Isolation and Characterization of a Vibrio Sp. Strain MA3 Possibly Associated with Mass Mortalities of the Pearl Oyster Pinctada fucata


 In the summers of 2019 and 2020, a previously undescribed disease occurred in both juvenile and adult shellfish, causing mass mortalities in cultured pearl production, characterized by the major symptom of extreme atrophy of the soft tissues, including the mantle. However, the causative organism was uncertain. We isolated Vibrio sp. strain MA3 from the mantles of diseased pearl oysters Pinctada fucata. Analyses of 16S rRNA gene and DNA gyrase sequence homologies and its biochemical and morphological characteristics suggested that strain MA3 is a new strain of Vibrio alginolyticus. In addition, a hemolysin gene (Vhe1) of strain MA3 was detected as one of the virulence factors, and the complete sequence was determined. BLAST searches showed that Vhe1 shares 99.8% nucleotide sequence identity with Vibrio alginolyticus strain A056 lecithin-dependent hemolysin (ldh) gene, complete cds. Experimental infection of healthy oysters via injection with strain MA3 indicated it could cause high mortalities of the typically affected oysters from which the strain was isolated. These results suggest that the newly isolated Vibrio sp. strain MA3 is a putative causal agent of the recent disease outbreaks in Akoya pearl oysters.


Introduction
The cultured pearl industry was started by Mikimoto Kokichi in 1893 [Nagai 2013], and pearl production has become an important export industry in Japan. However, in the summers of 2019 and 2020, a previously undescribed disease occurred in not only juveniles but also in adult shell sh, causing mass mortalities. The disease was characterized by its major symptom: an extreme atrophy of the soft tissues, including the mantle.
News reports stated that at least 2 million juvenile shell sh died in Mie Prefecture alone, resulting in a loss of more than 300 million yen. Although measures to prevent this disease were urgent, the cause was still unknown.
In this study, we describe the isolation and characterization of another pathogenic bacterium as a rst step towards resolving the recent disease outbreaks in cultured pearl oysters, as characterized by extreme atrophy of their soft tissues. Furthermore, we reproduced the disease by experimentally infecting healthy oysters, via injection, with the isolated Vibrio sp. strain MA3.

Materials And Methods
Isolation and characterization of the pathogenic bacterium Five diseased pearl oysters Pinctada fucata martensii (Akoya) (Fig. 1) and one healthy oyster (average body weight 50.2 g) were collected from the pearl oyster farm in Ago Bay in Shima City, Mie Prefecture, Japan. After external and internal observations of the oysters, the mantle, gills, and adductor muscle were punctured with a platinum loop and streaked on AO agar plates (5% tryptone, 0.5% yeast extract, 0.2% beef extract, 0.2% sodium acetate, and 1.5% agar) with 70% seawater [Pazos et al. 1996], and the plates were incubated at 25 °C for 48 h. The dominant bacterial colonies on the plates were then spread on new AO agar medium and isolated. Finally, any bacterial clones were preserved on AO agar slants or plates for use in this study.
First, the morphology and motility of the isolated bacteria were observed by phase-contrast microscopy (Olympus BX51; Tokyo, Japan). Next, the samples were prepared for examination by negative staining with 1% w/v phosphotungstic acid and then observed under a JEM-100SX electron microscope (JEOL, Tokyo, Japan).
The ampli ed products were puri ed using a QIAquick PCR Puri cation Kit (Qiagen K.K., Tokyo, Japan), followed by direct sequencing using a BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, Foster City, CA, USA) and an ABI Prism 3130xl Genetic Analyzer (Applied Biosystems). The following primers were used; 16S rRNA gene: r1L (GTATTACCGCGGCTGCTGG), r2L (CATCGTTTACGGCGTGGAC), r3L (TTGCGCTCGTTGCGGGACT), r4L (ACGGGCGGTGTGTACAAG), f1L (GAGTTTGATCCTGGCTCAG), f3L (GTCCCGCAACGAGCGCAAC), and 926f (AAACTCAAAGGAATTGACGG), GyrB: UP-1S (GAAGTCATCATGACCGTTCTGCA) and UP-2Sr (AGCAGGGTACGGATGTGCGAGCC). Sequence similarities between the isolates and the sequences available in public databases were then searched using the BLAST program on the National Center for Biotechnology Information (NCBI) website (http://www.ncbi.nlm.nih.gov). Gram-staining was performed using a FAVOR-G SET-S kit (Nissui Pharmaceutical Co. Ltd., Tokyo, Japan). Biochemical characterization was performed using API 20E and API ZYM tests (bioMérieux, Marcy-l'Etoile, France). Growth at 5-55 °C on AO medium was measured for 2 days to determine the optimal temperature and temperature range for growth. Growth at various concentrations of seawater (0-90% v/v) was investigated for 3 days by supplementing the medium with appropriate concentrations of seawater. The effect of antibiotics on cell growth was assessed after 2 days of incubation at 25 °C, using BD BBL Sensi-Disc Antimicrobial Susceptibility Test Discs (Becton, Dickinson and Company, Franklin Lakes, New Jersey, USA) on AO agar plates.

Isolation of hemolysin gene
The partial hemolysin gene fragment was PCR-ampli ed using the primer pair tlhF (TCTAGCGAACGAGAACGCAG) and tlhR (ACTGCCCAGTTGTATAGCGG), designed on the basis of Vibrio alginolyticus hemolysin [Lv et al. 2019]. The ampli cation products were puri ed and sequenced following the same method as gyrB gene sequencing. The full lengths of the hemolysin gene were sequenced by the primer walking [Szybalski 1993] and inverse PCR [Howard et al. 1988] methods. Restriction digests were performed using 3 μg of genomic DNA treated with 40 U of BamHI. One nanogram of digested DNA was then circularized with a Rapid DNA Dephos & Ligation Kit (Roche, Manheim, Germany). The ligated sample was treated with an equal volume of phenol-chloroform-isoamyl alcohol mixture (25:24:1) before the aqueous phase was recovered. The DNA was precipitated with ethanol and collected by centrifugation. The entire genes were ampli ed by nested PCR. The two sets of primers used were designed from the partially sequenced hemolysin gene. Primer sequences were as follows: 1st-f, TCTTTAACGCGTCCCAATGG; 1st-r, CTATCGCCTAAAGCCACTAC; 2nd-f, CTAACCCGAATAGCTGGTTC; 2nd-r, GATCTGGTTGCATTGCAGCA. The rst ampli cation reaction was performed in a 40-μl PCR reaction mixture containing 1 × PCR buffer for KOD FX (Toyobo, Osaka, Japan), 0.3 μmol l -1 of each primer, 0.4 mmol l -1 of dNTPs, 1 U of KOD FX (Toyobo), and 1 ng of precipitated DNA. The PCR ampli cation was performed using an initial denaturation step of 94 °C for 2 min, followed by 30 cycles at 98 °C for 10 sec, 50 °C for 30 sec, and 68°C for 4 min, before the temperature was maintained at 4°C. In the second PCR, 1 μl of the ampli ed product of the rst PCR was used as a template, and the reaction cycle was carried out under the same conditions. The resulting PCR products were also puri ed and sequenced as described for the gyrB gene. Nucleotide and deduced amino acid sequence analyses, open reading frame (ORF) searches, and molecular-mass and isoelectric-point calculations were performed using Genetyx Ver.8 software (Genetyx Corporation, Tokyo, Japan). A database homology search was performed using the BLAST program on the NCBI website.

Experimental infection of pearl oysters with the isolated bacterium
For the infection-challenge experiment, apparently healthy pearl oysters (average body weight 47.6 g), in which the present disease was not detected, were obtained from a private hatchery in Mie Prefecture. Oysters were allowed to adapt in arti cial seawater for 19 days at 25 °C or 6 days at 28°C. Prior to use, 10 oysters were randomly selected and bacteriologically tested to con rm the absence of Vibrio species. The bacterial strain originally isolated from a diseased pearl oyster in Ago Bay, Mie Prefecture, was selected as the representative strain. The bacteria were grown in AO medium at 25 °C for 24 h with shaking. After sterilization of the shell surface with 70% ethanol, these oysters were then divided into two test groups and two control groups (n = 10 each group). In the test groups, the adductor muscle of the oysters was injected with 100 μl of the bacterial suspension (1.2×10 8 or 1.7×10 8 CFU/oyster) in AO medium. The control groups were treated similarly but with sterile AO medium instead of bacterial suspension. After the infection challenge, the oysters in each group (n = 10) were maintained individually in 3-l polyethylene beakers containing arti cial seawater, at 25 °C for 30 days or at 28 °C for 14 days, without feeding. The water was partially changed (50%) twice weekly. Each group was monitored daily for mortalities over the experimental period. We collected the dead individuals during the experiment and the surviving individuals at the end of the experiment, and isolated the bacteria from the gills, mantle, and adductor muscles of individuals following a described above procedure of bacterial examination of naturally diseased oysters. Slide agglutination tests with rabbit anti-MA3 serum were conducted to identify the isolates.

Nucleotide sequence accession number
The GenBank/EMBL/DDBJ accession numbers of the nucleotide sequence of the 16S rRNA gene, gyrB gene and hemolysin gene (Vhe1) of strain MA3 reported in this paper are LC628646, LC619648 and LC619649, respectively. During periods of high seawater temperatures (summer) in Japan in 2019 and 2020, a serious disease outbreak occurred in Akoya pearl oyster Pinctada fucata farms, causing mass mortalities. The diseased oysters exhibited extremely atrophied soft tissues, including the mantle and gills, accompanied by blackish discoloration in part. We considered that this disease was caused by a bacterial infection and thus tried to isolate the causative organism from diseased oysters. Bacteria forming similar white colonies were isolated from the mantle, gills, and adductor muscles of all diseased oysters, but not from any of the healthy oysters. We characterized one of the isolated bacterial strains (MA3) in detail.

Results And Discussion
Strain MA3 is a motile gram-negative bacterium with an average 1.5-µm wide and 3.0-µm long, and has a single lateral agellum (Fig. 2). The 1489-bp 16S rRNA gene and 1197-bp gyrB gene sequence of strain MA3 were used to identify the strain to species. The sequences showed 99.9% similarity with Vibrio alginolyticus strain NBRC 15630 (GenBank accession number NR_121709) and 99.6% similarity with Vibrio alginolyticus strain VABZ0005 (GenBank accession number JQ698511), respectively. We designated the isolate as Vibrio sp. strain MA3. The 16S rRNA gene and gyrB gene sequences of MA3 were deposited in GenBank under accession no. LC628646 and LC619648, respectively.
Bacteria of genus Vibrio are abundant in marine waters [Thompson et al. 2004]. The Harveyi clade is also known as the Vibrio core group [Sawabe et al. 2007]. This clade includes the following species: V. harveyi, V. campbellii, V. rotiferianus, V. parahaemolyticus, V. alginolyticus, V. natriegens, and V. mytili [Sawabe et al. 2007]. To our knowledge, this is the rst report to isolate a species of Vibrio of the Harveyi clade from Pinctada fucata. Clam production is often affected by vibriosis, which leads to high mortality rates mainly in nursery cultures of juvenile bivalves [Castro et al. 1992 Strain MA3 could grow at 10-45°C (with optimal growth occurring at 25-45°C) and 10-90% (w/v) seawater. These results suggest that strain MA3 can survive in brackish water in winter. Strain MA3 was identi ed as a closely related species of V. alginolyticus. However, strain MA3 showed some important differences from the typestrain of V. alginolyticus ATCC 17749. Table 1 (Table 1). MA3 was positive for lipase (C14), valine arylamidase and α-glucosidase, and it was negative for α-chymotrypsin in the API ZYM system (Table 1).
Strain MA3 was susceptible to chloramphenicol but resistant to kanamycin, vancomycin, tetracycline, ampicillin, penicillin and streptomycin (Table 1). These results suggest that MA3 is a new strain of Vibrio alginolyticus. In the future, a more detailed characterization (the full genome, ANI analysis or MALDI-TOF analysis) will be needed.
The full-length hemolysin gene from Vibrio sp. MA3 (GenBank accession no. LC619649) was obtained by inverse PCR. The obtained gene was 2063-bp containing a 1254-bp ORF that was designated Vhe1. (Fig. 3). The deduced product of the Vhe1 gene is a protein of 417 amino acids, which has an estimated molecular mass of 47.2 kDa and a pI of 5.01 (Fig. 3). This protein has a typical signal peptide with a cleavage site between Ala19 and Glu20 at the N-terminal end of the protein sequence. A BLAST homology search of the GenBank database showed that Vhe1 shares 99.8% nucleotide sequence identity with Vibrio alginolyticus strain A056 lecithin-dependent hemolysin (ldh) gene, complete cds (JX064518). The similarity levels of the deduced amino acid sequence were 100% with thermolabile hemolysin [Vibrio alginolyticus NBRC 15630 = ATCC 17749] (accession no. AGV19473). In general, bacterial hemolysins have been suggested to be important factors of pathogenic vibrios by causing hemorrhagic septicemia in the host [Zhang et al. 2001]. Hemolysin produced by pathogens can lyse host cells and release ironcontaining compounds such as heme or hemoglobin that are bene cial for bacterial growth in the host [Stoebner and Payne 1988]. Because the strain MA3 possessed hemolysin gene, the possibility of causing vibriosis by infecting pearl oyster was considered. In the future, this hemolysin must be puri ed and characterized in detail.
The results of the experimental infection are summarized in Table 2. Mortality occurred only in the test group at 28°C, but not in the other test group at 25°C, nor in the control groups at both 25°C and 28°C. The rst dead individual in the test group at 28°C appeared 1 day after injection, after which the mortality rate increased dramatically, resulting in 90% at the end of the experimental period. The dead oysters usually exhibited the same symptoms as described above for naturally diseased oysters. Strain MA3 was isolated from the mantle, gill or adductor muscle tissue of all dead individuals in the test group at 28°C, with a high isolation rate of 88.9-100% for each tissue. We could not isolate strain MA3 from any surviving individuals in either the test groups or control groups. These results allowed us to infer that the present disease could be reproduced in healthy oysters by experimentally injecting strain MA3.
Note that the lethality of strain MA3 was found only at 28°C, but not at 25°C, although the strain grew optimally at both temperatures, as described above in its characterization. Therefore, the susceptibility of oysters to strain MA3 might depend on water temperature. This is supported by a previous report [Numaguchi 1994] in which the ltration rate of Akoya pearl oysters increased with increasing water temperatures up to 25-28°C, whereas their ltration rate was very low at 31°C. Furthermore, the rate diminished remarkably in summer at water temperatures above 28°C. Considering the physiological evidence, we speculate that the susceptibility of oysters to strain MA3 increases with increasing water temperatures from 25°C to 28°C, by possibly affecting the organism's physiological functions, such as its defense mechanisms against the pathogenic bacteria. Hence, it is probable that the higher temperatures in summer play an important role in affecting such defense mechanisms in cultured oysters.

Conclusion
We isolated a Vibrio sp. strain MA3 from the mantle and other soft tissues of diseased pearl oysters. Strain MA3 possessed a pathogenic factor, hemolysin (Vhe1), of vibriosis. Moreover, the results of the infection-challenge experiment in this study provide evidence that strain MA3 was one of the putative causal agents of mass mortalities of farmed pearl oysters in Japan in the summers of 2019 and 2020.