Physiological, biochemical and HSP70 and HSP90 gene expression profiles of tropical abalone Haliotis squamata in response to Vibrio alginolyticus infection

Vibrio spp. have been known responsible for fish diseases in marine and brackish‐water systems in the tropics regions. Heat shock proteins are a highly conserved protein group that is known for its rapid response to environmental stresses, including infection. This study aimed to investigate physiological and biochemical responses of tropical abalone Haliotis squamata to Vibrio alginolyticus infection. Abalones were infected with V. alginolyticus by intramuscular injection at a dose of 105, 106, 107 cfu/abalone. The expression of HSP70 and HSP90 genes, the activity of superoxide dismutase, phenol oxidase and catalase enzymes, histology, falling and mortality were observed at 12, 24, 48, 72, and 96 h post‐infection (hpi). The different expression of HSPs was found in this study. While the expression of HSP70was downregulated after infection, the expression of HSP90 was upregulated at 12 hpi and followed by downregulated after 24 hpi for 106 cfu infection, but expressed at a normal level for 105 infection treatment. The expression ofsuperoxide dismutase and catalase increased within 12 hpi, and the expression of phenol oxidase increased after 24 hpi. V. alginolyticus is virulent with LD50 of less than 105 cfu on H. squamata with an average weight of 5.13 g, and caused enlargement of hemolymph sinus and development intraepithelial and intramuscular abscesses.


Introduction
The development of aquaculture on increasing inten sification and commercialization of aquatic production will increase occurring major disease problems (Bondad Reantaso et al. 2005). Abalones or ear shells have a low shell, open spiral structure, and are characterized by several open respiratory pores in a row near the shell's outer edge. Abalone species are economically valuable for fishery production in the temperate or subtropical areas. Whereas, the large size of abalones are distributed in tem perate seas, the small size abalones are distributed at wide range geographical distribution in warm water, including Indonesia. The commercial aquaculture of the small size of abalone has already well developed, especially in East Asia (Hsu and Gwo 2017). The aquaculture industry has been overwhelmed with its share of diseases and prob lems caused by several pathogens (BondadReantaso et al. 2005). In Taiwan, the production of small abalone has been dramatically decreased in the past 15 years due to the lack of suitable diatom feed for larvae, poor water quality, habitat degradation, genetic problems, disease and infec tion problems (Hsu and Gwo 2017).
Given that bacteria can survive well in aquatic envi ronments independently of their hosts, bacterial diseases have become major impediments to aquaculture, espe cially when the water temperature is warm (Pridgeon and Klesius 2012). The most frequently encountered bacte rial agents associated with fish diseases in marine and brackishwater systems in the tropical environments are Vibrio spp. (Karunasagar et al. 2003). Vibrio harveyi is known to be pathogenic in a large range of vertebrates and invertebrates, including molluscs. Abalone diseases due to the pathogen V. harveyi have been described in Hali otis diversicolor, H. laevigata and H. tuberculata caus ing septicaemia. The evidence of vibriosis on abalone has outbroken in Taiwan in 2000 which was caused by V. parahaemolyticus and made a significant economic loss in abalone (H. diversicolor supertexta L.) industry (Cheng et al. 2004; Cai et al. 2006b). The V. alginolyticus H 11 strain has been isolated from a mass mortality out breaks of small abalone H. diversicolor supertexta with abscess/ulcers in the mantle that occurred in 1998 at Kao Hsiung, Taiwan. This strain and its extracellular products were virulent to small abalones with LD 50 values of 3.6 × 10 5 colony forming units (cfu) and 2.96 µg protein/g body weight, respectively (Liu et al. 2001). Vibrio alginolyticus also caused disease on post larvae and small juvenille H. diversicolor. This bacterium is virulent with LD 50 as 1.0 × 10 4 cfu/ml on post larvae (Cai et al. 2006a).
Animals have defense mechanisms against the pathogen, which is composed of innate and adaptive immune systems. The innate immune system is the only defense system that existed in invertebrates. This innate immune system is the first line defense against nonself pathogens and can be divided into physical barriers, cellular, and humoral components. Specifically, humoral parameters include growth inhibitors, various lytic en zymes and components of the complement pathways, agglutinins, precipitins, natural antibodies, cytokines, chemokines, and antibacterial peptides. Furthermore, several external and internal factors can also influence the activity of innate immune parameters. The cellular immune system is performed by several types of cells (Magnadóttir 2006).
Hemocytes in molluscs are consisted of agranular and granular hemocytes, which are involved in phagocytosis, an important process of eliminating microorganisms or foreign particles. During phagocytosis, several types of reactive oxygen intermediates (ROIs) are produced, such as: superoxide anion (O 2 ), hydrogen peroxide (H 2 O 2 ), sin glet oxygen, and hydroxyl radical. The release of superox ide anion is known as the respiratory burst, and it plays an essential role in antibacterial activity (Cheng et al. 2004). Acid phosphatase (ACP) and alkaline phosphatase (AKP) are important for innate immune defense in the small size of abalones. Superoxide dismutase (SOD) is a key antiox idant enzyme playing a firstline protective role against reactive oxygen species (ROS) by converting superoxide (O 2 ) into H 2 O 2 . The AKP, and SOD activities of diseased abalones were significantly lower than in the healthy group (Di et al. 2016). Several enzymes on abalone have been evaluated in response to pathogen infection on H. diversi color (Yao et al. 2019), and on greenlip abalone (H. laevi gata) in the high water stress (Buss et al. 2017).
Heat shock proteins (HSPs) are a group of highly con served chaperone proteins expressed by the cell that re spond to unfavorable environmental changes (Fang et al. 2019). The HSPs are considered as ubiquitous protein and widely preserved in prokaryotic and eukaryotic organ isms (Roberts et al. 2010). These proteins have functioned as cellular defenses, prevent protein denaturation, and as sist in the reintroduction and removal of denatured protein due to biotic and abiotic pressures (Wang et al. 2004). In aquatic organisms, expression of HSP genes was increased as a response to several stresses, such as heat (Park et al. 2015), organic pollutants (Paulino et al. 2014), correla tions between metals (Qian et al. 2012), and Vibrio infec tions (Rungrassamee et al. 2010).
Many studies on physiology and disease have been conducted on abalones from the temperate or subtropical zone (Rungrassamee et al. 2010; Di et al. 2016; Fang et al. 2019; Yao et al. 2019). However, only limited studies have been addressed on tropical abalone. The H. squamata is an indigenous species with an excellent taste and has been caught on the southern coast of Bali. This species was started to be cultured, especially in Bali. In this study, we investigated the biological responses of tropical abalone H. squamata in the response to V. alginolyticus infection. This study is the first investigation of V. alginolyticus in fection in H. squamata in Indonesia with a comprehensive evaluation of mortality, histology, enzymes activity, and HSP gene expression.

Animal collection and maintenance
The uniform and high quality abalone seeds are very essen tial for this study, then this research begins with the hatch ing of abalone in the Abalone Hatchery Unit, National Broodstock Center for Shrimps and Mollusc in Tigaron, Karangasem, Bali, Indonesia. The abalones with a normal morphological and appearance, agile movements, sticking firmly to the substrate, minimal size of the shell length 4 cm were selected as broodstocks for use in this study. The broodstocks were maintained in fiberglass tubs with PVC pipes as shelters and fed with Gracillaria sp. and Ulva sp. seaweed at the dosage of 1020% of biomass/day.
The stress treatment was applied to mature gonad abalones for inducing the spawning. The stress was ad dressed by lifting the basket of the broodstock from the wa ter tank for one hour then put it back into the water. Then the broodstock was maintained in a tank with a flowing water system until spawning. Eggs produced from spawn ing abalone were harvested using an egg collector. After 12 to 13 h of incubation, the eggs hatched into firststage swimming larvae, trochophores. The trochopores within a few hours become a veliger larvae. The veliger larvae were fed with attached diatoms (Nitzschia sp. and Navic ula sp.) which attached on the substrate rearing plate.
After one month rearing, the veliger larvae reached a juvenile stage at size diameter of shell more of than 0.6 cm. The juveniles were reared on the basket in the tank and fed with macroalgae Gracillaria sp. and Ulva sp. The grad ing was carried out every two months for continue rearing on relatively same size. After eight months of rearing, the juvenile abalone H. squamata with an average shell length of 32.97 ± 1.83 mm and an average weight of 5.13 ± 0.83 g were used for this study. The abalones from the hatchery were acclimatized to laboratory conditions for one week. During acclimatization, abalones were reared on pipe bas ket in the tank with seawater at a salinity of 34 g/L, the temperature at 2930°C and fed with seaweed Gracillaria sp. twice a day.

Vibrio alginolyticus infection
A pathogenic strain of V. alginolyticus was received from Fish Disease and Environmental Inspection Center at Serang, Banten, Indonesia. The bacterium was cultured on nutrient broth and incubated at 35°C for 48 h. The bacterium was harvested, washed and suspended PBS on at desired concentration for infection treatments.
The V. alginolyticus infection was conducted by intra muscular injection on pallial sinus using 25 gauge 1 mL sy ringe at a concentration of 10 5 , 10 6 , 10 7 cfu/abalone with a volume of 100 µL. For the control, abalones were injected with 100 µl of PBS. After injection, the abalones were kept on pipe baskets and observed on the superoxide dismutase (SOD), phenol oxidase (PO) and catalase (CAT) enzyme activity, Heat shock proteins (HSPs) expression, survival rate, falling rate, and histology.

SOD, PO, and CAT enzyme activity
The evaluation of enzyme activity was performed by sam pling at 0, 12, 24, 48, 72, and 96 h post V. alginolyticus infection. The hemolymph was collected and pooled from three animals for measuring the SOD, PO, and CAT ac tivities. The SOD activity was determined by measuring the ability to inhibit the reduction of photochemical nitrob lue tetrazolium chloride (NBT), as described previously (Datkhile et al. 2009) with SOD KitWST (watersoluble tetrazolium salt) Access (Dojindo, Japan). Briefly, 40 μL of hemolymph was added into 360 μL buffer phosphate, then centrifuged at 6000 g at 4°C for 7 min. The super natant was then heated up at 65°C for 5 min to obtain the crude extract. Finally, 150 µL of the crude extract was added with 50 µL of nitroblue tetrazolium (NBT) reagent (0.1 Mm EDTA, 13 μM methionine, 0.75 mM NBT and 20 μM riboflavin in 50 mM phosphate buffer, pH 7.8) and in cubated for 2 min. Then the optical density was measured at 450 nm using a spectrophotometer.
Phenol oxidase activity was measured spectrophoto metrically by recording the formation of dopachrome pro duced from Ldihydroxyphenylalanine (LDOPA) accord ing to (Hooper et al. 2014). One hundred microliters of hemolymph plasma were transferred in duplicate to 96 well microplate wells. The 100 μl of LDOPA (30 mM L 3,4dihydrophenylalanine, Sigma D9628, in HCl 0.2 M, pH 8) was added to each well and mixed for 10 s. The absorbance at 492 nm was recorded every 5 min at 20°C for over than 30 min, using a microplate reader Heales® MB580, (Shenzhen Huisong Technology China).
Catalase activity was measured colorimetrically by CAT activity Assay Kit (GeneWay, Biotech) according to the manufacture instruction. The level of H 2 O 2 loss was measured by reading absorbance with a microplate reader at 492 nm. One unit of enzyme was defined as the amount of enzyme required to convert 1 mol of H 2 O 2 to the prod uct in one min in pH 4.5 at 25°C.

HSPs expression
The hemolymph was collected from the animal using a syringe at 0, 12, 24, 48, 72, and 96 h post V. alginolyti cus infection. The hemolymphs from three animals were pooled in microtube then immediately proceed for RNA extraction or kept at 80°C until ready to be used. To tal RNA was extracted from hemolymph using Quick RNAMiniPrepPlus Kit (R1058) (Zymo Research) fol lowing manufacturer protocol. The integrity of RNA was assessed by electrophoresis on 1.2% agarose gel. The pu rity of RNA was verified by measuring absorbance at 260 nm and 280 nm with NDD 2000 (Nano Drop Technolo gies, USA). The cDNA was synthesized by mixing the 100 µg of RNA with others component of ReverTra Ace® qPCR RT Master Mix (Toyobo, Japan). The mixture was incubated at 37°C for 15 min and at 50°C for 5 min, then followed by incubation at 98°C for 5 min for enzyme in activation.
HSP gene expression was measured by realtime PCR using Thunderbird SYBR® qPCR kit with Ap plied Biosystem machine (ABI, USA). The 2 μL cDNA was used in each reaction and analyzed in triplicate. The HSP90 F (CCAGGAAGAATATGCC GAGT) and HSP90 R (CACGGAACTCCAACTGACC) primers were used to evaluate HSP90 expression, while HSP70 F (CCGCTCTAGAACTAGTGGAT) and HSP70 R (CCGCCAAGTGGGTGTCT) primers were used to evaluate HSP90 expression, and βactin F (GGGTGT GATGGTCGGTAT) and βactin F (AGCGAGGGCAGT GATTTC) primer pairs were used to determining the ex pression of βactin as an internal control (Farcy et al. 2007). The thermal cycling condition was 95°C for 30 s for the initial denaturation stage, followed by 40 cycles of 95°C for 5 s, 58°C for 30 s, and 72°C for 30 s for fi nal extension stage. At the end of reaction, the melting or dissociation curve analysis to ensure reaction specificity. This analysis was applied by increasing temperature from 65°C to 95°C, with rate increasing the temperature at 0.5°C sec1.

Falling rate
The abalones were injected intramuscularly with V. algi nolyticus at a dose of 10 5 , 10 6 , 10 7 cfu/abalone. Falling rate was conducted to evaluate the changes of adhesion ability of abalones on the PVC substrate. Thirty abalones were attached to vertical PVC pipe substrates in the aquar ium. The numbers of fallen abalones from the vertical sub strate was recorded every 12 h. This experiment was con ducted in triplicate.

Survival rate
Thirty abalones from each dose infection treatment were transferred to aquaria. The mortality of abalone was recorded daily. The death of abalone was indicated by fallen from the wall, laid at the bottom with upsidedown position or the shell at the floor. This experiment was con ducted in triplicate.

Histological analysis
Histology was conducted to observe the effects of V. al ginolyticus infection on the foot muscle structure. The abalones were collected at 96 hpi, the shells were removed, and the tissues were fixed in Bouin's solution. The tissues were proceeded on standard histology, then sectioned at a thickness of 5 µm and stained with H&E. The slides were observed under light microscopy (ZEISS Primovert P35 C). The level of histological alterations in the foot was de termined descriptively.

HSP70 and HSP90 gene expression of H. squamata
Heat shock proteins (HSPs) are a group of highly con served proteins which responsible for responding to dis ease infection. The expression profile of the two HSP genes in the abalone hemolymph after V. alginolyticus in fection was shown in Figure 1. HSP70 and HSP90 were expressed in a different pattern. The expression level of HSP70 was decreased rapidly in the first 12 h after infec tion (hpi), and remained in a low expression level until at the end of the experiment at 96 hpi ( Figure 1). In contrast, the expression of HSP90 gene was increased in all infec tion treatments at 12 hpi, with the HSP90 expression at 10 7 cfu infection treatment reached 4.5 times over the control. Next, the HSP90 expression was decreased in all infection treatment after 24 hpi. Moroever, the HSP90 expression at 10 6 and 10 7 cfu infection treatments were still in a low level until 96 hpi. Meanwhile, the HSP90 at 10 5 cfu infec tion treatment was expressed at the nearly same level with control ( Figure 1B).

Biochemical responses of H. squamata to V. alginolyticus infection
Hemocytes are involved in phagocytosis for the elimi nation of microorganisms or foreign particles. Several enzymes play important roles in phagocytosis process. Therefore, the superoxide dismutase (SOD), phenol oxi dase (PO) and catalase (CAT) enzyme activity were mea sured to determine abalone responses to V. alginolyticus infection at 0, 12, 24, 48, 72, and 96 hpi (Figure 2). Result showed that the SOD activity was increased in abalone in fected with V. alginolyticus (P<0.05) compared to the con trol after 12 h with the highest increased of SOD activity was observed in abalones with 10 6 cfu infection. More over, after 24 h, SOD activity was decreased significantly in the abalone with 10 5 and 10 5 cfu infection treatments   Figure 2B). The PO activity of infection treatments was significantly decreased at 12 hpi (P<0.05) compared to the control in 24 h. After 24 h, the PO activity tended to decreased and reached same level of expresssion with control at 48 hpi ( Figure 2C).

Physiologyical responses of H. squamata to V. alginolyticus infection
The adhesion of abalones to the substrate was an important endpoint of their health and protection from environmen tal threats. The falling rate of the substrate was observed in H. squamata after it was exposed to various level of bacterial densities (Figure 3). On the other hand, there was no statistically significant difference of the falling rate abalones H. squamata's substrate in the control group. However, abalones at the concentration of 10 5 cfu, 10 6 cfu, and 10 7 cfu treatments at 48 hpi showed falling rates as 50%, 70%, and 100% respectively. Furthermore, after 72 h, the falling rates were 80% at the concentration of 10 5 cfu and 90% for 10 6 cfu treatments. In this infection experiment of V. alginolyticus, at the concentration ranging from 10 5 to 10 7 cfu, the mortality was started at 24 hpi in all infected treatments. The sur vival rate of 10 7 cfu infection treatment was decreased sig nificantly and all of the abalones were deceased by 72 hpi. Moroever, at the concentration of 10 5 and 10 6 cfu infec tion treatments, 21.6% and 11.6% of animals still alive at 96 hpi. As shown in Figure 4, in the control group the mortality of abalones was not observed and a 96.6% of survival rate was achieved until 96 hpi. Taken together, all those data suggested that LD 50 of this V. alginolyticus in H. squamata with an average weight of 5.13 g was less than 10 5 cfu. This result indicated that V. alginoliyticus is a virulent bacterium against abalone H. squamata. Liu et al. (2000) reported that LD 50 of V. parahaemolyticus on abalone H. diversicolor supertexta weighing 10-14 g is 1.6 × 10 5 cfu, and mortalities occurred within 2 d of infection. While Liu et al. (2001) reported that LD 50 of V. alginolyti cus strain H11 on abalone H. diversicolor supertexta is 3.6 × 10 5 cfu. Therefore, the result of this study is in a good agreement with the previous report.

Histology changes of H. squamata in response to V. alginolyticus infection
The abalone attaches and moves using its foot muscles along to the substrate for feeding and other activities. Due to its vital role, the effect of V. alginolyticus infection on the foot muscle was investigated and the histology anal ysis was conducted in this study. The normal foot of H. squamata consisted of an epithelial layer (EL), connective tissue layer, and muscle layer (ML) in a crosssection view ( Figure 5A). The epithelial layer included mucous cells, eosinophilic granule cells (Egc), and melano granule cells (Mgc) ( Figure 5A). The muscle layer was broad and con sisted of muscle fiber bundles (Mfb) and hemolymph sinus (Hs) ( Figure 5A). Muscle fiber bundles distributed evenly to fulfilled the muscle layer as longitudinal fibers.
In the 1 × 10 5 cfu infection treatment, the structural was changed in the abalone foot included small abscess (Abs) in the muscle layer, vacuolation, and enlargement of hemolymph sinus (Hs) in the muscle layer ( Figure 5B). Whereas in the 1 × 10 6 cfu infection treatment, many ab scesses (Abs) both in the intra epithelial layer and muscle layer, and enlargement of hemolymph sinus (Hs) in the muscle layer were observed ( Figure 5C). In addition, in the 1 × 10 7 cfu infection treatment, the structural changes in the abalone foot included intrusion of hemolymph through the hemolymph sinus and moving closed to the epithelial layer post the enlargement of hemolymph sinus (Hs) and decreasing the density of muscle fiber bundles in the mus cle layer ( Figure 5D). Thus, the infection of V. alginolyti cus induced enlargement of hemolymph sinus, develop ment of abscess intra epithelial and intramuscular, and the intrusion of hemolymph closed to epithelial layer due to the disintegration of the epithelial layer and muscle layer of abalone foot tissues. The histological alterations of the foot in abalones were more severe with increasing bacte rial concentration.

Discussion
Abalone species are economically valuable for fishery pro duction in the temperate or subtropical areas, so that the commercial aquaculture of abalone has developed in many countries. However, this industry has faced several prob lems, and one of the most important problems is diseases (Hsu and Gwo 2017). Vibrio spp. bacteria have been iden tified as pathogenic bacteria that causes diseases in many species of abalone and it can lead to the economic signif icant losses (Liu et al. 2001; Cheng et al. 2004; Cai et al. 2006b). Several parameters of disease mechanism have been investigated on diseaserelated gene expression, en zyme activity of abalone (Rungrassamee et al. 2010; Di et al. 2016; Fang et al. 2019; Yao et al. 2019). However, the study on those subjects is very limited in Indonesia. This study is the first investigation the effect of V. al ginolyticus infection on Indonesian abalone H. squamata with a comprehensive evaluation of mortality, histology, enzymes activity, and HSP gene expression.
The large size of abalones are distributed in temperate seas, while the small size abalones are distributed at wide range geographical distribution in warm water, including Indonesia. The commercial aquaculture of the small size has developed well, especially in East Asia (Hsu and Gwo 2017). The aquaculture industry has been overwhelmed with its share of diseases and problems caused by several pathogens (BondadReantaso et al. 2005). In Taiwan, the production of small abalone has dramatically decreased in the past 15 years which caused by lack of suitable diatom feed for larvae, poor water quality, habitat degradation, and genetic problems, disease and infection problems (Hsu and Gwo 2017).
Heat shock protein (HSP) family forms the most an cient defense system in all living organisms, from bacteria to humans. Heat shock proteins are classified into six ma jor families: small HSPs, HSP40, HSP60, HSP70, HSP90, and HSP110 according to their molecular weight. Among these HSPs, HSP70 and HSP90 are common and widely studied heatrelated proteins (Wang et al. 2004; Xie et al. 2015. On abalone Haliotis diversicolor, the HSP is ex pressed in the mantle, mucous gland, muscle, gills, diges tive tract, hemocytes, and hepatopancreas tissues. How ever, the expression level of HSP differed among tissues with a significantly higher expression level being in hep atopancreas, followed by hemocytes (Fang et al. 2019). In this experiment, we studied the expression HSP on abalone form hemocytes. Taking samples as hemocytes has advan tages, due to its easiness to get hemocytes from animals and its low risk to abalone conditions.
In channel catfish (Ictalurus punctatus), Heat shock protein (HSP) genes are differentially expressed after Ed wardsiella ictaluri or Flavobacterium columnare bacterial infections. The expression of those genes exhibited both temporal and spatial regulation. The induction of HSP genes was observed soon after bacterial infection, suggest ing their distinct roles in immune responses and disease defenses (Xie et al. 2015). The expression level of HSP70 and HSP90 of Penaeus monodon genes have been reported significantly increased after a 3h exposure to V. harveyi (Rungrassamee et al. 2010). In this study, we evaluate the expression of HSP90 and HSP70 of abalone Haliotis squa mata in response to Vibrio alginolyticus infection at 12 hpi.
In this study, the expression of HSP90 was upregulated in 12 hpi in all doses infection and reached the highest up regulation more than four times at the treatment of 10 7 cfu infection compare to the control. The HSP90 expression was then downregulated after 24 hpi to one seventh to one seventeenth for infection treatments compared to the control (Figure 1). This expression pattern was similar to study of Wang et al. (2011) that transcription of HSP90 of disk abalone (H. discus) gene in response to bacterial LPS challenge significantly increased within 2 h and reached highest transcription at 4 hpi, then recovered to the normal level of transcription in 24 h finally. The low expression of HSP90 on high density infection (10 7 cfu) may occur due to the severe condition abalone, which leads to mortality.
Heatinducible forms of HSP70 play a central role in stress tolerance by the promotion of growth at moderately high temperature and/or protecting organisms from death at extreme temperature (Cheng et al. 2007). HSP70 has been reported exhibits physiological and ecological impor tance in response to pathogen infection and environmental stress. For example, heat shock in fish was the most effec tive stress stimuli to induce HSP70 response compared to other stressors including hypoxia and air exposure. In mol lusks, HSP70 transcripts increased significantly after acute heat stress. Upregulation of HSP70 was observed after V. parahaemolyticus infection in adult bay scallops Ar gopecten irradians. The expression of HSP70 in the zebra mussel Dreissena polymorpha showed a timedependent increase after lipopolysaccharide (LPS) stimulation (Fang et al. 2019).
The expression levels of HSP70 and HSP90 of Pe naeus monodon significantly increased after a 3h expo sure to V. harveyi (Rungrassamee et al. 2010). The bac terial challenge of V. anguillarum on Pacific abalone (H. discus hannai) showed a timedependent expression of the HSP gene with a significant increase in the expression of HSP70 mRNA and reach the highest at 124 h and expres sion level of HSP70 returned to about control levels fol lowing a 96h recovery period (Cheng et al. 2007). Dif ferent from the result of Cheng et al. (2007), the relative expression level of HSP70 in this study decreased rapidly in 12 h after V. alginolyticus infection. This result may be due to quick expression of the HSP70 and the peak of the espression was less than 12 h. Wang et al. (2011) noted that in response to the LPS challenge, the transcription of disk abalone HSP90 gene significantly increased within 2 hpi and approached maximum induction at 4 hpi. Due to the earliest analysis of of HSP in this study was at 12 hpi, the expression of the HSP at this time was already decreased. In this study, the superoxide dismutase (SOD) activ ity of H. squamata in response to V. alginolyticus infec tion was increased at 12 hpi and then decreases at 24 hpi followed with normal expression started on 48 hpi ( Figure  2A). Di et al. (2016) found activity SOD of H. diversicolor with the withering syndrome is significantly lower than in the healthy abalone. Catalase activities of infected abalone was started from 12 hpi then it was likely to decrease and there was no significant different among treatment after 72 hpi ( Figure 2B). Buss et al. (2017) found that catalase CAT activity of greenlip abalone (H. laevigata) is signifi cantly higher when reared at 25°C. Different from the ex pression of SOD and CAT which showed a significant in crease within 12 hpi, the expression of phenol oxidase was increased after 24 hpi ( Figure 2C). The increasing phenol oxidase (PO) activity in H. diversicolor is stimulated by a viral infection (Yao et al. 2019).
Several cases of mass mortality of abalone have been recorded from several countries. Mortality of Japanese abalone Sulculus (Haliotis) diversicolor supratexta in Kanawaga, Japan in June to October 1997 is caused by Vibrio carchariae (V. harveyi) (Nishimori et al. 1998). At the nearly same time, mass mortality of the abalone Hali otis tuberculata L. has occurred in the natural environ ment along the south coast of Brittany, French in 1997 also caused by V. carchariae (V. harveyi) (Nicolas et al. 2002). Mass mortality among cultured small abalone H. diversicolor supertexta with abscess/ulcers in the man tle in 1998 at KaoHsiung Taiwan was caused by V. al ginolyticus (Liu et al. 2001). In China, V. alginolyticus and V. parahaemolyticus were associated with a severe epidemic in farmed H. diversicolor supertexta in Fujian Province (Zhang et al. 2001), and a Vibrio harveyirelated species was linked with the mass mortality of farmed adult H. diversicolor in Fujian (Jiang et al. 2013). Those data supported that Vibrio spp caused diseases on abalones. In this study, we reported that V. alginolyticus caused disease on tropical abalone (H. squamata) (Figure 4). This is the first report on confirmation of the pathogenicity of V. al ginolyticus on H. squamata.
Vibrio spp. has been reported as virulent bacteria to abalone. Liu et al. (2000) reported that LD 50 of V. parahaemolyticus on abalone H. diversicolor supertexta weighing 10-14 g is 1.6 × 10 5 cfu, and mortalities occurred within 2 days of infection. Cai et al. (2006a) reported that V. alginolyticus Strain 19 was virulent to abalone post larvae with an LD 50 value of 1.00 × 10 4 cfu. Liu et al. (2001) reported that LD 50 of V. alginolyticus strain H11 on abalone H. diversicolor supertexta is 3.6 × 10 5 cfu. In this study, we also confirmed that V. alginolyticus was vir ulent with LD 50 on H. squamata with an average weight of 5.13 g is less than 10 5 cfu (Figure 4).
Vibrio spp. produced and released toxins from the cells as an extra cellular product (ECP). The LD 50 of ECP of V. alginolyticus strain H11 on small H. diversicolor super texta is 2.96 µg protein/g body weight (Liu et al. 2001), while Vibrio strain B4 has LD 50 of CPS as 7.58 µg pro tein g −1 bodyweight. This toxin caused several changes in abalone organs and tissues and led to mortality. In this study, the infection of V. alginolyticus caused histologi cal changes as enlargement of hemolymph sinus, develop ment of abscess intraephitelial and intramuscular, and the intrusion of hemolymph closed to epithelial layer ( Figure  5).

Conclusions
In this study, we showed that infection V. alginolyticus will be responsed by H. squamata with the rapid increas ing level of HSP70 and HSP90 expression, then it was fol lowed by decreasing level of HSP70 and HSP90 expres sion. Similar responses were occurred on antioxidant ac tivity of SOD and CAT enzymes with delay time. Those conditions caused histological change in the tissues and led to mortality.