2025 年 60 巻 3 号 p. 157-160
Lactococcus garvieae serotype III infections have been found in striped jack Pseudocaranx dentex since 2021. In the absence of a vaccine against serotype III, erythromycin (EM) is used to treat this infection. In 2024, five EM-resistant L.garvieae serotype III isolates were obtained from striped jack specimens. These isolates carried erm(B), an EM-resistance gene. The filter mating experiment confirmed that the EM-resistance trait was successfully transferred to L.formosensis as the recipient. All biased sinusoidal field gel electrophoresis (BSFGE) patterns of these strains were similar to the BSFGE type A of the previous study and coincided with each other.
Lactococcus garvieae serotype III, a newly emerging fish pathogen that causes lactococcal infection, has been detected in fish farms since 2021 (Minami et al., 2023; Araki et al., 2024a). A previous epidemiological study revealed that all strains of L. garvieae serotype III isolated from various fish farms between 2021 and 2023 were resistant to lincomycin (LCM) but not to erythromycin (EM) (Araki et al., 2024b). Currently, only antimicrobial chemotherapy is available to treat L. garvieae serotype III infections until the development of effective vaccines. EM has been effective in controlling lactococcal infections caused by L. garvieae serotype I and Lactococcus formosensis (formerly known as L. garvieae serotype II) in fish farms in the absence of EM-resistant lactococcal pathogens. However, EM-resistant strains possessing erm(B), an EM resistance gene, have been isolated from L. garvieae serotype I and L. formosensis and spread to fish farms (Maki et al., 2008; Akmal et al., 2023). Recently, it has become difficult to control L. formosensis infections using EM because of the prevalence of EM-resistant strains in fish farms.
In 2024, EM-resistant L. garvieae serotype III emerged in the striped jack Pseudocaranx dentex. In the future, the EM-resistant L. garvieae serotype III is expected to spread to fish farms. If EM-resistant L. garvieae serotype III spread in fish farms, it would be extremely difficult to treat this infection with EM medication. The aim of this study was to detect an EM-resistance gene and its transmissibility in the EM-resistance trait of L. garvieae serotype III. Furthermore, EM-resistant strains were analyzed to assess their relationships by using biased sinusoidal field gel electrophoresis (BSFGE).
EM-resistant strains (n = 5) and EM-sensitive strains (n = 5) used in this study were listed in Table 1. Five EM-resistant strains were isolated from the striped jack in a fish farm in Kumamoto Prefecture from September to October 2024. All EM-resistant strains (KMS2410SJ, KMS240927SJ, KMS241003SJ1, KMS241003SJ2, and KMS241003SJ3) were sampled at different days from same fish net cage. EM-resistant strains and EM-sensitive strains (KMS240716SJ, KMS240731A, MS240805SJ, MS240808A, and MS241016Y3) were cultured on Todd-Hewitt (Difco, Becton, Dickinson and Company) agar (THA) at 25°C for 2 days. All strains were used to identify L. garvieae serotype III with diagnostic antiserum prepared against L. garvieae serotype III strain MS210922A (Minami et al., 2023). Furthermore, diagnostic multiplex PCR (mPCR) assay was performed to identify serotype III, according to a previous study. This mPCR assay can simultaneously identify all three pathogens that cause lactococcosis (serotype I, L. formosensis, and serotype III) (Araki et al., 2024a).
| Strain | Source | Region of isolation | MIC (μg/mL) | |||||
|---|---|---|---|---|---|---|---|---|
| ABPC | EM | FF | OTC | LCM | TML | |||
| KMS2410SJ | Pseudocaranx dentex | Kumamoto | 3.13 | >400 | 6.25 | 3.13 | >400 | 400 |
| KMS240927SJ | Pseudocaranx dentex | Kumamoto | 1.56 | >400 | 6.25 | 3.13 | >400 | 400 |
| KMS241003SJ1 | Pseudocaranx dentex | Kumamoto | 3.13 | >400 | 6.25 | 1.56 | >400 | 400 |
| KMS241003SJ2 | Pseudocaranx dentex | Kumamoto | 3.13 | >400 | 6.25 | 3.13 | >400 | 400 |
| KMS241003SJ3 | Pseudocaranx dentex | Kumamoto | 3.13 | >400 | 6.25 | 3.13 | >400 | 400 |
| KMS240716SJ | Pseudocaranx dentex | Kumamoto | 3.13 | 0.39 | 6.25 | 3.13 | 200 | 200 |
| KMS240731A | Seriola dumerili | Kumamoto | 3.13 | 0.39 | 6.25 | 3.13 | 200 | 400 |
| MS240805SJ | Pseudocaranx dentex | Miyazaki | 3.13 | 0.39 | 6.25 | 3.13 | 200 | 400 |
| MS240808A | Seriola dumerili | Miyazaki | 3.13 | 0.39 | 6.25 | 3.13 | 200 | 400 |
| MS241016Y3 | Seriola quinqueradiata | Miyazaki | 3.13 | 0.39 | 6.25 | 3.13 | 200 | 400 |
ABPC: ampicilin, EM: erythromycin, FF: florfenicol, OTC: oxytetracycline, LCM: lincomycin, TML: tiamulin
The MICs for six chemotherapeutics, namely, EM, LCM, oxytetracycline (OTC), florfenicol (FF), ampicillin (ABPC), and tiamulin (TML) were examined using the broth microdilution method (Japanese Society of Antimicrobials for Animals, the Committee, 2003) with some modification. Briefly, bacterial concentration of approximately 104 CFU/well was inoculated into a microplate containing a drug and incubated at 25°C for 48 h.
Detection of erm(B) in EM-resistant L. garvieae serotype IIIEM-resistant strains (n = 5) grown on THA medium containing the concentration of EM (100 μg/mL) were used to detect erm(B), the EM-resistance gene. Bacterial DNA from all EM-resistant strains grown on the medium was extracted using a Kaneka Easy DNA Extraction Kit version 2 (Kaneka), according to the manufacturer’s protocol. The obtained DNA was used to detect erm(B) via PCR with primers and conditions as described previously (Maki et al., 2008). KGLA1504 ∆lsa(D) RFr (refer to “Conjugation transfer of EM-resistance trait, erm(B)” in Materials and methods) carrying a plasmid, pkh2101-encoded erm(B) (Akmal et al., 2023) was used as a positive control.
Biased sinusoidal field gel electrophoresis (BSFGE) analysisAll EM-resistant strains (n = 5) were used for BSFGE analysis, which was performed as described previously (Araki et al., 2024b). The restriction enzyme SmaI (Takara Bio) was used to separate the DNA molecules for BSFGE typing. BSFGE types A and B strains (MS210922A and KS220516A, respectively) of L. garvieae serotype III were used as references (Araki et al., 2024b).
Conjugation transfer of EM-resistance trait, erm(B)EM-resistant L. garvieae serotype II strains (n = 5) carrying erm(B) were used as donor strains for their transferable EM-resistance trait. The donor and recipient strains are listed in Table 2. The recipient strain L. formosensis; KGLA1504 ∆lsa(D) (Shi et al., 2021) was sub-cultured several times on THA in the presence of sub-inhibitory concentration of rifampicin (RF). After being transferred several times on medium containing increased sub-inhibitory concentration of RF, RF-resistant KGLA1504 ∆lsa(D) RFr (MIC = >400 μg/mL) was obtained as the recipient strain. Transferability of the EM-resistance trait was assessed using the filter mating method, as described in a previous study (Akmal et al., 2023). For screening the transconjugants, EM and RF were used at final concentrations of 100 and 300 μg/mL, respectively. Growth of all the EM-resistant L. garvieae serotype III strains was inhibited completely at 300 μg/mL RF. Representative transconjugants grown on the screening medium (100 μg/mL of EM and 300 μg/mL of RF) were also confirmed as the recipient via an agglutination test (n = 10) with antiserum against L. formosensis strain 121836 (Oinaka et al., 2015) and mPCR assay (n = 3) (Araki et al., 2024a). Furthermore, transconjugants (n = 3) were confirmed to detect erm(B) via PCR, as described previously. Two representative EM-resistant strains (KMS2410SJ and KMS240927SJ) were used to calculate transfer frequencies, according to a previous study (Akmal et al., 2023).
| Donors (L. garvieae serotype III) | Recipient (Lactococcus formosensis) | Transfer or transfer frequencies |
|---|---|---|
| KMS2410SJ | KGLA1504 ∆lsa(D) RFr | 3.6 × 10−3 |
| KMS240927SJ | 1.6 × 10−3 | |
| KMS241003SJ1 | + | |
| KMS241003SJ2 | + | |
| KMS241003SJ3 | + |
All EM-resistant strains (n = 5) showed strong agglutination with the clinical diagnostic antiserum (L. garvieae serotype III). In addition, the mPCR assay amplified around a 500 bp band to identify L. garvieae serotype III in all five EM-resistant strains (Fig. 1a).

a. Multiplex PCR (mPCR) results. The estimated PCR product sizes are 262 bp for L. garvieae serotype I, 1,007 bp for L. formosensis, and 500 bp for L. garvieae serotype III, respectively. The internal amplification control (16S rDNA) is 168 bp. Lane M: 100 bp DNA ladder marker. Lanes 1-5 correspond to EM-resistant strains (KMS2410SJ, KMS240927SJ, KMS241003SJ1, KMS241003SJ2, and KMS241003SJ3, respectively). Lane 6: L. garvieae serotype I reference strain (ATCC49156). Lane 7: L. formosensis reference strain (122061). Lane 8: L. garvieae serotype III reference strain (MS210922A).
b. Detection of erm(B) via PCR. The estimated PCR product of erm(B) is 738 bp. Lane M: 100 bp DNA ladder marker. Lanes 1-5 correspond to EM-resistant strains (KMS2410SJ, KMS240927SJ, KMS241003SJ1, KMS241003SJ2, and KMS241003SJ3, respectively). Lane 6: Positive control, KGLA1504 ∆lsa(D) RFr carrying pkh2101 plasmid-encoded erm(B).
The MICs for all five EM-resistant strains are listed in Table 1. All the strains showed high MIC values for EM and LCM (Table 1). A previous study showed the MIC values of EM ranged from 0.1 to 0.78 μg/mL for clinical strains (n = 153) isolated from 2021 to 2023 (Araki et al., 2024b). The MIC values of EM were more than 400 μg/mL in all five EM-resistant strains isolated in 2024. All EM-resistant strains with high MIC values were positive for erm(B) according to PCR amplification (Fig. 1b). All BSFGE patterns of the EM-resistant strains were similar to the type A profile (MS210922A) and coincided with each other (Fig. 2).

Selected transconjugants grown on the screening medium (100 μg/mL of EM and 300 μg/mL of RF) were confirmed to detect erm(B) via PCR. Furthermore, selected transconjugants were confirmed as recipients via mPCR to identify L. formosensis and agglutination tests with serotype II antiserum. The EM resistance trait of all EM-resistant strains was transferred to the recipient strain, L. formosensis KGLA1504 ∆lsa(D) RFr. The frequencies of conjugal transfer of the EM resistance trait in representative serotype III strains, KMS2410SJ and KMS240927SJ to a recipient cell were 3.6 × 10−3 and 1.6 × 10−3, respectively (Table 2).
The different fish pathogens causing lactococcosis, namely, Lactococcus garvieae serotype I, L. formosensis (formerly known as L. garvieae serotype II), and L. garvieae serotype III, have been isolated from fish farms in Japan. Emerging L. garvieae serotype III infections were first identified in 2021 and then spread to various fish farms (Minami et al., 2023). A previous study revealed that serotype III strains collected from various farms between 2021 and 2023 were all sensitive to EM (Araki et al., 2024b). In the fish farms of striped jack, EM has been used to control L. garvieae serotype III infection because it has been effectiveness since its first occurrence in fish farms. A previous study suggested the possible occurrence of EM-resistant L. garvieae serotype III in fish farms as well as other lactococcal infections in the future (Araki et al., 2024b).
In September 2024, EM-resistant strains of L. garvieae serotype III suddenly emerged in a striped jack farmed in Kumamoto Prefecture. In the future, EM-resistant strains are expected to spread to other fish farms. Thus, it is becoming difficult to control EM-resistant serotype III infections without the development of an effective vaccine for fish farms. This study provides basic and current information on EM-resistant L. garvieae serotype III.
A previous BSFGE analysis revealed that three BSFGE types of L. garvieae serotype III spread to fish farms. The main BSFGE types are types A and B. The study suggested some routes for spreading this emerging pathogen to fish farms (Araki et al., 2024b). In this study, BSFGE analysis of EM-resistant strains revealed that the electrophoretic patterns of all EM-resistant strains coincided with each other and were similar to that of type A. Therefore, the same genetic group of EM-resistant L. garvieae serotype III might spread to fish farms. Further MIC monitoring is strongly needed to prevent the spread of EM-resistance in fish farms.
The EM-resistance trait in serotype III can be transferred to recipient cells L. formosensis at a transfer frequency of approximately 10−3. Moreover, the EM-resistance trait was also transferred to a homologous EM-sensitive serotype III strain (data not shown). Genome analysis suggested that the EM-resistance gene erm(B) in the serotype III strains (KMS2410SJ and KMS240927SJ) was possibly located on a plasmid (data not shown). Further comparative genetic analysis between L. formosensis EM-resistance R-plasmid (Akmal et al., 2023) and serotype III EM resistance mobile genetic elements is in progress to clarify the evolution and transmittance of the EM-resistance trait.
Lactococcus garvieae serotype I infections have been observed in various fish species since 1974 (Kusuda et al., 1976). Aoki et al. (1983) performed a drug sensitivity test against 561 strains collected from 1974 to 1981 and revealed that no EM-resistant strains were found in fish farms. Subsequently, Aoki et al. (1990) revealed the strains resistant for macrolide antibiotics including EM, were first sampled from fish farms in 1986 and 1987. Based on these published records, around 12 years were needed to acquire the EM-resistance trait in L. garvieae serotype I since its first occurrence in fish farm. Furthermore, this resistance determinant was identified as two copies of erm(B) located on an R-plasmid, pKL0018 (Maki et al., 2009). On the other hand, EM-resistant L. formosensis R-plasmids were first identified in the previous study (Akmal et al., 2023) after the emergence of infection in fish farms in 2012 (Oinaka et al., 2015). EM resistant L. formosensis strains carried a resistant plasmid pkh2101 encoded with one copy of erm(B) (Akmal et al., 2023). This study revealed that serotype III required around 3 years to acquire the EM-resistance trait in fish farms; the drug resistance gene was also erm(B), which was the same as serotype I and L. formosensis. The EM-resistance of serotype III is thought to have spread to fish farms in near future. Vaccines are urgently needed to prevent EM-resistant L. garvieae serotype III from spreading in fish farms. It is important to continue the epidemiological surveillance of drug resistance to control the spread of EM-resistant L. garvieae serotype III in fish farms.
In conclusion, this is the first report on the emergence of EM-resistant L. garvieae serotype III isolated from diseased fish. The information provided by this study will be useful to monitor the EM-resistance of L. garvieae serotype III in fish farms.
This study was supported by Grants-in-Aid for Scientific Research from JSPS KAKENHI (grant number 21H02287) and Matsuoka Research Institute for Science. This study was also supported by the Food Safety and Consumer Affairs Bureau, Ministry of Agriculture, Forestry, and Fisheries of Japan. This work is partially supported by MAFF Commissioned project study on “Development of effective antibiotics administration methods for responding to the risk of new fish diseases which occur with promoting hatchery-reared juvenile in Seriola sp.” Grant Number JPJ 012045.