Nosema ceranae Fries is an intracellular parasite found in honeybees. Although it was first detected in the Asian honeybee Apis cerana in 1996 , it is now dominant in the European honeybee Apis mellifera, and was detected in all members of the colony, including adult bee and brood stages. Nosema ceranae is known to contribute to the winter mortality of honeybees worldwide . Consequently, the decline of healthy honeybee colonies has a negative influence on honey production as well as on agricultural crops and ecological plants due to the decrease in pollination in flowering plants . In Korea, Nosema ceranae infection was recorded for the first time in 1996 . The prevalence of nosema disease in apiaries was detected at 56% in 2003 , increased to 94% in 2013 , and the updated 2019 data from the Honeybee Disease Laboratory, Animal and Plant Quarantine Agency, Republic of Korea, showed that 24% of the surveyed apiaries were infected with Nosema ceranae .
Microscopy is one of the standard methods for detecting and quantifying Nosema . The advantages of this method are that it can be directly observed and infection intensity can be determined by counting spores with the aid of a hemocytometer and a light microscope at 400× magnification . However, this method is inadequate for identifying species because different Nosema species have similar spore sizes . Furthermore, the identification of all life stages using a conventional light microscope is difficult because of the differences in size and shape among the intracellular life stages of nosema . Therefore, microscopy is not an efficient method for the detection of nosema at a low intensity of infection.
Polymerase chain reaction (PCR) detection has been commonly used as a reliable method for diagnosing nosema infection (nosemosis) . PCR detection methods for nosema have been developed for N. ceranae and N. apis identification based on variations in the size of conventional PCR amplicons . A method to estimate spore counts by semi-quantitative triplex PCR assay was later developed , and quantitative real-time PCR was developed for accurate and sensitive detection of N. apis and N. ceranae simultaneously, using multiple primers in one reaction . Regrettably, all of these molecular-based methods suffer from the drawback of relying on time-intensive PCR. Recently, a new molecular-based method, known as loop-mediated isothermal amplification (LAMP), was developed for diagnosing nosemosis and was demonstrated to be less time-intensive than PCR detection . However, this method can only be used as an alternative method for conventional PCR because of the requirement of an electrophoresis step after polymerization to verify the specificity of detection . Since quantitative PCR has been used as an efficient tool for detecting and measuring the level of N. ceranae infection as well as for related studies, such as the evaluation of virulence of Nosema species , the influence of N. ceranae on its host , or interaction between N. ceranae and honeybee gut bacteria , it is important to develop a sensitive PCR system with stable amplification and less time consumption.
Accordingly, the current study aimed to introduce a molecular detection method that uses a newly developed real-time PCR method, known as ultra-rapid real-time quantitative PCR (UR-qPCR). The rapid heat transfer involved in this system markedly reduces the time required for PCR detection of N. ceranae, and it is expected to be an effective tool for rapidly diagnosing low levels of infection and for evaluating the efficacy of drug candidates for mitigating nosemosis in honeybees.
The isolation and purification of N. ceranae spores were based on the standard method for nosema research , with some modifications. Fifty adult bees were ground in 20 mL of distilled water using a mortar and pestle, followed by the addition of 30 mL of water. The homogenate was filtered through a fine mesh gauze (Two Guys & Tools Ltd., Korea) to remove large bee particles. The solution was vortexed and filtered again using a 70-µm nylon cell strainer (BD Falcon, USA) to remove large particles that originated from the honeybee alimentary canals. Next, the flow-through solutions were collected in a new 50-mL conical tube and centrifuged at 5,000 × g for 10 min. The supernatant was discarded, and the centrifugation was repeated twice. Finally, the pellets were re-suspended in 5 mL of distilled water and stored at 4°C until further use.
For the evaluation of N. ceranae in caged bees, the whole abdomen parts of five bees were ground together with a micropestle in a 1.5 mL tube containing distilled water (0.5 mL). After grinding, the homogenate was vortexed, and only the solution was transferred to a new 1.5 mL tube using a pipette. To minimize the loss of spores, another 0.5 mL of water was added to the remaining substrate. After vortexing, the solution was transferred again to the same new tube. The new tube was centrifuged at 12,000 × g for 1 min using a tabletop centrifuge, and the supernatant was discarded. Next, the pellet was suspended in 1 mL of water and centrifuged again at 12,000 × g for 1 min. After discarding the supernatant, the pellet was resuspended in distilled water to a final volume of 1 mL (corresponding to 0.2 mL solution/bee), and 0.2 mL of the solution was diluted in 0.8 mL) and used for microscopic counting. The remaining 0.8 mL spore solution was centrifuged at 12,000 × g for 1 min, and the supernatant was discarded. The resulting pellets were used for DNA extraction and PCR quantification.
The spore pellet prepared as described in the section on sample preparation was ground in liquid nitrogen using a micropestle from a sample grinding kit (Sigma Aldrich, Korea), and DNA was extracted using the DNeasy Blood & Tissue Kit (Qiagen, Germany) according to the manufacturer's protocol. Finally, 100 µL of DNA solution was acquired from each sample. DNA concentration was measured using a BioPhotometer (Eppendorf, Germany).
A recombinant vector (pCR2.1-Nosema) carrying a 590 bp ribosomal DNA fragment of N. ceranae was used as a positive control for N. ceranae detection . In addition, N. ceranae-specific primers, NO-DC-F3 (5-AGGCAGTTATGGGAAGTAATATTATA-3) and NO-DC-R2 (5-CAGGGTCGTCACATTTCATCTTTC-3), were designed based on the alignment of the N. ceranae rRNA gene sequences. The rRNA gene sequence of N. ceranae (GenBank accession No.: JX205151.1), which included a partial 5S sequence, complete sequence small subunit and internal transcribed spacer, and partial large subunit sequence, was used to search for similar sequences in GenBank using the National Center for Biotechnology Information Nucleotide BLAST tool. The rRNA gene sequences of N. ceranae and the complete sequences of the small subunit ribosomal RNA gene, internal transcribed spacer, and large subunit ribosomal RNA gene of N. apis, a closely related species of N. ceranae , were used for alignment using the program Clustal X version 2.0  to design a specific primer for the detection of N. ceranae (Fig. 1). Using these two primers, a 216-bp fragment of the small subunit rRNA gene was amplified.
The specificity of N. ceranae primers was evaluated by testing cross-amplification of total DNA extracted from honeybee (Apis mellifera) and other honeybee pathogens, Melissococcus plutonius (ATCC35311), Varroa destructor, Tropilaelaps spp., and Paenibacillus larvae (ATCC9545).
Microscopic counts of N. ceranae spores were conducted according to the standard method described by Human et al. . Briefly, 1 mL of spore solution was prepared for counting. The enumeration was conducted using a Neubauer improved bright-line hemocytometer (Paul Marienfeld GmbH & Co. KG, Germany) and a light microscope at 400× magnification. Two of the 16 (0.025 × 0.025) cm2 squares in each of the four corners of the hemocytometer were randomly selected for a total of eight selected squares in each count, and the spores within each randomly selected square were counted. After counting, the hemocytometer was carefully cleaned using distilled water. A new solution of the same sample was added to the hemocytometer for the second enumeration with eight other 8 squares. In total, 16 squares were counted twice for 2 times in each sample. The number of spores per mL was calculated using the following formula: total counted spores/(16 × 0.025 × 0.025 × 0.01 cm3).
UR-qPCR was performed using a GENECHECKER PCR machine UF150 (Genesystem Co., Ltd., Korea) and 2× Rapi Master Mix (Genesystem Co., Ltd.). Each 10-µL reaction included 5 µL 2× Rapi Mix, 2 µL primers (10 pmol of each primer), and a maximum of 3 µL with ≤ 10 ng DNA template. SYBR green, which is included in the Rapi Mix, was used as a fluorescent dye for detecting UR-qPCR amplification. The PCR conditions were as follows: 95°C for 30 s, followed by 50 cycles of 95°C for 3 s, 55°C for 3 s, and 72°C for 3 s.
The standard curve for calculation of DNA copy number was established based on amplification using recombinant DNA with 10-fold serial dilution (from 2.05 × 108 to 2.05 × 100 copies per reaction). Linear regression representing the relationship between the log10 initial DNA copy number and the corresponding threshold cycle (Ct) was created from triplicate PCR runs.
The same serially diluted recombinant DNA and specific primers of N. ceranae were also used for other real-time PCR systems (The CFX96 Touch Real-time PCR Detection System, BIO-RAD, USA) to compare the sensitivity of N. ceranae detection with the UR-qPCR. Twenty-µL reaction mix was prepared consisting of 10 µL 2× Rapi Master Mix (Genesystem Co., Ltd.), 1 µL (10 pmol) of each primer, 1 µL of DNA template, and 7 µL of ddH2O. PCR conditions were 95°C (3 min), followed by 45 cycles of 95°C for 15 s, 55°C for 15 s, and 72°C for 15 s.
The spore concentration of the prepared solution was determined to be 2.40 × 107 spores/mL through microscopy. The spore solution was divided into two 1.5-mL Eppendorf tubes (1 mL/tube). The spore solution in the first tube was serially 10-fold diluted using 0.1 mL spore solution plus 0.9 mL of ddH2O until reaching a concentration of 2.40 × 100 spores/mL and was used for microscopic enumeration. In addition, 10% (0.1 mL) of each dilution (2.40 × 107 to 2.40 × 101 spore/mL), which contained 2.40 × 106 to 2.40 × 100 spores, respectively, was set aside for DNA isolation, in order to generate a spore dilution series. The DNA solution extracted from each spore was 100 µL. Meanwhile, the spores (2.40 × 107 spores) in the second tube were subjected to DNA extraction to obtain 100 µL of DNA solution. The DNA solution was then serially 10-fold diluted by adding 10 µL of DNA solution plus 90 µL of ddH2O until reaching a concentration that corresponded to the concentration of DNA obtained from 2.40 × 100 spores, in order to generate a DNA dilution series.
To identify the limit of detection of the two counting methods, UR-qPCR was performed using the DNA solution that originated from the DNA and spore dilutions. Microscopic counting was conducted using a dilution series of the spore solution.
The relationship between the two methods was evaluated based on the quantitative results from samples of infected honeybees. N. ceranae levels were measured by molecular quantification and microscopic counting from caged bees (n = 300), A. mellifera, which were divided into three cages (100 bees/cage). The cages were designed according to the instructions of Williams et al. . After starvation for 24 h, the material named Bee happy (Bioforce Co., Ltd., Korea) with 100× and 300× dilution rates in 50% (w/v) sucrose solution were supplied to the bees in each cage. The bees in the control cage were fed with a 50% sucrose solution. The volume of the feeding solution supplied to each cage was based on an estimated consumption of 50 µL per bee per day. The severity of N. ceranae infection in the bees from each cage was assessed before feeding and every 2 days after feeding until the 8th day. Five bees in each cage were randomly removed for molecular quantification using UR-qPCR and microscopic enumeration at each time point. Caged bees were kept in a dark room at 25°C during the infection period.
The detection of N. ceranae using the primer pair NO-DC-F3/R2 showed a limit of 2.05 × 100 copies of recombinant DNA under PCR conditions of 95°C (30 s) and 50 cycles of 95°C (3 s), 55°C (3 s), and 72°C (3 s) (Fig. 2A). The linear regression of log10 values of initial DNA copies and the corresponding Ct values was y = −3.2283x + 37.796; R2 = 0.9971, where x and y represent log10 DNA copy number and Ct value, respectively (Fig. 2B). The amplification efficiency calculated from the slope of the standard curve was 104.06%. The sensitivity of N. ceranae detection using the primer pair NO-DC-F3/R2 was also seen with a limit of 2.05 × 100 copies of recombinant DNA can be detected using other conventional real-time PCR systems (Fig. 2D).
The specificity of the N. ceranae UR-qPCR system was confirmed by amplification of only N. ceranae DNA among the assessed DNA templates of honeybee (Apis mellifera), Melissococcus plutonius, Varroa destructor, Tropilaelaps spp., and Paenibacillus larvae (Fig. 2C).
The results showed that the Bee happy solution had possible repressive effects against the development of N. ceranae in infected bees. The 100× dilution of Bee happy solution (100× DBS) and 300× dilution of Bee happy solution (300× DBS) induced the increase of N. ceranae after 2 days from 3.42 × 106 ± 6.77 × 105 to 7.84 × 106 ± 9.00 × 105 and 2.00 × 107 ± 1.97 × 106 in the 2 feeding, respectively. In the 100× BDS feeding, the efficacy of N. ceranae inhibition was seen after 4 days of feeding and the lowest number of spore was seen at the 6th day with 2.26 × 105 ± 1.70 × 105 spores/bee before increasing to 5.47 × 106 ± 8.25 × 105 spore/bee at the 8th day of feeding period. Meanwhile, the inhibitory effect of 300×BDS diminished after the 4th day, and the spore numbers reached levels comparable to those of the negative control by the end of the feeding period (Table 1). The results were confirmed by molecular quantification using UR-qPCR, which showed the same trend of N. ceranae development as the microscopic counting results (Table 2).
|Treatment||N. ceranae spore/bee|
|0 day*||2 days||4 days||6 days||8 days|
|Control||3.42 × 106||4.28 × 106||1.54 × 107||1.71 × 107||1.29 × 107|
|± 6.77 × 105||± 5.94 × 105||± 1.82 × 106||± 1.18 × 106||± 1.29 × 106|
|100× DBS||3.42 × 106||7.84 × 106||3.88 × 106||2.26 × 105||5.47 × 106|
|± 6.77 × 105||± 9.00 × 105||± 8.30 × 105||± 1.70 × 105||± 8.25 × 105|
|300× DBS||3.42 × 106||2.00 × 107||0||6.80 × 106||1.13 × 107|
|± 6.77 × 105||± 1.97 × 106||± 8.70 × 105||± 1.24 × 106|
DBS, dilution of Bee happy solution.
*N. ceranae spores were counted before feeding (0 day) and every two days during the feeding period.
|Treatment||N. ceranae DNA copy/bee|
|0 day*||2 days||4 days||6 days||8 days|
|Control||2.92 × 107||2.74 × 107||1.13 × 108||1.35 × 108||1.27 × 108|
|± 9.44 × 106||± 9.22 × 106||± 4.05 × 107||± 3.82 × 107||±5.59 × 107|
|100× DBS||2.92 × 107||4.44 × 107||2.27 × 107||1.42 × 106||6.82 × 107|
|± 9.44 × 106||± 1.89 × 107||± 1.03 × 107||± 1.55 × 104||± 2.81 × 105|
|300× DBS||2.92 × 107||1.85 × 108||2.74 × 104||5.68 × 107||1.11 × 108|
|± 9.44 × 106||± 8.26 × 107||± 1.62 × 103||± 1.79 × 107||± 3.38 × 107|
DBS, dilution of Bee happy solution.
*N. ceranae DNA was quantified before feeding (0 day) and every two days during the feeding period.
The results of N. ceranae quantification from the different caged bees showed that molecular detection was more stable and sensitive than the microscopy approach. The microscopy method negatively detected N. ceranae at 2.74 × 104 DNA copies/bee (Tables 1 and 2). In the medium (106 spores/bee) and high levels (107 spores/bee) of N. ceranae infection , stable detection was achieved by microscopy, and DNA copy numbers were approximately 8-fold higher than spore counts from the same sample (Table 3).
|Treatment||N. ceranae DNA/spore*||Average|
|0 day*||2 days||4 days||6 days||8 days|
|Control||8.54||6.40||7.34||7.89||9.84||8.00 ± 1.29|
|100× DBS||8.54||5.66||5.85||6.28||12.47||7.76 ± 2.87|
|300× DBS||8.54||9.25||-†||8.35||9.82||8.99 ± 0.68|
DBS, dilution of Bee happy solution.
*Ratio of the mean value of DNA copy to spore number from counting every 2 days; †Unidentified result due to the negative detection of microscopic count.
Microscopic enumeration of the serial spore dilution (from 2.40 × 107 to 2.40 spores/mL) revealed that a minimum of 2.40 × 104 spores/mL are needed for quantitative detection. However, the number of spores at this concentration exhibited the greatest variation from the mean value (Fig. 3). The limit of detection for molecular quantification showed that 3 of 100 µL DNA solution isolated from 24 N. ceranae spores equal to 0.72, spores could be detected in each PCR reaction, and this result was confirmed by the counts of both the DNA dilution series solution and DNA isolated from the spore dilution series (Fig. 4).
DNA copy calculated from each DNA concentration of the DNA dilution series and spore dilution series showed that the quantification result by handling on DNA solution was more stable than on spore solution, as demonstrated by the DNA copy calculated from the DNA dilution series and spore dilution series in comparison with serially diluted spore number. DNA copy number calculated from DNA dilution series at all concentrations was approximately 8-fold higher (7.89 ± 1.24 on average) than spore number, and the relationship between DNA copy and spore number was represented by a linear regression, y = 0.9656x + 1.0213; R2 = 0.9973; where x and y are log10 of spore number and the corresponding DNA copy, respectively. Meanwhile, DNA copy calculated from DNA solution originating from spore dilution series at a concentration ≥ 2.40 × 103 spores/mL was also approximately 8-fold higher than the spore number (8.29 ± 1.52 on average). However, at lower spore concentrations (2.40 × 102 and 2.40 × 101 spores/mL), the dilution of spore solution results in the variation of detected result, which was demonstrated by the increase of DNA copy compared to the expected spore numbers to around 29.42 and 37.97 times, respectively.
Quantification of N. ceranae by UR-qPCR improved the sensitivity and stability of detection in comparison with microscopic enumeration. The microscopy approach is limited by the ability to observe and count spores, but the visible spores account for only a fraction of the life cycle of the microsporidia that live outside the intestinal cells of the host, and other stages remain invisible . Consequently, low levels of infection can be difficult to detect using microscopic observation. In contrast, molecular quantification relies on PCR performance; the total amount of Nosema DNA in the sample was detected and calculated according to a standard curve. Therefore, PCR detection using specific primers has been demonstrated to be useful for the quantitative detection of intracellular parasites .
At stable detection levels between 105 and 107 spores/bee from infected samples and ≥ 2.40 × 103 spores/mL from the purified spores, the DNA copy numbers were approximately 8-fold higher than the spore counts determined by microscopy. These results were consistent with those of a previous study that considered the Nosema genome to contain 10 copies of the targeted ribosomal RNA gene .
It has been previously claimed that the limit of detection in one microscopic field is 5 × 104 spores/mL, and the sensitivity is 2.5 × 103 spores/mL when observing 10 fields while counting Nosema bombycis spores . However, when one spore was detected in the area of 0.025 × 0.025 cm2 and 0.01 cm depth, the limit of detection was 1.6 × 105 spores/mL (1/0.01 × 0.025 × 0.025 cm3), and when 10 fields were screened, the sensitivity of detection reached 1.6 × 104 spores/mL. The optimal concentrations of purified spore solution for microscopic enumeration were 2.40 × 105 and 2.40 × 106 spores/mL, considering the time and accuracy of quantification. At these concentrations, the number of spores observed in each field was approximately 2 and 20, respectively. At a concentration of 2.40 × 107 spores/mL, more than 100 spores were observed in each area. Therefore, more time was required to count the spores, and it was determined that acquiring images of the observation field and then counting the spores in the image was a more efficient approach. Meanwhile, molecular detection showed stability of detection at 2.40 × 101 spores.
Grinding the spores in liquid nitrogen prior to DNA isolation using DNeasy Blood & Tissue Kit (Qiagen) and followed by UR-qPCR detection showed a sensitivity of N. ceranae detection with a limit of 24 spores per bee could be detected when the volume of DNA solution extracted from 24 spores was 100 µL (equal to 0.24 spore/µL), and 3 of the 100 µL was used for each PCR corresponding to 0.72 spore per PCR reaction can be detected. The results of quantitative PCR compared to spore counting using microscopic methods showed that the number of DNA copies was approximately eight times higher than spore number, multiple copies of the target gene in each spore . Therefore, approximately 5.76 DNA copies (0.72 spore × 8 copies = 5.76 copies) were used in each PCR reaction. The UR-qPCR and DNA isolation in this study were much more sensitive than the HBRC DNA isolation method and conventional PCR , which had a limit of detection of 100 spores per bee and 10 spores per PCR reaction, and DNA isolation using the DNeasy Plant Mini kit (Qiagen) followed by conventional real-time PCR with a limit of 10 spores per PCR reaction can be detected .
The N. ceranae-specific UR-qPCR developed in this study showed specificity and was as sensitive as conventional real-time PCR. However, this UR-qPCR system showed the advantages of mobility, simple protocol, and time saving for PCR performance. The time required to complete 50 cycles of UR-qPCR (3 s for each PCR step) was 20 min and 36 s. PCR can be limited by the half-life of Taq polymerase (approximately 45 min at 95°C ); therefore, the rapidity of UR-qPCR allows for more cycles before degradation of Taq polymerase. The UR-qPCR was more rapid than other real-time PCR systems , and the LAMP methods, which require 30 min to 1 h , have been developed for the detection of N. ceranae.
An appropriate method for N. ceranae enumeration using UR-qPCR was evaluated in this study. This method improves upon quantitative detection methods that use real-time PCR. The advantages of this method are the stability and sensitivity of quantification when compared to microscopic enumeration and rapidity when compared to previous molecular-based methods. Additionally, evaluation of the development of N. ceranae in honeybees fed with the Bee happy solution showed the repressive potential of this material against N. ceranae. These findings provide the basis for further studies that may demonstrate the efficacy and safety of this formulation for the control of nosemosis in honeybee populations.