402 Burkholderia bacterial strains from diseased rice plants exhibiting panicle blight symptoms in the United States. B. glumae, B. gladioli, B. multivorans, and B. plantarii accounted for 90% of these isolates, highlighting the significant role of B. glumae and B. plantarii in this rice bacterial disease [7]. Among them, B. glumae is the dominant pathogen responsible for RBPB [8].

The typical symptoms of RBPB include browning of the leaf sheath of the sword leaf and the leaf tongue, death of the panicle, and a clear demarcation between diseased and healthy grains. The upper part of infected grains is water-soaked, grayish-white, or yellowish-brown, whereas the lower part remains asymptomatic and normal. The diseased grains are empty or not full, with an increase in shriveled grains and incomplete glume closure [9]. In addition, this pathogen can cause seedling rot or even death in the rice seedling stage. This latent infection explains why Uematsu et al. termed the disease “bacterial seedling rot of rice” [10]. Surviving seedlings may harbor the pathogen in a dormant state until the later growth stages (booting stage to heading stage).

The rice-growing season, with its characteristic high temperature and humidity, provides a favorable environment for the outbreak of RBPB. For example, the optimal growth temperature for B. glumae is between 30 °C and 35 °C. Consequently, this disease is more prevalent in tropical and subtropical countries and during growing seasons with above-average temperatures. In addition, the coinfection of B. glumae and B. gladioli in infected samples poses additional challenges for controlling this disease [11].

The development of farmer-friendly, field-deployable diagnostic tools is critical for timely and effective management of this important rice disease, as they enable early detection and timely intervention. Symptom-based diagnosis of RBPB is inaccurate and untimely for the prevention and control of RBPB, as other pathogens or abiotic stresses can also cause similar symptoms [12]. Conventional diagnosis of RBPB mainly involves pathogen isolation and cultivation, pathogenicity analysis, Biolog microbial identification, and fatty acid profiling. Current molecular diagnostic approaches for RBPB employ various polymerase chain reaction (PCR)-based techniques, including conventional and duplex PCR, immunocapture PCR and real-time fluorescence quantitative PCR. These methods target genomic loci such as the internal transcribed spacer (ITS) region between 16S rRNA and 23S rRNA genes [10, 13] and the gyrB gene [14], as well as specific protein-coding genes [15, 16]. Nevertheless, these current diagnostic techniques mentioned above are labor-intensive and require professional technicians and instruments. Therefore, establishing sensitive, specific, efficient, and practical rapid detection techniques is crucial for curbing the spread of RBPB and mitigating its impact. Compared with molecular biological diagnostics, serological diagnostic techniques offer the advantages of simplicity, speed, low cost, and practicality [17–19]. Advances in technology have enabled the commercial-scale production of monoclonal antibodies (mAbs), paving the way for the development of mAb-based testing tools that are increasingly applied in clinical medicine [20], animal science [21], pesticide science [22], and plant pathogen detection and quarantine [23].

Both enzyme-linked immunosorbent assay (ELISA) and colloidal gold immunochromatographic strip (CGICS) are widely used serological techniques for detecting plant pathogens that are based on the principle of specific antigen–antibody binding [24, 25]. ELISA enables sensitive, laboratory-level quantitative detection by employing enzymatic amplification. Conversely, CGICS allows for one-step, on-site qualitative detection based on the visible color band developed by colloidal gold-labeled antibodies [17]. However, to date, serological techniques for the detection of RBPB pathogens remain unavailable.

Effective serological techniques depend on the specificity and sensitivity of the detection antibodies [19, 26]. To develop visual, simple, and rapid serological detection techniques for B. glumae and B. plantarii of RBPB, in this study, we successfully prepared a pair of specific monoclonal antibodies (mAbs) against B. glumae and a pair of specific mAbs against B. plantarii. Based on these prepared mAbs, two serological techniques, Dot-ELISA and CGICS, were developed for the detection of B. glumae and B. plantarii. These two serological assays can accurately and quickly detect B. glumae and B. plantarii, respectively. We believe that these two established serological techniques will be particularly useful for large-scale epidemiological studies of RBPB and for quarantine inspection aimed at preventing the local and international spread of B. glumae and B. plantarii.

Material and Methods

Bacterial strains, bacterium culture and preparation

Bacterial strains (Table 1) were initially streaked on nutrient agar medium comprising 10 g/L tryptone, 3 g/L beef extract, 2.5 g/L glucose, 5 g/L NaCl, and 15 g/L agar (pH 7.0) and incubated at 30 °C for 2 days. After polymerase chain reaction (PCR) verification, a single positive colony was transferred to liquid nutrient medium without agar and cultured at 30 °C for 24 h in a shaking incubator at 200 rpm. The cultures (1 ml per sample) were centrifuged at 13,500 × g for 5 min. The pellets were resuspended in 1 ml of sterile water. The bacterial concentration in each suspension was quantified as colony-forming units (CFU) per milliliter using the standard plate count method, as described previously [24]. The bacterial strains used in this study are detailed in Table 1. RBPB pathogen-infected and uninfected rice plants were collected from rice fields in Zhejiang and Yunnan provinces and Chongqing Municipality in China during the 2025 rice-growing season.

Table 1.

Bacterial strains used in this work

Bacterial strains	Hosts	Origins	Provided organizations
Burkhloderia glumae Os1	Rice	China	Institute of Biotechnology, Zhejiang University
B. glumae Os2	Rice	China	Institute of Biotechnology, Zhejiang University
B. glumae Os48	Rice	China	Institute of Biotechnology, Zhejiang University
B. glumae XY	Rice	China	Institute of Biotechnology, Zhejiang University
B. glumae ZJ178	Rice	China	Institute of Biotechnology, Zhejiang University
B. plantarii ZJ171	Rice	China	Institute of Pesticide and Environmental Toxicology, Zhejiang University
B. plantarii B-23	Rice	China	Institute of Pesticide and Environmental Toxicology, Zhejiang University
B. plantarii B-24	Rice	China	Institute of Pesticide and Environmental Toxicology, Zhejiang University
B. gladioli Os6	Rice	China	Institute of Biotechnology, Zhejiang University
B. gladioli ZJ164	Rice	China	Institute of Pesticide and Environmental Toxicology, Zhejiang University
B. vietnamiensis Os13	Rice	China	Institute of Biotechnology, Zhejiang University
B. ambifaria Os40	Rice	China	Institute of Biotechnology, Zhejiang University
B. cenocepacia M292	Maize	China	Institute of Biotechnology, Zhejiang University
B. pyrrocinia M318	Maize	China	Institute of Biotechnology, Zhejiang University
B. cepacian LMG1222	onions	Belgium	Belgian Co-ordinated Collections of Micro-organisms
Xanthomonas oryzae pv. oryzae PXO86	Rice	Philippines	Shanghai Entry-Exit Inspection and Quarantine Bureau
X. oryzae pv. oryzicola RS105	Rice	China	Shanghai Entry-Exit Inspection and Quarantine Bureau
Acidovorax oryzae CGMCC 1.1728	Rice	Japan	China General Microbiological Culture Collection Center

Production and characterization of murine monoclonal antibodies (mAbs) against B. glumae and B. plantarii

Bacterial strains B. glumae Os48 and B. plantarii ZJ171 were cultured in liquid nutrient medium for 24 h and then centrifuged at 13,500 × g for 5 min to obtain the precipitate. The precipitate was resuspended in a physiological saline solution containing 0.5% formaldehyde for bacterial inactivation, as described previously [27]. The inactivated bacteria were washed three times with physiological saline to remove formaldehyde. The obtained bacterial solution was emulsified with an equal volume of Freund's complete adjuvant or Freund's incomplete adjuvant (Sigma-Aldrich, St. Louis, Missouri, USA). Emulsified bacteria (10⁶ CFU per mouse) were used as antigens to immunize BALB/c mice, with an interval of 3 weeks. For the third booster immunization, the bacteria resuspended in physiological saline solution (2 × 10⁶ CFU per mouse) were intraperitoneally injected. Three days later, spleen lymphocytes were isolated from the mice with the highest serum titers and fused with Sp2/0 myeloma cells using a 50% polyethylene glycol (PEG) solution (Sigma-Aldrich, MW = 3500) as previously described [28]. Hybridoma cell lines secreting anti-B. glumae mAbs and anti-B. plantarii antibodies were screened by indirect ELISA, and then cell clones were obtained by the limited dilution method as described by Li et al. [29]. The obtained hybridoma cells were intraperitoneally injected into BALB/c mice to produce ascitic fluid with mAb [30]. The titer of the ascitic fluid containing mAb was determined by indirect ELISA using B. glumae and B. plantarii as the coating antigen. IgG of mAb was purified from ascites fluid using the saturated ammonium sulfate precipitation method, as described previously [28]. According to the manufacturer's protocol, the classes and subclasses of mAbs were identified using a mouse mAb isotyping kit (Biodragon Company, Suzhou, China).

Establishment of Dot-ELISAs for detecting B. glumae and B. plantarii

A Dot-ELISA was developed to detect B. glumae or B. plantarii, following a previously reported assay [30] with minor modifications. Briefly, each rice grain sample (approximately 100 mg) was ground into powder in liquid nitrogen and then homogenized in 3 mL of 0.01 M phosphate-buffered saline (PBS) at a 1:30 (w/v) ratio to prepare a homogenate. The diluted bacterial suspension and rice grain homogenate (2 μL per sample) were loaded separately onto nitrocellulose membranes (GE Healthcare, Bucks, UK). The dried membranes at room temperature (RT) were blocked in PBS containing 0.05% Tween-20 (PBST) and 5% skim milk for 30 min, followed by incubation with anti-B. plantarii or anti-B. glumae mAb diluent at 37 °C for 1 h. After three washes with PBST, the membranes were incubated in a 1:8000 (v/v) dilution of alkaline phosphatase (AP)-conjugated goat anti-mouse IgG (Sigma-Aldrich) for 1 h. Following three additional washes with PBST, the membranes were incubated in nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate (NBT/BCIP) substrate solution at RT for 10~20 min, and then color development was observed. The appearance of a purple dot on the nitrocellulose membrane was interpreted as a positive result, whereas no color change was interpreted as a negative result.

Production of colloidal gold nanoparticle (CGNP)‑labeled mAb

Colloidal gold nanoparticles (CGNPs) with a diameter of 30 nm were synthesized by reducing HAuCl₄·3H₂O with sodium citrate as described previously [25, 31]. Briefly, 1 mL of 1% HAuCl₄·3H₂O solution was added to 100 mL of distilled water and heated to boiling under vigorous stirring. Then, 3 mL of 1% sodium citrate was immediately added to the mixture, and the mixture was boiled for 10 min until the solution turned red. After stirring for an additional 15 min, CGNPs with a diameter of 30 nm were successfully prepared for subsequent conjugation with mAbs. For mAb labeling with CGNPs, the pH of the gold nanoparticle solution was adjusted with 0.2 M K₂CO₃, followed by the dropwise addition of 1.0 mg of purified mAb into 100 mL CGNP solution under constant stirring. The mixture was then gently stirred at room temperature (RT) for 30 min. Subsequently, 5 mL of 0.01 M borate buffer containing 10% BSA was introduced for the final blocking step, followed by another 30 min of gentle stirring at RT. The mixture was centrifuged at 20,000 × g for 15 min at 4 °C. The resulting CGNP-labeled mAb precipitate was resuspended in 10 mL of 0.02 M PBS (pH 7.4) containing 0.02% NaN₃, 2% BSA and 3% sucrose and stored at 4 °C until use [32].

Establishment of a colloidal gold immunochromatographic strip (CGICS) for B. glumae or B. plantarii detection

The CGICS for detecting B. glumae or B. plantarii was developed as previously reported [28]. The strip consists sequentially of a sample pad, conjugation pad, and nitrocellulose membrane (NC) (Sartorius AG. Millipore, Billerica, MA, USA), and an absorbent pad. Briefly, the glass fiber sample pad was soaked in 0.01 M PBS (pH 7.4) containing 1% BSA for 15 min to minimize nonspecific binding, followed by drying at 37 °C for 5~7 h. Using the Bio-Dot XYZ-3000 dispenser (BioDot, CA, USA), capture mAb and goat anti-mouse secondary antibody were separately dispensed onto the test (T) line and control (C) line of the nitrocellulose membrane. Subsequently, all components were assembled in sequence onto an adhesive polyvinyl chloride (PVC) backing card with a 2 mm overlap at each junction. Finally, the assembled plate was cut longitudinally using a guillotine cutter to produce 3 mm-wide strips [25].

The test samples included positive bacterial suspensions, control bacterial suspensions, and homogenates from infected or uninfected rice grains (0.1 g rice grain was ground and suspended in 1 mL of 0.01 M PBS, pH 7.4). The prepared samples were then added dropwise into the sample pad. Following a 5~10 min reaction, the results were interpreted as follows: the development of two red lines at both the T and C positions indicated a positive result; a single red line at the C position indicated a negative result; and the absence of a red line at the C position rendered the test invalid.

PCR for monitoring B. glumae and B. plantarii

To evaluate the accuracy and effectiveness of Dot-ELISA and CGICS developed in this study, 14 randomly field-collected rice samples were tested for the presence of RBPB pathogens employing these two serological assays. Conventional PCR was used to confirm the results of Dot-ELISA and CGICS. Briefly, two specific primer pairs based on 16S-23S rRNA spacer regions of RBPB pathogens were designed and synthesized. B. glumae-forward primer: 5′-ACACGGAACACCTGGGTA-3′ and B. glumae-reverse primer: 5'-TCGCTCTCCCGAAGAGAT-3' (product length is 400 bp), B. plantarii-forward primer: 5'-AGCCAGTCAGAGGATAAGTC-3' and B. plantarii-reverse primer: 5'-CAATTGAGCCGAACATTTAAG-3' (product length is 200 bp) [13]. For rice grains, total DNA was extracted using the Fast Pure Plant DNA Isolation Mini Kit-BOX2 (Vazyme, Nanjing, China) according to the manufacturer's instructions and used as the PCR template. A bacterial suspension of 1 × 10⁷ CFU/mL was serially two-fold diluted, and 1 μL of each diluted bacterial suspension was used as the PCR template. Each PCR system consisted of 1 μL of each primer (10 μmol/L), 1 μL of DNA/bacterial suspension, 10 μL of 2 × Green Taq Mix (Vazyme), and 7 μL of sterile deionized water. The reaction cycles were as follows: initial denaturation at 94 °C for 5 min, 30 cycles of denaturation at 94 °C for 30 s, annealing at 55 °C (for B. glumae primers) or 53 °C (for B. plantarii primers) for 30 s, extension at 72 °C for 10 s, and a final extension at 72 °C for 5 min.

Results

Development and characteristics of mAbs against B. glumae or B. plantarii

Using B. glumae strain Os48 and B. plantarii strain ZJ171 as immunogens, two murine hybridoma cell lines (12B5 and 14B3) secreting anti-B. plantarii mAbs and two murine hybridoma cell lines (4A7 and 8C5) secreting anti-B. glumae mAbs were separately obtained through cell fusion, hybridoma cell screening, antibody detection and cell cloning. Subsequently, the hybridoma cells were intraperitoneally injected into BALB/c mice pretreated with 200 μL of norphytane to induce ascites production. After 7 to 10 days, ascites containing anti-B. glumae or anti-B. plantarii mAbs were collected via abdominal paracentesis using blood lancets. The immunoglobulin types and subclasses of mAbs were identified as IgG1 and қ light chain (Table 2). The indirect ELISA results demonstrated that all four mAbs exhibited a high titer of 10^–7 (Table 2). The concentration of IgG in mAb ascites ranged from 2.59 mg/mL to 3.66 mg/mL (Table 2).

Table 2.

characteristics of the mAbs against Burkholderia. glumae or B. plantarii

mAbs	Isotypes	Ascites titers	IgG yield in ascites (mg/mL)
4A7	IgG1, κ	10−7	3.36
8C5	IgG1, κ	10−7	2.87
12B5	IgG1, κ	10−7	2.59
14B3	IgG1, κ	10−7	3.66

Dot-ELISA for detecting B. glumae or B. plantarii

Four Dot-ELISAs for the detection of B. glumae or B. plantarii were established using mAbs 4A7, 8C5, 12B5, or 14B3 as the detection antibody with a working concentration of 1:5000 (v/v) dilution and AP-conjugated goat anti-mouse IgG as the secondary antibody with a working concentration of 1:8000 (v/v) dilution. As shown in Fig. 1, the two Dot-ELISAs based on mAbs 4A7 and 8C5 specifically and broad-spectrum detected five B. glumae strains (B. glumae Os48, B. glumae Os1, B. glumae Os2, B. glumae XY, and B. glumae ZJ178) with a visual limit of detection (LOD) of 1.96 × 10⁴ CFU/mL, exhibiting no cross-reactivity with eight other RBPB pathogens (B. plantarii ZJ171, B. gladioli Os6, B. gladioli ZJ164, B. vietnamiensis Os13, B. ambifaria Os40, B. cenocepacia M292, B. pyrrocinia M318, and B. cepacian LMG1222) or three control bacterial strains (Xanthomonas oryzae pv. oryzae PXO86, X. oryzae pv. oryzicola RS105, and Acidovorax oryzae CGMCC 1.1728) (Fig. 1a, c). The two Dot-ELISAs based on mAbs 12B5 and 14B3 specifically and broad-spectrum detected three B. plantarii strains (B. plantarii ZJ171, B. plantarii B-23, and B. plantarii B-24) with a visual LOD of 1.96 × 10⁴ CFU/mL, exhibiting no cross-reactivity with ten other RBPB pathogens (B. glumae Os48, B. glumae XY, B. glumae ZJ178, B. gladioli Os6, B. gladioli ZJ164, B. vietnamiensis Os13, B. ambifaria Os40, B. cenocepacia M292, B. pyrrocinia M318, and B. cepacian LMG1222) and three control bacterial strains (X. oryzae pv. oryzae PXO86, X. oryzae pv. oryzicola RS105, and A. oryzae CGMCC 1.1728) (Fig. 1b, d). The specificity analysis results of Dot-ELISAs were consistent with those of traditional PCR (Fig. 1g, h). The sensitivity of PCR was also determined using serial dilutions of bacterial suspensions, and the results revealed that a specific 400 bp DNA band was amplified from B. glumae suspensions diluted from 1 × 10⁷ to 3.91 × 10⁴ CFU/mL (Fig. 1i), and a specific 200 bp DNA band was amplified from B. plantarii suspensions diluted from 1 × 10⁷ to 7.81 × 10⁴ CFU/mL (Fig. 1j). These above findings indicate that the sensitivities of Dot-ELISAs are approximately 2 and 4 times higher than that of PCR.

Fig. 1.

The specificity, broad-spectrum and sensitivity analysis results of the developed Dot-ELISAs and PCR for detecting BurkholderiaB glumae or B. plantarii. a, b The specificity and broad-spectrum analysis results of the two developed Dot-ELISAs for detecting B. glumae or B. plantarii suspension. c, d The sensitivity analysis results of the two developed Dot-ELISAs for detecting B. glumae or B. plantarii suspension. e, f Sensitivity analysis results of the developed Dot-ELISAs for detecting B. glumae- or B. plantarii-infected rice grain homogenates. Homogenates extracted from infected or uninfected rice grains were serially two-fold-diluted, and each dilution (2 µL) was loaded onto NC membranes for Dot-ELISA. g, h Specificity and broad-spectrum analysis results of PCR for detecting B. glumae or B. plantarii suspension. i, j Sensitivity analysis results of PCR for detecting B. glumae or B. plantarii suspension. B. glumae Os48 and B. plantarii ZJ171 suspensions were serially diluted in PBS, and the diluted suspensions were used as PCR templates. The marker was a 1 kb DNA ladder. k, l Sensitivity analysis results of PCR for detecting B. glumae- or B. plantarii-infected rice grain homogenates

Additionally, to determine the detection sensitivity of the developed Dot-ELISAs for B. glumae- or B. plantarii-infected rice grains, homogenates from infected or non-infected rice grains were serially two-fold-diluted in PBS. Then, the presence of the pathogen in each dilution was tested by Dot-ELISAs. The test results showed that the lowest detection limit of the Dot-ELISA was a 1:7680 dilution (w/v, g/mL) for B. glumae- or B. plantarii-infected rice grain homogenates (Fig. 1e, f). The low detection limit of PCR was a 1:3840 dilution (w/v, g/mL) for B. glumae-infected or B. plantarii-infected rice grains (Fig. 1k, l). These findings indicate that the developed Dot-ELISAs were 2 times more sensitive than PCR.

Two point-of-care CGICSs for B. glumae and B. plantarii detection

Preliminary antibody-pairing experiments revealed that CGICS based on the self-paired mAb 4A7 exhibited superior detection performance for B. glumae, and CGICS employing mAb 12B5 as the CGNP-labeled antibody and mAb 14B3 as the capture antibody at the test line demonstrated optimal performance for detecting B. plantarii. The working principle of the test strip is shown in Fig. 2.

Fig. 2.

Schematic diagram of colloidal gold immunochromatographic test strip for detecting Burkholderia plantarii

To achieve the best detection result, we optimized the test strips. The color intensity of the test line of the strip was gradually enhanced with increasing concentrations of the capture antibody (Fig. 3a, b). Furthermore, no obvious difference in test line color intensity was observed when the concentration of the capture antibody (mAb 4A7 or 14B3) was increased from 2.5 to 3.0 mg/mL (Fig. 3a, b). Based on a balance between detection signal strength and antibody consumption, 2.5 mg/mL was selected for the working concentrations of the capture antibodies (mAb 4A7 and 14B3). We then labeled different concentrations of mAb 4A7 or 12B5 with colloidal gold solution and assessed the test performance of the resulting CGNP-labeled antibodies under the working concentration of the capture antibody. The analysis results revealed that the optimal working concentrations, defined by a clear, intense test line, were determined to be 5 µg/mL for mAb 4A7 (B. glumae) and 12.5 µg/mL for mAb 12B5 (B. plantarii) (Fig. 3c, d). Additionally, the pH value also affects the conjugation of CGNPs and antibodies and the stability of CGNPs, and an unsuitable pH value may lead to the aggregation of CGNPs [19]. Thus, a series of CGNP solutions were adjusted to pH values with different concentrations of K₂CO₃ (0.0%−0.3%) to identify the optimal concentration of K₂CO₃. The CGICS test results demonstrated that the CGNP solution adjusted with 0.2% K₂CO₃ solution generated the optimal colorimetric signal intensity in the CGICS (Fig. 3e, f). Finally, using the three optimized parameters above, two CGICSs were developed for the detection of B. glumae and B. plantarii.

Fig. 3.

Optimization of CGICSs for the detection of B. glumae and B. plantarii. CK + represents B. glumae Os48 or B. plantarii ZJ171, and CK- represents the PBS solution without any pathogenic bacteria. a, b The effect of capture antibody concentration on the strip detection signal intensity. c, d The effect of colloidal gold-conjugated antibody concentration on the strip detection signal intensity. e, f The effect of pH value of CGNP solution on the test strip detection signal intensity. In this experiment, the CGNP solution was treated with 0.2%, 0.3%, or 0.4% K₂CO₃ prior to conjugation with mAb

Specificity, broad-spectrum and sensitivity analysis results of the two developed CGICSs

To evaluate the specificity and broad-spectrum detection capability of the two test strips established in this study, the strips were applied to test five different B. glumae strains (Os48, Os1, Os2, XY, and ZJ178), three different B. plantarii strains (ZJ171, B-23, and B-24), and ten non-target control bacterial strains (i.e., B. gladioli Os6, B. gladioli ZJ164, B. vietnamiensis Os13, B. ambifaria Os40, B. cenocepacia M292, B. pyrrocinia M318, B. cepacian LMG1222, X. oryzae pv. oryzae PXO86, X. oryzae pv. oryzicola RS105, and A. oryzae CGMCC 1.1728). The monitoring results demonstrated that the B. glumae-specific test strip successfully detected all five analyzed B. glumae strains and exhibited negative reactions with all 11 non-target control bacteria, including other RBPB pathogens (i.e., B. plantarii ZJ171, B. gladioli Os6, B. gladioli ZJ164, B. vietnamiensis Os13, B. ambifaria Os40, B. cenocepacia M292, B. pyrrocinia M318, B. cepacian LMG1222, X. oryzae pv. oryzae PXO86, X. oryzae pv. oryzicola RS105, and A. oryzae CGMCC 1.1728) (Fig. 4a). The B. plantarii-specific test strip successfully detected all three analyzed B. plantarii strains (B. plantarii ZJ171, B. plantarii B-23, and B. plantarii B-24) and exhibited negative reactions with all 13 non-targetcontrol bacteria, including other RBPB pathogens (i.e., B. glumae Os48, B. glumae XY, B. glumae ZJ178, B. gladioli Os6, B. gladioli ZJ164, B. vietnamiensis Os13, B. ambifaria Os40, B. cenocepacia M292, B. pyrrocinia M318, B. cepacia LMG1222, X. oryzae pv. oryzae PXO86, X. oryzae pv. oryzicola RS105, and A. oryzae CGMCC 1.1728) (Fig. 4b). These findings suggest that the two developed test strips were highly specific and broad-spectrum for B. glumae and B. plantarii detection, respectively.

Fig. 4.

Specificity, broad-spectrum, and sensitivity analysis results of the two CGICS for the detection of Burkholderia glumae and B. plantarii. (a, b) Specificity and broad-spectrum analysis results of the two CGICS for the detection of B. glumae and B. plantarii, respectively. (c, d) Sensitivities of the two developed CGICSs were determined using serial two-fold-diluted B. glumae or B. plantarii bacterial suspensions. e, f Sensitivities of the two developed CGICSs for the detection of B. glumae- or B. plantarii-infected rice grain homogenates. Homogenates were prepared from infected rice grains and subjected to serial two-fold dilutions. Each dilution (100 μL) was loaded onto the sample pad of the CGICS for testing

To determine the sensitivity of the two test strips, bacterial suspensions with an initial concentration of 10⁷ CFU/mL were subjected to serial two-fold dilutions, and each diluted bacterial suspension was monitored using test strips. The results showed that CGICS could detect B. glumae or B. plantarii in bacterial suspensions diluted from 1 × 10⁷ CFU/mL to 9.78 × 10³ CFU/mL (Fig. 4c, 4 d), which indicates that the sensitivities of CGICSs are approximately 4 and 8 times higher than that of PCR.

Additionally, as shown in Fig. 4e and 4f, homogenates from B. glumae- or B. plantarii-infected rice grains were serially two-fold-diluted and tested with the test strips. The test results revealed that the detection limit of the two strips for the infected rice grain homogenates was a dilution of 1:7680 (w/v, g/mL), indicating that the sensitivities of CGICSs are 2 times higher than that of PCR (Fig. 1k and 1 l). The above data further indicate that the developed two CGICSs have ultra-high sensitivity for the detection of B. glumae and B. plantarii.

Detection of B. glumae and B. plantarii in paddy‑collected rice grains and leaves using Dot-ELISA and CGICS, as well as PCR

To further evaluate the accuracy and effectiveness of the Dot-ELISA and CGICS developed in this study for the detection of B. glumae and B. plantarii, we collected 14 rice samples suspected of being infected with RBPB pathogens from rice fields in Yunnan Province, Zhejiang Province and Chongqing Municipality of China in 2025 and tested them using both Dot-ELISA and CGICS. The test results of both serological techniques showed that of the 14 rice samples, nine and five were infected with B. glumae and B. plantarii, respectively, whereas all remaining field samples and uninfected controls yielded negative results (Fig. 5c and d). Additionally, samples 6 and 11 were co-infected with B. glumae and B. plantarii. Furthermore, the detection results of Dot-ELISA and CGICS were consistent with those of PCR analyses (Fig. 5e, f). These findings demonstrate that the newly developed Dot-ELISA and CGICS are highly efficient and accurate for detecting B. glumae and B. plantarii infection and their coinfection in rice grain samples.

Fig. 5.

Detection results of Burkholderia glumae (a, c, e) and B. plantarii (b, d, f) infection in field-collected rice grains using the developed Dot-ELISA, CGICS, and conventional PCR. Numbers 1~14 represent 14 field-collected rice samples. “ + ” represents rice grains infected with B. glumae or B. plantarii, used as positive controls. “-” represents uninfected rice grains, used as negative controls. a, b Detection results of B. glumae and B. plantarii infection in 14 field-collected rice grain samples using the developed Dot-ELISA. c, d Detection results of B. glumae and B. plantarii infection in 14 field-collected rice grain samples using the CGICS. e, f Detection results of B. glumae and B. plantarii infection in 14 field-collected rice gain samples using PCR

Moreover, we collected six rice leaf samples suspected of RBPB infection from fields in Zhejiang Province, China, in 2025. These samples were then tested using the newly developed Dot-ELISA and CGICS, as well as conventional PCR. The results from both serological techniques and PCR were consistent, showing co-infection of one leaf sample by B. glumae and B. plantarii. The other five field leaf samples and the uninfected control all yielded negative results (Fig. S1). These data indicate that both Dot-ELISA and CGICS are effective and accurate for detecting B. glumae and B. plantarii infection and their coinfection in rice leaf samples.

Discussion

Rice bacterial panicle blight (RBPB) is an important bacterial disease that severely constrains rice yield and quality. Its management remains difficult because of the wide variety of causative pathogens, the limited availability of both effective controls and disease-resistant rice varieties. Both B. glumae and B. plantarii are primary pathogens responsible for RBPB [33, 34]. Notably, the pathogenicity of B. glumae extends beyond plants and poses a threat to human health. In 2007, Weinberg et al. reported that B. glumae can cause chronic granulomatous disease in children's lungs and exhibited greater pathogenicity in rice than the common strain [35]. This discovery highlights the cross-species pathogenic potential of B. glumae, which provides new insights for unraveling its molecular mechanisms, tracking transmission routes, and developing control strategies [36]. Therefore, rapid and reliable quarantine tools are urgently required to prevent the large-scale spread of RBPB in China and worldwide.

Current diagnostic techniques for RBPB pathogens mainly include symptom observation, isolation and cultivation of pathogenic bacteria combined with morphological and biochemical analyses, and PCR-based molecular biological detection [33, 37]. Among them, PCR-based molecular biological techniques have become commonly used tools for RBPB pathogen detection [38, 39]. For instance, Luo and colleagues developed a PCR assay using a specific primer pair targeting the gyrB gene of B. glumae, with a sensitivity of 1.0 × 10⁵ CFU/mL [40]. Maeda and colleagues established a multiplex PCR to detect B. plantarii in rice seeds by targeting the DNA gyrase gene gyrB and σ⁷⁰ factor gene rpoD [14]. However, this multiplex PCR exhibited a limited sensitivity of 1.0 × 10⁸ CFU/mL, which is insufficient for detecting low-concentration pathogen samples [41]. Moreover, these PCR methods require thermal cycler and electrophoresis equipment and multiple operation steps, making them cumbersome, costly, and unsuitable for rapid on-site and high-throughput detection at ports or in the field. Therefore, we urgently need a simple and efficient RBPB quarantine technology to precisely detect RBPB and ensure the safe global circulation of sterile rice seeds. In comparison, serological techniques are one of the best detection techniques. Serological techniques have become important tools widely used in plant pathogen detection owing to their significant advantages of simple operation, low cost, and high-throughput [19, 23, 26]. The accuracy of serological techniques depends on the quality of the detection antibodies. Currently, antibodies for plant pathogen detection primarily comprise polyclonal antibodies (pAbs) and mAbs. PAbs are heterogeneous mixtures of antibodies secreted by multiple B-cell clones, targeting different epitopes on the antigen. Consequently, their ability to recognize multiple epitopes can result in cross-reactivity and non-specific signals. In contrast, mAbs are homogeneous antibodies secreted by a single B-cell clone, recognizing a unique epitope on the antigen. This inherent characteristic minimizes cross-reactions and yields cleaner, more specific signals. Therefore, mAbs are often preferred for diagnostic applications due to their higher specificity and more reliable results.

In this study, two super-sensitive and highly specific mAbs against B. glumae or B. plantarii were first prepared using B. glumae Os48 and B. plantarii ZJ171 bacteria as immunogens. The high antigenic homology among Burkholderia species [42] poses a challenge for mAb specificity, as it can lead to cross-reactivity of mAbs between multiple different Burkholderia species. Remarkably, specificity analyses demonstrated that the two mAbs (4A7 and 8C5) developed in this study were highly specific to B. glumae, showing no cross-reactivity with B. gladioli, B. plantarii, or other tested Burkholderia species (Fig. 1a). The two mAbs (12B5 and 14B3) specifically recognize B. plantarii, showing no cross-reactivity with B. gladioli, B. glumae, or other tested Burkholderia species (Fig. 1b). Moreover, the four Dot-ELISAs developed with these four mAbs showed a sensitivity of 1.96 × 10⁴ CFU/mL, which is 2 to 4 times more sensitive than that of conventional PCR (Fig. 1; Table S1).

CGICS is by far the fastest and simplest tool for detecting various pathogens [43]. In this work, we developed two specific test strips for the detection of B. glumae and B. plantarii, achieving detection limits of 9.78 × 10³ CFU/mL, which is 4 or 8 times more sensitive than that of conventional PCR (Fig. 4; Table S1). Therefore, we believe that the two new serological techniques developed in this study offer effective tools for the rapid, on-site detection and quarantine of B. glumae and B. plantarii in rice grains, thereby helping to prevent their further spread.

However, this study still has areas that require further investigation, such as exploring the impact of different storage conditions on the performance of Dot-ELISA and CGCIS, analyzing their effectiveness in detecting coinfected samples with extremely low bacterial loads, and examining the influence of the operator's subjectivity in the visual observation of the test results.

Conclusion

In this study, we successfully prepared four highly specific, broad-spectrum, and ultra-sensitive mAbs (two against B. glumae and two against B. plantarii). Using these mAbs as detection antibodies, we further established Dot-ELISA and CGICS technologies for the specific and ultra-sensitive detection of B. glumae and B. plantarii. The four mAbs and two serological tools developed in this work provide reliable detection tools for the quarantine and control of RBPB and have promising applications in the detection of B. plantarii and B. glumae.

Abbreviations

AP

Alkaline phosphatase

RBPB

Rice bacterial panicle blight

Burkholderia glumae

B. glumae

Burkholderia plantarii

 B. plantarii

CGNP

Colloidal gold nanoparticle

CGICS

Colloidal gold immunochromatographic strip

CFU

Colony-forming unit

CGMCC

China General Microbiological Collection Center

Dot-ELISA

Dot-enzyme-linked immunosorbent assay

LMG

Laboratorium voor Microbiologie

LOD

Limit of detection

mAb

Monoclonal antibody

NC

Nitrocellulose membrane

PAb

Polyclonal antibody

PBS

Phosphate-buffered saline

PCR

Polymerase chain reaction

PEG

Polyethylene glycol

PVC

Polyvinyl chloride

Authors’ contributions

JW, JL and ZA conceived and designed the experiments; JD, WM, CZ and BL performed the experiments; JD and JW analyzed the data and wrote the paper. All the authors have discussed the results of the manuscript.

Funding

This work was funded by the Shanghai Agricultural Science and Technology Innovation Project (No. T2023101), the Hainan Province Science and Technology Special Fund (No. ZDYF2022XDNY240), and the Earmarked Fund for Modern Agro-industry Technology Research System (No. nycytx-001).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

The animal study protocol was approved by the Animal Experimentation Ethics Committee of Zhejiang University, Hangzhou, China (protocol code: ZJU20210215, 30 August 2021).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.