Abstract
American foulbrood (AFB) and European foulbrood (EFB), caused by Paenibacillus larvae and Melissococcus plutonius, respectively, are severe bacterial diseases that significantly affect honey bee health and productivity worldwide. Rapid and accurate diagnosis is essential for effective disease management in apiaries. We developed and validated a multiplex point-of-care (POC) quantitative real-time PCR (qPCR) assay that enables simultaneous and rapid detection of P. larvae and M. plutonius directly in apiaries. Our POC qPCR assay, using the XQ station (Postbio), had diagnostic sensitivity and specificity comparable to laboratory-based qPCR assays. For P. larvae detection, the POC qPCR assay had a sensitivity of 93% (56 of 60 samples) and specificity of 97% (97 of 100 samples). Similarly, for M. plutonius, sensitivity was 95% (59 of 62 samples) and specificity was 97% (97 of 100 samples). Our POC qPCR assay had analytical sensitivity comparable to that of laboratory-based qPCR assays, with a limit of detection of ~102 copies/reaction for both pathogens. Additionally, the assay had high analytical specificity, with no cross-reactivity against other common honey bee pathogens, including deformed wing virus (DWV), black queen cell virus (BQCV), and Vairimorpha ceranae. Our POC qPCR assay consistently had lower cycle quantification values than the laboratory-based qPCR on the same samples, indicating robust detection capability even at low pathogen concentrations. Although not compared to a gold standard, our POC qPCR assay had reliable diagnostic performance. Our multiplex POC qPCR assay enables rapid, accurate pathogen detection in apiaries, thereby enhancing disease management and supporting sustainable, productive apiculture.
Keywords: American foulbrood, Apis mellifera, European foulbrood, point-of-care testing, qPCR
Honey bees (Apis mellifera) are critical to agricultural productivity and ecosystem stability because of their essential role in pollination, directly facilitating the enhancement of the yield and quality of various crops. 23 However, global honey bee populations have been declining significantly in recent decades. For example, in the United States, managed honey bee colonies declined by ~30–40% annually in the decade following 2006, which was largely attributed to colony collapse disorder (CCD). 24 CCD is associated with various factors, including environmental stressors, pesticide exposure, and infectious diseases. 17 Among these infectious diseases, American foulbrood (AFB) and European foulbrood (EFB) are severe bacterial diseases that primarily affect honey bee larvae and cause substantial economic losses in apiculture worldwide.14,16
AFB is caused by Paenibacillus larvae, a gram-positive bacterium that forms highly resistant endospores capable of surviving for prolonged periods in the environment, facilitating rapid transmission within and between colonies. 15 Clinically, larvae infected with P. larvae degrade into a viscous, brown mass, with a distinctive rope-like consistency. 2 EFB is caused by another gram-positive bacterium, Melissococcus plutonius, which is transmitted primarily through the ingestion of contaminated food by larvae. 4 Larvae infected with M. plutonius have delayed growth, yellow-white discoloration, and a loss of their characteristic C-shaped posture. 3 Adult honey bees, although typically subclinical, frequently act as mechanical vectors for P. larvae and M. plutonius by transferring bacterial spores or cells from contaminated colonies to healthy ones during foraging, robbing behaviors, and hive maintenance activities. 18
To date, detection methods for P. larvae and M. plutonius include visual inspection, microscopic examination, bacterial culture, and PCR-based molecular testing. Visual inspection of brood frames can lead to a high rate of missed early or subclinical infections; moreover, the subjective interpretation of clinical signs often requires additional confirmation. 6 Microscopic examination of larval smears is operator-dependent and may lead to the underestimation of low bacterial loads that can be insufficient for reliable detection. 25 Although bacterial culture is widely regarded as the gold standard, 5 this method necessitates specialized media and several days for P. larvae to grow, rendering it impractical for rapid decision making in the field. 19 PCR-based assays, including conventional and quantitative real-time PCR (qPCR), offer high sensitivity and specificity, and are widely used in laboratory assays. 11 However, their reliance on laboratory equipment, skilled personnel, and sample transport are factors that introduce practical limitations for use in the field. 20
In South Korea, AFB and EFB are designated as class III notifiable infectious diseases, necessitating strict regulatory control and underscoring the urgent need for the development and implementation of rapid and reliable field-based detection methods. 21 Loop-mediated isothermal amplification (LAMP) and lateral flow immunoassays (LFIAs) have been introduced as alternative foulbrood detection methods. LAMP can generate results in <30 min without a thermocycler, but it often lacks multiplexing capability and may demand extra steps for readout, increasing complexity. 8 Some LAMP assays also have variable performance in field settings, especially where environmental contaminants can hinder amplification. 1 On the other hand, LFIA offers quick, easy-to-interpret “yes/no” results on-site but trades off some sensitivity and specificity. 7 These tests may yield false-negative results when antigen levels are low (e.g., early infection) or if larval tissues have decomposed, thus degrading the target antigen. 12 Consequently, while LAMP and LFIA are valuable for rapid screening, their constraints highlight the need for a more robust, multiplexed field assay.
In contrast, a point-of-care (POC) qPCR assay combines the analytical power of qPCR with the convenience of a portable, field-deployable format. P. larvae and M. plutonius could be detected simultaneously in a single reaction, which would be a significant advantage over single-target assays. Moreover, cycle quantification (Cq) values allow for pathogen load estimation, which can be useful for monitoring disease progression and treatment efficacy. Furthermore, the POC integrated sample preparation, amplification, and detection system reduces contamination risks and allows non-specialists to perform tests with minimal training.
Given its high transmissibility and regulatory implications, early and accurate detection of AFB is essential to prevent colony loss and further spread. 13 In contrast, EFB is less devastating in its transmission and is commonly managed using antibiotics such as oxytetracycline. 27 However, the widespread prophylactic use of antimicrobials in beekeeping—sometimes administered without laboratory confirmation of EFB—has contributed to the emergence of antimicrobial-resistant (AMR) bacterial strains, ultimately compromising treatment efficacy and necessitating alternative control measures. 22 By enabling rapid and accurate on-site detection of foulbrood agents, a POC qPCR assay will provide a targeted approach for foulbrood management, ensuring that antimicrobials are administered only to colonies with confirmed bacterial infections. This will reduce unnecessary antimicrobial use and mitigate the risk of AMR development while also minimizing antimicrobial residues in honey and other hive products.
We developed and evaluated a multiplex POC qPCR assay for the simultaneous and rapid detection of P. larvae and M. plutonius on-site in apiaries. We selected the XQ station (Postbio) as the device for this assay due to its portability and ease of use, making it ideal for field applications. We assessed the performance of our novel POC assay and compared it with that of a standard laboratory-based qPCR assay, focusing on its sensitivity, specificity, and practical applicability.
Materials and methods
Design and selection of primers and probes for POC qPCR detection of P. larvae and M. plutonius
We used DNAstar software to design primers and probes targeting P. larvae (transposase gene; GenBank CP003355) and M. plutonius (16S ribosomal RNA gene; GenBank NR_043240). TaqMan probe–based assays utilize a hydrolysis-based quenching mechanism, which enables highly sensitive and specific detection of both bacterial pathogens.
For P. larvae–specific sequences, the sense primer was 5′-TACGCT TTTCGATTTCTG-3′, and the anti-sense primer was 5′-GTCTGTACTGAACCAAGTC-3′. The TaqMan probe sequence was 5′-ATCTGCTTCCACTTGTTCACTCACCA-3′; the probe was labeled with the fluorescent reporter dye FAM (6-carboxyfluorescein) at the 5′-end. A 161-bp product was expected when these primers were used.
For M. plutonius–specific sequences, the sense primer was 5′-TGTTGTTAGAGAAGAATAGGGGAA-3′, and the anti-sense primer was 5′-CGTGGCTTTCTGGTTAGA-3′. The TaqMan probe sequence was 5′-AGAGTAACTGTTTTCCTCGTGACGGT-3′; the probe was labeled with the fluorescent reporter dye ROX (6-carboxyfluorescein) at the 5′-end. A 161-bp product was expected when these primers were used.
Samples for analysis
We collected 60 P. larvae–positive and 62 M. plutonius–positive samples and 100 samples that were negative for both agents, confirmed using a validated laboratory-based qPCR assay that had been jointly developed in 2018 by the Animal and Plant Quarantine Agency (APQA; Gimcheon, Republic of Korea)—the official governmental agency responsible for honey bee disease testing in the Republic of Korea—and Postbio, an institution affiliated with one of the authors (D-S Cheon). Following its official patent registration in 2020, this assay was designated as the standard detection protocol for honey bee bacterial diseases by APQA, and has since been used routinely for official testing, subject to annual proficiency assessments. 10
Sample preparation for P. larvae and M. plutonius gene amplification
Each sample, consisting of 2 or 3 adult honey bees and 2 or 3 larvae, was ground in 2.5 mL of 10 mM phosphate buffer for tissue disruption. Coarse homogenates were transferred to microtubes containing 0.1–0.25-mm ceramic beads for fine homogenization (Omni bead ruptor 12; Revvity) at 1,000 rpm for 1 min to disrupt honey bee cells and release bacterial DNA. After brief centrifugation (5 min, 3,000 × g, 4°C), the supernatant was used for nucleic acid purification using silica-coated magnetic beads and a total nucleic acid purification kit (Postbio) according to the manufacturer’s instructions, including proteinase K lysis at 56°C for 10 min, using 10 µL per sample. For POC qPCR, 50 µL of diluent from the tissue homogenates of the samples was directly added to the POC qPCR cassette for the integrated process comprising nucleic acid extraction and subsequent qPCR in the cassette within 60 min.
Laboratory qPCR test for the detection of P. larvae and M. plutonius
An aliquot (140 µL) of each sample homogenate in PBS was used to extract nucleic acids for qPCR-based P. larvae and M. plutonius gene amplification and identification. The qPCR with P. larvae– and M. plutonius–specific primers, and the fluorescent probe listed in the patent, was used as the reference test. The QuantiNova Probe PCR kit (Qiagen) was used for gene amplification. The qPCR mixture contained 12.5 pmol of each primer (500 nM final concentration), 6.25 pmol of the probe (250 nM final concentration), 12.5 µL of 2× qPCR buffer, and 5 µL of the extracted nucleic acids in a total reaction volume of 25 µL. Pre-denaturation, amplification via PCR, and fluorescence detection were performed (Linegene 9600 Plus system; Bioer) with the following cycling parameters: 95°C for 5 min, followed by 45 cycles of 95°C for 10 s, and 60°C for 30 s for gene amplification. Fluorescence readings were obtained during annealing and extension. Additionally, based on standard operating protocols, a Cq <35 was considered positive based on preliminary tests to determine the limit of detection (LOD) for all pathogens.
POC qPCR test for the detection of P. larvae and M. plutonius
The POC qPCR device (XQ station) automatically runs an integrated test in a single-use microfluidic cassette for the dual detection of P. larvae and M. plutonius (XQ AFB and EFB Dx kit; Postbio). Loading the components and initiating the test required ~5 min and involves 3 main steps: 1) preparing the diluent (1:20) using rough homogenates from honey bee samples, 2) loading a 50-µL diluent sample with 10 µL of silica-coated magnetic beads and 10 µL of processing control into the XQ AFB and EFB Dx kit, and 3) inserting the XQ AFB and EFB Dx kit into the XQ station. The results were interpreted 60 min after the completion of 45 cycles of amplification. A Cq <35 was considered positive, and invalid results were retested.
Determination of LOD
We determined the LOD for P. larvae and M. plutonius by using serial dilutions of P. larvae– and M. plutonius–positive specimens of known concentrations (104–100 copies/reaction), with 8 replicates for each concentration in triplicate. The lowest concentration that achieved a positive detection rate of 95% was considered to be the LOD.
Evaluation of analytical specificity and cross-reactivity of the P. larvae and M. plutonius POC qPCR assays
We chose the 12 honey bee pathogens for the critical analytical specificity testing based on their clinical and epidemiologic relevance, as well as their detection frequency in field samples. These included major viral, fungal, and bacterial pathogens that are commonly encountered in apicultural testing and are known to affect honey bee health globally. 16 The panel included 7 viral pathogens (acute bee paralysis virus [ABPV], black queen cell virus [BQCV], chronic bee paralysis virus [CBPV], deformed wing virus [DWV], Israeli acute paralysis virus [IAPV], Kashmir bee virus [KBV], sacbrood virus [SBV]), 3 fungal pathogens (Ascosphaera apis, Aspergillus flavus, Vairimorpha ceranae), and the 2 bacterial pathogens (P. larvae, M. plutonius). Cross-reactivity was assessed for both the POC qPCR and laboratory qPCR assays. Specifically, the cross-reactivity between the 2 target pathogens (P. larvae, M. plutonius) was assessed by testing the P. larvae–specific assay against M. plutonius–positive samples and vice versa.
Performance evaluation of POC qPCR against laboratory qPCR
We tested 222 samples (60 P. larvae–positive, 62 M. plutonius–positive, and 100 negative for both agents) using POC qPCR and laboratory qPCR assays. Concordance between the 2 methods was assessed using laboratory qPCR as the reference standard. In cases of discordant results between the 2 assays, duplicate retesting was performed. The final interpretation was based on the majority outcome of 3 independent tests to ensure consistency and reliability in result determination.
Results
LOD for P. larvae and M. plutonius
Our POC qPCR detected 24 of 24 (100%) replicates of the samples at 102 copies/reaction; the laboratory qPCR detected 23 of 24 (96%) replicates at the same concentration for P. larvae–positive samples (Table 1). For M. plutonius–positive samples, both the POC qPCR and the laboratory qPCR assays detected 23 of 24 (96%) replicates at 102 copies/reaction (Table 2). Standard curves generated from these dilution series were also used to calculate amplification efficiencies. The efficiencies of the POC qPCR and laboratory qPCR assays were 92–95% for both P. larvae and M. plutonius, with high linearity for all assays (R2 >0.99; Table 3).
Table 1.
Limit of detection determination for Paenibacillus larvae using laboratory quantitative real-time PCR (qPCR) and point-of-care (POC) qPCR assays.
| P. larvae titers, copies/reaction | Laboratory qPCR | POC qPCR |
|---|---|---|
| No. positive (%) | No. positive (%) | |
| 100 | 2/24 (8) | 0/24 (0) |
| 101 | 12/24 (50) | 8/24 (33) |
| 102 | 23/24 (96) | 24/24 (100) |
| 103 | 24/24 (100) | 24/24 (100) |
| 104 | 24/24 (100) | 24/24 (100) |
Boldface values indicate the determined limit of detection, defined as the lowest concentration at which ≥95% of the 24 replicates were positive.
Table 2.
Limit of detection determination for Melissococcus plutonius using laboratory quantitative real-time PCR (qPCR) and point-of-care (POC) qPCR assays.
| M. plutonius titers, copies/reaction | Laboratory qPCR | POC qPCR |
|---|---|---|
| No. positive (%) | No. positive (%) | |
| 100 | 3/24 (13) | 2/24 (8) |
| 101 | 11/24 (46) | 13/24 (54) |
| 102 | 23/24 (96) | 23/24 (96) |
| 103 | 24/24 (100) | 24/24 (100) |
| 104 | 24/24 (100) | 24/24(100) |
Boldface values indicate the determined limit of detection, defined as the lowest concentration at which ≥95% of the 24 replicates were positive.
Table 3.
Amplification efficiencies, linearity, and raw Cq values derived from standard curves for laboratory quantitative real-time PCR (qPCR) and point-of-care (POC) qPCR assays targeting Paenibacillus larvae and Melissococcus plutonius.
| Pathogen/Assay | Titers, copies/reaction | Cq | Slope | Correlation coefficient, R2 | Efficiency, % |
|---|---|---|---|---|---|
| P. larvae | |||||
| Laboratory qPCR | 105 | 20.3 | 3.66 | 0.999 | 92.5 |
| 104 | 24.1 | ||||
| 103 | 27.8 | ||||
| 102 | 31.3 | ||||
| 101 | 35.0 | ||||
| POC qPCR | 105 | 19.9 | 3.83 | 0.999 | 94.7 |
| 104 | 23.7 | ||||
| 103 | 27.5 | ||||
| 102 | 31.0 | ||||
| 101 | 35.4 | ||||
| M. plutonius | |||||
| Laboratory qPCR | 105 | 18.5 | 4.06 | 0.994 | 93.1 |
| 104 | 21.9 | ||||
| 103 | 25.4 | ||||
| 102 | 30.2 | ||||
| 101 | 34.6 | ||||
| POC qPCR | 105 | 15.6 | 4.54 | 0.991 | 94.9 |
| 104 | 19.5 | ||||
| 103 | 22.9 | ||||
| 102 | 28.7 | ||||
| 101 | 33.6 | ||||
Efficiency (%) was calculated using the formula: E(%) = (10(−1/slope) − 1) × 100%.
Analytical specificity and cross-reactivity of POC qPCR assay for P. larvae and M. plutonius
No cross-reactivity was observed with the 7 non-target viral (ABPV, BQCV, CBPV, DWV, IAPV, KBV, SBV) and 3 fungal (A. apis, A. flavus, V. ceranae) pathogens in either the POC qPCR or laboratory qPCR assays, as they all tested negative for P. larvae and M. plutonius. Additionally, the P. larvae–specific assay did not detect M. plutonius, and the M. plutonius–specific assay did not detect P. larvae. Each non-target pathogen was confirmed using pathogen-specific qPCR or reverse-transcription qPCR assays, with Cq values of 23.5–28.5.
Performance and Cq value of POC qPCR for P. larvae and M. plutonius
We evaluated the performance of our POC qPCR assay for P. larvae detection using 60 P. larvae–positive and 100 P. larvae–negative samples, confirmed by laboratory qPCR assays. Our POC qPCR assay detected 56 of 60 (93% sensitivity) P. larvae–positive samples and correctly identified 97 of 100 (97% specificity) P. larvae–negative samples. The positive percentage agreement (PPA) was 95%, and the negative percentage agreement (NPA) was 96%. The overall agreement (OA) between the POC qPCR and the laboratory qPCR assays for P. larvae detection was 96%.
Among the 56 samples that tested P. larvae–positive via POC qPCR assay, 47 had lower Cq values in the POC qPCR assay than in the laboratory qPCR assay (Table 4). The largest difference in Cq values was observed in sample 7 (adult honey bees), in which Cq values were 26.3 for POC qPCR and 33.0 for laboratory qPCR assays, a Cq difference of 6.7. In contrast, 4 samples (18, 26, 33, 41) that were positive in laboratory qPCR but undetected by POC qPCR had high Cq values that were close to the LOD.
Table 4.
Comparison of Cq values between laboratory quantitative real-time PCR (qPCR) and point-of-care (POC) qPCR assays for Paenibacillus larvae detection.
| ID | Honey bee type | Laboratory qPCR | POC qPCR | ||
|---|---|---|---|---|---|
| Cq | Interpretation | Cq | Interpretation | ||
| 1–6 | Adult | 25.4–33.2 | + | 22.6–28.3 | + |
| 7 | Adult | 33.0 | + | 26.3 | + |
| 8–17 | Larvae or adult | 24.6–34.5 | + | 24.5–34.2 | + |
| 18 | Adult | 35.0 | + | ND | – |
| 19–25 | Larvae or adult | 23.6–33.9 | + | 21.3–31.1 | + |
| 26 | Adult | 33.3 | + | ND | – |
| 27–32 | Adult | 29.2–33.1 | + | 26.1–28.1 | + |
| 33 | Adult | 33.4 | + | ND | – |
| 34–40 | Larvae or adult | 14.5–30.7 | + | 14.2–29.1 | + |
| 41 | Adult | 34.2 | + | ND | – |
| 42–60 | Larvae or adult | 24.7–34.1 | + | 22.1–30.1 | + |
| 61 | Adult | ND | – | 31.3 | + |
| 62 | Adult | ND | – | 30.2 | + |
| 63 | Adult | 39.0 | – | 31.2 | + |
| 64–160 | Larvae or adult | ND | – | ND | – |
ND = not detected. Full sample-level data (for samples 1–63) are available upon request.
Similarly, for M. plutonius detection, 62 M. plutonius–positive and 100 M. plutonius–negative samples, confirmed by laboratory qPCR, were tested using our POC qPCR. The POC qPCR assay detected 59 of 62 (95% sensitivity) M. plutonius–positive samples and correctly identified 97 of 100 (97% specificity) M. plutonius–negative samples. The PPA and NPA for M. plutonius detection were 95% and 97%, respectively. The OA between the POC qPCR and laboratory qPCR assays for M. plutonius detection was 96%.
Among the 59 samples that tested M. plutonius–positive via POC qPCR, 38 had lower Cq values in POC qPCR than in the laboratory qPCR assay (Table 5). The largest difference in Cq values was observed in sample 47 (larvae), in which the Cq values were 24.8 for POC qPCR and 31.9 for laboratory qPCR assays, a Cq difference of 7.1. Similar to the results with P. larvae, 3 samples (32, 49, 50) that were M. plutonius–positive in the laboratory qPCR but undetected by the POC qPCR assay had high Cq values close to the LOD.
Table 5.
Comparison of Cq values between laboratory quantitative real-time PCR (qPCR) and point-of-care (POC) qPCR assays for Melissococcus plutonius detection.
| ID | Honey bee type | Laboratory qPCR | POC qPCR | ||
|---|---|---|---|---|---|
| Cq | Interpretation | Cq | Interpretation | ||
| 1–31 | Larvae or adult | 20.0–34.9 | + | 20.3–34.2 | + |
| 32 | Adult | 34.1 | + | ND | – |
| 33–46 | Larvae or adult | 20.3–33.2 | + | 18.3–31.2 | + |
| 47 | Adult | 31.9 | + | 24.8 | + |
| 48 | Adult | 32.6 | + | 34.2 | + |
| 49–50 | Adult | 33.2–33.4 | + | ND | – |
| 51–62 | Adult | 28.5–34.5 | + | 25.5–33.0 | + |
| 63 | Adult | 37.9 | – | 28.8 | + |
| 64 | Adult | 37.0 | – | 31.2 | + |
| 65 | Adult | 38.9 | – | 26.0 | + |
| 66–162 | Larvae or adult | ND | – | ND | – |
ND = not detected. Full sample-level data (for samples 1–65) are available upon request.
Discussion
Our multiplex POC qPCR assay enabled the simultaneous detection of P. larvae and M. plutonius, and had analytical and test performance comparable to conventional laboratory qPCR assays. The POC qPCR assay LOD of 102 copies/reaction for both P. larvae and M. plutonius confirmed that its analytical sensitivity was comparable to that of laboratory qPCR assays. Notably, this level of sensitivity ensures the reliable detection of the agents even at low pathogen loads, which is particularly critical for identifying early or subclinical infections in field settings. 26
Our POC qPCR assay yielded lower Cq values (i.e., faster detection) than the laboratory qPCR assays in 47 of 56 P. larvae–positive samples and 38 of 59 M. plutonius–positive samples, thereby reinforcing its high analytical sensitivity. This trend suggests that the integrated sample processing of the POC qPCR platform may enhance DNA recovery efficiency or improve amplification kinetics compared with the multistep procedure used in the laboratory qPCR assay. Notably, the largest observed differences were 6.7 (P. larvae, sample 7) and 7.1 (M. plutonius, sample 47). To rule out the possibility of technical artefacts, testing was repeated, and consistent results were obtained, which supported the robust analytical performance of our POC qPCR system even for samples with low pathogen concentrations, which is an essential feature for field applications.
Although strong concordance was noted between POC qPCR and laboratory qPCR assays for most samples, some discrepancies and variability were observed, particularly near the LOD. Among the tested samples, 4 P. larvae–positive and 3 M. plutonius–positive samples detected by the laboratory qPCR assay were not identified by the POC qPCR assay. These samples had Cq values near or beyond the reliable detection threshold (>35) in the laboratory qPCR assay, indicating that pathogen DNA was present at extremely low levels. Such conditions are susceptible to the so-called Monte Carlo effect, which is a form of statistical variability that occurs during amplification of low-copy-number templates (<100 copies). Additionally, the minor variability observed in the POC qPCR assay may be attributable to differences in sample input and preparation between the 2 assays. Whereas laboratory qPCR uses purified nucleic acids, our POC qPCR processes minimally treated, crude samples directly in the field, which may reduce dilution linearity and introduce variation due to sample homogeneity. Despite these factors, the PPA and NPA, which were 95% and 96% for P. larvae and 95% and 97% for M. plutonius, respectively, indicate a high level of agreement between the 2 assays.
Our POC qPCR assay had high analytical specificity, with no cross-reactivity with various non-target honey bee pathogens. No false-positive amplifications were observed for P. larvae or M. plutonius. Importantly, our multiplex assay reliably differentiated P. larvae and M. plutonius, with no cross-reactivity. Such high analytical specificity is crucial in field conditions, where honey bee colonies are often affected by multiple simultaneous infections. 9 Accurate differentiation of P. larvae and M. plutonius from other coinfecting pathogens prevents misdiagnosis and ensures proper disease management.
Although our POC qPCR assay offers several advantages, certain practical limitations must be considered. First, the assay requires a power source, which limits its deployment in remote or off-grid apiary locations where electricity is not readily available. Future advancements in low-power design and the integration of long-lasting battery modules could enhance the usability of the POC qPCR device itself in field settings. However, certain sample-preparation steps, such as homogenization, also require electrical equipment. Therefore, full field deployment of this assay may still require additional portable tools or infrastructure to support the entire workflow. To address the sample homogenization challenge in the field, we propose a simple manual method using a flexible plastic dropping bottle to physically crush the sample without the need for electrical equipment. Incorporating a pre-filter at the outlet of the bottle may further improve the clarity of the homogenized extract and reduce the risk of inhibition, thereby enhancing test reliability under field conditions. Second, we did not evaluate the potential impact of environmental factors, such as temperature and humidity, on the performance of the POC qPCR XQ Station. Further studies are required to assess its stability and reliability under various field conditions. Third, additional optimization is required to further reduce processing time and increase throughput for large-scale screening.
Although our POC qPCR assay demonstrated strong concordance with the laboratory qPCR assay in terms of PPA, NPA, and OA, it is important to acknowledge a key limitation in the study design. The laboratory qPCR assay used for comparison, although routinely employed in national testing and based on a government-recognized protocol, was not independently validated against classical gold standard methods such as bacterial culture. Consequently, the performance metrics presented in our study should be interpreted with caution. To assess the clinical utility of the POC qPCR assay more robustly, future studies incorporating culture-confirmed samples or comparisons with internationally validated testing platforms are warranted.
Additionally, our POC qPCR assay relies on a dedicated device (XQ Station) and the proprietary XQ AFB and EFB Dx kit, necessitating a stable and continuous supply of these components. The XQ Station can process up to 4 samples simultaneously using single-use microfluidic cassettes, delivering results within ~60 min. The equipment costs ~20,000 USD for a 4-unit configuration and 7,000 USD for a single-unit setup, with a per-test cost of ~20 USD, including consumables. Considering this cost structure, institutional or governmental financial support may be essential to ensure the system’s long-term availability and facilitate its widespread adoption for field-based testing. Addressing these limitations will further enhance the practical applicability of POC qPCR technology in field testing. Our rapid field assay could enable immediate pathogen detection and facilitate timely management, thereby significantly contributing to disease-control efforts and the protection of honey bee populations.
Acknowledgments
We thank the beekeepers affiliated with the Korea Apicultural Agriculture Cooperative for their help with obtaining the samples used in our study.
Footnotes
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: Our study was supported by the Research of Animal and Plant Quarantine Agency (APQA), South Korea (fund Z-154308-2023-24-01), and by the Scale-up Technology Commercialization Program of the Korea Institute for Advancement of Technology (KIAT) under the Ministry of Trade, Industry, and Energy (MOTIE), South Korea, through the research and development project Development of mycoplasma analysis materials and diagnostic kits for biopharmaceutical production quality control (Project P0023228).
ORCID iDs: Juhaeng Heo 
https://orcid.org/0009-0003-8121-494X
Dae-Yong Kim
https://orcid.org/0000-0002-3168-3938
Contributor Information
Juhaeng Heo, Department of Veterinary Pathology, College of Veterinary Medicine, Seoul National University, Seoul, Republic of Korea.
Doo-Sung Cheon, Postbio, Gyeonggi-do, Republic of Korea.
Nyun-Ki Chung, Honeybee Clinic, Daejeon, Republic of Korea.
Yongrae Kim, Korea Apicultural Agriculture Cooperative, Seoul, Republic of Korea.
Dae-Yong Kim, Department of Veterinary Pathology, College of Veterinary Medicine, Seoul National University, Seoul, Republic of Korea.
References
- 1. Ackerly D, et al. The development of a loop-mediated isothermal amplification (LAMP) assay to detect American foulbrood in managed honey bee populations. Apidologie 2024;55:38. [Google Scholar]
- 2. Alburaki M, et al. Occurrence and distribution of two bacterial brood diseases (American and European foulbrood) in US honey bee colonies and resistance to antibiotics from 2015 to 2022. J Apic Res 2024;63:701–710. [Google Scholar]
- 3. Anderson KE, et al. A high-throughput sequencing survey characterizing European foulbrood disease and Varroosis in honey bees. Sci Rep 2023;13:1162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Arai R, et al. Diversity of Melissococcus plutonius from honeybee larvae in Japan and experimental reproduction of European foulbrood with cultured atypical isolates. PLoS One 2012;7:e33708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Bakonyi T, et al. Development and evaluation of PCR assays for the detection of Paenibacillus larvae in honey samples: comparison with isolation and biochemical characterization. Appl Environ Microbiol 2003;69:1504–1510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Budge GE, et al. The occurrence of Melissococcus plutonius in healthy colonies of Apis mellifera and the efficacy of European foulbrood control measures. J Invertebr Pathol 2010;105:164–170. [DOI] [PubMed] [Google Scholar]
- 7. Budge GE, et al. Standard methods for European foulbrood research 2.0. J Apic Res 2025;64:403–442. [Google Scholar]
- 8. Cameron TC, et al. Current status of loop-mediated isothermal amplification technologies for the detection of honey bee pathogens. Front Vet Sci 2021;8:659683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Chen Y, et al. Multiple virus infections in the honey bee and genome divergence of honey bee viruses. J Invertebr Pathol 2004;87:84–93. [DOI] [PubMed] [Google Scholar]
- 10. Cho YS, et al. Inventors; Animal and Plant Quarantine Agency; Postbio, assignees. Primer for diagnosis of parasitic and bacterial diseases of bees and diagnostic kit using the same. Republic of Korea patent KR 1021463590000. 2020. [Google Scholar]
- 11. De Graaf DC, et al. Diagnosis of American foulbrood in honey bees: a synthesis and proposed analytical protocols. Lett Appl Microbiol 2006;43:583–590. [DOI] [PubMed] [Google Scholar]
- 12. De Graaf DC, et al. Standard methods for American foulbrood research. J Apic Res 2013;52:1–28. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Ebeling J, et al. Biology of Paenibacillus larvae, a deadly pathogen of honey bee larvae. Appl Microbiol Biotechnol 2016;100:7387–7395. [DOI] [PubMed] [Google Scholar]
- 14. Forsgren E. European foulbrood in honey bees. J Invertebr Pathol 2010;103:S5–S9. [DOI] [PubMed] [Google Scholar]
- 15. Genersch E. American foulbrood in honeybees and its causative agent, Paenibacillus larvae. J Invertebr Pathol 2010;103:S10–S19. [DOI] [PubMed] [Google Scholar]
- 16. Genersch E. Honey bee pathology: current threats to honey bees and beekeeping. Appl Microbiol Biotechnol 2010;87:87–97. [DOI] [PubMed] [Google Scholar]
- 17. Heo J, et al. Prevalence of pathogens in abnormal honey bees in South Korea, 2020–2023. J Vet Diagn Invest 2025;37:104–113. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lindström A, et al. Horizontal transmission of Paenibacillus larvae spores between honey bee (Apis mellifera) colonies through robbing. Apidologie 2008;39:515–522. [Google Scholar]
- 19. Martínez J, et al. A real-time PCR-based strategy for the detection of Paenibacillus larvae vegetative cells and spores to improve the diagnosis and the screening of American foulbrood. Lett Appl Microbiol 2010;50:603–610. [DOI] [PubMed] [Google Scholar]
- 20. Milbrath MO, et al. Validation of diagnostic methods for European foulbrood on commercial honey bee colonies in the United States. J Insect Sci 2021;21:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Na H-H, Kim K-C. Identification of diagnostic PCR markers for honeybee foulbrood disease from specific genes of Paenibacillus larvae. J Life Sci 2017;27:67–71. [Google Scholar]
- 22. Reybroeck W, et al. Antimicrobials in beekeeping. Vet Microbiol 2012;158:1–11. [DOI] [PubMed] [Google Scholar]
- 23. Rollin O, et al. Impacts of honeybee density on crop yield: a meta-analysis. J Appl Ecol 2019;56:1152–1163. [Google Scholar]
- 24. Steinhauer N, et al. Drivers of colony losses. Curr Opin Insect Sci 2018;26:142–148. [DOI] [PubMed] [Google Scholar]
- 25. Tomkies V, et al. Development and validation of a novel field test kit for European foulbrood. Apidologie 2009;40:63–72. [Google Scholar]
- 26. Waite R, et al. Control of American foulbrood by eradication of infected colonies. Apiacta 2003;38:134–136. [Google Scholar]
- 27. Waite RJ, et al. Controlling European foulbrood with the shook swarm method and oxytetracycline in the UK. Apidologie 2003;34:569–575. [Google Scholar]