Occurrence of tetracycline resistance genes tet(M) and tet(S) in bacteria from marine aquaculture sites


*Corresponding author. Tel.: +81-89-9278552; fax: +81-89-9278552, E-mail address: ssuzuki@agr.ehime-u.ac.jp


Occurrence of tetracycline resistance genes encoding ribosomal protection proteins was examined in 151 tetracycline-resistant bacterial isolates from fish and seawater at coastal aquaculture sites in Japan and Korea. The tet(M) gene was detected in 34 Japanese and Korean isolates, which included Vibrio sp., Lactococcus garvieae, Photobacterium damsela subsp. piscicida, and unidentified Gram-positive bacteria. The majority of these bacterial isolates displayed high-level resistance with a minimum inhibitory concentrations (MICs) equal to or greater than 250 μg/ml of oxytetracycline and only four isolates had MICs less than 31.3 μg/ml. 16S rDNA RFLP typing of tet(M)-positive Vibrio isolates suggests that these are clonal populations of the same phylotype specific to a particular location. One Vibrio clone (phylotype III), however, is widely disseminated, being detected during different sampling years, at different locations, and in different fish species in both Japan and Korea. The tet(S) gene was detected in L. garvieae from yellowtail in Japan and in Vibrio sp. from seawater in Korea. This is the first report of tet(S) occurrence in Gram-negative facultative anaerobes. These results suggest that tet(M) and tet(S) genes are present in fish intestinal and seawater bacteria at aquaculture sites and could be an important reservoir of tetracycline resistance genes in the marine environment.


Since its discovery in 1948, tetracycline has been widely used in human and veterinary medicine, in food animal production for growth promotion and prophylaxis, and in horticulture [1]. One of the derivatives, oxytetracycline is widely used in aquaculture systems to treat and prevent bacterial diseases of fish and other marine animals. This wide use of oxytetracycline, however, has increased the occurrence of tetracycline resistant fish pathogens [2,3]. Presently, more than 40 different tetracycline resistance determinants have been reported [1,4,5]. Resistance to tetracyclines occurs via two primary mechanisms – one is the energy-dependent efflux of tetracycline from the cell and the other involves synthesis of an alternative elongation factor that protects the ribosome from tetracycline binding, termed ribosomal protection proteins (RPP) [1,6].

The occurrence of genes encoding efflux pumps has been demonstrated in various aquaculture environments [7–9], however, there are presently no reports of RPP genes associated with aquaculture. Among the RPP-encoding genes, the tet(M) gene is the most common and is frequently encountered in both Gram-positive and Gram-negative pathogens and mycoplasmas in clinical and terrestrial environments [1,10]. The only report of tet(M) incidence in the aquatic environment is described by Andreas and Anders [11], who detected tet(M) in Enterococcus spp. isolated from integrated and traditional fish-farm ponds in Thailand. Widespread presence of tetracycline resistance genes is thought to be due to broad-host-range conjugative plasmids and transposons that may play a significant role in dissemination of resistance among clinically important species [12,13]. The tet(M) gene is often associated with conjugative transposons belonging to the Tn1545–Tn916 family. The Tn1545–Tn916 family is a broad-host-range transposon, which can transfer between bacteria of at least 52 species in 24 different genera [1,12,13]. The integrase (Int) gene product of the Tn1545–Tn916 family is essential for its excision and integration into the chromosome [12,13], and provides a convenient genetic marker that can be used to determine the presence of these transposons in a particular bacterial isolate.

Another RPP gene, tet(S), has been detected in Gram-positive bacteria such as Listeria sp., Enterooccus sp. and Lactococcus lactis[14–17], and in the Gram-negative bacterium Veillonella spp. [18].

In this study, we report the detection of the RPP genes tet(M) and tet(S) and Int-Tn genes in bacterial isolates from fish and seawater at aquaculture sites in Japan and Korea.

2Materials and methods

2.1Isolation of tetracycline-resistant marine bacteria

A total of 151 tetracycline-resistant bacterial strains were isolated from fish and seawater at commercial aquaculture sites in Japan and Korea. The aquaculture sites in Japan used tetracycline in feed (30 mg/kg feed) for 20 days in the case of disease outbreaks (vibriosis, edwardsiellosis, aeromonasis, pseudomonasis, and streptococcosis), otherwise no antibiotics were fed. No antibiotic use information was available from the Korean producers. Fish for analysis were taken directly from cages and seawater was sampled directly on a farm. Samples were placed on ice and transported to the laboratory. All isolates were randomly chosen from independent colonies growing on a marine broth (MB) (Difco, Detroit, MI, USA) agar plate [19] supplemented with 30 μg/ml of oxytetracycline (Sigma Chemical Co, St. Louis, MO, USA). Isolation was performed along with a viable bacterial count from the samples. Table 1 lists 60 isolates from the intestines of yellowtail Seriola quinqueradiata and seawater sampled during 1999 in Japan. The isolates included 34 Vibrio sp., 3 Pseudomonas sp., 19 Moraxella sp., 3 Gram-positive bacteria, and one unknown isolate. From fish with bacterial infections, 33 isolates, including 15 Vibrio sp., 10 Edwaresiella tarda, 7 Lactococcus garivieae, and one Photobacterium damsela subsp. piscicida, were obtained (Table 2) [20]. These bacterial species are known pathogens of marine fish and were identified by slide agglutination test using antiserum specific to each species at the Kagawa Prefectural Fisheries Experimental Station, Japan. Table 3 shows 58 isolates from Korea, which were all identified as Vibrio sp.; 31 isolates were collected from several fish at different farms and 27 were from seawater [21]. All isolates listed in Tables 1 and 3 were identified according to Muroga et al. [22]. Bacteria were cultured at 25 °C on nutrient salt agar (NSA) plates [19] containing 5 g of polypepton (Wako Pure Chemical Industries Ltd., Osaka, Japan), 1 g of yeast extract (Difco), 1 g of protease peptone (Difco), 1 g of beef extract (Wako Pure Chemical industries Ltd.), 7.5 g of Bacto agar (Difco), and 20 g of NaCl per liter (pH 7.5) of distilled water.

Table 1.  MIC of oxytetracycline and detection of tet(M), Int-Tn and tet(S) in tetracycline-resistant bacteria isolated from intestine of healthy yellowtail and seawater in Japan in 1999
IsolateGroupFish no.OriginMIC (μg/ml)tet(M)Int-Tntet(S)RFLP profile
  1. aThe fish numbers represents the same individual fish.

  2. bND, not determined.

6Vibrio/Aeromonas3Yellowtail62.5NDPattern I
9Vibrio/Aeromonas3Yellowtail62.5NDPattern I
19Vibrio/Aeromonas3Yellowtail62.5NDPattern I
21Vibrio /Aeromonas4Yellowtail62.5NDPattern I
22Vibrio /Aeromonas4Yellowtail62.5NDPattern I
23Vibrio/Aeromonas4Yellowtail62.5NDPattern I
24Vibrio/Aeromonas4Yellowtail62.5NDPattern I
25Vibrio/Aeromonas4Yellowtail<31.3NDPattern I
26Vibrio/Aeromonas4Yellowtail62.5NDPattern I
27Vibrio/Aeromonas4Yellowtail62.5NDPattern I
29Vibrio/Aeromonas4Yellowtail62.5NDPattern I
30Vibrio/Aeromonas4Yellowtail62.5NDPattern I
31Vibrio/Aeromonas4Yellowtail<31.3NDPattern 1
33Vibrio/Aeromonas4Yellowtail62.5NDPattern I
34Vibrio/Aeromonas4Yellowtail62.5NDPattern I
35Vibrio/Aeromonas4Yellowtail62.5NDPattern I
37Vibrio/Aeromonas4Yellowtail62.5NDPattern I
44Vibrio/Aeromonas6Yellowtail250++Pattern II
48Vibrio/Aeromonas6Yellowtail250++Pattern II
50Vibrio/Aeromonas6Yellowtail250++Pattern II
52Vibrio/Aeromonas6Yellowtail500++Pattern II
54Vibrio/Aeromonas6Yellowtail250++Pattern II
56Vibrio/Aeromonas6Yellowtail500++Pattern II
58Vibrio/Aeromonas7Yellowtail500++Pattern II
59Vibrio/Aeromonas7Yellowtail250++Pattern II
62Vibrio/Aeromonas8Yellowtail250++Pattern II
71Vibrio/Aeromonas8Yellowtail250++Pattern II
74Vibrio/Aeromonas8Yellowtail250++Pattern II
156Vibrio/Aeromonas Seawater<31.3ND 
157Vibrio/Aeromonas Seawater<31.3ND 
158Vibrio/Aeromonas Seawater<31.3ND 
159Vibrio/Aeromonas Seawater<31.3ND 
77Moraxella Seawater<31.3ND 
80Moraxella Seawater<31.3ND 
81Moraxella Seawater<31.3ND 
82Moraxella Seawater<31.3ND 
224Moraxella Seawater<31.3ND 
225Moraxella Seawater<31.3ND 
227Moraxella Seawater<31.3ND 
228Moraxella Seawater<31.3ND 
230Moraxella Seawater<31.3ND 
232Moraxella Seawater<31.3ND 
288Unknown Seawater<31.3ND 
Table 2.  MIC of oxytetracycline and detection of tet(M), Int-Tn and tet(S) in tetracycline-resistant pathogenic bacteria isolated from diseased fish in Japan during 1997–2000
  1. aStriped beakperch (Oplegnathus fasciatus), red sea bream (Pagrus major), gold striped amberjack (Seriola lalnndi), tiger puffer (Takifugu rubripes), great amberjack (Seriola dumerili), sea bass (Lateolabraxjaponicus), Schlegel's black rockfish (Sebastes schlegeli), Japanese flounder (Paralichthys olivaceus). K, kidney; S, spleen; I, intestinal content; B, brain; L, liver.

  2. bNSA, nutrient salt agar.

  3. cBHI, brain heart infusion broth.

  4. dND, not determined.

IsolateGroup or speciesOriginaYearMIC (μg/ml)tet(M)Int-Tntet(S)RFLP profile
KIDX-97076Vibrio sp.Striped beakperch (K)1997<62.5NDdND 
KOHX-98043Vibrio sp.Yellowtail (K)1998250ND++Pattern II
KHOX-98101Vibrio sp.Yellowtail (K)1998125NDND 
KTX-98024Vibrio sp.Red sea bream (L)1998125ND+Pattern III
KIDX-98074Vibrio sp.Striped beakperch (L)1998<62.5NDND 
KHRX-98123Vibrio sp.Gold striped amberjack (K)1998250NDND 
KFX-98118Vibrio sp.Tiger puffer (L)1998<62.5NDND 
KKAX-99017Vibrio sp.Great amberjack (L)1999<62.5NDND 
KCSX-99104Vibrio sp.Sea bass (B)1999<62.5NDND 
KKUX-99102Vibrio sp.Schlegel's black rockfish (K)1999<62.5NDND 
KFX-99065Vibrio sp.Tiger puffer (K)1999125NDND 
KHX-00105Vibrio sp.Yellowtail brain2000<62.5NDND 
KFX-00014Vibrio sp.Tiger puffer (L)2000<62.5NDND 
KFX-00019Vibrio sp.Tiger puffer (K)2000250ND+Pattern III
KRX-00017Vibrio sp.Japanese flounder (L)2000250ND+Pattern III
KRE-97013Edwardsiella tardaJapanese flounder (K)1997ND250ND 
KRE-97068Edwardsiella tardaJapanese flounder (K)1997ND250ND 
KRE-97120Edwardsiella tardaJapanese flounder (K)1997ND<31.3ND 
KRE-98063Edwardsiella tardaJapanese flounder (L)1998ND125ND 
KRE-98161Edwardsiella tardaJapanese flounder (L)1998ND<31.3ND 
KRE-99017Edwardsiella tardaJapanese flounder (L)1999ND250ND 
KRE-99090Edwardsiella tardaJapanese flounder (L)1999ND250ND 
KRE-99122Edwardsiella tardaJapanese flounder (L)1999ND125ND 
KRE-00049Edwardsiella tardaJapanese flounder (K)2000ND250ND 
KRE-00115Edwardsiella tardaJapanese flounder (K)2000ND250ND 
KHS-97051Lactococcus garvieaeYellowtail (B)1997ND250+++ 
KHS-98032Lactococcus garvieaeYellowtail (B)1998ND250+++ 
KHS-98057Lactococcus garvieaeYellowtail (B)1998ND250++ 
KHS-99008Lactococcus garvieaeYellowtail (B)1999ND250+++ 
KHS-99047Lactococcus garvieaeYellowtail (B)1999ND250+++ 
KHS-00005Lactococcus garvieaeYellowtail (K)2000ND250+++ 
KHS-00083Lactococcus garvieaeYellowtail (B)2000ND250+++ 
KHOPA-00030Photobacterium damsela subsp. piscicidaYellowtail (L)2000ND<31.3++ 
Table 3.  MIC of oxytetracycline and detection of tet(M), Int-Tn and tet(S) in Vibrio sp. isolated from cultured fish in Korea during 2000–2002
  1. aRock fish (Sebastes schlegeli), rock bream (Oplegnathus fasciatus), striped mullet (Mugil cephalus). K, kidney, S, spleen, I, intestinal content.

  2. bND, not determined.

IsolateOriginaYearAreaMIC (μg/ml)tet(M)Int-Tntet(S)RFLP profile
FK0001Flounder (K)2000Yosu<31.3NDb 
SK0102Sea bass (K)2001Yosu<31.3ND 
RFK01041Rock fish (K)2001Yosu<31.3ND 
RFS01042Rock fish (S)2001Yosu<31.3ND 
FK01073Flounder (K)2001Wando250+Pattern III
FK0103Flounder (K)2001Youngkwang<31.3ND 
FK01071Flounder (K)2001Youngkwang<31.3ND 
FK01072Flounder (K)2001Youngkwang125ND 
FI0203Flounder (I)2002Youngkwang<31.3ND 
FI0204Flounder (I)2002Youngkwang125ND 
FI0205Flounder (I)2002Youngkwang250ND 
FI0206Flounder (I)2002Youngkwang125ND 
FI0207Flounder (I)2002Youngkwang<31.3ND 
FI0208Flounder (I)2002Youngkwang<31.3ND 
FI0209Flounder (I)2002Youngkwang250++Pattern IV
RI0210Rock bream (I)2002Youngkwang<31.3ND 
RI0211Rock bream (I)2002Youngkwang62.5ND 
RI0212Rock bream (I)2002Youngkwang<31.3ND 
FI0213Flounder (K)2002Wando<31.3ND 
FI0214Flounder (K)2002Jindo<31.3ND 
FI0215Flounder (K)2002Jindo<31.3ND 
FI0216Flounder (I)2002Jindo<31.3+Pattern V
FI0217Flounder (I)2002Jindo<31.3ND 
FI0218Flounder (I)2002Jindo<31.3++Pattern VI
FI0219Flounder (I)2002Jindo<31.3++Pattern VI
FI0220Flounder (I)2002Jindo<31.3ND 
FI0221Flounder (I)2002Jindo<31.3ND 
FI0222Flounder (I)2002Jindo<31.3ND 
FI0223Flounder (I)2002Jindo<31.3ND 
SMI0224Striped mullet (I)2002Mokpo<31.3ND 
SMI0225Striped mullet (I)2002Mokpo<31.3ND 
SW01071Sea water2001Youngkwang<31.3ND 
SW01072Sea water2001Youngkwang250ND 
SW01073Sea water2001Youngkwang500ND 
SW01074Sea water2001Youngkwang<31.3ND 
SW01075Sea water2001Youngkwang<31.3ND 
SW01076Sea water2001Youngkwang250ND 
SW0201Sea water2002Youngkwang62.5ND 
SW0202Sea water2002Youngkwang250ND 
SW0203Sea water2002Youngkwang125++Pattern VII
SW0204Sea water2002Youngkwang<31.3ND 
SW0205Sea water2002Youngkwang<31.3ND 
SW0206Sea water2002Youngkwang250ND 
SW0207Sea water2002Youngkwang<31.3ND 
SW0208Sea water2002Youngkwang<31.3ND 
SW0209Sea water2002Youngkwang<31.3ND 
SW0210Sea water2002Youngkwang250ND 
SW0211Sea water2002Youngkwang<31.3ND
SW0212Sea water2002Youngkwang250+Pattern III
SW0213Sea water2002Youngkwang<31.3ND
SW0214Sea water2002Mokpo<31.3ND
SW0215Sea water2002Mokpo<31.3ND
SW0216Sea water2002Jindo125+Pattern VIII
SW0217Sea water2002Jindo<31.3ND
SW0218Sea water2002Jindo<31.3ND
SW0219Sea water2002Yosu<31.3ND
SW0220Sea water2002Yosu<31.3ND
SW0221Sea water2002Yosu<31.3ND

2.2Determination of minimum inhibitory concentration

The minimum inhibitory concentration (MIC) of oxytetracycline for all isolates was determined as follows. NSA plates were used for the Vibrio, Pseudomonas and Moraxella isolates. Brain heart infusion (BHI) agar plates supplemented with 1.5% NaCl were used for the E. tarda, L. garvieae, P. damsela subsp. piscicida, and Gram positive isolates. Bacterial cell suspensions were prepared in phosphate-buffered saline (PBS) and the cell optical density was adjusted to MacFarland No. 0.5 with PBS. One microliter portions of the suspension were then spotted on agar plates containing 31.3, 62.5, 125 and 250 μg/ml of oxytetracycline(Sigma), respectively. The plates were incubated at 25 °C for 48 h, and the lowest concentration needed for inhibition was recorded.

2.3DNA extraction

For chromosomal DNA extraction, cultures were plated on NSA agar and incubated for 24 h. After harvesting, cells were washed and suspended in 200 μl of lysozyme solution (0.15 M NaCl, 0.1 M EDTA, 0.5 μg/ml RNase A, 50 μg/ml lysozyme). The mixture was incubated at 37 °C for 30 min, and then 500 μl of lysis buffer (0.15 M NaCl, 0.1 M EDTA, 0.5% SDS) was added. DNA was purified by extraction with phenol saturated with TE buffer [10 mM Tris–HCl (pH 8.0) and 1 mM EDTA], then with TE-saturated phenol:chloroform:iso-amyl alcohol (25:24:1, v/v/v). The mixture was centrifuged at 15,000 rpm for 5 min, and the water phase was extracted once with chloroform. The DNA was finally precipitated with ethanol, washed with 70% ethanol, dried, and suspended in TE buffer.

Plasmid DNA was isolated from E. coli JM109 harboring pFD310 and pAT451 [6] as positive controls for tet(M) and tet(S), respectively, by the method of Sambrook and Russell [23]. Cells were grown in LB medium at 37 °C for 14 h, centrifuged, re-suspended in 100 μl of glucose buffer (50 mM glucose, 25 mM Tris–HCl (pH 8.0), and 10 mM EDTA), and then 200 μl of 0.2 M NaOH and 1% SDS solution were added. The tube containing the mixture was placed on ice, and 150 μl of 3 M sodium acetate (pH 5.2) was added. The contents were thoroughly mixed by inverting and placed on ice for 5 min. The mixture was then centrifuged at 15,000 rpm for 5 min, an equal volume of phenol:chloroform:iso-amyl alcohol (25:24:1, v/v/v) solution was added to the supernatant, and the mixture was further centrifuged at 15,000 rpm for 2 min. The plasmid DNA was finally precipitated with cold ethanol. Plasmid DNA was suspended in TE buffer (pH 8.0) containing 20 μg/ml RNase A. Each sample of purified chromosomal and plasmid DNAs was resuspended in 50 μl of TE buffer and stored at −20 °C until analyzed.

2.4Polymerase chain reaction

First, polymerase chain reaction (PCR) primers (Ribo2-FW and Ribo2-RV) targeting the RPP gene family were used [6], to generate amplicons which were then sequenced. The sequences obtained were more than 99% identical to those of the tet(M) and tet(S) genes published in GenBank (http://www.ncbi.nlm.nih.gov/Genbank/index.html) (data not shown). Primers specific for tet(M) and tet(S) were used to confirm the results obtained by sequencing. For positive and negative controls, plasmids pFD310 [tet(M)], pGEM-tetO [tet(O)], pJIR667 [tetB(P)], pAT451 [tet(S)], pBT-1 [tet(Q)] and pGEM-tetW [tet(W)] were used. PCR primers for tet(M) were tet(M)-FW; 5′-GTTAAATAGTGTTCTTGGAG-3′ and tet(M)-RV; 5′-CTAAGATATGGCTCTAACAA-3′, which gave a 656-bp specific amplicon [24]. For tet(S), tet(S)-FW; 5′-CATAGACAAGCCGTTGACC-3′ and tet(S)-RV; 5′-ATGTTTTTGGAACGACAGAG-3′ were used, which gave a 667-bp specific amplicon [25]. Plasmids pFD310 and pAT451 were used as positive controls for tet(M) and tet(S), respectively [6]. PCR primers for the integrase gene of transposon (Int-Tn) were Int1: 5′-TGACACTCTGCCAGCTTTAC-3′ and Int2: 5′-CCATAGGAACTTGACGTTCG-3′, which gave a 579-bp fragment of the Int-Tn gene of Tn1545–Tn916. [26].

A typical PCR reaction mixture for detection of tet(M), tet (S) and Int-Tn contained 20 pmol of each primer, 2.5 μl of 10× ExTaq reaction buffer (TaKaRa, Kyoto, Japan), 2 μl of 100 μM dNTP solution from the same kit, 1.0 U of ExTaq DNA polymerase (TaKaRa), and 1 μl (50–100 ng) of template DNA, adjusted to 25 μl with sterile distilled water. PCR amplification was conducted in a GeneAmp PCR system 9700 (Applied Biosystems, Foster City, CA, USA) thermal cycler and utilized 25 cycles [denaturation at 95 °C for 30 s, annealing at 56 °C (tet(M)) or 59 °C (tet(S) and Int-Tn) for 30 s, and extension at 72 °C for 1 min]. A final extension was performed at 72 °C for 5 min. The amplified products were analyzed by electrophoresis on a 1.5% agarose gel and stained with ethidium bromide.

2.5Restriction fragment length polymorphism

To investigate the relatedness of isolates in the Vibrio/Aeromonas group, all tet(M)-positive Vibrio/Aeromonas isolates were analyzed by PCR-fragment length polymorphism (RFLP) of 16S rDNA. In addition, tet(M)-negative Vibrio/Aeromonas isolates from fish Nos. 3 and 4 were also analyzed. Primers described by Weisburg et al. [27] were used for PCR amplification, which gave a ∼1.5 kb (between 8 and 1512; E. coli numbering) amplicon of bacterial 16S rDNA. The fD1 primer was 5′-AGAGTTTGATCCTGGCTCAG-3′ and rP2 primer was 5′-ACGGCTACCTTGTTACGACTT-3′. Conditions consisted of 25 cycles of denaturation at 95 °C for 2 min, annealing at 45 °C for 30 s, and extension at 72 °C for 4 min, followed by a final 7 min extension at 72 °C.

Approximately 0.5–1 μg of PCR products were digested with restriction endonucleases at 37 °C for 60 min according to the manufacturer's instructions. The following enzymes recommended for typing of this group [28] were used: HhaI (TaKaRa,), MspI (TaKaRa) and RsaI (Invitrogen, Carlsbad, CA, USA). Digested DNA fragments were analyzed by horizontal electrophoresis on a 3.5% NuSieve agarose gel (Cambrex Bio Science, Inc. Rockland, ME, USA) in TAE electrophoresis buffer (40 mM Tris, 20 mM acetate, 2 mM EDTA) and staining with ethidium bromide.



The MIC values for the bacterial isolates are shown in Tables 1–3. Of 45 isolates from the intestines of healthy yellowtail, 33 had MIC greater than 62.5 μg/ml of oxytetracycline (Table 1). Among 33 isolates from the diseased fish, 22 had MIC values greater than 62.5 μg/ml, but others were more sensitive to oxytetracycline (Table 2). For the Korean strains, which were uniformly identified as Vibrio, 7 out of 31 isolates from fish and 10 out of 27 isolates from seawater had MIC values greater than 62.5 μg/ml (Table 3) of oxytetracycline.

3.2Detection of tet(M), tet(S), and Int-Tn

Fifteen strains isolated from the intestines of healthy yellowtail were positive for tet(M) (Table 1); they included 12 Vibrio/Aeromonas and 3 Gram-positive isolates. All tet(M) positive isolates had MIC values greater than 250 μg/ml of oxytetracycline. Most of the isolates from fish Nos. 6, 7 and 8, which were sampled simultaneously at the same site, tested positive for tet(M) and had the same 16S rDNA RFLP profile (pattern II in Fig. 1(a)). All tet(M)-negative isolates from fish Nos. 3 and 4, isolated at the same time and location, showed a different RFLP profile (pattern I in Fig. 1(a)). The Int-Tn gene was detected in 13 out of 15 tet (M)-positive isolates (Table 1). The Japanese isolates with low MIC values, from both fish and seawater, tested negative for tet(M). The tet(S) gene was not detected in any Gram-negative isolate from Japan.

Figure 1.

16S rDNA RFLP profiles of Vibrio sp. produced with HhaI (H), MspI (M) and RsaI (R). (a) Isolates from the intestine of yellowtail (Table 1). (b) Isolates from diseased fish (Table 2). (c) Isolates from Korean fish (Table 3). (d) Isolates from Korean seawater (Table 3). Lane m: DNA size marker (100-bp DNA Ladder, New England Biolabs, Beverly, MA, USA). Roman numbering in the bottom of gels indicates a specific restriction profile.

In isolates from the diseased fish, 12 isolates tested positive for tet(M); they included four Vibrio strains, seven L. garvieae strains, and one P. damsela subsp. piscicida strain (Table 2). Results of 16S rDNA genotyping of the four Vibrio sp. isolates demonstrated the presence of two phylotypes (patterns II and III, Fig. 1(b)). The strain KOHX-98043 isolated from yellowtail in 1998 (Table 2) showed the same RFLP pattern (II) observed in Vibrio/Aeromonas isolates in 1999 (Table 1). The Int-Tn was gene detected in 8 out of 11 tet(M)-positive isolates. The tet(M)-positive strains were isolated from different fish species and in different years (Table 2). The tet(S) gene was detected only in L. garvieae that originated from diseased yellowtail sampled during 1997–2000 (Table 2).

Among the Korean isolates, 8 isolates were tet(M) positive; they included 5 fish and 3 seawater isolates (Table 3). The 16S rDNA genotyping of the tet(M)-positive Vibrio isolates produced six patterns (phylotypes III, IV, V, VI, VII, and VIII in Fig. 1(c) and (d)). Strains FK01073 and SW0212, isolated from flounder and seawater in Korea, respectively, showed the same pattern (pattern III) observed in Japanese Vibrio isolates (KTX-98024, KFX-00019, and KRX-00017 in Table 2). The Int-Tn gene was detected in 4 out of 8 tet (M)-positive isolates. The tet(S) gene was detected in one isolate from seawater, which had the greatest MIC (Table 3).


We examined the occurrence of tet(M) and tet(S) genes among 151 tetracycline-resistant bacterial isolates from healthy and diseased fish and seawater at coastal aquaculture sites in Japan and Korea between 1997 and 2002. These genes have not been previously detected in marine fish or in the marine environment, although tet(M) has been reported in fish from a freshwater pond [11]. This study is the first to show that these genes are present in healthy and diseased fish and in seawater, suggesting that marine aquaculture sites could serve as a reservoir of antibiotic resistance genes, in particular for tetracycline resistance genes.

At the Japanese sampling sites, the majority of bacterial isolates from healthy fish demonstrated intermediate and high level of resistance to oxytetracycline while among the seawater isolates the MIC values were generally lower (Table 1). We focused our efforts on characterization of bacteria with high levels of resistance and found that majority of them carried tet(M) (Table 1). In particular, isolates from fish Nos. 6, 7, and 8 were tet(M)-positive, except for a Pseudomonas isolate from fish No. 6 and a Vibrio/Aeromonas isolate from fish No 7. We examined 16S rDNA RFLP profiles in the Vibrio/Aeromonas group to determine if this is a clonal distribution of a tet (M)-positive strain or if tet(M) is present in a genetically unrelated populations. All tet(M)-positive Vibrio/Aeromonas isolates demonstrated the same RFLP pattern (pattern II in Fig. 1) suggesting that this a clonal distribution of the same strain in several fish sampled at the same location. In addition, nearly all of these isolates were positive for Int-Tn (Table 1). It should be noted that despite having the same RFLP pattern, this is not necessarily the same clone in all instances. For example, we were unable to amplify the Int-Tn-specific sequence from isolate No. 59 and the phenotypic characteristic of MIC showed a degree of heterogeneity within the group exemplified by the elevated MIC values in isolates Nos. 52, 56, and 58 (Table 1). Unidentified Gram-positive bacteria carrying tet(M) and positive for Int-Tn were also isolated suggesting that at least two taxonomically distinct species in healthy fish intestines may share the same tet marker, possibly through a transposon-mediated transfer event. No attempts to confirm the possibility of conjugal transfer between these species under laboratory or model conditions were undertaken during this study but this will be tested in our future research.

In diseased fish, tet(M) was detected in a broader range of bacteria including Vibrio sp., L. garvieae, and P. damsela subsp. piscicida (Table 2). This may be reflective of the broader range of fish species and organ sampling points used in this analysis. In pathogenic Vibrio species, simultaneous detection of tet(M) and Int-Tn occurred less frequently than in healthy fish suggesting that acquisition of tet (M) in pathogenic bacteria may involve different routes of transmission than in bacteria inhabiting the intestines of healthy fish. In L. garvieae, in addition to tet(M), all isolates carried tet (S) and, with one exception, also tested positive for Int-Tn. It is also noteworthy that these bacteria were consistently isolated during a four-year period, from 1997 to 2000, thus being not only a reservoir of tet resistance genes but also an endemic source of fish streptococcosis, for which L. garvieae is one of the principal agents. In addition, L. garvieae is both a fish and mammalian pathogen [31], causing mastitis in cows [32]. These bacteria are widely distributed in various environments and have been isolated from human blood, urine, skin and liver abscesses [32]. Therefore, given its broad host range, L. garvieae could be an antibiotic resistance vector between clinical, terrestrial and marine environments. The Vibrio strain KTX-98024 exhibiting phylotype II and carrying tet(M) and Int-Tn markers was originally isolated from the kidney of a diseased yellowtail in 1998 and the same phylotype was detected in the intestines of healthy yellowtails in 1999 in different locations (Table 1) suggesting possible dissemination of a particular Vibrio clone among geographically diverse cultured yellowtail populations.

At the Korean sampling sites, the MIC levels of oxytetracycline were generally lower, which may reflect a different antibiotic uptake regimen at these sites. It is interesting to note that all Korean tetracycline-resistant isolates were identified as belonging to the Vibrio genus, while at the Japanese sites the taxonomic range of tetracycline-resistant bacteria from healthy fish included Pseudomonas sp., Moraxella sp., and Gram-positive bacteria. However, tet(M) was detected in a broader range of healthy fish species and in seawater samples (Table 3). Interestingly, the RFLP patterns of the Korean tet(M)-positive Vibrio isolates are more diverse (six different phylotypes), than that of the Japanese Vibrio isolates from healthy fish which are dominated by just two phylotypes, with no similarity to the Korean isolates. At the same time, a similar phylotype (pattern III, Fig. 1) was detected among the Vibrio isolates from healthy fish (FK01073) and seawater (SW0212) at two locations in Korea (Table 3) and in diseased fish (KTX-98024, KFX-00019, and KRX-00017) at the Japanese sampling sites (Table 2). The three other studied loci are also similar with regard to this phylotype, e.g., it carries the tet(M) marker and is negative for tet (S) and Int-Tn. This may be the case of a clone that has disseminated to different geographical locations and different fish species within the marine environment.

Also at the Korean location, tet(S) was detected in tet(M) negative Vibrio isolated from seawater (Table 3). Historically, the tet(S) gene has been found predominantly in Gram-positive bacteria such as Listeria monocyto g enes and Enterococcus faecalis isolated from humans [16,17] and L. lactis isolated from cheese [16]. In Gram-negative bacteria, only Veilonella, an oral bacterium, is known to possess tet(S) [18]. Now this range is extended and includes a marine Vibrio species as well.

These preliminary results clearly indicate that tet(M) can be consistently encountered in the gastrointestinal tract (GIT) microflora of healthy fish, in isolates from other organs in diseased fish as well as in the water of aquaculture sites from different geographic locations. Tet(M) was originally described in streptococci [29,30] and subsequently in a broad variety of both Gram-positive and Gram-negative bacteria [1,10]. Villedieu et al. [18] reported that tet(M) was the most common tetracycline resistance gene in several bacterial species. Our results are consistent with this observation – while tet(M) was widely distributed among bacteria isolated from aquaculture sites, screening of these isolates for other RPP genes with primers targeting tet(M), tet(O), tetB(P), tet(S), tet(Q), and tet(W) yielded only tet(M) and tet(S)-specific amplicons. Frequently, these two genes were associated with Int-Tn encouraging speculation that conjugative transposons of the Tn1545–Tn916 family could be the vehicles for dissemination of tet(M) and tet(S) in marine bacteria, but this hypothesis requires further research to be validated. To our knowledge, this is the first study to report the occurrence of the tet(M) and tet(S) genes in a range of GIT and marine bacteria, isolated from healthy fish intestines, various diseased fish organs and seawater.

Representatives of the Vibrio genus are among the predominant gut microflora in marine fish [33,34] and in the marine environment. To date, tet(A), tet(B), tet(C), tet(D), tet(E), tet(G) and tet(34) have been detected in the genus Vibrio in marine environments [1,9,19,21], primarily genes that encode tetracycline efflux pumps. Now, with the addition of the RPP genes tet(M) and tet(S) to this list, it becomes apparent that Vibrio species may be an important reservoir of a diverse group of tetracycline resistance genes conferring resistance by different mechanisms. Uncovering the genetic mechanisms behind the transfer and dissemination of these genes by Vibrio and other marine species would be an important direction for future research.


We thank Dr. R. I. Aminov from the Rowett Research Institute, Aberdeen, UK, for providing tet plasmids and for editorial help and Dr. T. Isshiki, Kagawa Prefectural Fisheries Experimental Station, Japan, for providing the isolates in Table 2. Editorial help of this manuscript reviewer is greatly appreciated. We also thank Dr. J. Bower for his critical review of this paper. This work was partly supported by Grants-in-Aid from MEXT, Japan (14208063 and 15201006), and the 21st Century COE Program.