Detection of macrolide resistance genes, ermC and ermB, in Japanese honey using real-time PCR assays

Abstract

American foulbrood (AFB) is a honeybee disease caused by Paenibacillus larvae, and tylosin is used as the prophylactic in Japan. Honey contains macrolide-resistant bacteria that are a potential source of genes that may confer tylosin resistance to P. larvae. To investigate the potential risk of such genes in Japanese honey, we developed real-time PCR assays for the detection of important macrolide resistance genes, ermC and ermB, and analyzed 116 Japanese honey samples with known contamination status of P. larvae. Consequently, 91.38% of samples contained ermC and/or ermB, and 71.55% of samples contained both ermC and P. larvae, suggesting the possible emergence of tylosin-resistant P. larvae in Japan. Therefore, judicious use of the prophylactic is essential in maintaining its effectiveness.

Paenibacillus larvae is a spore-forming bacterium that causes a honeybee brood disease: American foulbrood (AFB) [3, 4]. In Japan, beekeepers have to burn diseased colonies that develop AFB. Tylosin, a macrolide antibiotic, is the only commercially available approved prophylactic agent for AFB in Japan. Although tylosin-resistant P. larvae are yet to be found in Japan [11], the only prophylactic could lose its effectiveness if P. larvae acquires tylosin resistance. Therefore, judious use of tylosin is essential to prevent the emergence and selection of tylosin-resistant P. larvae strains.

Oxytetracycline is used for control of AFB in North America, and oxytetracycline-resistant P. larvae that acquired a small plasmid possessing a tetracycline resistance gene (tet(L)) has already been reported [1, 6, 7]. In other reports, Bacillus strains isolated from Argentinian honey possessed an oxytetracycline resistance gene similar to tet(L) [5], and the transmission of a plasmid containing the oxytetracycline resistance gene from P. larvae to Bacillus subtilis was demonstrated in vitro [2]. These studies indicate that bacteria in honey are potential source of antibiotic resistance genes that could confer resistance to P. larvae. Tylosin-resistant P. larvae may also arise via a similar mechanism. Indeed, bacteria possessing macrolide resistance genes, such as ermC and ermB, have been isolated from Japanese honey, and an ermC-carrying putative mobilizable plasmid conferred tylosin resistance to P. larvae [9]. Although the effect of ermB on P. larvae was less than that of ermC, the ermB gene also decreased the susceptibility of P. larvae to macrolides [9]. As P. larvae was suggested to be widely distributed in apiaries in Japan [10], tylosin-resistant P. larvae could easily develop if such macrolide-resistant bacteria are already present in beehives and apiaries, and the developed resistant P. larvae would be selected by using the prophylactic. However, bacteria possessing ermC and ermB genes have only been isolated from one of each of the fifty-three honey samples analyzed using cultivation methods in a previous study [9]. The distribution of these macrolide resistance genes in Japanese apiaries remains unclear. Therefore, in this study, we developed real-time PCR assays to detect ermC and ermB genes and analyzed their distribution in Japanese honey.

To develop real-time PCR assays, multiple ermC and ermB gene sequences were retrieved from the GenBank database (Table 1), and specific primers were designed for the conserved regions of the retrieved genes (Table 2). Primer specificity was confirmed using a BLAST search (http://blast.ncbi.nlm.nih.gov). QuantiTect SYBR Green PCR kits (QIAGEN, Hilden, Germany) were used to perform real-time PCR assays using a final reaction volume of 25 µL containing 12.5 µL of 2x QuantiTect SYBR Green PCR Master Mix, 0.3 µM of each primer, and 1 µL of template DNA. The PCR cycling conditions are listed in Table 2. All amplifications were performed using the QuantiStudio^TM 3 Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). Each sample was tested in duplicates.

Table 1. Accession numbers of the reference sequences of ermC and ermB used for primer design.

Gene	Accession number	Locus tag	Bacterial species
ermC	NC_001376	H4K10_RS00010	Bacillus subtilis
	M13761	-	Bacillus subtilis
	NG_047806	A7J11_00443	Bacillus subtilis
	MIOQ01000118	BHF98_11650	Corynebacterium diphtheriae
	NG_047817	A7J11_02291	Lactobacillus reuteri
	FJ489650	PA16_11	Lactobacillus reuteri
	NZ_MH423314	HTS84_RS00035	Mammaliicoccus lentus
	NZ_ATPV01000221	M703_RS06465	Neisseria gonorrhoeae
	LC586958	-	Oceanobacillus oncorhynchi subsp. incaldanensis
	NZ_AUPS01000034	M397_RS14135	Staphylococcus aureus
	NC_001395	HTN12_RS00010	Staphylococcus aureus
	NZ_LFXH01000034	AC235_RS12515	Staphylococcus aureus
	CP054554	FOB69_13025	Staphylococcus hominis
	NC_016139	HS754_RS00010	Staphylococcus hyicus
	LTIO01000074	HMPREF2695_03090	Staphylococcus sp.
ermB	AF480459	-	Bacillus cereus
	PNK22492	-	Bacillus thuringiensis
	CP052842	HIR78_19720	Bacillus subtilis subsp. subtilis
	CP041081	FJR70_32845	Bacillus tropicus
	AISD01000042	SC9_03208	Enterococcus faecalis
	NZ_LC597664	JNG29_RS00185	Paenibacillus sp.
	MH785229	BJL72_k00065	Staphylococcus aureus
	WXZD01000018	GT923_05810	Streptococcus pyogenes

Table 2. PCR primers and cycling conditions used in this study.

Target	Primer name	Sequence (5′-3′)	PCR conditions
ermC	ermC_qF	ATCGTGGAATACGGGTTTGC	95°C 15 min–94°C 15 sec, 55°C/56°Ca 30 sec, 72°C 30 sec (45 cycles)
	ermC_qR	CTGATAAGYGAGCTATTCAC
ermB	ermB_qF	CCGCCATACCACAGATGTTC	95°C 15 min–94°C 15 sec, 55°C 30 sec, 72°C 30 sec (45 cycles)
	ermB_qR	ACTTTGGCGTGTTTCATTGC

^a Presence of the ermC gene was assessed at two different annealing temperatures during the PCR cycle. All PCR results from the two conditions were used to calculate the ermC detection rates.

Oceanobacillus oncorhynchi subsp. incaldanensis (strain J18TS1) and Paenibacillus sp. (strain J45TS6) are honey-derived bacteria that carry plasmids containing ermC and ermB genes, respectively [9]. To determine the detection limit of these representative bacteria in honey samples using our assays, we prepared bacterial spore-spiked honey using ermC- and ermB-free honey harvested in Thailand. The absence of these genes in the honey sample was confirmed using our assays. J18TS1 and J45TS6 spore suspensions were prepared according to Ohashi et al. [8], with certain modifications to the culture conditions (Supplementary Table 1). The number of spores in each suspension was calculated using hemocytometers under an inverted microscope (CKX41, OLYMPUS, Tokyo, Japan). The spore suspensions were then added to honey samples to obtain final solutions of 1,000, 100, 10, and 1 spore(s)/mL honey. Genomic DNA was extracted from 5 mL of each spore-spiked honey sample using the Johne Pure Spin kit (FASMAC Co., Ltd., Atsugi, Japan), according to Okamoto et al. [10]. Real-time PCR assays were performed as described earlier, and the detection limit for both genes was found to be 10 spores/mL honey. The melting point temperatures of the ermC and ermB amplicons were 75.0°C and 75.6°C, respectively.

To study the distribution of ermC and ermB in Japanese honey, 116 honey samples collected from 2017 to 2020, were analyzed in this study (Supplementary Table 2). DNA was extracted from 5 mL of each honey sample as previously described [10]. Genomic DNA of J18TS1 and J45TS6 (positive controls) was extracted using InstaGene™ Matrix (Bio-Rad Laboratories, Inc., Hercules, CA, USA), according to the manufacturer’s instructions. Positive controls and non-template negative controls were placed on each reaction plate.

In this study, the ermB and ermC genes were detected in 30.17% and 86.21% of honey samples, respectively, and 25% of samples contained both genes (Fig. 1A, Supplementary Fig. 1, and Supplementary Table 2). The ermB amplicons showed a single specific melting temperature at 75.5 ± 0.2°C indicating negligible sequence variations. Since ermC showed an erratic melting temperature (71.7–75.2°C), we determined the sequences of all ermC amplicons and found that ermC amplicons exhibited sequence variations (Fig. 2). The higher detection rate of ermC than ermB and the sequence variations in ermC imply that ermC is carried by a wide variety of bacteria in honey as compared to ermB. However, the possibility that different ermC genes are carried by a single or limited bacterial species cannot be excluded. As reported from a previous study of our research group [9], Japanese honey contains a wide variety of bacteria of different genus, such as Bacillus, Paenibacillus and Alkalihalobacillus. Among the bacterial isolates analyzed in the previous study, the ermC and ermB genes have only been identified in O. oncorhynchi subsp. incaldanensis and Paenibacillus sp., respectively [9]. However, information on other bacterial species in honey that carry erm genes remains unclear. Therefore, further investigation is essential to elucidate the bacterial species that carry macrolide resistance genes and their distribution in Japanese honey.

Fig. 1.

Detection rates of (A) macrolide resistance genes, ermC and ermB, and (B) Paenibacillus larvae and ermC from Japanese honey. DNA was extracted from the sediments of 5 mL of honey using the Johne Pure Spin kit (FASMAC Co., Ltd., Atsugi, Japan), and 1 µL of the extracted DNA from each sample was used as a template in real-time PCR assay to detect ermC/ermB. The detection rate of P. larvae from the honey samples was retrieved from a previous study conducted by our research group [10]. N. D., not detected.

Fig. 2.

Sequence variations of the ermC gene fragments detected from Japanese honey samples. The ermC gene sequence of Oceanobacillus oncorhynchi subsp. incaldanensis (strain J18TS1) was retrieved from the GenBank database (accession no. LC586958) and compared with the amplified gene sequences from Japanese honey. The ermC gene sequence of J18TS1 is shown at the bottom. The numbers indicate the position of nucleotides in the ermC gene of strain J18TS1. Nucleotide differences are shown with colored backgrounds. R=A or G; Y=C or T.

P. larvae becomes tylosin-resistant if it acquires the ermC gene [9]; therefore, the co-contamination of honey with ermC and P. larvae suggests the potential risk of the emergence of macrolide-resistant P. larvae in beehives and apiaries. To further investigate this risk, we retrieved the P. larvae-positive/negative data of the same 116 honey samples analyzed in an earlier study conducted by our research group [10] and compared it with the ermC-positive/negative data obtained from the present study. Surprisingly, both ermC and P. larvae were detected in 71.55% of honey samples (Fig. 1B, Supplementary Fig. 1, and Supplementary Table 2). Since most samples analyzed in this study were bulk honey, the detection rate did not represent the status of a single honey bee colony. In addition, as our assays do not provide insight into the transferability of ermC, the detected gene may not have the potential to transfer to P. larvae and subsequently confer tylosin resistance to the bacterium. However, in apiaries contaminated with both ermC and P. larvae, judicious use of tylosin is recommended to prevent the selection of macrolide-resistant P. larvae strains. Although we cannot deny the possibility that bacteria in honey may contain other antibiotic resistance genes that could confer macrolide resistance to P. larvae, tylosin can be efficiently used in the control of AFB with a low risk of tylosin-resistant P. larvae selection in the apiaries where only P. larvae is detected.

To the best of our knowledge, this is the first study that focuses on erm genes in honey. The real-time PCR assays developed in this study can be applied in apicultural industries to ensure judicious use of antimicrobials. Future improvement of the assays for specific detection of transferable ermC/ermB genes would further help efficient and appropriate control of AFB.

CONFLICT OF INTEREST

The authors have no conflicts of interest to declare.