Journal of Aquatic Animal Health Screening Medicinal Plants for Use

Home , Dactylogyrus, Dactylogyrus vastator, Novirhabdovirus

This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:47 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Screening Medicinal Plants for Use against Dactylogyrus intermedius (Monogenea) Infection in Goldfish Yang Hua, Jie Jib, Fei Lingb, Yunhe Chenb, Lin Lub, Qizhong Zhangc & Gaoxue Wangab a College of Science, Northwest A&F University, Yangling 712100, China b College of Animal Science and Technology, Northwest A&F University, Yangling 712100, China c Institute of Hydrobiology, Jinan University, Engineering Research Center of Tropical and Subtropical Aquatic Ecological Engineering Minister of Education, Key Laboratory of Aquatic Eutrophication and Control of Harmful Algal Blooms of Guangdong Higher Education Institutes, Guangzhou 510632, China Published online: 03 Jul 2014.

To cite this article: Yang Hu, Jie Ji, Fei Ling, Yunhe Chen, Lin Lu, Qizhong Zhang & Gaoxue Wang (2014) Screening Medicinal Plants for Use against Dactylogyrus intermedius (Monogenea) Infection in Goldfish, Journal of Aquatic Animal Health, 26:3, 127-136, DOI: 10.1080/08997659.2014.902872 To link to this article: http://dx.doi.org/10.1080/08997659.2014.902872

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:127–136, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.902872

ARTICLE

Screening Medicinal Plants for Use against Dactylogyrus intermedius (Monogenea) Infection in Goldﬁsh

Yang Hu College of Science, Northwest A&F University, Yangling 712100, China Jie Ji, Fei Ling, Yunhe Chen, and Lin Lu College of Animal Science and Technology, Northwest A&F University, Yangling 712100, China Qizhong Zhang* Institute of Hydrobiology, Jinan University, Engineering Research Center of Tropical and Subtropical Aquatic Ecological Engineering Minister of Education, Key Laboratory of Aquatic Eutrophication and Control of Harmful Algal Blooms of Guangdong Higher Education Institutes, Guangzhou 510632, China Gaoxue Wang* College of Science, Northwest A&F University, Yangling 712100, China; and College of Animal Science and Technology, Northwest A&F University, Yangling 712100, China

Abstract Methanol extracts of 24 traditional medicinal plants with potential anthelmintic activity against Dactylogyrus intermedius (Monogenea) in Goldfish Carassius auratus were investigated. Abrus cantoniensis, Citrus medica, Dioscorea collettii,andPolygonum multiflorum exhibited 100% activity and were selected for further evaluation by applying five solvents (petroleum ether, chloroform, ethyl acetate, methanol, and water) for the extraction of the samples, followed by an in vivo bioassay. Among the plants tested, water, methanol, and ethyl-acetate extracts of P. multiflorum showed the highest efficacies; EC50 values (median concentration that results in 50% of its maximal effect) were 1.9, 5.4, and 9.1 mg/L, respectively, and extracts showed 100% efficacy against Dactylogyrus intermedius at 100, 12.5, and 25 mg/L. This was followed by ethyl-acetate, chloroform, and methanol extracts of Dioscorea collettii, which demonstrated 100% efficacy at 80, 80, and 120 mg/L and had EC50 values of 19.7, 27.1, and 37.8 mg/L, respectively, after 48 h of exposure. Chloroform and ethyl-acetate extracts of C. medica, which exhibited 100% efficacy against Dactylogyrus intermedius at 100 and 125 mg/L, revealed similar activity and had EC50 values of 58.7 and 51.3 mg/L, respectively. The ethyl-acetate and methanol extracts of A. cantoniensis exhibited the lowest activity and had EC50 values of 279.4

Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 and 64.3 mg/L. Acute toxicities of these active extracts were investigated on Goldﬁsh for 48 h. The ﬁndings indicated that extracts of the four plants can be developed as a preferred natural antiparasitic for the control of D. intermedius.

According to the Food and Agriculture Organization of the world’s largest producer of ﬁsh, which accounts for two-thirds of United Nations (FAO 2008), global aquaculture has developed the global production in 2009 (FAO 2008), and thus plays a vital rapidly in recent years and grown at an annual rate of 6.4% from and decisive role in aquaculture. However, intensive ﬁsh pro- 2002 to 2006. In addition, aquaculture in developing countries duction often results in increased incidences of diseases caused will continuously grow in the next 10 years (Food Business by viruses, bacteria, fungi, parasites, and other undiagnosed and Network 2008). China, the biggest developing country, is the emerging pathogens (Bondad-Reantaso et al. 2005).

*Corresponding authors: [email protected]; [email protected] Received June 15, 2013; accepted February 11, 2014 127 128 HU ET AL.

Dactylogyrus intermedius is an important monogenean para- imens have been deposited at the herbarium of the College of site that invades the skin and gills of cyprinid fishes and thereby Life Science, Northwest A&F University. The plant materials causes serious problems, such as gill inflammation, excessive were washed thoroughly with water and oven-dried at 45◦Cfor mucous secretions, respiratory distress, and mixed infections 48 h, then they were crushed and reduced to a fine powder using with other parasites and secondary bacterial infections (Reed a strainer (30–40 mesh) with an electrical disintegrator (model et al. 2009). Fish infected with D. intermedius can be treated FW177, Tianjin Taisite Instrument Company). The powdered effectively with a wide spectrum of parasiticides, such as for- samples were freeze-dried at −45◦C to ensure complete removal malin (Marshall 1999), trichlorfon (Goven and Amend 1982), of water. a triazine derivative (Schmahl 1993), and the two most effec- Screening experiment.—A 50-g dry powder sample of each tive treatments in practice, praziquantel (Schmahl and Mehlhorn of 24 different plants was extracted three times with methanol 1985) and mebendazole (Buchmann et al. 1993). However, these (500 mL) for 48 h. To get more or less solidified crude extracts, chemical parasiticides can present serious drawbacks through the methanol filtrates were separately filtered and evaporated frequent use, including the threats of anthelmintic resistance, under reduced pressure in a vacuum rotary evaporator (model risk of residue, environmental contamination, and toxicity to the R-201, Shanghai Shenshen) until the solvents were completely host (Goven et al. 1980; Klinger and Floyd 2002). Therefore, evaporated. The resulting extracts of different plants were dis- there is an urgent need for alternative therapies, including the solved in dimethyl sulfoxide (DMSO) and diluted with distilled use of natural products from medicinal plants, which are more water to obtain 0.6 g/mL (sample/solvent) of stock solutions, environmentally acceptable and generally have lower toxicity which were used for the preparations of the desired concentra- than chemicals. tions for the anthelmintic efficacy assay. Recently, medicinal plants and their extracts were tested for An in vivo study was designed to determine the anthelmintic their ability to control D. intermedius, and different extracts activity of extracts to D. intermedius according to the previous from several medicinal plants had significant killing effects in method of Ji et al. (2012). Five fish previously infected with the vivo (Wang et al. 2009, 2010a, 2010b; Liu et al. 2010; Wu et al. parasite were placed in plastic basins (N = 120) of 5 L capac- 2011; Ji et al. 2012; Lu et al. 2012). In our research, we exploited ity that were filled with 2 L of aerated groundwater. Extracts the crude extracts of different medicinal plants extracted by five with designed concentration gradients of 100, 200, 300, 400, solvents (petroleum ether, chloroform, ethyl acetate, methanol, 500, and 600 mg/L were added to the basins. Negative control and water) for anthelmintic activity against D. intermedius in groups containing no plant extract were set up under the same Goldfish Carassius auratus. conditions as used for the test groups. Another control containing the highest percentage of DMSO was included to assess the possible effects of DMSO on the parasites. After 48 h, the sur- METHODS viving fish in all of the treatment and control groups were killed Parasite and host.—Goldfish weighing 5.8 ± 1.9 g (mean by spinal severance and biopsied under a light microscope at ± SD) and without any record of previous infestation with 40 × magnification (Table 1). parasites were used throughout the study. All fish were col- Experiment with selected anthelmintic plants.—Through lected from a Changxing fish farm, Xianyang, Shaanxi, China previous experiments, material from four plant (Abrus can- (34◦3215N, 108◦518E). The fish were then acclimated in toniensis, Citrus medica, Dioscorea collettii, and Polygonum glass aquaria (200 fish per aquarium) containing 180 L ground- multiflorum) that had 100% anthelmintic efficacy were selected water at 25 ± 1◦C (controlled by automatic aquarium heater) from 24 different plants. Each plant material sample (50 g) with aeration for 7 d and were fed once daily at 2% of their was extracted with petroleum ether, chloroform, ethyl acetate,

Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 body weight with commercial pellet feed. One week later, all methanol, or water for 48 h for complete extraction, and the the fish were cohabitated with fish infected with D. intermedius, process was repeated three times. The ratio of sample to solvent which were retained in our laboratory; the infection procedure was 1:10 (w/v). All the extracts were filtered, combined, and was described in our previous study (Wang et al. 2008). Three evaporated under reduced pressure in a vacuum rotary evapora- weeks later, 10 fish were randomly sampled and killed by spinal tor (model R-201, Shanghai Shenshen). The resulting extracts severance, and the biopsies of eight gill filaments of each fish of different plants were dissolved in DMSO and diluted with were used to determine the infestation level and intensity of distilled water to obtain 0.6 g/mL (sample/solvent) of stock so- adult D. intermedius under a light microscope (Olympus BX41, lutions, which were used for the assay. Tokyo, Japan) at 40 × magnification. Fish were chosen for the Tests were conducted in each plastic basin (N = 100) (5 L assays when the infection rate was 100% and the mean number capacity) that were filled with 2 L aerated groundwater. Each of D. intermedius on gills was 40–50 parasites per fish. basin contained plant extract samples and five previously in- Preparation of plant materials.—The plant materials from fected fish. Water pH ranged from 7 to 7.5, dissolved oxygen each of the selected species (Table 1) were collected in Septem- was between 6.2 and 7.8 g/mL (72–85% saturation), and the wa- ber 2012 and the taxonomic identification was made by X. P. ter temperature was constant at 24 ± 1◦C. The negative control Song, Northwest A&F University, Shaanxi. The voucher spec- groups containing no plant extract were set up under the same ANTHELMINTIC ACTIVITY OF MEDICINAL PLANTS 129

TABLE 1. Plants used in this study, plant part used, highest anthelmintic efficacy, concentration of the highest anthelmintic efficacy treatment, and concentration at which Goldfish died with Dactylogyrus intermedius infection. NA = not analyzed.

Best Concentration having Concentration at Plant part used for anthelmintic the best anthelmintic which Goldfish Species extract efficacy (%) efficacy (mg/L) died (mg/L) Polygonum multiflorum Thunb. Stem 100 60 125 Asparagus cochin-chinensis Root tuber 0 NA 500 (Lour.) Merr. Sinapis alba L. Seed 30 500 500 Piper kadsura (Choisy) Ohwi. Rattan 0 NA 30 Aucklandia lappa Decne. Root 0 NA 30 Notopterygium incisum Ting ex Root 0 NA 15 H. T. Chang Lycium chinense Mill. Root bark 0 NA 400 Citrus medica L. Fruit 100 250 300 Abrus cantoniensis Hance Herb 100 200 250 Perilla frutescens (L.) Britt. var. Leaf 90 350 400 crispa (Thumb.) Hand.-Mand.-Mazz. Calvatia lilacina (Mont. et Berk.) Sporocarp 89 500 >1,000 Lloyd Typha angustifolia L. Pollen 83 25 30 Buddleja officinalis Maxim. Alabastrum 96 80 125 Platycladus orientalis (L.) Leaf 0 NA 125 Franco. Ilex pubescens Hook. et Arn. Leaf and root 60 500 >1,000 Citrus reticulata Blanco Pericarp of immature 79 400 500 fruit Gynostemma pentaphyllum Rhizome 70 450 500 (Thunb.) Makino Dioscorea collettii Hook. f. var. Rhizome 100 45 62.5 hypoglauca (Palibin) C. Pei and C. T. Ting Trichosanthes kirilowii Maxim., Root 88 500 >1,000 T. roswthornii Harms Dalbergia odorifera T. Chen Trunk and root 70 25 30 Arnebia euchroma (Royle) Root 58 60 125 Johnst.

Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 Aster tataricus L. F. Root 0 NA >800 Amomum villosum Lour. Fruit 76 230 240 Morus alba L. Twig 64 225 250

conditions as the test groups and included the DMSO control following formula: group. During the experiments, no food was offered to the fish. The death of each fish was recorded when opercular movement and tail beat stopped and the fish no longer responded to phys- AE = (B − T )/B × 100, ical stimulus. To avoid deterioration of the water quality, the observed dead fish were removed from the water immediately. After 48 h, the surviving fish in all the treatments were killed where AE is anthelmintic efficacy, B is the average number by spinal severance and biopsied under a light microscope at of surviving Dactylogyrus intermedius in the negative control, 40 × magnification. Anthelmintic efficacy of each treatment and T is the average number of surviving D. intermedius in the and the negative control group was calculated according to the treatment groups. 130 HU ET AL.

FIGURE 1. Anthelmintic efficacy of different extracts of Abrus cantoniensis against Dactylogyrus intermedius after 48 h. PEE = petroleum ether extract, CLE = chloroform extract, EAE = ethyl-acetate extract, MEE = methanol extract, WAE = water extract; star indicates when fish mortality firstly occurred.

Acute toxicity test.—Aqueous static renewal 48-h bioassays 95% CI with upper and lower confidence limits (Finney 1971). were conducted to determine the acute toxicity of crude extracts Since a chi-square test for goodness of fit was significant, a to Goldfish. Healthy Goldfish were placed into several 5-L plas- heterogeneity factor was used in the calculation of confidence tic basins (10 fish/basin) and crude extract concentrations for limits. Goldfish- were added (a preliminary study had been performed to establish the concentrations [ranging from 0% to 100%] at which mortalities occurred). Control groups were set using the RESULTS same test conditions but without extracts. Another control group The results of an in vivo study on the anthelmintic efficacies containing the highest percentage of DMSO was also included. of selected plants against D. intermedius are shown in Table 1. Mortality was observed in each aquarium every day. The fish Among the screened plants, A. cantoniensis, C. medica, D. col- were not fed during the exposure. This test was conducted two lettii, and P. multiflorum had 100% anthelmintic efficacy at the times. concentrations of 200, 150, 45, and 60 mg/L. As a control the Statistical analysis.—The homogeneity of the replicates of solvent DMSO showed no anthelmintic activity when used at the samples was checked by the Mann–Whitney U-test. Probit the highest concentration. analysis was used for calculating the median lethal concentra- The anthelmintic efficacies of different extracts of A. can- tions resulting in 50% and 90% mortality (LC50 and LC90) toniensis, C. medica, D. collettii, and P.multiflorum are depicted and the median effective concentrations resulting in 50% and in Figures 1–4, and the EC50 and EC90 values are shown in Ta- 90% of its maximal effect (EC50 and EC90) and included the ble 2. The water extract of P. multiflorum was the most effective Downloaded by [Department Of Fisheries] at 23:47 16 November 2014

FIGURE 2. Anthelmintic efficacy of different extracts of Citrus medica against Dactylogyrus intermedius after 48 h. PEE = petroleum ether extract, CLE = chloroform extract, EAE = ethyl-acetate extract, MEE = methanol extract, WAE = water extract; star indicates when fish mortality firstly occurred. ANTHELMINTIC ACTIVITY OF MEDICINAL PLANTS 131

TABLE 2. Anthelmintic efficacy (EC50 and EC90 concentrations) of extracts from Abrus cantoniensis, Citrus medica, Dioscorea collettii,andPolygonum multiflorum against Dactylogyrus intermedius after 48 h of exposure. LCL and UCL are lower and upper 95% confidence limits.

Extraction EC50 (mg/L) EC90 (mg/L) Plants solvent (LCL–UCL) (LCL–UCL) χ2-value P-value A. cantoniensis Methanol 64.33 (57.23–71.23) 86.93 (78.96–99.93) 0.41 0.94 Ethyl acetate 279.37 (264.03–294.14) 335.57 (317.04–367.93) 3.22 0.20 C. medica Ethyl acetate 51.34 (26.13–62.96) 97.12 (84.21–128.60) 0.64 0.72 Chloroform 58.70 (55.89–61.44) 78.00 (74.14–83.22) 1.77 0.41 D. collettii Methanol 37.77 (7.36–74.07) 78.90 (54.60–273.52) 6.50 0.04 Ethyl acetate 19.69 (18.10–21.29) 34.83 (32.23–38.23) 0.004 1.00 Chloroform 27.12 (15.81–33.84) 40.14 (33.45–55.17) 3.95 0.14 P. multiﬂorum Water 1.87 (12.47–7.70) 27.03 (19.76–48.67) 9.16 0.06 Methanol 5.40 (4.70–6.10) 7.10 (6.35–8.73) 0 1.00 Ethyl acetate 9.14 (6.12–11.64) 16.51 (13.40–26.57) 0.1 0.95

FIGURE 3. Anthelmintic efficacy of different extracts of Dioscorea collettii against Dactylogyrus intermedius after 48 h. PEE = petroleum ether extract, CLE = chloroform extract, EAE = ethyl-acetate extract, MEE = methanol extract, WAE = water extract star indicates when fish mortality firstly occurred. Downloaded by [Department Of Fisheries] at 23:47 16 November 2014

FIGURE 4. Anthelmintic efficacy of different extracts of Polygonum multiflorum against Dactylogyrus intermedius after 48 h. PEE = petroleum ether extract, CLE = chloroform extract, EAE = ethyl-acetate extract, MEE = methanol extract, WAE = water extract; star indicates when fish mortality firstly occurred. 132 HU ET AL.

TABLE 3. Acute toxicity (LC50 and LC90 concentrations) of extracts from Abrus cantoniensis, Citrus medica, Dioscorea collettii,andPolygonum multiflorum to Goldfish during a 48-h exposure. LCL and UCL are lower and upper 95% confidence limits.

Plants Extraction solvent LC50 (mg/L) (LCL–UCL) LC90 (mg/L) (LCL–UCL) χ2-value P-value A. cantoniensis Methanol 338.33 (308.58–361.83) 387.30 (363.40–455.15) 1.34 0.25 Ethyl acetate 546.48 (474.11–593.59) 643.30 (595.60–816.77) 0.47 0.50 C. medica Ethyl acetate 134.60 (103.37–152.50) 158.07 (144.84–323.13) 0 1.00 Chloroform 145.19 (62.74–194.06) 249.55 (199.18–443.33) 0.71 0.40 D. collettii Methanol 221.83 (192.97–243.57) 261.50 (240.66–334.07) 0.09 0.77 Ethyl acetate 103.33 (82.79–111.96) 119.55 (111.14–161.50) 0.02 0.88 Chloroform 87.87 (83.64–92.21) 95.06 (91.08–106.82) 0.31 0.58 P. multiﬂorum Water 218.70 (1.35–273.99) 349.14 (290.51–709.80) 1.08 0.30 Methanol 100.27 (39.84–171.66) 281.29 (198.81–740.15) 1.06 0.59 Ethyl acetate 64.52 (32.00–87.01) 100.00 (80.16–196.04) 0 1.00

with EC50 and EC90 values of 1.9 and 27 mg/L, respectively. The results of the acute toxicity assay for methanol and ethyl- After a 48-h exposure, this extract exhibited 100% efficacy acetate extracts of A. cantoniensis, the ethyl-acetate and chlo- against D. intermedius at 100 mg/L. High anthelmintic activity roform extracts of C. medica, the methanol, ethyl-acetate, and against D. intermedius was also observed for the methanol and chloroform extracts of D. collettii, and the water, methanol, ethyl-acetate extracts with EC50 and EC90 values of 5.4 and and ethyl-acetate extracts of P. multiflorum are summarized in 7.1 mg/L (methanol), and 9.1 and 16.5 mg/L (ethyl acetate), re- Table 3. The results indicated thatthese extracts had low toxic- spectively. The chloroform and petroleum ether extracts, how- ity to Goldfish. The 48-h LC50 values of methanol extracts of ever, exhibited weak activity with the maximum anthelmintic A. cantoniensis, ethyl-acetate and methanol extracts of D. col- efficacy of 81.8% and 77%, respectively, at 100 mg/L. lettii, and the ethyl-acetate, methanol, and water extracts of P. For D. collettii, the ethyl-acetate extract was the most effec- multiflorum were approximately fivefold (A. cantoniensis), five- tive, and had EC50 and EC90 values of 19.7 and 34.8 mg/L, and sixfold (D. collettii), and 7-, 20-, and 100-fold (P. multiflo- respectively, after 48 h of treatment. The extracts of methanol rum) higher than the corresponding EC50s, which had values of and chloroform exhibited 100% anthelmintic efficacy at 120 338.3 (A. cantoniensis), 103.3 and 221.8 (D. collettii), and 64.5, and 80 mg/L, and had EC50 and EC90 values of 37.8 and 100.3, and 218.7 mg/L (P. multiflorum), respectively. Results 78.9 mg/L (methanol), and 27.1 and 40.1 mg/L (chloroform), of the acute toxicity tests are provided in the Appendix (Tables respectively. However, the extracts with water showed no an- A.1–A.4). thelmintic activity at 120 mg/L when the fish began to die. Fish mortality occurred when the concentration reached 100 mg/L for petroleum-ether extract, which exhibited 4.9% anthelmintic DISCUSSION efficacy. Dactylogyrus, a genus of gill parasites that causes serious The methanol and ethyl-acetate extracts of A. cantoniensis damage to fish, has become a common parasite in Chinese displayed the optimal anthelmintic activity and showed 100% aquaculture (Ogawa 2002). Traditionally, a number of effec- efficacy at the doses of 200 and 400 mg/L, respectively, while tive chemotherapeutic agents for the treatment of Dactylogyrus

Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 the C. medica extracts with ethyl acetate and chloroform demon- disease in aquaculture have been gradually prohibited because strated equivalent efficacy at 100 and 125 mg/L, respectively. of their side effects, such as accumulation of drugs in tissues, The EC50 and EC90 values for A. cantoniensis were 64.3 development of drug resistance, and the potential deleterious ef- and 86.9 mg/L for the methanol extract, and were 279.4 and fects on the environment and human consumers (e.g., formalin, 335.6 mg/L for ethyl-acetate extract. The EC50 and EC90 val- malachite green). Therefore, it is logical to look for novel aqua- ues for ethyl-acetate and chloroform extracts of C. medica were culture medicines that are both environmentally friendly and 51.3 and 97.1 mg/L (ethyl acetate) and 58.7 and 78 mg/L (chlo- highly efficient. Extracts from medicinal plants that could offer roform), respectively. The highest activity was found in the re- possible alternatives to control D. intermedius infection have maining extracts of these two plants before fish began to die. Wa- been reported (Liu et al. 2010; Wu et al. 2011; Ji et al. 2012; Lu ter, chloroform, and petroleum-ether extracts of A. cantoniensis et al. 2012). In the present study, 24 medicinal plants were eval- showed 87.4, 77, and 20% activity at 600, 90, and 600 mg/L, uated for the in vivo anthelmintic activity against D. intermedius respectively, and water, methanol, and petroleum-ether extracts (Monogenea) in Goldfish. Extracts from four plants—A. can- of C. medica exhibited 13.6, 76.5, and 0% activity at 400, 300, toniensis, C. medica, D. collettii, and P. multiflorum—resulted and 300 mg/L, respectively. in 100% parasite elimination rate at low concentrations. As far ANTHELMINTIC ACTIVITY OF MEDICINAL PLANTS 133

as we know, this is the first report on anthelmintic activity of A. et al. (1999) found that the oil of C. medica exhibited a high cantoniensis, C. medica, D. collettii, and P. multiflorum. content of limonene (70.4%) and the proportions of geranial Among the four potent plants, extracts from P. multiflorum and neral were 14.4% and 7.6%, respectively. In our study, the exhibited the strongest efficacy with the lowest EC50 and EC90 chloroform extract of C. medica showed 100% activity at the values. Dried stem of P. multiflorum, known as Ye Jiao Teng, concentration of 100 mg/L1 and had relatively low EC50 and is one of the most popular traditional medicinal herbs in China EC90 values. Therefore, the main component limonene might and is officially listed in the Chinese Pharmacopoeia (China be the anthelmintic compound in these plants. Although there Pharmacopoeia Committee 2004). We have previously reported have been no attempts to identify the anthelmintic compounds that the root of P. multiflorum showed no anthelmintic activ- in this plant, some of the substances mentioned earlier are be- ity against D. intermedius (Ji et al. 2012). Wong et al. (2006) lieved to contribute jointly or independently to its anthelmintic found that the water extract from the stem of P. multiflorum activity. showed a potent antioxidant activity and might exhibit a greater Abrus cantoniensis, or Ji gu cao in Chinese, has long been antioxidant-mediated, anti-aging effect than its extract from the used in southern China and southeastern Asia as an effective root. This finding may explain the high anthelmintic activity folk medicine for the treatment of infectious hepatitis (Kiangsu from the extract of the stem and the lack of activity from the Hsin Yi Medical College 1977). According to Li et al. (2005), extract from the root of this plant. Stems of P. multiflorum also the ethanol extract from the aerial parts of A. cantoniensis ex- have a history of use in the treatment of traumatic arthritis, hibited the strongest growth inhibitions (minimum inhibitory aching of the loins and knees, and traumatic injuries (Tu et al. concentration, ∼40 mg/mL) against six test strains of Heli- 1992). Recently, Wong et al. (2006) and Li et al. (2007) found cobacter pylori among 30 Chinese herbal medicines. Mean- that P. multiflorum has the highest antioxidant capacity because while, the ethanol extracts of A. cantoniensis can have an obvi- of their rich phenolic content (including gallic acid, catechin, ous inhibitory effect on the hepatitis B surface antigen (HBsAg) and stilbene glycosides). Additionally, it has been suggested that and the hepatitis B e antigen (HBeAg) in serum and against 2,3,5,4-tetra-hydroxystilbene-2-O-β-D-glucoside, which is iso- Escherichia coli and Pseudomonas aeruginous (Cheng et al. lated from the aqueous extract of P. multiflorum, could be con- 2006; Chen et al. 2009). Phytochemical results ascertained that sidered, among future therapeutic drugs, for the treatment of A. cantoniensis contains triterpenic saponins, anthraquinones, Alzheimer’s disease (Um et al. 2006; Zhang et al. 2006). Con- alkaloids, and flavonoids (Xiao et al. 2002). The methanol ex- sidering the major bioactive constituents of P.multiflorum,some tract that showed 100% anthelmintic activity at the concentra- of the substances mentioned above may contribute to the effi- tion of 200 mgL in this study contains β-sitosterol, stigmas- cacy of this extract against D. intermedius independently or in terol, tannins, anthraquinone glycosides, and saponin. There- combination. fore, triterpenic saponins and alkaloids in the plant could be The rhizome of D. collettii is a traditional Chinese herb used the anthelmintic substances since trierpenic saponins were the for the treatment of cervical carcinoma, carcinoma of the uri- main compounds, and some of the alkaloids were found to be nary bladder, and renal tumor in China. He et al. (2006) reported effective against D. intermedius in Goldfish (Wang et al. 2010c). that methyl protodioscin, a furostanol saponin, is a preclinical The results of the acute toxicity assays for the extracts of drug that has potent antiproliferative activities against most cell A. cantoniensis, C. medica, D. collettii, and P. multiflorum in- lines from leukemia and solid tumors. Kobayashi et al. (1996) dicated that these extracts had low toxicity to Goldfish. The showed that 14 steroidal saponins, which were isolated from the 48-h LC50 values of these extracts were higher than the corre- rhizome of D. collettii, performed antifungal activity inducing sponding EC50 values. In summary, these four plants have the morphological deformation of mycelia and conidia of Pyric- potential for the development of a novel therapy for the treatment

Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 ularia oryzae (a phytopathogenic fungus responsible for rice against D. intermedius infection. However, more investigations, blast). Among them, 11 steroidal saponins showed cytotoxicity such as pharmacological evaluations before clinical trials, as- against the human acute myeloid leukemia K562 (AML) cell sessment of ecological risk posed by practical usage, and the line in vitro (Hu et al. 1996, 1997). Meanwhile, according to our detailed mechanisms of anthelmintic activity (i.e., against D. previous study, steroidal saponins from Paris polyphylla exhib- intermedius) must be performed. Further bioassay-guided isola- ited high anthelmintic activity (Wang et al. 2010a). Accordingly, tion and purification of compounds responsible for the observed anthelmintic activity of the methanol extract of D. collettii in anthelmintic efficacy are in progress. this study might be related to the presence of steroidal saponins. Essien et al. (2008) reported that the essential oil of C. medica exhibited a wide spectrum of fungitoxicity (inhibiting 14 ACKNOWLEDGMENTS isolated fungus species) at the minimum inhibitory concentra- Authors Yang Hu and Jie Ji contributed equally to this tion of 500 ppm. Moreover, the n-hexane extract of C. medica work. This work was supported by the National High Tech- has significant antioxidant activity, hypoglycaemic activity, and nology Research and Development Program of China (863 an anticholinesterase effect which may benefit the treatment of Program; number 2011AA10A216), the National Natural diabetes and Alzheimer’s disease (Conforti et al. 2007). Lota Science Foundation of China (number 31072242), and the 134 HU ET AL.

National Training Programs of Innovation and Entrepreneur- Li, Y.,C. Xu, Q. Zhang, J. Y.Liu, and R. X. Tan. 2005. In vitro anti-Helicobacter ship for Undergraduates. pylori action of 30 Chinese herbal medicines used to treat ulcer diseases. Journal of Ethnopharmacology 98:329–333. Liu, Y. T., F. Wang, G. X. Wang, J. Han, Y. Wang, and Y. H. Wang. 2010. In vivo anthelmintic activity of crude extracts of Radix angelicae pubescentis, REFERENCES Fructus bruceae, Caulis spatholobi, Semen aesculi,andSemen pharbitidis Bondad-Reantaso, M. G., R. P. Subasinghe, J. R. Arthur, K. Ogawa, S. Chinabut, against Dactylogyrus intermedius (Monogenea) in Goldfish (Carassius au- R. Adlard, Z. L. Tan, and M. Shariff. 2005. Disease and health management ratus). Parasitology Research 106:1233–1239. in Asian aquaculture. Veterinary Parasitology 132:249–272. Lota, M. L., D. De Rocca Serra, F. Tomi, J. M. Bessiere, and J. Casanova. 1999. Buchmann, K., H. C. Slotved, and D. Dana. 1993. Epidemiology of gill parasite Chemical composition of peel and leaf essential oils of Citrus medica L. and infections in Cyprinus carpio in Indonesia and possible control methods. C. limonimedica Lush. Flavour and Fragrance Journal 14:161–166. Aquaculture 118:9–21. Lu, C., H. Y. Zhang, J. Ji, and G. X. Wang. 2012. In vivo anthelmintic activity of Chen, X, Y. Han, and P. Xu. 2009. The inhibitory effect of Abrus cantoniensis Dryopteris crassirhizoma, Kochia scoparia,andPolygala tenuifolia against Hance on Hepatitis B virus in vitro. Herald of Medicine 28:418–420. Dactylogyrus intermedius (Monogenea) in Goldfish (Carassius auratus). Par- Cheng, Y.K., Y.Chen, L. Wang, M. Li, L. L. Zhong, and L. R. Teng. 2006. Study asitology Research 110:1085–1090. on the antibacterial activities of the ethanol extracts of Abrus cantoniensis. Marshall, C. J. 1999. Use of SupavermR for the treatment of monogenean Research and Practice on Chinese Medicines 2:39–42. infestation in koi carp (Cyprinus carpio). Fish Veterinary Journal 4:33–37. China Pharmacopoeia Committee. 2004. Pharmacopoeia of the People’s Re- Ogawa, K. 2002. Impacts of diclidophorid monogenean infections on fisheries public of China (the first division of 2005 edition). China Chemical Industry in Japan. International Journal Parasitology 32:373–380. Press, Beijing. Reed, P., R. Francis-Floyd, R. Klinger, and D. Petty. 2009. Monogenean para- Conforti, F., G. A. Statti, R. Tundis, M. R. Loizzo, and F. Menichini. 2007. In sites of fish. University of Florida, Institute of Food and Agricultural Sciences vitro activities of Citrus medica L. cv. Diamante (Diamante citron) relevant Extension FA28, Gainesville. Available: http://edis.ifas.ufl.edu/FA033. (May to treatment of diabetes and Alzheimer’s disease. Phytotherapy Research 2009). 21:427–433. Schmahl, G. 1993. Treatment of fish parasites 10. Effects of a new triazine Essien, E. P., J. P. Essien, B. N. Ita, and G. A. Ebong. 2008. Physicochemical derivative, HOE 092V, on Monogenea: a light and transmission electron properties and fungitoxicity of the essential oil of Citrus medica L. against microscopy study. Parasitology Research 79:559–566. groundnut storage fungi. Turkish Journal of Botany 32:161–164. Schmahl, G., and H. Mehlhorn. 1985. Treatment of fish parasites. 1. Praziquantel FAO (Food and Agriculture Organization of the United Nations). 2008. The effective against Monogenea (Dactylogyrus vastator, Dactylogyrus extensus, state of world fisheries and aquaculture. FAO, Rome. Diplozoon paradoxum). Zeitschrift fur Parasitenkunde 71:727–737. Finney, D. J. 1971. Probit analysis, 3rd edition. Cambridge University Press, Tu, G. S., Q. Z. Fang, and S. C. Yuan. 1992. Pharmacopoeia of the People’s Cambridge, UK. Republic of China, English edition. Gaungdong Science and Technology Food Business Network. 2008. World aquaculture development prospects Press, Guangzhou, China. and international status of aquaculture for China. (In Chinese.) Available: Um, M. Y., W. H. Choi, J. Y. Aan, S. R. Kim, and T. Y. Ha. 2006. Protective http://www.21food.cn/html/news/12/260400.htm. (March 2011). effect of Polygonum multiflorum Thunb. onamyloid β-peptide 25–35 induced Goven, B. A., and D. F. Amend. 1982. Mebendazole/trichlorfon combination: cognitive deficits in mice. Journal of Ethnopharmacology 104:144–148. a new anthelmintic for removing monogenetic trematodes from fish. Journal Wang, G. X., J. Han, T. T. Feng, F. Y. Li, and B. Zhu. 2009. Bioassay-guided of Fish Biology 20:373–378. isolation and identification of active compounds from Fructus arctii against Goven, B., J. Gilbert, and J. Gratzek. 1980. Apparent drug resistance to the Dactylogyrus intermedius (Monogenea) in Goldfish (Carassius auratus). Par- organophosphate dimethyl (2, 2, 2-trichloro-1-hydroxyethyl) phosphonate asitology Research 106:247–255. by monogenetic trematodes. Journal of Wildlife Disease 16:343–346. Wang, G. X., J. Han, L. W. Zhao, D. X. Jiang, Y. T. Liu, and X. L. Liu. He, X. J., A. M. Qiao, X. L. Wang, B. Liu, M. M. Jiang, L. N. Su, and X. 2010a. Anthelmintic activity of steroidal saponins from Paris polyphylla. S. Yao. 2006. Structural identification of methyl protodioscin metabolites in Phytomedicine 17:1102–1105. rats’ urine and their antiproliferative activities against human tumor cell lines. Wang, G. X., D. X. Jiang, J. Li, J. Han, Y. T. Liu, and X. L. Liu. 2010b. An- Steroids 71:828–833. thelmintic activity of steroidal saponins from Dioscorea zingiberensis C. H. Hu, K., A. J. Dong, X. S. Yao, H. Kobayashi, and S. Iwasaki. 1996. Antineo- Wright against Dactylogyrus intermedius (Monogenea) in Goldfish (Caras- plastic agents. I. Three spirostanol glycosides from rhizomes of Dioscorea sius auratus). Parasitology Research 107:1365–1371. collettii var. hypoglauca, Planta Medica 62:573–575. Wang, G. X., Z. Zhou, C. Cheng, J. Y. Yao, and Z. W. Yang. 2008. Osthol and Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 Hu, K., A. J. Dong, X. S. Yao, H. Kobayashi, and S. Iwasaki 1997. A furostanol isopimpinellin from Fructus cnidii to control Dactylogyrus intermedius in glycoside from rhizomes of Dioscorea collettii var. hypoglauca. Phytochem- Carassius auratus. Veterinary Parasitology 158:144–151. istry 44:161–165. Wang, G. X., Z. Zhou, D. X. Jiang, J. Han, and J. F. Wang. 2010c. In vivo Ji, J., C. Lu, Y. J. Kang, G. X. Wang, and P. Chen. 2012. Screening of 42 anthelmintic activity of five alkaloids from Macleaya microcarpa (Maxim) medicinal plants for in vivo anthelmintic activity against Dactylogyrus inter- Fedde against Dactylogyrus intermedius in Carassius auratus. Veterinary medius (Monogenea) in goldfish (Carassius auratus). Parasitology Research Parasitology 171:305–313. 111:97–104. Wong, C. C., H. B. Li, K. W. Cheng, and F. Chen. 2006. A systematic survey of Kiangsu Hsin Yi Medical College. 1977. Dictionary of Chinese drugs. Shanghai antioxidant activity of 30 Chinese medicinal plants using the ferric reducing People’s Press, Shanghai. (In Chinese.) antioxidant power assay. Food Chemistry 97:705–711. Klinger, R., and R. F. Floyd. 2002. Introduction to freshwater fish parasites. Wu, Z. F., B. Zhu, Y. Wang, C. Lu, and G. X. Wang. 2011. In vivo evaluation of University of Florida, Institute of Food and Agricultural Science, Document anthelmintic potential of medicinal plant extracts against Dactylogyrus inter- CIR716, Gainesville. medius (Monogenea) in Goldfish (Carassius auratus). Parasitology Research Kobayashi, H., M. Namikoshi, T. Yoshimoto, and T. Yokochi. 1996. A screening 108:1557–1563. method for antimitotic and antifungal substances using conidia of Pyricularia Xiao, P. G., D. P. Li, and S. L. Yang. 2002. Modern Chinese materia medica. oryzae, modification and application to tropical marine fungi. Journal of Chemical Industry Press, Beijing. Antibiotics 49:873–879. Zhang, L., Y. Xing, C. F. Ye, H. X. Ai, H. F. Wei, and L. Li. 2006. Learning- Li, H. B., Y. Jiang, C. C. Wong, K. W. Cheng, and F. Chen. 2007. Evaluation memory deficit with aging in APP transgenic mice of Alzheimer’s disease of two methods for the extraction of antioxidants from medicinal plants. and intervention by using tetrahydroxystilbene glucoside. Behaviour Brain Analytical and Bioanalytical Chemistry 388:483–488. Research 173:246–254. ANTHELMINTIC ACTIVITY OF MEDICINAL PLANTS 135

Appendix: Acute Toxicities of Plant Extracts to Goldﬁsh

TABLE A.1. Acute toxicity of different extracts of Abrus cantoniensis to Goldﬁsh during 48 h exposure.

Number dead Extraction solvent Concentration (mg/L) Total number tested 12 h 24 h 48 h Survival (%) None Control 10 0 0 0 100 None DMSO (0.3%) 10 0 0 0 100 Ethyl acetate 500 10 1 2 0 70 600 10 3 0 4 30 700 10 1 5 4 0 Methanol 300 10 0 0 2 80 360 10 0 4 2 40 400 10 7 3 0 0

TABLE A.2. Acute toxicity of different extracts of Citrus medica to Goldﬁsh during 48 h exposure.

Number dead Extraction solvent Concentration (mg/L) Total number tested 12 h 24 h 48 h Survival (%) None Control 10 0 0 0 100 None DMSO (0.3%) 10 0 0 0 100 Ethyl acetate 125 10 0 0 3 70 150 10 0 2 6 20 250 10 0 9 1 0 Chloroform 100 10 0 3 0 70 200 10 2 5 0 30 300 10 6 3 1 0

TABLE A.3. Acute toxicity of different extracts of Dioscorea collettii to Goldﬁsh during 48 h exposure.

Number dead Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 Extraction solvent Concentration (mg/L) Total number tested 12 h 24 h 48 h Survival (%) None Control 10 0 0 0 100 None DMSO (0.3%) 10 0 0 0 100 Ethyl acetate 100 10 1 0 3 60 120 10 3 4 2 10 140 10 4 5 1 0 Methanol 200 10 0 0 2 70 250 10 0 8 0 20 300 10 7 0 3 0 Chloroform 80 10 0 0 1 90 90 10 0 5 1 40 100 10 0 6 4 0 136 HU ET AL.

TABLE A.4. Acute toxicity of different extracts of Polygonum multiﬂorum to Goldﬁsh during 48 h exposure.

Number dead Extraction solvent Concentration (mg/L) Total number tested 12 h 24 h 48 h Survival (%) None Control 10 0 0 0 100 None DMSO (0.3%) 10 0 0 0 100 Ethyl acetate 50 10 3 0 0 70 100 10 5 4 0 10 200 10 3 3 4 0 Water 200 10 0 0 5 50 300 10 0 1 6 30 400 10 0 9 1 0 Methanol 50 10 0 2 1 70 100 10 0 4 2 40 200 10 0 3 4 30 400 10 4 3 3 0 Downloaded by [Department Of Fisheries] at 23:47 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:50 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Identification of Genes Involved in Taura Syndrome Virus Resistance in Litopenaeus Vannamei I. Boubea, J. M. Lotza, A. E. Pozhitkovb, S. Lic & R. J. Griffitta a Department of Coastal Sciences, University of Southern Mississippi, 703 East Beach Drive, Ocean Springs, Mississippi 39564, USA b Department of Periodontology, University of Washington, Box 357444, Seattle, Washington 98195, USA c Emory Vaccine Center, 954 Gatewood Road, Atlanta, Georgia 30329, USA Published online: 07 Jul 2014.

To cite this article: I. Boube, J. M. Lotz, A. E. Pozhitkov, S. Li & R. J. Griffitt (2014) Identification of Genes Involved in Taura Syndrome Virus Resistance in Litopenaeus Vannamei, Journal of Aquatic Animal Health, 26:3, 137-143, DOI: 10.1080/08997659.2013.860058 To link to this article: http://dx.doi.org/10.1080/08997659.2013.860058

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:137–143, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2013.860058

ARTICLE

Identiﬁcation of Genes Involved in Taura Syndrome Virus Resistance in Litopenaeus Vannamei

I.BoubeandJ.M.Lotz Department of Coastal Sciences, University of Southern Mississippi, 703 East Beach Drive, Ocean Springs, Mississippi 39564, USA A. E. Pozhitkov Department of Periodontology, University of Washington, Box 357444, Seattle, Washington 98195, USA S. Li Emory Vaccine Center, 954 Gatewood Road, Atlanta, Georgia 30329, USA R. J. Grifﬁtt* Department of Coastal Sciences, University of Southern Mississippi, 703 East Beach Drive, Ocean Springs, Mississippi 39564, USA

Abstract The goal of the present research was to identify the genes that are differentially expressed between two lineages of Pacific white shrimp Litopenaeus vannamei displaying different susceptibilities to Taura syndrome virus (TSV) and to understand the molecular pathways involved in resistance to the disease. An oligonucleotide microarray was constructed and used to identify several genes that were differentially expressed in the two L. vannamei lineages following infection with TSV. Individual L. vannamei from either resistant or susceptible lineages were exposed via injection to TSV. Individuals were removed at 6 and 24 h postinfection, and gene expression was assessed with the in- house microarray. The microarray data resulted in the selection of a set of 397 genes that were altered by TSV exposure between the different lineages. Significantly differentially expressed genes were subjected to hierarchical clustering and revealed a lineage-dependent clustering at 24 h postinoculation, but not at 6 h postinoculation. Discriminant analysis resulted in the identification of a set of 11 genes that were able to correctly classify Pacific white shrimp as resistant or susceptible based on gene expression data. Downloaded by [Department Of Fisheries] at 23:50 16 November 2014

The Pacific white shrimp Litopenaeus vannamei is one of lost production in shrimp aquaculture (OIE 2009). The first the most economically important shrimp species in the world, known instance of a TSV outbreak was in 1992 when the virus with an estimated economic value in excess of US$11 billion impacted the shrimp population near the Taura River in Ecuador annually (Wyban 2009). Production of L. vannamei has been (Jimenez 1992). The disease quickly spread to other countries. seriously impacted by several viral outbreaks around the world The presence of TSV in shrimp farms along the Mexican and (Lightner et al. 2012). One of the most damaging of these viruses Guatemalan borders was first noticed in 1995 (Zarain-Herzberg is Taura syndrome virus (TSV), which has caused economic et al. 2003). Incidences of TSV have been observed in Columbia, damages in excess of $1.2 billion (Hasson 1995; Lightner 1996). Peru, Brazil, El Salvador, Honduras, Nicaragua, Costa Rica, The Office International des Epizooties (OIE) has stated that Panama, and Venezuela (Hasson et al. 1999; Cotˆ e´ et al. 2008). TSV outbreaks are responsible for the second-most cause of By 1998, TSV was introduced into southeastern Asia through

*Corresponding author: joe.grifﬁ[email protected] Received June 21, 2013; accepted October 24, 2013 137 138 BOUBE ET AL.

infected stocks of L. vannamei intended for aquaculture and has 3 d. The TSV inoculate was prepared by homogenizing mori- spread through much of the region (Tu et al. 1999; Yu and Song bund Kona stock L. vannamei that were infected with TSV iso- 2000; Nielsen et al. 2005; Lightner et al. 2012). lated from an outbreak in Belize (Erickson 2002; Erickson et al. Taura syndrome virus is a 31–32-nm, nonenveloped, icosa- 2002). The homogenized tissue was diluted 1:10 with a saline hedral virus (Lightner 1996) with a single-stranded, positive- solution prior to injection. Each of the 16 L. vannamei were sense RNA genome (Bonami et al. 1997). The genome con- injected intramuscularly with 20 µL of inoculate per gram body sists of a linear, positive-sense, single-stranded RNA (ssRNA) weight and returned to their individual baskets. Four shrimp of of 10,205 bases (Mari et al. 2002), a 5 open reading frame both resistant and susceptible lineages were removed at 6 and (ORF), or (ORF1), that encodes the nonstructural proteins, and 24 h postinoculation. The shrimp were immediately sacrificed ORF2 that encodes three major and one minor capsid proteins and 40 mg of cephalothorax tissue was removed and immedi- (CPs). Currently, there are believed to be four different strains of ately snap frozen in liquid nitrogen for microarray analysis. TSV (Wertheim et al. 2009). How these strain differences affect Viral load quantification.—At each time point each sacri- pathogenicity is as yet unknown. Interestingly, exposure to the ficed individual was assessed for TSV tissue load using a quan- cyclodiene pesticide endosulfan has been shown to significantly titative PCR (qPCR) method described previously (Tang et al. increase L. vannamei mortality (Tumburu et al. 2012), implying 2004). Briefly, this was performed by amplifying a 72-bp DNA that responsiveness to TSV exposure is at least in part controlled fragment with a TaqMan probe selected from the ORF1 of the by shrimp-specific responses. TSV genome. A plasmid (pTSV-1) containing the target TSV While the development of biosecurity protocols and specific sequence was used to generate a positive control and standard pathogen-free (SPF) shrimp has reduced the reliance of farmers curve. A 10-fold dilution series of this plasmid DNA was used to on wild-caught L. vannamei, and therefore reduced the impact demonstrate that the real-time PCR (RT-PCR) assay has a detec- of viruses on farmed populations (Flegel et al. 2012; Lightner tion limit of 100 copies and a log–linear range up to 108 copies et al. 2012; Moss et al. 2012), there is still a need to understand of TSV RNA and has CV values (CV = SD/mean) that were the interaction between shrimp and the viruses that affect them. previously estimated to be 0.04–8.9 and 0.05–3.7 for replicates In this research, a global transcriptomic approach was used to within and among assays, respectively (Tang et al. 2004). identify, at multiple time points, genes that were significantly Microarray construction.—Briefly, microarray construction differentially expressed between two lineages of L. vannamei included retrieving all available sequences of L. vannamei from that displayed different susceptibilities that resulted in mortal- the National Center for Biotechnology Information (NCBI), ity due to TSV (Moss et al 2005). A microarray (25,815 unique which comprised a total of 156,952 sequences as of June 13, elements) specific to L. vannamei was used to investigate the 2008. All sequences were retrieved in extensible markup lan- expression of differentially expressed genes in resistant versus guage (XML) batches and stored locally (785 XML files, each susceptible L. vannamei infected with TSV.Global gene expres- under 1 MB). All nucleotide sequences were parsed out of the sion profiles were assessed to identify differentially expressed XML files and compiled into a FASTA file (76 MB). Only genes that have an important role in the resistance to TSV. unambiguous stretches from the 12,947 consensus sequences The microarray expression data were analyzed to identify the were used for designing probes. Probes were designed as significantly affected genes, and hierarchical clustering and dis- 60-mers, selected by criteria of uniqueness, information content, criminate analysis techniques were utilized to identify lineage and secondary structures, using the program ArrayOligoSelec- and time-specific patterns in the gene expression profiles. tor (Bozdech et al. 2003). Because most expressed sequence tags (EST) do not have accurate strand definition, probes were designed for both “sense” and “antisense” versions of sequences.

Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 METHODS The “sense” design led to 12,918 probes for 5,137 transcripts, Experimental organisms.—Litopenaeus vannamei weighing and “antisense” led to 12,897 probes for 5,141 putative tran- 2–3 g were obtained from the Oceanic Institute in Waimanalo, scripts. In total, 25,815 unique probes were designed and made Hawaii. This stock is part of the U.S. Marine Shrimp Farming to 5,499 putative shrimp transcripts. Microarrays were printed Program (USMSFP) (Garcia et al. 1994; Pruder et al. 1995). using a Nimblegen platform containing 25,815 probes for L. Of the16 L. vannamei individuals used in this experiment, eight vannamei (for 5,499 transcripts). were from a breeding program of shrimp known to be resistant to Microarray analysis.—Total RNA from the tissue below the TSV (TSV resistant; i.e, stocks are periodically evaluated over cuticle was extracted using Stat-60 (Tel Test, Friendswood, a 24-month period by PCR and histological examination for the Texas) following the manufacturer’s protocol. A ND-1000 Nan- presence of viruses and bacteria listed by the Oceanic Institute) oDrop spectrophotometer (NanoDrop Technologies, Wilming- (Garcia et al. 1994) and eight were from USMSFP Kona stock ton, Delaware) was used to check RNA for purity and concen- that are known to be highly susceptible to TSV infection (Brock tration. The integrity of the RNA was assessed with the Agilent et al. 1995; Pruder et al. 1995; Moss et al. 2005, 2012; J. M. 2100 bioanalyzer (Agilent Technologies, Palo Alto, California) Lotz, personal observation). using the RNA 6000 Nano Assay kit (Agilent Technologies). TSV exposure.—The 16 shrimp were placed in individual First- and second-stranded complementary DNA (cDNA) baskets contained in 100-L tanks and allowed to acclimate for was synthesized using the Fermentas Reverse Aid First Strand GENES INVOLVED IN TSV RESISTANCE 139

cDNA Synthesis kit (Thermo Scientific) according to the manufacturer’s protocol using 1 µg total RNA and dT primers. The cDNA was then quantified with a Nanodrop ND 1000 spectrophotometer. The double-stranded cDNA (1 µg) was labeled with Cy3 (TriLink BioTechnologies, San Diego, California) following the labeling protocol provided. The labeled cDNA was quantified as described above, and 300 ng of labeled cDNA per sample was sent to the Fred Hutchinson Cancer Research Insti- tute, Seattle, Washington, where the samples were hybridized to the microarrays and scanned. Scanned data were returned to University of Southern Mississippi for analysis. The raw data files were analyzed using JMP Genomics 3.1 (SAS), as a one-color data set. The raw expression data were LOESS nor- malized prior to ANOVA. Differentially expressed genes were identified by performing a two-way ANOVA with lineage and time postinoculation as the factors. Benjamini and Hockenburg False detection rate (FDR) multiple test correction was used and the significance was assigned at a P-value of 0.05. The genes that were identified as significant were subjected to hierarchi- FIGURE 1. Levels of TSV present in the hemolymph of exposed Litope- naeus vannamei measured by qPCR. Viral levels were quantified at 6 and 24 h cal clustering and discriminant analysis in JMP Genomics 3.1. postinoculation. Data are presented as mean + SD of three replicates. Hierarchical clustering was performed using the Fast Ward algorithm in JMP Genomics 3.1, starting with the subset of genes tion, while susceptible individuals had only four genes showing that were significantly differentially expressed in at least one a significant change in expression between 6 and 24 h. of the comparisons. Discriminant analysis was performed using the Discriminant Analysis function in JMP Genomics 3.1, using Hierarchical Clustering the Linear Common Covariance Method, starting with the en- The clustering revealed an interesting pattern in the global tire list of significantly altered genes, and iterated until a stable gene expression profiles. The samples tended to cluster pri- subset was identified. marily on the basis of time, with the 6-h and 24-h samples

RESULTS TSV Infection There was no statistically signiﬁcant difference in viral loads between the resistant or susceptible lineages at either 6 or 24 h postinoculation (Figure 1). Viral loads were generally lower in both lineages at 24 h postinoculation than at 6 h postinoculation, although the differences were not statistically signiﬁcant.

Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 Microarray Analysis The microarray analysis resulted in the identification of a subset of 397 genes that were differentially expressed. Of the 397 genes, 130 were found to have significant homologies to known genes in Genbank. A set of 265 genes of the total were differentially expressed between the main variables (i.e., between time points within a lineage or between lineages at a time point) and placed into subsets in a Venn diagram that shows the number of unique probes that were significantly affected within each treatment × time sampling point (Figure 2). The factor comparison demonstrating the largest effect was the difference between resistant and susceptible lineages at 6 h postinoculation, where 92 genes were uniquely affected; 59 genes were FIGURE 2. Venn Diagram showing relatedness between subsets of Litope- affected at 24 h postinoculation. Resistant L. vannamei had 55 naeus vannamei genes that were significantly differentially expressed at 6 or genes differentially expressed between 6 and 24 h postinocula- 24 h postinoculation with TSV. [Color figure available online.] 140 BOUBE ET AL.

FIGURE 3. Hierarchical clustering of the individual microarrays. In the heatmap, the darkest cells indicate genes expressed at relatively low levels, the next darkest cells genes expressed at relatively high levels. Resistant refers to TSV-resistant Pacific white shrimp, susceptible refers to TSV-susceptible shrimp. The numbers 6 and 24 refer to the time point (hours postinoculation) at which that shrimp was sampled. A color version of this figure is available in the onlineversion of this article. [Color figure available online.]

producing clearly distinct clusters (Figure 3). Within the two DISCUSSION time-dependant clusters, there were marked differences in the The aim of this experiment was to characterize gene expres- clustering of the lineages. At 6 h, there was no clear separa- sion patterns in strains of L. vannamei that were either suscepti- tion of the resistant and susceptible lineages, indicating that ble or resistant to TSV-induced mortality following inoculation there was no clear lineage-specific transcriptional profile pro- of the virus. To address this question, we used a species-specific duced by the TSV exposure. However, at 24 h there was a much microarray to identify genes that were expressed differentially sharper distinction between the lineages, indicating that by 24 h in the two lineages following TSV inoculation. Hierarchical postinoculation, TSV was producing different expression pat- clustering and discriminant analysis were used to identify gene terns between the resistant and susceptible lineages. expression patterns informative of lineage. The lack of clear separation of the transcriptome profiles be- Discriminant Analysis tween resistant and susceptible individuals at the 6-h time point A stepwise discriminate analysis was employed to identify provided evidence that the TSV inoculation had not had an im- the smallest possible subset of genes that was capable of ac- mediately observable effect on the transcriptional profile of the curately distinguishing between survival lineages based solely exposed organisms. By 24 h postinoculation, however, there upon the gene expression data. This procedure was performed in was a strongly divergent response in the transcriptional profiles JMP Genomics 3.1, and identified a subset of 11 genes (Table 1) that were used to accurately classify the microarray samples into the correct lineage with 100% accuracy (Figure 4). TABLE 1. Genes identified by discriminant analysis as predictive of lineage based on microarray analysis. NA = not available.

Description E-score Litopenaeus vannamei mitochondrion, 0.0 complete genome Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 est l vannamei7537 hCdit2 Litopenaeus 0.0 vannamei cDNA, mRNA sequence Haemaphysalis longicornis VCPa gene 5.9908 × 10−36 for valosin No significant homology NA No significant homology NA PREDICTED: eukaryotic translation 1.104 × 10−12 initiation factor 3 subunit 8 Penaeus monodon ribophorin I 0.0 Cu/Zn-superoxide dismutase (sod1) 9.99 × 10−10 No significant homology NA FIGURE 4. Results of discriminant analysis identifying the smallest subset of No significant homology NA genes that can be used to accurately classify the exposed Pacific white shrimp as No significant homology NA either susceptible or resistant based on expression values. Each dot represents gene expression data from one shrimp. [Color figure available online.] a Valosin containing protein. GENES INVOLVED IN TSV RESISTANCE 141

between resistant and susceptible individuals, indicating that the survival lineage of L. vannamei based solely on gene ex- the different lineages were responding to the TSV inoculation pression data. The advantages to this are obvious and offer the in different ways at the transcriptional level. This clustering has possibility of predicting L. vannamei survival by surveying this several possible explanations. It is possible that this difference limited subset of genes. It should be noted that this technique in gene expression profiles is reflective of genes involved in relies on identifying those genes that are most discriminating resistance to TSV exposure in Pacific white shrimp, and this between lineages; that is, ones that display the highest consis- divergence is due to expression levels of the genes involved in tent difference in expression values between lineages. That does this resistance being altered. Alternatively, it is possible that not mean that these genes are necessarily the ones that are most the observed differences are a function of the differential ef- involved in resistance or susceptibility to TSV infection; rather fect of the virus on the two L. vannamei survival lineages, and they are diagnostic of the potential for survival following TSV the differences in gene expression profiles between the lineages exposure. Clearly, these results need to be retested and vali- at 24 h postinoculation are a function of the differing suscep- dated in another experiment, but that is beyond the scope of this tibilities, rather than evidence of resistance. Previous research research. has found that the expression profile of different organs can be Veloso et al. (2011) also examined a similar question to the discriminated based on cluster analysis following white spot one examined here. Those investigators also used a microar- syndrome virus (WSSV) exposure (Robalino et al. 2007). This ray to demonstrate that L. vannamei lineages could accurately provides evidence that survival of L. vananmei exposed to TSV be distinguished based on gene expression profiles. Our re- is in part controlled by lineage-specific transcriptional expres- search differs in several ways from that presented by Veloso sion patterns. et al. (2011). The array used in our study was larger (∼5,000 The viral loads were generally depressed in both lineages genes versus ∼2,500 genes in Veloso et al. 2011), we examined at 24 h postinoculation compared with 6 h postinoculation, the responses at two time points (6 and 24 h) whereas Veloso although the difference was not statistically significant. This et al. (2011) performed a 14-d exposure, and we used hierarchi- shows that the same load of virus was present in the shrimp cal clustering and discriminant analysis to identify those genes at the different time points, and so the transcriptional profiles most important in identifying or conferring resistance to TSV observed were due to differences in lineage, rather than to a in L. vannamei. The fact that Veloso et al. (2011) found that different amount of virus being present. This is in accordance TSV exposure elicits a distinctive transcriptional pattern and with previous work on TSV loads in shrimp, where the TSV our current work showed a distinct response at 24 h but not at copy number reaches its highest level by 12 h postinoculation 6 h indicates that there is a relatively rapid (between 6 and 24 h and decreases thereafter (Cao et al. 2010; George et al. 2010). postinoculation) and stable (between 24 h and 14 d postexpo- Although Lotz et al. (2010) presented evidence of shrimp resis- sure) transcriptional response to TSV exposure. In addition, the tance by control of viral loads, our data suggest that tolerance results of both our research and that of Veloso et al. (2011) are (reduced pathology at the same viral loads) may also be a fac- largely complementary in that in both cases the different lin- tor (see Flegel 2007). It is possible that TSV proliferates in eages are sufficiently different in the gene expression profiles infected shrimp through a similar mechanism to that observed elicited by exposure to TSV that they can easily be distinguished in WSSV. Viral loads of WSSV in infected shrimp display a based on gene expression data and, therefore, it is highly likely broadly similar pattern to that observed here, in that within that alterations in gene expression are responsible in some part 24 h postexposure, WSSV viral loads begin to decrease, after for conferring resistance to TSV. silencing an immune-relevant gene like the antilipopolysaccha- This research represents one of the first gene expression stud- ride factor (ALF) found in the crayfish Pacifastacus leniusculus ies in L. vannamei at early stages (6 and 24 h) of postviral TSV

Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 (Liu et al. 2006). White spot syndrome virus significantly affects infection between lineages. The microarray data analysis was the proportion of circulating hemocytes in black tiger shrimp Pe- used to demonstrate that both survival lineage and the time naeus monodon (Van de Braak et al. 2002; Wongprasert et al. postinfection affected gene expression profiles, and that by 24 h 2003), and it is known that immune-relevant molecules from there was a strong lineage-dependent effect. This effect was not hemocytes are necessary for the crustacean’s defense mech- present at 6 h, indicating that this difference is not innate but is anism during infection. Circulating hemocytes in crustaceans a result of the TSV exposure. Discriminant analysis was used to vary in response to environmental stress, during the molt cycle, identify a subset of genes diagnostic of lineage, indicating that and with infection, and the decrease in hemocyte levels that is a relatively small number of genes were involved in conferring caused by a nonself exposure will likely cause the death of the resistance to TSV. This research offers a potential for better un- animal (Johansson et al. 2000; Jiravanichpaisal et al. 2006). derstanding the cellular and molecular mechanisms underlying The discriminant analysis approach that was used in this re- resistance or susceptibility to TSV in Pacific white shrimp and search is a powerful tool for identifying genes that are highly will serve as a jumping-off point for future research aimed at diagnostic or predictive of state of the organism. In this case, elucidating the relationship between survival and TSV exposure it was used to identify a set of 11 genes capable of identifying in L. vannamei. 142 BOUBE ET AL.

ACKNOWLEDGMENTS Johansson, M. W., P. Keyser, K. Sritunyaluksana, and K. Soderh¨ all.¨ 2000. Crus- This research was supported by U.S. Department of Agricul- tacean hemocytes and haematopoiesis. Aquaculture 191:45–52. ture National Institute of Food and Agriculture, startup funds Lightner, D. V. 1996. A handbook of shrimp pathology and diagnostic procedures for disease of cultured penaeid shrimp. World Aquaculture Society, from University of Southern Mississippi to R.J.G., and the Na- Baton Rouge, Louisiana. tional Oceanic and Atmospheric Administration Northern Gulf Lightner D. V., R. M. Redman, C. R. Pantoja, K. F. J. Tang, B. L. Noble, P. Institute Diversity Internship Program. The authors gratefully Schofield, L. L. Mohney, L. M. Nunan, and S. A. Navarro. 2012. Historic acknowledge the support of VerleeBreland, Rachel Ryan, Marie emergence, impact and current status of shrimp pathogens in the Americas. Mullen, Will Thompson, and Tina Sibuea. Journal of Invertebrate Pathology 110:174–183. Liu, H., P. Jiravanichpaisal, I. Soderhall, L. Cerenius, and K. Soderhall. 2006. Antilipopolysaccharide factor interferes with white spot syndrome virus replication in vitro and in vivo in the crayfish Pacifastacus leniusculus. Journal of REFERENCES Virology 80:10365–71. Bonami, J. R., K. W. Hasson, J. Mari, B. T. Poulos, and D. V. Lightner. 1997. Lotz, J. M. 2010. Evolutionary principles applied to disease control and health Taura syndrome of marine penaeid shrimp: characterization of the viral agent. management in shrimp aquaculture. Pages 679–694 in V. Alday-Sanz, editor. Journal of General Virology 78:313–319. The shrimp book. Nottingham University Press, Nottingham, UK. Bozdech Z., J. Zhu, M. P. Joachimiak, F. E. Cohen, B. Pulliam, and J. L. Mari, J., B. T. Poulos, D. V. Lightner, and J. R. Bonami. 2002. Shrimp Taura DeRisi. 2003. Expression profiling of the schizont and trophozoite stages syndrome virus: genomic characterization and similarity with members of the of Plasmodium falciparum with a long-oligonucleotide microarray. Genome genus Cricket paralysis-like viruses. Journal of General Virology 83:915–26. Biology 4(2):R9. Moss, S. M., R. W. Doyle, and D. V.Lightner. 2005. Breeding shrimp for disease Brock, J. A., R. Gose, D. V. Lightner, and K. W. Hasson. 1995. An overview on resistance. Challenges and opportunities for improvement. Pages 379–393 in Taura syndrome, an important disease of farmed Penaeus vannamei. Pages P. Walker, R. Lester, and M. G. Bondad-Reantaso, editors. Diseases of Asian 84–94 in C. L. Browdy and J. S. Hopkins, editors. Swimming through troubled aquaculture V. Asian Fisheries Society, Fish Health Section, Manila, The water: proceedings of the special section on shrimp farming, Aquaculture ‘95. Philippines. World Aquaculture Society, Baton Rouge, Louisiana. Moss, S. M., D. R. Moss, S. M. Arce, D. V. Lightner, and J. M. Lotz. 2012. Cao, Z., S. Y.Wang, V.Breeland, A-M. Moore, J. M. Lotz. 2010. Taura syndrome The role of selective breeding and biosecurity in the prevention of disease in virus loads in Litopenaeus vannamei hemolymph following infection are penaeid shrimp aquaculture. Journal of Invertebrate Pathology 110(2):247– related to differential mortality. Diseases of Aquatic Organisms 91:97–103. 50. Cotˆ e,´ I., and S. A. Navarro, et al. 2008. Taura syndrome virus from Venezuela Nielsen, L., W. Sang-oum, S. Cheevadhanarak, and T. W. Flegel. 2005. Taura is a new genetic variant. Aquaculture 284:62–67. syndrome virus (TSV) in Thailand and its relationship to TSV in China and Erickson, H. S. 2002. Characterization of Taura syndrome virus (TSV) isolates the Americas. Diseases of Aquatic Organisms 63:101–106. from penaeid shrimp: pathology, virulence, structural protein analysis and OIE (Office International des Epizooties). 2009. OIE Manual of diagnostic genetic diversity. Doctoral thesis. University of Arizona, Tucson. tests for aquatic animal diseases, 6th edition. World Organization for Animal Erickson, H. S., M. Zarain-Herzberg, and D. V. Lightner. 2002. Detection of Health, Paris. Taura syndrome virus (TSV) strain differences using selected diagnostic Pruder G. D. P., C. L. Brown, J. N. Sweeney, and W. H. Carr. 1995. High methods: diagnostic implications in penaeid shrimp. Diseases of Aquatic health shrimp systems: seed supply—theory and practice. Pages 40–52 in C. Organisms 52:1–10. L. Browdy, and J. S. Hopkins, editors. Swimming through troubled water: Flegel, T. W. 2007. Update on viral accommodation, a model for host-viral proceedings of the special session on shrimp farming, Aquaculture ‘95 1– interaction in shrimp and other arthropods. Developmental and Comparative 4 February 1995, San Diego. World Aquaculture Society, Baton Rouge, Immunology 31:217–231. Louisiana. Flegel, T. W. 2012. Historic emergence, impact and current status of Robalino, J., J. S. Almeida, D. Mckillen, J. Colglazier, H. F. Trent, Y. A. Chen, shrimp pathogens in Asia. Journal of Invertebrate Pathology 110:166– M. E. Peck, C. L. Browdy, R. W. Chapman, G. W. Warr, and P. S. Gross. 2007. 173. Insights into the immune transcriptome of the shrimp Litopenaeus vannamei: Garcia, D. K., M. A. Faggart, L. Rhoades, A. A. Alcivar-Warren, J. A. Wyban, tissue-specific expression profiles and transcriptomic responses to immune W. H. Carr, J. N. Sweeney, and K. M. Ebert. 1994. Genetic diversity of challenge. Physiological Genomics 29:44–56. cultured Penaeus vannamei shrimp using three molecular genetic techniques. Tang, K. F., J. Wang, and D. V. Lightner. 2004. Quantitation of Taura syndrome Molecular Marine Biology and Biotechnology 3:270–80. virus by real-time RT-PCR with a TaqMan assay. Journal of Virological Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 George, S. K., K. N. Kaizer, Y. M. Betz, and A. K. Dhar. 2010. Multiplication Methods 115:109–14. of Taura syndrome virus in primary hemocyte culture of shrimp (Penaeus Tu, C., H-T. Huang, S-H. Chuang, J-P. Hsu, S-T. Kuo, N-J. Li, T-L. Hsu, vannamei). Journal of Virological Methods 172:54–9. M-C. Li, and S-Y. Lin. 1999. Taura syndrome in Pacific white shrimp Pe- Hasson K. W., D. V. Lightner, L. L. Mohney, R. M. Redman, and B. M. naeus vannamei cultured in Taiwan. Diseases of Aquatic Organisms 38:159– White. 1999. Role of lympohoid organ spheroids in chronic Taura syndrome 161. virus (TSV) infections in Penaeus vannamei. Diseases of Aquatic Organisms Tumburu L., E. F. Shepard, A. E. Strand, and C. L. Browdy. 2012. Effects of 38:93–105. endosulfan exposure and Taura syndrome virus infection on the survival and Hasson, K. W., D. V. Lightner, B. T. Poulos, R. M. Redman, B. L. White, J. molting of the marine penaeid shrimp, Litopenaeus vannamei. Chemosphere A. Brock, and J. R. Bonami. 1995. Taura syndrome in Penaeus vannamei: 86:912–918. demonstration of a viral etiology. Diseases of Aquatic Organisms 23:115– Van De Braak, C. B., M. H. Botterblom, E. A. Huisman, J. H. Rombout, and 126. W. P. Van Der Knaap. 2002. Preliminary study on haemocyte response to Jiravanichpaisal, P., K. Soderhall, and I. Soderhall. 2006. Characterization of white spot syndrome virus infection in black tiger shrimp Penaeus monodon. white spot syndrome virus replication in in vitro-cultured haematopoietic Diseases of Aquatic Organisms 51:149–155. stem cells of freshwater crayfish, Pacifastacus leniusculus. Journal of General Veloso, A., G. W. Warr, C. L. Browdy, and R. W. Chapman. 2011. Virology 87:847–54. The transcriptomic response to viral infection of two strains of shrimp Jimenez, R. 1992. Syndroma de Taura (Resumen). [Taura syndrome (summary).] (Litopenaeus vannamei). Developmental and Comparitive Immunology 35: Aquacultura de Ecuador 1:1–16. 241–6. GENES INVOLVED IN TSV RESISTANCE 143

Wertheim, J. O., K. F. Tang, et al. 2009. A quick fuse and the emergence of proceeding of the special session on sustainable shrimp farming. World Aqua- Taura syndrome virus. Virology 390:324–9. culture Society, Baton Rouge, Louisiana. Wongprasert, K., Khanobdee, S. S. Glunukarn, P. Meeratana, and B. Withy- Yu, C. I., and Y. L. Song. 2000. Outbreaks of Taura syndrome in Pacific achumnarnkul. 2003. Time-course and levels of apoptosis in various tissues white shrimp Penaeus vannamei cultured in Taiwan. Fish Pathology 32: of black tiger shrimp Penaeus monodon infected with white-spot syndrome 21–24. virus. Diseases of Aquatic Organisms 55:3–10. Zarain-Herzberg, M. 2003. Biological characterization of a less virulent Wyban, J. A. 2009. World shrimp farming revolution: industry impact of do- Taura syndrome in Pacific white shrimp Litopenaeus vannamei (Crus- mestication, breeding and widespread use of specific pathogen free Penaeus tacea:Decapoda): gross signs, histopathological lesions, and mortalities. Jour- vannamei. Pages 12–21 in E. Jory, J. C. L. B. A. D. editor. The rising tide: nal of the World Aquaculture Society 34:99–105. Downloaded by [Department Of Fisheries] at 23:50 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:52 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Dietary Supplementation with Allspice Pimenta dioica Reduces the Occurrence of Streptococcal Disease during First Feeding of Mozambique Tilapia Fry a a Sevdan Yılmaz & Sebahattin Ergün a Department of Aquaculture, Faculty of Marine Sciences and Technology, Çanakkale Onsekiz Mart University, Çanakkale 17100, Turkey Published online: 10 Jul 2014.

To cite this article: Sevdan Yılmaz & Sebahattin Ergün (2014) Dietary Supplementation with Allspice Pimenta dioica Reduces the Occurrence of Streptococcal Disease during First Feeding of Mozambique Tilapia Fry, Journal of Aquatic Animal Health, 26:3, 144-148, DOI: 10.1080/08997659.2014.893459 To link to this article: http://dx.doi.org/10.1080/08997659.2014.893459

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:144–148, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.893459

COMMUNICATION

Dietary Supplementation with Allspice Pimenta dioica Reduces the Occurrence of Streptococcal Disease during First Feeding of Mozambique Tilapia Fry

Sevdan Yılmaz* and Sebahattin Ergun¨ Department of Aquaculture, Faculty of Marine Sciences and Technology, C¸ anakkale Onsekiz Mart University, C¸ anakkale 17100, Turkey

outbreaks, antibiotics and chemotherapeutants are often used Abstract indiscriminately, which in turn leads to residual problems in Allspice Pimenta dioica as a feed additive was studied for its ef- the surrounding environment, affecting higher animals and also fects on growth performance and disease resistance in Mozambique humans (Caruso et al. 2013). Herbal additives have been used Tilapia Oreochromis mossambicus. Five isonitrogenous (36% crude protein) and isocaloric (18.5 kJ/g) diets were formulated to contain recently instead of antibiotics and chemotherapeutic agents 0 (control), 5, 10, 15, or 20 g of allspice/kg of fish feed. In a 50-d in aquaculture (Vaseeharan and Thaya 2013) and constitute feeding trial, 15 plastic tanks (21 L) were stocked with 35 fish fry an environmentally friendly approach to disease management (0.012 g) each. After the feeding trial, fish were exposed to Strep- (Raa 1996; Bilen et al. 2013). Moreover, herbs reportedly have tococcus iniae and mortalities were recorded. The second-order a broad spectrum of benefits, including growth promotion, polynomial regression indicated that the dietary allspice level of 10 g/kg provided the best growth performance and feed utilization. appetite stimulation, and antimicrobial, immunostimulant, The greatest survival after pathogen challenge was also obtained anti-inflammatory, and anti-stress properties in humans and from the diet supplemented with allspice at 10 g/kg. Therefore, animals (Bulfon et al., in press). Herbal supplements and allspice acts as a growth promoter to improve feed utilization and immunostimulants have also been shown to enhance immunity, weight gain in Mozambique Tilapia fry and acts an antimicro- generate resistance to pathogens (Harikrishnan et al. 2011), and bial agent to enhance disease resistance during first feeding of fry. These results suggest that allspice can be used as an alternative to improve other characteristics of fish health (Chakraborty and antibiotics in controlling streptococcal disease in tilapia culture. Hancz 2011; Yılmaz and Ergun¨ 2012; Yılmaz et al. 2013c). Allspice Pimenta dioica (Myrtaceae) has been used in traditional folklore medicine for several hundreds of years The Mozambique Tilapia Oreochromis mossambicus is an (Nayak and Abhilash 2008) to remedy poor appetite, chills, di- economically important fish species possessing characteristics arrhea, dyspepsia, flatulence, indigestion, high blood sugar, and that allow these fish to do well in cultivation, such as high growth rheumatism (Mars 2007: 28–29). Allspice has also been shown Downloaded by [Department Of Fisheries] at 23:52 16 November 2014 rate, good-quality flesh, ability to reproduce in captivity, and to have antioxidant (Kikuzaki et al. 1999) and antimicrobial (Du feeding at lower trophic levels (El-Sayed 2006; Nguyen et al. et al. 2009) properties. Moreover, some studies indicate that 2009). Nevertheless, the production of Mozambique Tilapia fry allspice has significant cytoprotective activities (Al-Rehaily is often hampered by high mortality rates due to outbreaks of in- et al. 2002; Ramos et al. 2003; Nayak and Abhilash 2008). fectious disease (Yılmaz et al. 2012b) caused by pathogens such Therefore, it may potentially be used for disease prevention as Streptococcus sp., Flavobacterium columnare, Aeromonas and growth promotion in fish. hydrophila, and Edwardsiella tarda (El-Sayed 2006; Amal and Previous studies have reported that spices such as garlic Zamri-Saad 2011). Allium sativum (Aly et al. 2008), clove Syzygium aromaticum Streptococcal disease causes significant economic losses to (Rattanachaikunsopon and Phumkhachorn 2009), rosemary the tilapia culture industry. The disease is mainly controlled Rosmarinus officinalis (Abutbul et al. 2004; Ergun¨ et al. 2011), by antibiotics (Abutbul et al. 2005). To control disease thyme Thymus vulgaris, and fenugreek Trigonella foenum

*Corresponding author: [email protected] Received October 19, 2012; accepted January 20, 2014

144 COMMUNICATION 145

graecum (Ergun¨ et al. 2011) can be successfully used in tilapia Experimental design and feeding trial.—The experiment was culture. However, there is no clear information regarding the conducted in triplicate for each diet. Fifteen 21-L-capacity plas- effects of herbal supplements on larval culture, especially fish tic tanks were stocked with 525 fry (16 d old; 35 fry/tank); initial performance and disease resistance. There have been only a few average weight of the fry was 0.012 g. The fry were fed in three previous reports on the effects of herbs on growth performance daily feedings (0900, 1300, and 1700 hours) at a rate of 12% of and disease resistance in tilapia fry (Aly et al. 2010; Yılmaz body weight (BW) per day for 15 d, 10% BW/d for 15 d, and et al. 2012b, 2013c) or the fry of other fish species (Chakrabarti 8% BW/d for 20 d. Feeding rates during the 50-d experimental and Srivastava 2012). However, the effects of dietary allspice period were determined according to the protocol described by on growth rate and resistance to the bacterium Streptococcus Al Hafedh (1999), with some minor modifications. Each tank iniae have not previously been studied in Mozambique Tilapia was provided with sponge filters connected via air tubing to fry. Therefore, the purpose of this study was to investigate the a Resun GF-120 air pump. During the experiment, water was effect of an allspice-supplemented diet on growth performance exchanged daily at approximately 10% of the total volume. and disease resistance in first-feeding Mozambique Tilapia fry. Growth trial.—Growth performance and feed utilization were calculated according to the method of Yılmaz et al. (2012a). The following growth metrics were calculated (fish METHODS weights are in grams): Fish and experimental conditions.—Mozambique Tilapia fry were produced at Ç anakkale Onsekiz Mart University, Weight gain (%) Marine Sciences and Technology, Ç anakkale, Turkey. Wa- = 100 × [(final fish weight − initial fish weight) ter quality characteristics (mean ± SD) throughout the exper- /(initial fish weight)], iments were as follows: temperature was 28.4 ± 0.7◦C, pH was / 7.2 ± 0.5, dissolved oxygen was 7.1 ± 0.6 mg/L, conductivity Specific growth rate (SGR; % d) ± µ ± = ×{ was 595 45 S, total ammonia was 0.09 0.02 mg/L, nitrite 100 [loge(final fish weight) was 0.03 ± 0.02 mg/L, and nitrate was 1.2 ± 0.2 mg/L. Fish − / }, loge(initial fish weight)] (experimental days) experiments were performed in accordance with the guidelines Feed conversion ratio (FCR) = (feed intake)/(weight gain), for fish research from the animal ethics committee at Ç anakkale Onsekiz Mart University. Condition factor Experimental herb and diets.—Allspice seed meal powder = 100 × [(final fish weight)/(final fish TL, cm)3]. (Kotanyi GmbH, Istanbul, Turkey) was added to the basal diet (see below) at the following concentrations: 0 (control), 5, 10, Bacteria and challenge experiment.—Streptococcus iniae 15, and 20 g of allspice/kg of feed (hereafter, the diets are re- was obtained from diseased Mozambique Tilapia, and samples ferred to as A-0, A-5, A-10, A-15, and A-20, respectively). were collected aseptically from the brain and anterior kidney of Diets were adjusted as isocaloric and isonitrogenous (Yılmaz the fish. The tissues were cultured directly on sheep blood agar et al. 2013b). Basal diets contained approximately 36% crude at 28◦C for 24–48 h. Gram-positive, beta-hemolytic, catalase- protein and 10% crude lipid. The feed components of the total negative coccus colonies were subcultured onto blood agar and mixed diet (in g/kg of feed) were 280.0 g of fish meal, 320.0 g then identified by APIStrep (bioMerieux). The isolated S. iniae of soybean meal, 262.0 g of wheat flour, 65.0 g of fish oil, cultures were kept in aliquots consisting of 15% glycerol and 40 g of vitamin–mineral mix, and 33.0 g of starch. The same 85% brain–heart infusion (BHI) broth and were frozen at –70◦C diet was modified by replacing starch with varying amounts of until use.

Downloaded by [Department Of Fisheries] at 23:52 16 November 2014 allspice to produce the 5-, 10-, 15-, and 20-g/kg diets. The in- After the 50-d growth trial, 90 Mozambique Tilapia fry from gredients were mixed, and pellets (2-mm diameter) were made each dietary treatment were kept in aerated, 3,000-mL glass in a mincing machine and dried in a drying cabinet (40◦C) until jars (30 fry/jar) for 6 d at 28◦C (Yılmaz et al. 2013c) and were the moisture content declined to approximately 10%. Moisture exposed to the bacteria. Fish were not fed during this time. The was determined with an instant moisture analyzer (IR-35; Den- bacterial culture was prepared by inoculating 250 mL of BHI ver Instrument, Bohemia, New York). The pellets were crushed broth in a 500-mL culture ﬂask with a thawed, 1-mL aliquot of into desirable particle sizes (150–250 µm) and were stored at the frozen S. iniae isolate and then incubating at 28◦C for 24 h. −20◦C until use. Proximate analyses of the diets (% wet matter) The broth culture was centrifuged at 5,000 rotations/min at were performed using standard methods (AOAC 1998). Dietary 15◦C for 10 min (Abutbul et al. 2004). Aliquots of the resultant components were determined as follows: moisture was deter- S. iniae pellets were resuspended in phosphate-buffered saline mined by drying to a constant weight at 105◦Cinanovenfor solution (30 mL). Cell density was calculated from optical den- 24 h; crude fat was determined by ether extraction; crude pro- sity values at 600 nm and was correlated with colony-forming tein was determined by the Kjeldahl method; and crude ash was unit counts by using serial dilution and spread plating on BHI determined by incineration at 525◦C in a mufﬂe furnace for 12 h agar. Streptococcus iniae cultures with an initial concentration (AOAC 1998). of 9.0 × 109 colony-forming units/mL of BHI broth were 146 YILMAZ AND ERGUN¨

prepared. A 30-mL injection of the dilution was added to each experimental glass jar containing 2,970 mL of water, and the temperature was maintained at 28◦C throughout the challenge experiment (Yılmaz et al. 2013c). The trial was conducted in triplicate, and fish mortalities were recorded daily for 6 d. The challenge was terminated after 6 d, at which time mortality had ceased. Streptococcus iniae was re-isolated from the brain and anterior kidney on blood agar to confirm that mortality of experimental fish was due to the bacterial infection. Statistics.—Results for each of the measured variables are expressed as mean ± SD. The statistical significance (P < 0.05) of growth variables and survival during the feeding experiment was tested using one-way ANOVA followed by Tukey’s multiple comparison test conducted in SPSS version 17.0. Survival percentages from the growth phase of the experiment were arcsine transformed prior to analysis. Second-order polynomial regression analysis was introduced to determine the optimum allspice requirement of Mozambique Tilapia fry (Ahmad and Abdel-Tawwab 2011; Abdel-Tawwab 2012). The survival of fish in each S. iniae-challenged treatment group was estimated using Kaplan–Meier analysis, and differences between groups FIGURE 1. Kaplan–Meier survivorship curves (cumulative survival [%] over time [h]) for Mozambique Tilapia fry after challenge with Streptococcus iniae; were assessed with the log-rank (Mantel–Cox) test for pairwise the fish were fed allspice-supplemented diets (0, 5, 10, 15, or 20 g of allspice/kg comparisons. of feed; diets A-0, A-5, A-10, A-15, and A-20, respectively) prior to bacterial challenge. Asterisks denote significant differences in survival between the control group (A-0) and the indicated treatment group. RESULTS All diets were accepted by the fish and were avidly consumed At the end of the disease challenge experiment, survival of during the experiment. Growth results are presented in Table 1. fish that received the A-0, A-5, A-10, A-15, and A-20 diets was At the end of the growth trial, the final weight gain (%), SGR, 38, 52, 80, 73, and 49%, respectively (Figure 1). The maximum and condition factor of fish that were fed diet A-10 were higher survival (80%) was observed in fish that were given the A-10 than those of fish that received diets A-0, A-5, A-15, and A-20. diet, and the lowest survival (38%) was recorded among fish The FCR was significantly lower in fish that were given the A- that were fed the A-0 diet. Survival in fish that received the A-0 10 diet compared with fish that were fed the A-0 and A-20 diets. diet was significantly lower than survival in the A-5, A-10, and During the feeding experiment, survival in the treatment groups A-15 dietary treatment groups (Figure 1). was similar to survival in the control (A-0) group. The second- order polynomial regression between dietary allspice level and fish weight gain (%), SGR, and FCR indicated that the most DISCUSSION suitable allspice level for maximum growth of fry was 10 g/kg The present study demonstrates that dietary allspice levels (Table 1). can markedly affect growth and feed utilization in Mozambique Downloaded by [Department Of Fisheries] at 23:52 16 November 2014 TABLE 1. Growth performance, feed utilization, and survival in Mozambique Tilapia fry that were fed diets containing different levels of allspice (0, 5, 10, 15, or 20 g/kg of feed; diets A-0, A-5, A-10, A-15, and A-20, respectively) for 50 d (SGR = specific growth rate; FCR = feed conversion ratio). Values are reported as mean ± SD. Within a row, means with differing letters are significantly different (P < 0.05). Quadratic effects were as follows: weight gain = 3,998 + 255.24x – 13.631x2 (R2 = 0.73); SGR = 7.4277 + 0.1094x − 0.0059x2 (R2 = 0.76); and FCR = 1.136 – 0.0724x + 0.0041x2 (R2 = 0.84).

Diet

Variable A-0 A-5 A-10 A-15 A-20 Initial weight (g) 0.012 0.012 0.012 0.012 0.012 Final weight (g) 0.502 ± 0.044 yx 0.566 ± 0.034 y 0.686 ± 0.036 z 0.552 ± 0.034 y 0.457 ± 0.013 x Weight gain (%) 4,083.333 ± 368.650 yx 4,619.444 ± 280.143 y 5,616.865 ± 297.086 z 4,500.860 ± 284.110 y 3,708.333 ± 106.393 x SGR (%/d) 7.462 ± 0.181 yx 7.706 ± 0.118 y 8.090 ± 0.104 z 7.655 ± 0.122 y 7.279 ± 0.056 x FCR 1.118 ± 0.100 z 0.927 ± 0.085 zy 0.780 ± 0.049 y 0.987 ± 0.080 zy 1.337 ± 0.16 z Condition factor 1.761 ± 0.291 y 1.974 ± 0.209 y 2.580 ± 0.125 z 1.887 ± 0.113 y 1.532 ± 0.060 y Survival (%) 95.557 ± 7.696 100.000 ± 0.000 97.777 ± 3.851 97.777 ± 3.851 88.890 ± 19.243 COMMUNICATION 147

Tilapia fry by improving nutrient digestibility, which in turn promoter and antimicrobial agent during first feeding of Mozam- leads to improved nutrient utilization (fat and nitrogen) and ul- bique Tilapia fry. timately better growth. Al-Rehaily et al. (2002) reported that different active compounds of allspice stimulate digestion and ACKNOWLEDGMENTS improve protein assimilation in the upper intestinal tract of hu- We thank Nergiz Soytas¸ for her assistance while conducting mans. Yılmaz et al. (2013a) also demonstrated that other herbs, the experiment. such as rosemary, thyme, and fenugreek, can stimulate digestion in Sea Bass Dicentrarchus labrax via increased bile production as well as possible stimulation of the pancreas and increased REFERENCES secretion of digestive enzymes. Abd El Hakim, N. F., M. H. Ahmad, E. S. Azab, M. S. Lashien, and E. S. The published literature contains no previous reports of di- Baghdady. 2010. Response of Nile Tilapia, Oreochromis niloticus to diets supplemented with different levels of fennel seeds meal (Foeniculum vulgare). etary allspice’s effects on the growth rates of fish. However, Abbassa International Journal of Aquaculture 3:215–230. studies have reported that different spice additives were associ- Abdel-Tawwab, M. 2012. The use of American ginseng (Panax quinquefolium) ated with improved growth performance in tilapia. For example, in practical diets for Nile Tilapia (Oreochromis niloticus): growth perfor- Ahmad and Abdel-Tawwab (2011) conducted an experiment mance and challenge with Aeromonas hydrophila. Journal of Applied Aqua- with Nile Tilapia O. niloticus fingerlings that were fed diets culture 24:366–376. Abutbul, S., A. Golan-Goldhirsh, O. Barazani, R. Ofir, and D. Zilberg. 2005. containing 5, 10, 15, or 20 g of caraway Carum carvi seed meal Screening of desert plants for use against bacterial pathogens in fish. Israeli per kilogram of feed for 12 weeks; those authors found that Journal of Aquaculture Bamidgeh 57:71–80. a 12.5-g/kg concentration improved fish performance. Abd El Abutbul, S., A. Golan-Goldhirsh, O. Barazani, and D. Zilberg. 2004. Use of Hakim et al. (2010) also evaluated the use of fennel Foeniculum Rosmarinus officinalis as a treatment against Streptococcus iniae in tilapia vulgare seed (0, 1, 2, and 3%) in diets fed to Nile Tilapia fin- (Oreochromis sp.). Aquaculture 238:97–105. Ahmad, M. H., and M. Abdel-Tawwab. 2011. The use of caraway seed meal as gerlings for 14 weeks, and they found that 1% fennel produced a feed additive in fish diets: growth performance, feed utilization, and whole- the maximum fish performance. However, some studies demon- body composition of Nile Tilapia, Oreochromis niloticus (L.) fingerlings. strated that supplementation with garlic at 4% (Aly et al. 2010) Aquaculture 314:110–114. or with rosemary, thyme, or fenugreek at 1% (Yılmaz et al. Al Hafedh, Y.S. 1999. Effects of dietary protein on growth and body composition 2013c) did not affect growth performance in tilapia fry. Simi- of Nile Tilapia, Oreochromis niloticus. Aquaculture Research 30:385–393. Al-Rehaily, A. J., M. S. Al-Said, M. A. Al-Yahya, J. S. Mossa, and S. larly, Yılmaz et al. (2012a) conducted an experiment in which Rafatullah. 2002. Ethnopharmacological studies on allspice (Pimenta dioica) Mozambique Tilapia fingerlings were fed a basal diet containing in laboratory animals. Pharmaceutical Biology 40:200–205. 0.5, 1.0, 1.5, or 2.0% cumin Cuminum cyminum for 75 d; how- Aly, S. M., N. M. A. Attı, and M. F. Mohamed. 2008. Effect of garlic on ever, the use of cumin-supplemented diets did not affect growth the survival, growth, resistance and quality of Oreochromis niloticus. Pages performance. 277–295 in 8th International symposium on tilapia in aquaculture. American Tilapia Association, Pine Bluff, Arkansas. In this study, Mozambique Tilapia fry that were fed allspice- Aly, S. M., G. O. El-Naggar, F. M. Mohamed, and W. E. Mohamed. 2010. Effect supplemented diets A-5, A-10, and A-15 for 50 d exhibited of garlic, echinacea, organic green and vet-yeast on survival, weight gain, and reduced mortality when challenged with S. iniae. Seeds of all- bacterial challenge of overwintered Nile Tilapia fry (Orechromis niloticus). spice contain various chemical components, such as flavonoids, Journal of Applied Aquaculture 22:210–215. phenolic acids, catechins, and several phenylpropanoids (Krish- Amal, M. N. A., and M. Zamri-Saad. 2011. Streptococcosis in tilapia (Ore- ochromis niloticus): a review. Pertanika Journal of Tropical Agricultural Sci- namoorthy and Rema 2004). Eugenol is reported to be the major ence 34:195–206. constituent of allspice oil, and it also contains cineol, phellan- AOAC (Association of Official Analytical Chemists). 1998. Official methods of drene, caryophyllene, and pimentol (Shylaja and Peter 2004), analysis, 15th edition. AOAC, Arlington, Virginia.

Downloaded by [Department Of Fisheries] at 23:52 16 November 2014 which may have antioxidant and anti-inflammatory properties Bilen, S., S. Yılmaz, and A. M. Bilen. 2013. Influence of Tetra (Cotinus cog- and may improve immune response. Allspice also has antimicro- gygria) extract against Vibrio anguillarum infection in koi carp, Cyprinus carpio with reference to haematological and immunological changes. Turk- bial, antiviral, antifungal, and antiparasitic effects (Duke et al. ish Journal of Fisheries and Aquatic Sciences 13:527–532. 2002). Nayak and Abhilash (2008) reported that in mice, oral Bulfon, C., D. Volpatti, and M. Galeotti. In press. Current research on the − − doses of allspice (250, 500, and 750 mg·kg 1·d 1) provided use of plant-derived products in farmed fish. Aquaculture Research. DOI: significant protection from myelosuppression; they observed 10.1111/are.12238. that the increase in the white blood cells and hemoglobin con- Caruso, D., A. M. Lusiastuti, J. Slembrouck, O. Komarudin, and M. Legendre. 2013. Traditional pharmacopeia in small scale freshwater fish farms in West centration might be due to the immunostimulant properties of Java, Indonesia: an ethnoveterinary approach. Aquaculture 416:334–345. allspice. Enhancement of the immune system by an allspice- Chakrabarti, R., and P. K. Srivastava. 2012. Effect of dietary supplementation supplemented diet may be an important factor in protecting with Achyranthes aspera seed on larval Rohu Labeo rohita challenged with Mozambique Tilapia against bacterial challenge and in decreas- Aeromonas hydrophila. Journal of Aquatic Animal Heath 24:213–218. ing their percentage mortality. Chakraborty, S. B., and C. Hancz. 2011. Application of phytochemicals as immunostimulant, antipathogenic and antistress agents in finfish culture. Re- In the present study, a dietary allspice level of 10 g/kg pro- views in Aquaculture 3:103–119. vided the best survival, growth performance, and feed utilization Du, W. X., C. W. Olsen, R. J. Avena-Bustillos, T. H. McHugh, C. E. Levin, R. in Mozambique Tilapia fry. Allspice can be used as a growth Mandrell, and M. Friedman. 2009. Antibacterial effects of allspice, garlic, and 148 YILMAZ AND ERGUN¨

oregano essential oils in tomato films determined by overlay and vapor-phase Rattanachaikunsopon, P., and P. Phumkhachorn. 2009. Prophylactic effect of methods. Journal of Food Science 74:390–397. Andrographis paniculata extracts against Streptococcus agalactiae infection Duke, J. A., M. J. Bogenschutz-Godwin, J. duCellier, and P. A. K. Duke. 2002. in Nile Tilapia (Oreochromis niloticus). Journal of Bioscience and Bioengi- Allspice. Page 13 in J. A. Duke, M. J. Bogenschutz-Godwin, J. duCellier, neering 107:579–582. and P. A. K. Duke, editors. Handbook of medicinal herbs, 2nd edition. CRC Shylaja, M. R., and K. V. Peter. 2004. The functional role of herbal spices. Press, Boca Raton, Florida. Pages 11–21 in K. V. Peter, editor. Handbook of herbs and spices. Woodhead El-Sayed, A.-F. M. 2006. Tilapia culture. CAB International, Wallingford, UK. Publishing, Boca Raton, Florida. Ergun,¨ S., S. Yılmaz, and M. Yigit. 2011. Effects of thyme, rosemary and Vaseeharan, B., and R. Thaya. 2014. Medicinal plant derivatives as immunos- fenugreek on some hematological and immunological parameters of tilapia timulants: an alternative to chemotherapeutics and antibiotics in aquaculture. Oreochromis mossambicus. Pages 18–21 in Mediterranean Aquaculture 2020. Aquaculture International 22:1079–1091. Aquaculture Europe, Rhodes, Greece. Yılmaz, S., and S. Ergun.¨ 2012. Effects of garlic and ginger oils on hematolog- Harikrishnan, R., C. Balasundaram, and M. S. Heo. 2011. Impact of plant prod- ical and biochemical variables of Sea Bass Dicentrarchus labrax. Journal of ucts on innate and adaptive immune system of cultured finfish and shellfish. Aquatic Animal Health 24:219–224. Aquaculture 317:1–15. Yılmaz, S., S. Ergun,¨ and E. S¸. Ç elik. 2012a. Effects of herbal supplements Kikuzaki, H., S. Hara, Y. Kawai, and N. Nakatani. 1999. Antioxidative phenyl- on growth performance of Sea Bass (Dicentrarchus labrax): change in body propanoids from berries of Pimenta dioica. Phytochemistry 52:1307–1312. composition and some blood parameters. Journal of Bioscience and Biotech- Krishnamoorthy, B., and J. Rema. 2004. Allspice. Pages 117–139 in K. V. nology 1:217–222. Peter, editor. Handbook of herbs and spices. Woodhead Publishing, Boca Yılmaz, S., S. Ergun,¨ and E. S¸.Ç elik. 2013a. Effect of dietary herbal supplements Raton, Florida. on some physiological conditions of Sea Bass Dicentrarchus labrax. Journal Mars, B. 2007. The desktop guide to herbal medicine: the ultimate multidis- of Aquatic Animal Health 25:98–103. ciplinary reference to the amazing realm of healing plants. Basic Health Yılmaz, S., S. Ergun,¨ and N. Soytas¸. 2013b. Dietary supplementation of cumin Publications, Laguna Beach, California. (Cuminum cyminum) preventing streptococcal disease during first-feeding of Nayak, Y., and D. Abhilash. 2008. Protection of cyclophosphamide induced Mozambique Tilapia (Oreochromis mossambicus). Journal of Bioscience and myelosuppression by alcoholic extract of Pimenta dioica leaves in mice. Biotechnology 2:117–124. Pharmacologyonline [online serial] 3:719–723. Yılmaz, S., S. Ergun,¨ and N. Soytas¸. 2013c. Herbal supplements: useful tools Nguyen, T. N., D. A. Davis, and I. P. Saoud. 2009. Evaluation of alternative for preventing streptococcal disease during first–feeding of tilapia fry, Ore- protein sources to replace fish meal in practical diets for juvenile tilapia, ochromis mossambicus. Israeli Journal of Aquaculture Bamidgeh [online Oreochromis spp. Journal of the World Aquaculture Society 40:113–121. serial] 65:article IJA 65.2013.833. Raa, J. 1996. The use of immunostimulatory substances in fish and shellfish Yılmaz, S., S. Ergun,¨ and N. Turk.¨ 2012b. Effects of cumin-supplemented farming. Reviews in Fisheries Science 4:229–288. diets on growth and disease (Streptococcus iniae) resistance of tilapia (Ore- Ramos, A., A. Visozo, J. Piloto, A. Garcia, C. A. Rodriguez, and R. Rivero. 2003. ochromis mossambicus). Israeli Journal of Aquaculture Bamidgeh [online Screening of antimutagenicity via antioxidant activity in Cuban medicinal serial] 64:article IJA 64.2012.768. plants. Journal of Ethnopharmacology 87:241–246. Downloaded by [Department Of Fisheries] at 23:52 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:53 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 The Effects of a Sublethal Dose of Botulinum Serotype E on the Swimming Performance of Channel Catfish Fingerlings Rachel Beechama, Torri Thomasa, Dana X. Gaob & Patricia S. Gauntb a Department of Natural Science, Mississippi Valley State University, 14000 Highway 82 West, Itta Bena, Mississippi 38941, USA b College of Veterinary Medicine, Mississippi State University, Post Office Box 197, 127 Experiment Station Road, Stoneville, Mississippi 38776, USA Published online: 05 Aug 2014.

To cite this article: Rachel Beecham, Torri Thomas, Dana X. Gao & Patricia S. Gaunt (2014) The Effects of a Sublethal Dose of Botulinum Serotype E on the Swimming Performance of Channel Catfish Fingerlings, Journal of Aquatic Animal Health, 26:3, 149-153, DOI: 10.1080/08997659.2014.902873 To link to this article: http://dx.doi.org/10.1080/08997659.2014.902873

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:149–153, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.902873

ARTICLE

The Effects of a Sublethal Dose of Botulinum Serotype E on the Swimming Performance of Channel Catﬁsh Fingerlings

Rachel Beecham* and Torri Thomas Department of Natural Science, Mississippi Valley State University, 14000 Highway 82 West, Itta Bena, Mississippi 38941, USA Dana X. Gao and Patricia S. Gaunt College of Veterinary Medicine, Mississippi State University, Post Ofﬁce Box 197, 127 Experiment Station Road, Stoneville, Mississippi 38776, USA

Abstract Visceral toxicosis of catfish (VTC) is a disease of cultured Channel Catfish Ictalurus punctatus in the Mississippi Delta region and surrounding states. The etiology of VTC is associated with botulinum serotype E (BoNT/E), which causes blockage of acetylcholine release at the neuromuscular junction, leading to weakness and paralysis of skeletal muscles (including those involved in swimming). This study attempted to determine if sublethal exposure to purified BoNT/E caused reductions in swimming performance and metabolism of Channel Catfish. Catfish swimming performance was assessed on stocker-sized Channel Catfish (mean weight ± SD, 62.35 ± 2.5 g) with 10 sham-injected fish and 10 fish injected with a sublethal dose of BoNT/E. A modified Blazka-type swim chamber was used to assess swimming performance. We injected Channel Catfish with either 0.015% trypsin or 400 pg purified BoNT/E digested with 0.015% trypsin intracoelomically, then acclimated an individual catfish in the swim chamber for 17 h prior to the swimming trial. Water temperature was maintained at ∼28◦C, and dissolved oxygen (DO) was between 4 and 7 mg/L. A critical swimming speed (Ucrit) protocol was followed, and DO and temperature were monitored every 2 min throughout the swim trial. Cost of transport was calculated from the oxygen consumption at each test speed (10–70 cm/s). There was a statistical difference between the Ucrits (P = 0.0034), but no differences were found between the cost of transports (P = 0.67) of the sham-injected and BoNT/E groups. There was a difference in the cost of transport as it relates to the speeds tested (P < 0.0001), cost of transports being highest at low speeds and decreasing as speed increased. These results indicate that botulinum E interferes with the swimming speed of the catfish, which could contribute to the mortality from the disease of VTC and potentially make the fish more susceptible to predation. Downloaded by [Department Of Fisheries] at 23:53 16 November 2014

Visceral toxicosis of catfish (VTC) is a disease that causes the neuromuscular junction, leading to weakness and paraly- high mortality of market-sized cultured Channel Catfish Ictalu- sis of skeletal muscles (including those involved in swimming; rus punctatus in the Mississippi Delta region and surrounding Chatla et al. 2012). southeastern states (Wise et al. 2004). Fish infected with VTC Death from BoNT intoxication in humans and other mam- show erratic swimming and progressive muscular weakness, mals is attributed to suffocation resulting from loss of muscular leading to paralysis, lethargy, and death (Gaunt et al. 2007). control of the diaphragm (Roder 2004). Presumably, a lethal The etiology of this syndrome was shown to be associated with dose of botulinum in a catfish would result in a loss of muscular botulinum serotype E (BoNT/E). Signs and lesions of fish in- control of the branchial pump in its effort to force water over the jected with BoNT/E resembled those of VTC (Chatla et al. gills, ending in respiratory failure. In our field investigations, we 2012). Botulinum causes blockage of acetylcholine release at noted that not all VTC-affected fish died (Gaunt et al. 2007). It

*Corresponding author: [email protected] Received September 20, 2013; accepted February 14, 2014 149 150 BEECHAM ET AL.

is assumed that sublethal doses of botulinum indirectly increase mortality in fish by decreasing control over skeletal muscles controlling swimming. A number of toxins are known to affect swimming performance in fishes. Experimental exposure of Japanese Medaka Oryzias latipes to polyamine biosynthesis inhibitors demonstrated that the time to fatigue in the swimming endurance test was the most sensitive indicator of exposure (Ortego et al. 1994) by showing a decrease in the time to fatigue with increased dosage. A sublethal exposure caused reductions in both swim- FIGURE 1. Schematic of the swim tunnel respirometer as described in ming performance and metabolism, which affected the fish’s Beecham (2004). ability to grow, avoid predators, compete for food, and later reproduce. The goal of this study was to determine if sublethal exposure to purified BoNT/E caused reductions in swimming maintained at room temperature (27.27 ± 0.2◦C and 5.04 ± performance and metabolism of channel catfish. 0.1 mg/L dissolved oxygen [DO]) prior to testing. Swimming trials were conducted using a modified swim chamber respirometer (Blazka et al. 1960; Beecham 2004). The 100-L swim METHODS chamber system consisted of a swimming tube inside a con- Experimental design.—This project was implemented un- tainer chamber (Figure 1). A Dayton DC variable-speed motor der the approval of Mississippi State University Animal Care (Model 2Z846D) was used to turn the propeller located inside and Use Committee and the Institutional Biosafety Committee. the swimming tube. The fish was placed inside the swimming Two groups each of 11 Channel Catfish stockers (sham-injected tube, and an end cap was attached to keep the fish from leaving fish: TL = 21.3 ± 0.4 cm, weight 58.0 ± 3.4 g; BoNT/E- the tube. Water flowed through the swimming tube and then back injected fish: TL = 21.8 ± 0.5 cm, weight 66.7 ± 3.4 g) were around to the front of the tube in the container chamber. Flow tested to determine the effect of BoNT/E activated with 0.015% rate (cm/s) was calibrated with a Marsh McBirney flowmeter trypsin (GIBCO 10 × ) on swimming performance. Group 1 and then equated to the speed of the propeller shaft (in rpm). (injected) was injected intracoelomically (IC) with a sublethal Oxygen consumption and water temperature were monitored us- dose of BoNT/E, and group 2 (sham injected) was injected with ing a calibrated YSI polarographic DO meter (Model 52; YSI, 0.015% trypsin only. Fish were netted arbitrarily without bias Yellow Springs, Ohio; accuracy: ± 0.03 mg/L oxygen, ± 0.1◦C and assigned to each group. temperature). The oxygen probe was mounted into the top of the Botulinum neurotoxin dilutions.—Botulinum was handled swim chamber to ensure that the tip was constantly exposed to within a class 2 biosafety cabinet equipped with HEPA filters. moving water. Oxygen and temperature readings were measured The BoNT/E complex (Metabiologics, Madison, Wisconsin) at 2-min intervals for each 30-min swimming period. was provided at a concentration of 1 mg/mL with specific toxi- Following acclimation, the test fish was swum slowly for 1 h city of 6 × 107 MLD50/mg posttrypsin activation. We diluted at 10 cm/s to allow adjustment to swimming against a current. BoNT/E with sterile gel phosphate buffer (Fisher Scientific, After 1 h, water velocity was increased by 10 cm/s every 30 min Fairlawn, New Jersey; Becton Dickerson, Sparks, Maryland) to until the fish fatigued. Fatigue was defined as the point at which yield 104 pg/µL of stock BoNT/E. We further diluted 60 µLof the fish could no longer maintain position in the chamber and the stock BoNT/E with gel phosphate buffer (pH 6.2) then acti- collided against the rear grating of the swim chamber, at which

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 vated with trypsin to yield 4,000 pg/1,000 µL. The dilution was time the run was terminated. incubated at room temperature (∼22–23◦C) for 30 min prior to Standard length, FL, and TL measurements were recorded to fish injections. the nearest 0.1 cm at the conclusion of each swimming trial to Experimental fish.—One stocker-size Channel Catfish and be used for comparison to ensure that fish in each group were of two fingerling Channel Catfish (5.0 g) were arbitrarily netted a similar size. Mass was measured to the nearest 0.01 g, and test from a holding vat and injected IC with 100 µL of the activated fish were placed in a holding tank for observation of any clin- BoNT/E. The stocker fish was acclimated in the swim chamber ical signs previously reported in fish with BoNT/E, including for 17 h prior to the swimming trial (no water flow during hyperactivity, lethargy, paresis, paralysis, inability to right itself this acclimation). The two fingerling catfish were placed in a in water column, exophthalmia, distention of the coelomic area, 60-L free-flow aquaria and were used as positive controls for and death (Yule et al. 2006; Gaunt et al. 2007). After 48 h, surviv- the assessment of the BoNT/E potency as previously described ing fish were euthanized in tricaine methanesulfonate (MS-222) (Chatla et al. 2012). The fingerling catfish were observed for anesthetic (Western Chemical). signs of VTC, including mortality for 96 h. Swimming calculations.—Swimming performance (Ucrit) Swimming protocol.—Fish were not fed for 36 h prior to and cost of transport (metabolic cost of swimming) were mea- testing. Water in the holding tanks and swim chamber was sured for each stocker-sized fish. Critical swim speed (Ucrit) EFFECT OF BOTULINUM SEROTYPE E ON CHANNEL CATFISH 151

was calculated using Brett’s (1964) equation the swim speed (cm/s) to determine the total cost of transport as described by Schmidt-Nielsen (1972). Statistical methods.—Data were analyzed using SAS 8.02 = + / , Ucrit U1 (U2)(T 1 T 2) (SAS Institute 2013). Regression analysis was used to describe the relationships between the swim speed and both metabolic where U1 was the highest velocity maintained for the prescribed rate and cost of transport. We used ANOVA to compare the time (30 min), U2 was the velocity increment (10 cm/s), T1was metabolic rates and cost of transport across all the swimming the time (min) fish swim at fatigue velocity, and T2wasthe speeds and between the two test groups of fish. Least-squares prescribed period of swimming (30 min). means was used to determine where differences occur. A t-test The oxygen consumption data were used to calculate the to- was used to compare the critical swimming speeds of the two tal cost of transport (TCOT, cal/g/km) for each Channel Catfish test groups of fish. at each swimming speed. This was done using the methods described by Brafield and Solomon (1972) and Schmidt-Nielsen RESULTS (1972), where the oxygen used at a particular speed was de- The critical swimming speeds of the sham-injected (54.76 ± termined by subtracting the oxygen value at the end of the 2.8 cm/s; N = 11) and BoNT/E-injected (40.75 ± 3.1 cm/s; prescribed swim velocity from the oxygen value at the begin- N = 11) Channel Catfish were significantly different (P = ning of the prescribed swim velocity. That oxygen consumed 0.003). No significant differences in cost of transport were found value was then divided by the prescribed time period (30 min between the two groups (P = 0.67). However, significant dif- or 0.5 h for swimming speeds from 20 to 70 cm/s). That value ferences were found across the swimming speeds tested (10– was multiplied by 100 L and divided by the weight of the fish 60 cm/s, P < 0.0001; Figure 2), cost of transports being highest in grams, which gives the oxygen consumption (mg O2/g/h). at low speeds and decreasing as speed increased. An exponen- Oxygen consumption is multiplied by the oxycalorific coeffi- tial regression line was fit to the cost of transport data for each e–0.037x 2 cient of 3.25 cal/mg O2 (Brafield and Solomon 1972), giving group (sham-injected Y = 10.653 , R = 0.61; injected Y = the metabolic rate (cal/g/h). Metabolic rate is then divided by 7.7865e–0.032x, R2 = 0.58). The size (P = 0.46) and weight Downloaded by [Department Of Fisheries] at 23:53 16 November 2014

FIGURE 2. The cost of transport (cal/g/km) of sham-injected and injected (injected with a sublethal dose of BoNT/E) stocker-size Channel Catfish over a range of swimming speeds (10–60 cm/s). An exponential regression line was fit to each group. [Color figure available online] 152 BEECHAM ET AL.

(P = 0.09) of the stocker-sized catfish used in this study were not Because metabolism accounts for the largest portion of the different between the sham-injected and the BoNT/E-injected energetic budget of an individual fish (Webb 1978; Brett 1979; catfish. Elliott 1979), understanding the way fish partition energy among The two fingerlings that were injected as positive controls their metabolic demands helps us to determine how fish parti- when the stockers were injected with BoNT/E died within 48 h tion their available energy (energy budget) resources among of injection with lesions previously described for VTC (Chatla their metabolic demands and various activities such as feeding, et al. 2012). Two stocker-sized BoNT/E Channel Catfish also excretion, reproduction, growth, and swimming. The laws of died after injection and swimming, but those two fish were not thermodynamics mandate that energy resources used in one as- used in swimming trial data. pect of metabolism will not be available for use in others. For example, the energy used to process and digest food will not be available for locomotion or growth. Obviously, trade-offs must DISCUSSION be made among the various components of metabolism (Calow There are only a few published studies on swimming perfor- 1985). Understanding the cost of transport (the amount of energy mance and energetics of Channel Catfish. The mean swimming used per gram of fish per unit distance) for the Channel Catfish performance value (Ucrit) of our sham-injected stocker catfish is important. Though the Ucrit (maximum sustained swimming compared favorably with those previously reported in Channel speed) was reduced in the BoNT/E-injected catfish, no differ- Catfish (Table 1). These results demonstrate that our sham- ences in cost of transport were found. This lack of difference injected fish performed similarly to Channel Catfish in other was unexpected, but indicated that the BoNT/E was affecting studies under similar environmental conditions and indicates the neuromuscular physiology of the fish. Further studies are that the differences in performance between the sham-injected needed to determine why there was no effect on the metabolic and BoNT/E-injected fish in this study are due to the effects of physiology of these intoxicated fish. the toxin. The decreased Ucrit of the injected fish likely reflected Because BoNT/E is a biological toxin, we needed to ascer- the paresis induced by BoNT/E (Gaunt et al. 2007). tain its potency by injecting fingerlings with a known lethal Swimming requires the integration of numerous neuro- dose (400 pg/fingerling; Chatla et al. 2012). From pilot studies muscular processes (Thomas and Donahoo 1977; Taylor and (unpublished), this dose was not lethal to stocker-sized catfish McPhail 1985; Hawkins and Quinn 1996; Adams et al. 2003) (∼66.7 g) but produced the sublethal, lethargic effect required and is considered a relevant measure of an organism’s fitness for implementation of this study. The stockers appeared less and physiological compensation to its environment (Huey and susceptible to the effects of BoNT/E because the dose that the Stevenson 1979; Arnold 1986; Parsons 1990; Ortego et al. 1994; stockers received was ∼6.0 pg/g fish compared with the median Parsons and Carlson 1998). Swimming performance tests have lethal dose (LD50) of 2.6 pg/g in Channel Catfish fingerlings re- been shown to be sensitive indicators of sublethal exposure to ported previously (Chatla et al. 2012). Future studies will help contaminants. The swimming endurance of Rainbow Trout On- determine whether the difference in response to BoNT/E be- corhynchus mykiss was decreased by copper exposure at 12% tween the two sizes of fish is related to the fishes’ physiologic of the LD50 (Waiwood and Beamish 1978), while the swim- stages (stocker versus fingerling size) or other experimental ming performance of Brook Trout Salvelinus fontinalis was factors. decreased by 33% of the LD50 of malathion (Post and Leasure Brett and Groves (1979) indicated that as much as 25% of 1974). Therefore, the difference in swimming performance be- ingested energy may be invested in active metabolism (swim- tween BoNT-injected and sham-injected fish measured in this ming) by carnivorous fishes. While commercially raised Chan- experiment was expected. nel Catfish are usually considered omnivorous fish, they could

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 be feeding on each other, especially during the cold months TABLE 1. A comparison of the mean critical swimming speeds (Ucrit), mean when they are fed less than one time a week. These cold months fish weight, mean water temperature, and mean fish length of Channel Catfish are also when VTC outbreaks typically occur. It was hypoth- determined in various published studies for comparison with the data collected esized that during a VTC outbreak, BoNT killed fish not only from sham-injected fish in this study. directly but also indirectly by decreasing the ability to swim and Mean flee from predators (Chatla et al. 2012). Studies are underway to Mean Mean Mean temperature quantify BoNT/E in field outbreaks of VTC and correlate these Study Ucrit weight (g) length (◦C) doses with BoNT/E from our experimental studies. The difference in the critical swimming speeds between Current study 54.8 58 21.3 TL 27 the BoNT/E-injected and the sham-injected fish indicates that Hocutt (1973) 58.5 14.0–15.4 TL 25 this toxin interfered with the swimming speed of the Chan- Sylvester (1992) 55.2 191.5 20 nel Catfish. This could contribute to the mortality from VTC Bartlett (1998) 52.1 131 16–27 SL 25 and potentially make the fish more susceptible to predation. Beecham (2004) 59.28 100.2 19.4 SL 19–21 Future research into the influence of BoNT/E on swimming EFFECT OF BOTULINUM SEROTYPE E ON CHANNEL CATFISH 153

performance in fish is needed to further explain lack of its phys- Gaunt, P., S. R. Kalb, J. R. Barr. 2007. Detection of botulinum type E toxin in iologic effects on metabolic rates or cost of transport. Channel Catfish with visceral toxicosis syndrome using catfish bioassay and endopep mass spectrometry. Journal of Veterinary Diagnostic Investigation 19:349–354. ACKNOWLEDGMENTS Hawkins, D. K., and T. P. Quinn. 1996. Critical swimming velocity and asso- This project was funded by U.S. Department of Agricul- ciated morphology of juvenile coastal Cutthroat Trout (Oncorhynchus clarki clarki), steelhead trout (Oncorhynchus mykiss), and their hybrids. Canadian ture (USDA) ARS-58-6402-7-190 and USDA NIFA 2011- Journal of Fisheries and Aquatic Sciences 53:1487–1496. 67015 30174. Additional funding was provided through MS- Hocutt, C. H. 1973. Swimming performance of three warmwater fishes exposed INBRE and the U.S. Department of Education Title III, Plan to a rapid temperature change. Chesapeake Science 14:11–16. B Program. A special thanks to Rodgers Polk for his technical Huey, R. B., and R. D. Stevenson. 1979. Integrating thermal physiology and ecol- assistance. ogy of ectotherms: a discussion of approaches. American Zoologist 19:357– 366. Ortego, L. S., A. C. Nimrod, W. T. Brehm, G. R. Parsons, and W. H. Benson. 1994. Early life stage effects in Medaka (Oryzias latipes) following in ovo REFERENCES exposure to polyamine biosynthetic inhibitors. Ecotoxicology and Environ- Adams, S. R., G. L. Adams, and G. R. Parsons. 2003. Critical swimming mental Safety 28:329–339. speed and behavior of juvenile Shovelnose Sturgeon and Pallid Sturgeon. Parsons, G. R. 1990. Metabolism and swimming efficiency of the Bonnethead Transaction of the American Fisheries Society 132:392–397. Shark, Sphyrna tiburo. Marine Biology 104:363–367. Arnold, S. J. 1986. Laboratory and field approaches to the study of adaptation. Parsons, G. R., and J. K. Carlson. 1998. Physiological and behavioral responses Pages 157–180 in M. E. Feder and G. V. Lauder, editors. Predator-prey to hypoxia in the Bonnethead Shark, Sphyrna tiburo: routine swimming and relationships. University of Chicago Press, Chicago. respiratory regulation. Fish Physiology and Biochemistry 19:189–196. Bartlett, J. H. W. 1998. The effect of dissolved oxygen and absorptive state Post, G., and R. A. Leasure. 1974. Sublethal effect of malathion to three salmonid on the swimming performance and swimming respiration of juvenile Chan- species. Bulletin of Environmental Contamination Toxicology 12:312–319. nel Catfish, Ictalurus punctatus (Rafinesque). Master’s thesis. University of Roder, J. D. 2004. Biotoxins. Pages 98–99 in K. H. Plumlee, editor. Clinical Mississippi, Oxford. veterinary toxicology. Mosby, St. Louis, Missouri. Beecham, R. V. 2004. A study of the swimming capabilities of Blue, Ictalurus SAS Institute. 2013. SAS/STAT user’s guide, version 9.3. SAS Institute, Cary, furcatus, and Channel, I. punctatus, catfish. Doctoral dissertation. University North Carolina. of Mississippi, Oxford. Schmidt-Neilsen, K. 1972. Locomotion: energy cost of swimming, flying, and Blazka, P. M., M. Volf, and M. Cepela. 1960. A new type of respirometer for running. Science 177:222–228. the determination of the metabolism of fish in an active state. Physiologia Sylvester, J. L. Jr. 1992. Swimming performance and energetics in Bluegill Bohemoslovaca 9:553–558. sunfish, Lepomis macrochirus, Largemouth Bass, Micropterus salmoides, Brafield, A. E., and D. J. Solomon. 1972. Oxycalorific coefficients for animals and Channel Catfish, Ictalurus punctatus: a laboratory and field analysis. respiring nitrogenous substrates. Comparative Biochemistry and Physiology Master’s thesis. University of Mississippi, Oxford. 443:837–841. Taylor, E. B., and J. D. McPhail. 1985. Variation in burst and prolonged swim- Brett, J. R. 1964. The respiratory metabolism and swimming performance of ming performance among British Columbia populations of Coho Salmon, young Sockeye Salmon. Journal of the Fisheries Research Board of Canada Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences 21:1183–1226. 42:2029–2033. Brett, J. R. 1979. Environmental factors and growth. Pages 599–675 in W. S. Thomas, A. E., and M. J. Donahoo. 1977. Differences in swimming performance Hoar and D. J. Randall, editors. Fish physiology, volume VIII. Academic among strains of Rainbow Trout (Salmo gairdneri). Journal of the Fisheries Press, New York. Research Board of Canada 34:304–306. Brett, J. R., and T. D. D. Groves. 1979. Physiological energetics. Pages 279– Waiwood, K. G., and F. W. H. Beamish. 1978. The effects of copper, pH, and 352 in W. S. Hoar and D. J. Randall, editors. Fish physiology volume VIII. hardness on the critical swimming performance of Rainbow Trout, Salmo Academic Press, New York. gairdneri. Water Research 12:611–619. Calow, P. 1985. Adaptive aspects of energy allocation. Pages 13–31 in P. Tyler Webb, P. W. 1978. Partitioning of energy into metabolism and growth. Pages and P. Calow, editors. Fish energetics: new perspectives. Johns Hopkins 184–214 in S. D. Gerking, editor. Ecology of freshwater fish production. University Press, Baltimore, Maryland. Blackwell Scientific Publications, Oxford, UK. Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 Chatla, K., P. Gaunt, L. Hanson, D. Gao, and R. Wills. 2012. Determina- Wise, D., A. Camus, T. Schwedler, and J. Terhune. 2004. Health management. tion of the median lethal dose of botulinum serotype E in Channel Cat- Pages 444–502 in C. S. Tucker and J. A. Hargreaves, editors. Biology and fish (Ictalurus punctatus) fingerlings. Journal of Aquatic Animal Health 24: culture of Channel Catfish, 2nd edition, Elsevier, Amsterdam. 105–109. Yule, A. M., I. K. Barker, J. W. Austin, and R. D. Moccia. 2006. Toxicity of Elliott, J. M. 1979. Energetics of freshwater teleosts. Symposium of the Zoo- Clostridium botulinum type E neurotoxin to Great Lakes fish: implications logical Society of London 44:29–61. for avian botulism. Journal of Wildlife Diseases 42:479–493. This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:53 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Microbiological Investigation of Persistent Mortalities in Litopenaeus vannamei Grown in Low Saline Waters in India Hirekudel Sanathkumara, Charan Ravia, Suresh Babu Padinhatupurayila, Mini Mola, Jilagam Krishna Prasada & Binaya Bhusan Nayaka a Central Institute of Fisheries Education, Seven Bungalows, Versova, Andheri W, Mumbai 400061, India Published online: 05 Aug 2014.

To cite this article: Hirekudel Sanathkumar, Charan Ravi, Suresh Babu Padinhatupurayil, Mini Mol, Jilagam Krishna Prasad & Binaya Bhusan Nayak (2014) Microbiological Investigation of Persistent Mortalities in Litopenaeus vannamei Grown in Low Saline Waters in India, Journal of Aquatic Animal Health, 26:3, 154-159, DOI: 10.1080/08997659.2014.902875 To link to this article: http://dx.doi.org/10.1080/08997659.2014.902875

PLEASE SCROLL DOWN FOR ARTICLE

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:154–159, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.902875

ARTICLE

Microbiological Investigation of Persistent Mortalities in Litopenaeus vannamei Grown in Low Saline Waters in India

Hirekudel Sanathkumar,* Charan Ravi, Suresh Babu Padinhatupurayil, Mini Mol, Jilagam Krishna Prasad, and Binaya Bhusan Nayak Central Institute of Fisheries Education, Seven Bungalows, Versova, Andheri W, Mumbai 400061, India

Abstract Microbial diseases are a serious hindrance to successful shrimp aquaculture. The Paciﬁc white shrimp Litopenaeus vannamei is an exotic species recently introduced in India to supplement the failing aquaculture of the Asian tiger shrimp Penaeus monodon due to viral diseases. However, after a brief initial success, the aquaculture of L. vannamei is also experiencing serious problems due to microbial diseases. In this study, we conducted a microbiological investigation into the problem of persistent mortalities in selected L. vannamei farms on the southeastern coast of India. The infected shrimps were positive for the white spot syndrome virus (WSSV) by a nested PCR, though no visible white spots were present on the animals. The shrimps were heavily colonized by Vibrio parahemolyticus, which were isolated from the hepatopancreas, gills, and the body surface. The pond water, despite being low saline groundwater, harbored large numbers of V. parahemolyticus and other Vibrio species, and V. parahemolyticus isolated from L. vannamei were resistant to β-lactam antibiotics and cephalosporins. Our results strongly suggest that the persistent mortalities of L. vannamei were due to a co-infection by V. parahemolyticus and WSSV.

The Paciﬁc white shrimp Litopenaeus vannamei was intro- of microbial disease outbreaks. To date, no detailed scientiﬁc duced in India in the late 2000s because the aquaculture of the study has been done to understand the problems of L. vannamei Asian tiger shrimp Penaeus monodon was failing due to devas- aquaculture in India, particularly the diseases affecting L. van- tating viral diseases and the consequent economic losses. Sev- namei. eral attributes of L. vannamei, such as the faster growth rate and Though L. vannamei was introduced to India and other its ability to tolerate wider ranges of salinity (1–40‰) (Davis Asian countries as a hardy and disease-tolerant species, several et al. 2004), pH, and dissolved oxygen, made it popular, and it reports from all over the world suggest that a number of

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 was quickly adopted in India as an alternative species to P. mon- microbial diseases adversely affect its aquaculture, prominent odon. The aquaculture of this exotic species was allowed but among them being the viral diseases. At least four different with strict guidelines to prevent its indigenization, its escape to viruses are considered to be highly pathogenic to L. vannamei: the natural ecosystem, and the outbreak and spread of diseases. white spot syndrome virus (WSSV), infectious hypodermal The seeds, which are produced in licensed hatcheries using im- and hematopoietic necrosis virus (IHHNV), yellow head virus ported broodstock, are screened and quarantined on arrival in (YHV), and Taura syndrome virus (TSV) (Lightner and Redman India by the Aquatic Quarantine Facility (AQF) for seven vi- 1998; Seibert and Pinto 2012). In India, WSSV, hepatopancre- ral pathogens listed by the Ofﬁce International des Epizooties´ atic parvo virus (HPV), and Monodon baculovirus (MBV) have (OIE) (Remany et al. 2010). A spectacular initial success of L. been responsible for large scale production losses of P.monodon vannamei aquaculture in India led to its rapid expansion, vir- (Umesha et al. 2008). A few studies from India have reported tually replacing P. monodon aquaculture. However, the rapidly the occurrence of WSSV and IHHNV infections in L. vannamei, growing L. vannamei culture is being threatened by a number which either individually or together have been responsible for

*Corresponding author: [email protected] Received December 16, 2013; accepted February 14, 2014

154 INVESTIGATION OF PERSISTENT MORTALITIES IN LITOPENAEUS VANNAMEI 155

TABLE 1. Oligonucleotide primers used in this study. F = forward, R = reverse.

Target virus Primers (5–3) Product size (bp) Reference IHHNV 389F–CGGAACACAACCCGACTTTA 389 Tang et al. (2000) 389R–GGCCAAGACCAAAATACGAA WSSV F1–ATCATGGCTGCTTCACAGAC 982 Kimura et al. (1996) F2–GGCTGGAGAGGACAAGACAT F3–TCTTCATCAGATGCTACTGC 570 (nested) F4–TAACGCTATCCAGTATCACG HPV F–ATAGAACGCATAGAAAACGCT 265 Manjanaik et al. (2005) R–GGTGGCGCTGGAATGAATCGCT

mortalities in culture systems (Balakrishnan et al. 2011; Moger using previously described primers and protocols (Bej et al. et al. 2011; Otta et al. 2014). With the expansion of L. vannamei 1999). aquaculture, unregulated procurement of seeds from unlicensed PCR detection of viruses.—For PCR detection of viruses, sources and poor culture practices have resulted in the outbreak DNA samples were extracted from the walking legs, gills, and and spread of microbial diseases in L. vannamei. In this study, muscle and combined for the detection of WSSV and from the we sought to determine the bacterial composition of infected hepatopancreas for the detection of HPV using a previously L. vannamei from ponds that showed similar disease symptoms described DNA extraction protocol (Otta et al. 2003). For the and a persistent but moderate level of shrimp mortalities in detection of WSSV, a nested PCR protocol previously described Andhra Pradesh on the east coast of India. Further, all these by Kimura et al. (1996) was used. For IHHNV and HPV, PCR ponds are unique in terms of the culture practices, since low assays described by Tang et al. (2000) and Manjanaik et al. saline groundwater was used for growing L. vannamei.The (2005), respectively, were used. The primers and the expected results of this study showed that both V. parahemolyticus and ampliﬁcation products are shown in the Table 1. The prod- WSSV were involved in the mortalities of L. vannamei. ucts of PCR were separated on 1.6% agarose gels, stained with ethidium bromide (0.5 µg/mL), and photographed using a gel documentation system (BioRad). METHODS Determination of antibiotic resistance of V.parahemolyticus Sample collection and microbiological examination.— isolates.—The antibiotic resistance of V. parahemolyticus iso- Samples of L. vannamei were collected from six different ponds lates was determined by the standard antibiotic disk diffusion that were using low saline groundwater of about 6‰. Ponds method according to the Clinical and Laboratory Standards In- belonging to different farmers were located in different loca- stitute guidelines (CLSI 2006). The antibiotics tested were ampi- tions within a radius of 5 km and were not physically connected cillin, amoxicillin–clavulanic acid, cephalothin, nalidixic acid, with each other. All these ponds exhibited similar patterns of norﬂoxacin, erythromycin, tetracycline, and chloramphenicaol. ∼ mortalities that started with initial high mortality ( 20–30%) The production of extended spectrum β-lactamase (ESBL) was followed by persistent but moderate levels (100–200 shrimps tested on Hi-Chrome ESBL agar (Hi-Media, Mumbai, India). per day; ∼0.5–1%) of mortalities. For microbiological examination, swabs were collected from

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 the hepatopancreas, gills, and appendages of the affected RESULTS shrimps and streaked on thiosulphate cirate bile salt sucrose The patterns of L. vannamei mortalities in all six ponds stud- (TCBS) agar and the Zobell marine agar (ZMA). For enumer- ied were similar. A detailed description of the observed mor- ation, 1 g of the hepatopancreas was homogenized in 9 mL of talities and the results of bacteriological study are provided in buffered peptone water with 1% NaCl and a 0.1-mL aliquot Table 2. Some of the common features of all these L. vannamei of this was spread-plated on TCBS and ZMA plates and incu- farms were the high stocking densities (200–400 post larvae bated at 30◦C for 24 h. The green colonies were presumptively [PL]/m2), use of low saline (5–6‰) tube-well water, use of mi- identified as V. parahemolyticus by a series of biochemical tests cronutrients, and antimicrobials. All the ponds were stocked for oxidase, oxidation/fermentation (O/F), salt tolerance, amino with specific pathogen free (SPF) post larvae. Following ini- acid decarboxylase and carbohydrate fermentation and were tial high mortalities of few thousand shrimps per day, a gradual subjected to a toxR-specific PCR for confirmation (Kim et al. decline in the mortality rates was noted, which stabilized at 100– 1999). The isolates were also subjected to PCR for the de- 200 shrimps dying every day during our investigation. In pond tection of virulence genes commonly associated with human 1, the initial mass mortality within 5 d of stocking wiped out pathogenic strains of V. parahemolyticus, namely thermostable the entire stock and the juveniles that survived the outbreak still direct hemolysin (tdh) and tdh-related hemolysin (trh) genes showed symptoms of infection. The moribund shrimps did not 156 SANATHKUMAR ET AL.

TABLE 2. Description of L. vannamei culture ponds and the bacteriological characteristics of shrimps and the water; DOC = days of culture.

Growth on TCBS Pond number Description Hepatopancreas Water 1 20 DOC: Large scale mortality started5dafter All green colonies, >106/g. 103/mL yellow and green stocking. Culture continued with regular colonies; majority (two-thirds) mortalities of about 200 shrimps/d. were yellow colonies. 2 28 DOC: Daily mortality of 100–200 shrimps. All green colonies, >106/g. >105/mL tiny yellow colonies; Moribund shrimps had a pinkish body and tail 102/mL green colonies. color. 3 28 DOC: Following initial mortalities of 500–1,000 All green colonies, 104/g. <1/mL green colonies. shrimps/d, the stock stabilized with mortalities of 50–100 shrimps/d. 4 28 DOC: Regular mortalities of 100 shrimps/d. All yellow colonies, All yellow colonies, >105/mL. Infected shrimps had a pinkish body and tail >105/g. color. 5 40 DOC: Persistent mortalities were as above, and All green colonies, 105/g. All green colonies, 6 × 102/mL. moribund shrimps had a pink body color. The hepatopancreas had turned a purple color. The tail also showed dark coloration. 6 61 DOC: After initial high mortalities, the culture All green colonies, >105/g. Yellow and green colonies, recovered and was continued with persistent 103/mL. mortalities of 50–100 shrimps/d.

have visible white spots on their body, but had a characteristic were purified and sequenced and were confirmed to be from reddish body discoloration (Figure 1A). The hepatopancreas of the WSSV genome by BLAST analysis (data not shown). The the shrimps also showed distinct reddish discoloration, as did samples were also tested for the presence of IHHNV and HPV, the tail fins (Figure 1B, C). In some shrimp samples (pond 5), but none of the samples was positive for either of these viruses. greenish discoloration of the walking legs was noticed. Micro- biological examination of the infected shrimps was focused on the isolation of Vibrio spp. and hence TCBS and ZMA were DISCUSSION used. The TCBS agar streaked with swabs taken from the cara- The devastating disease outbreaks in P. monodon in India pace, gills, and legs grew green and yellow colonies. Similarly, during the 1990s, involving mainly WSSV, forced the industry large numbers of green and yellow colonies grew on TCBS to look for alternate species of shrimp for aquaculture. The suc- spread-plated with hepatopancreas and water samples (Table 2) cess of L. vannamei in the Americas and many Asian countries and were identified as V. parahemolyticus by biochemical tests where it was introduced made it an attractive species for culture and toxR PCR. Some of the infected L. vannamei yielded only

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 V. parahemolyticus. None of these V. parahemolyticus harbored the virulence genes (tdh and trh) generally found in strains as- TABLE 3. Antimicrobial resistance patterns of V. parahemolyticus isolated from L. vannamei (n = 43), expressed as number and percent (in parentheses) sociated with human infections (data not shown). The results of of isolates resistant. antimicrobial susceptibility testing of V. parahemolyticus isolates are shown in Table 3. All V. parahemolyticus isolates (n = Number (% of 43) tested in this study were resistant to ampicillin, amoxicillin– Antibiotic (µg) isolates resistant) clavulanic acid, and cefepime. Apart from this, 58% of the iso- Ampicillin (10) 43 (100) lates were resistant to erythromycin, 11% to both nalidixic acid Amoxicillin/clavulanate (30) 43 (100) and tetracycline, and 6.9% to norfloxacin. All the isolates were Cefepime (30) 43 (100) sensitive to chloramphenicol. Nalidixic acid (30) 5 (11.6) The nested PCR used in this study amplified WSSV DNA Chloramphenicol (30) 0 (0) from all the samples in the second step suggesting there were Tetracycline (30) 5 (11.6) low levels of WSSV infections (Figure 2). The two control Erythromicin (15) 25 (58.1) samples of healthy L. vannamei as well as the negative controls Norfloxacin (10) 3 (6.9) were both negative in both steps. The amplification products INVESTIGATION OF PERSISTENT MORTALITIES IN LITOPENAEUS VANNAMEI 157

FIGURE 2. Detection of WSSV in L. vannamei by nested PCR.

several avenues for viral contamination exist, especially in the form of latent carriers of viruses (Lo et al. 1996; Chou et al. 1998; Flegel 2012). The infection of pathogen-free larvae can occur at any stage from seed production to grow out if the culture system is not biosecure. Litopenaeus vannamei is highly susceptible to WSSV infections and the virus has been responsible for severe economic losses in the western hemisphere in the past (Martorelli et al. 2010; Lightner 2011). Several natural hosts or carriers of WSSV have been reported and include copepods, worms, gastropods, insects, and polychaete worms (Hossain et al. 2001; Vijayan et al. 2005; Sanchez-Mart´ ´ınez et al. 2007), which are always present in the environment where L. vannamei is cultured. It is only in the recent past that WSSV caused large-scale infections in India and reservoirs of this virus, which can infect L. vannamei stocks, still exist in the environment. The coastal environment in general and wild shrimps and other crustaceans in particular have become the natural carriers of this virus due to the long-term contaminations of the aquatic bodies with WSSV from aquaculture pond efﬂuents. As long as its natural hosts are available, the virus cannot be eliminated from the environment, though the introduction of a new viral pathogen can be prevented by strictly implementing biosecurity measures. Though the detection of WSSV DNA in the second step suggests there was a low viral load in these shrimps, this does not preclude the possibility that the virus is responsible for their

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 FIGURE 1. Infected L. vannamei showing (A) discolored hepatopancreas mortality. It is possible that the viral infection might have pre- from the exterior, (B) red discoloration of the tail, and (C) red discoloration of disposed shrimps to bacterial infection by V. parahemolyticus, the hepatopancreas. or vice versa. In any case, the presence of WSSV in shrimps from different ponds with similar mortality patterns suggests in India. Within 5 years of its introduction, L. vannamei had that this virus will be a serious threat to L. vannamei aquacul- virtually replaced P. monodon on the east coast of India. The ture in India. The infection with WSSV generally causes mass introduction of L. vannamei also came with several biosafety mortality of up to 100% within 2–7 d, though the mortality rates recommendations to prevent viral infections. These included vary widely in different species of crustaceans. Sudden changes biologically securing farms from all sides as well as the top in salinity and temperature predispose shrimps to infections by and the purchase of seeds from licensed hatcheries that use SPF WSSV (Vidal et al. 2001; Liu et al. 2006). broodstock. However, despite using SPF larvae and following The focus of this study was the grossly infected hepatopan- precautionary measures to keep pathogens away, the infections creas of the shrimps, which showed varying coloration from of shrimps by opportunistic bacteria and viruses do occur for purple to dark pink. Our hypothesis of a Vibrio infection was various reasons. It is unrealistic to assume that the SPF lar- proved to be correct, since a large number of V. parahemolyti- vae will remain uninfected in a grow-out environment in which cus were isolated from the hepatopancreas (Table 2). The farms 158 SANATHKUMAR ET AL.

under investigation were using low saline (5–6‰) groundwa- resistant to nalidixic acid, tetracycline, and norfloxacin. Resis- ter for L. vannamei culture in the region and these farms were tance to β-lactam antibiotics with the lactamase inhibitors as physically disconnected from coastal water sources. It was in- well as to cephalosporins is worrying. deed surprising that V. parahemolyticus were present in shrimps Though the shrimp farmers in India are advised to adopt and pond water, and we were unable to precisely identify their biosecurity measures such as bird netting, crab fencing, wa- sources. However, some of the likely sources were infected PL, ter treatment, and reduced stocking density to prevent the en- water in which PL were transported from the hatchery to the try and spread of viruses, they are not strictly followed by all farms, contaminated farm equipment, or the groundwater itself. farmers primarily due to practical and economic considerations. Vibrios are naturally present in coastal waters and hence could Lessons must be learned from similar situations in other coun- be found in shrimp culture environments as well as in associ- tries, for example Taiwan, where L. vannamei replaced P. mon- ation with the shellfish as opportunistic pathogens. Infections odon completely in early 1990s following serious crop losses of shrimps take place when the animals are under stress due to due to WSSV infections in P. monodon. However, by late 1998, sudden and drastic physicochemical changes in the water or in- L. vannamei aquaculture collapsed completely due to outbreaks fections by viruses. Several Vibrio species can infect shrimps at of diseases and resulting mass mortalities (Tu et al. 1999). The various stages from larval to grow out and include V. harveyi, V. rapidly expanding L. vannamei aquaculture industry in India alginolyticus, V. penaeicida, and V. anguillarum (Karunasagar requires careful management to make it sustainable. To accom- et al. 1994; Lavilla-Pitogo et al. 1998; Vandenberghe et al. 1999). plish this, strict enforcement of good culture practices, promo- Luminescent vibrios are a major problem as causative agents of tion of alternatives to antibiotics such as immunostimulants and severe infections and mortalities of shrimps in their early life vaccines, and surveillance of ponds for outbreak of diseases are stages. In a recent study by Zhou et al. (2012), a nonluminescent necessary. but highly virulent strain of V. harveyi (HLB0905) was found to be responsible for the bacterial white muscle disease (BWMD) ACKNOWLEDGMENTS that caused mass mortalities of pond-cultured L. vannamei in The authors are grateful to W. S. Lakra, Director and Vice- China. Infections by vibrios require the presence of large num- Chancellor, Central Institute of Fisheries Education, for help bers of them, as has been experimentally shown in the case of and support. V. campbelli by Phuoc et al. (2008), who found that mortalities occurred only when >106 cells of V. campbelli were injected into juvenile L. vannamei. However, the co-infection of L. van- REFERENCES namei with both WSSV and V. campbelli resulted in accelerated Balakrishnan, G., S. Peyail, R. Kumaran, A. Theivasigamani, K. S. Anil, S. B. mortalities compared with the shrimps infected with WSSV or Jitesh, and N. Srinivasan. 2011. First report on white spot syndrome virus V. campbelli alone. Those authors concluded that when shrimps (WSSV) infection in white leg shrimp Litopenaeus vannamei (Crustacea, Penaeidae) under semi intensive culture condition in India. Aquaculture, already infected with WSSV are colonized by V. campbelli the Aquarium, Conservation & Legislation Bioflux 4:301–305. animals fail to clear the bacterial infection efficiently, leading Bej, A. K., D. P. Patterson, C. W. Brasher, M. C. Vickery, D. D. Jones, and to rapid multiplication of the bacterium in the hemolymph. It C. A. Kaysner. 1999. Detection of total and hemolysin-producing Vibrio is likely that several other Vibrio species are able to infect and parahemolyticus in shellfish using multiplex PCR amplification of tl, tdh and cause of mortalities of shrimps preinfected with WSSV or other trh. Journal of Microbiological Methods 36:215–225. Chou, H. Y., C. Y.Huang, C. F. Lo, and G. H. Kou. 1998. Studies on transmission viruses. of white spot syndrome associated baculovirus (WSBV) in Penaeus monodon Vibrio parahemolyticus is being increasingly recognized as and P. japonicus via waterborne contact and oral ingestion. Aquaculture an important pathogen of L. vannamei. Early mortality syn- 164:263–276.

Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 drome (EMS) or an acute form of this disease known as acute CLSI (Clinical and Laboratory Standards Institute). 2006. Performance stan- hepatopancreatic necrosis syndrome (AHPNS) has turned out dards for antimicrobial susceptibility testing. CLSI, Informational Supple- ment M100-S16, Wayne, Pennsylvania. to be a major disease of L. vannamei in Asia. The causative Davis, D. A., T. M. Samocha, and C. E. Boyd. 2004. Acclimating Paciﬁc agent of AHPNS was not known until Tran et al. (2013) iso- white shrimp, Litopenaeus vannamei, to inland, low-salinity waters. Southern lated V. parahemolyticus from infected L. vannamei. Immersion Regional Aquaculture Center, Publication 2601, Stoneville Mississippi. challenge tests using V.parahemolyticus caused 100% mortality Flegel, T. W. 2012. Historic emergence, impact and current status of shrimp with typical APHN pathology in experimental shrimps. In our pathogens in Asia. Journal of Invertebrate Pathology 110:166–173. Hossain, M. S., A. Chakraborty, B. Joseph, S. K. Otta, and I. Karunasagar. 2001. study, we believe that V. parahemolyticus in association with Detection of new hosts for white spot syndrome virus of shrimp using nested WSSV could be responsible for the mortalities of L. vannamei. polymerase chain reaction. Aquaculture 198:1–11. Further studies are required to establish the virulence factors of Karunasagar, I., R. Pai, G. R. Malthi, and I. Karunasagar. 1994. Mass mor- V. parahemolyticus isolated in this study and the physicochem- tality of Penaeus monodon larvae due to antibiotic-resistant Vibrio harveyi ical and biological factors that trigger shrimp mortalities. All V. infection. Aquaculture 128:203–209. Kim, Y. B., J. Okuda, C. Matsumoto, N. Takahashi, S. Hashimoto, and M. parahemolyticus isolated in our study were resistant to ampi- Nishibuchi. 1999. Identiﬁcation of Vibrio parahemolyticus strains at the cillin, amoxicillin–clavulanic acid, and cefepime (Table 3), but species level by PCR targeted to the toxR gene. Journal of Clinical Mi- were all sensitive to chloramphenicol. A few isolates were also crobiology 37:1173–1177. INVESTIGATION OF PERSISTENT MORTALITIES IN LITOPENAEUS VANNAMEI 159

Kimura, T., K. Yamano, H. Nakano, K. Momoyama, M. Hiraoka, and K. Inouye. Remany, M. C., D. Cyriac, S. Nagaraj, B. Rao, A. K. Panda, J. Kumar, and 1996. Detection of penaeid rod-shaped DNA virus (PRDV) by PCR. Fish Y. C. T. Samraj. 2010. Specific pathogen-free assurance of imported Pacific Pathology 31:93–98. white shrimp Litopenaeus vannamei (Boone, 1931) in the aquatic quarantine Lavilla-Pitogo, C. R., E. M. Leano, and M. G. Paner. 1998. Mortalities of facility, Chennai. Current Science 99:1656–1658. pond-cultured juvenile shrimp, Penaeus monodon, associated with domi- Sanchez-Mart´ ´ınez, J. G., G. Aguirre-Guzman,´ and H. Mej´ıa-Ru´ız. 2007. White nance of luminescent vibrios in the rearing environment. Aquaculture 164: spot syndrome virus in cultured shrimp: a review. Aquaculture Research 337–349. 38:1339–1354. Lightner, D. V. 2011. Virus diseases of farmed shrimp in the western hemi- Seibert, C. H., and A. R. Pinto, 2012. Challenges in shrimp aquaculture due sphere (the Americas): a review. Journal of Invertebrate Pathology 106: to viral diseases: distribution and biology of the five major penaeid viruses 110–130. and interventions to avoid viral incidence and dispersion. Brazilian Journal Lightner, D. V., and R. M. Redman. 1998. Shrimp diseases and current diag- of Microbiology 43:857–864. nostic methods. Aquaculture 164:201–220. Tang, K. F. J., S. V. Durand, B. L. White, R. M. Redman, C. R. Pantoja, and Liu, B., Z. Yu, X. Song, Y. Guan, X. Jian, and J. He. 2006. The effect of acute D. V. Lightner. 2000. Postlarvae and juveniles of a selected line of Penaeus salinity change on white spot syndrome (WSS) outbreaks in Fenneropenaeus stylirostris are resistant to infectious hypodermal and hematopoietic necrosis chinensis. Aquaculture 253:163–170. virus infection. Aquaculture 190:203–210. Lo,C.F.,C.H.Ho,S.E.Peng,C.H.Chen,H.C.Hsu,Y.L.Chiu,C.F.Chang, Tran, L., L. Nunan, R. M. Redman, L. L. Mohney, C. R. Pantoja, K. Fitzsimmons, K. F. Liu, M. S. Su, C. H. Wang, and G. H. Kou. 1996. White spot syndrome and D. V.Lightner. 2013. Determination of the infectious nature of the agent of baculovirus (WSBV) detected in cultured and captured shrimp, crab and other acute hepatopancreatic necrosis syndrome affecting penaeid shrimp. Diseases arthropods. Diseases of Aquatic Organisms 27:215–225. of Aquatic Organisms 105:45–55. Manjanaik, B., K. R. Umesha, I. Karunasagar, and I. Karunasagar. 2005. De- Tu, C., H. T. Huang, S. H. Chuang, J. P. Hsu, S. T. Kuo, N. J. Li, T. L. Hus, tection of hepatopancreatic parvovirus (HPV) in wild shrimp from India by M. C. Li, and S. Y.Lin. 1999. Taura syndrome in Pacific white shrimp Penaeus nested polymerase chain reaction (PCR). Diseases of Aquatic Organisms vannamei cultured in Taiwan. Diseases of Aquatic Organisms 38:159–161. 63:255–259. Umesha, K. R., A. Chakraborty, M. N. Venugopal, I. Karunasagar, and I. Martorelli, S. R., R. M. Overstreet, and J. A. Jovonovich. 2010. First report Karunasagar. 2008. Occurrence of multiple viruses in Penaeus monodon of viral pathogens WSSV and IHHNV in argentine crustaceans. Bulletin of shrimp ponds and their effects on shrimp production. Pages 389–398 in M. Marine Science 86:117–131. G. Bondad-Reantaso, C. V. Mohan, M. Crumlish, and R. P. Subasinghe, edi- Moger, R., S. A. Mohan, M. N. Venugopal, and I. Karunasagar. 2011. Presence tors. Diseases in Asian aquaculture VI. Asian Fisheries Society, Fish Health of multiple viruses in newly introduced white leg shrimp, Penaeus vannamei Section, Manila. and assessment of their risk to aquaculture in India. Page 158 in K. M. Vandenberghe, J., L. Verdonck, R. Robles-Arozarena, G. Rivera, A. Bolland, Shanker, I. Karunasagar, M. N. Venugopal, and M. Shekar, editors. Eighth M. Balladares, B. Gomez-Gil, J. Calderon, P. Sorgeloos, and J. Swings. 1999. symposium on diseases in Asian aquaculture. Asian Fisheries Society, Fish Vibrios associated with Litopenaeus vannamei larvae, postlarvae, broodstock, Health Section, Selangor, Malaysia. and hatchery probionts. Applied and Environmental Microbiology 65:2592– Otta, S. K., R. Arulraj, P. E. Praveena, R. Manivel, A. Panigrahi, T. 2597. Bhuvaneswari, P. Ravichandran, K. P. Jithendran, and A. G. Ponniah. 2014. Vidal, O. M., C. B. Granja, F. Aranguren, J. A. Brock, and M. Salazar. 2001. Association of dual viral infection with mortality of Pacific white shrimp A profound effect of hyperthermia on survival of Litopenaeus vannamei (Litopenaeus vannamei) in culture ponds in India. Indian Journal of Virology juveniles infected with white spot syndrome virus. Journal of the World 25:63–68. Aquaculture Society 32:364–372. Otta, S. K., I. Karunasagar, and I. Karunasagar. 2003. Detection of monodon Vijayan, K. K., V. Stalin Raj, C. P. Balasubramanian, S. V. Alavandi, V. baculovirus and whitespot syndrome virus in apparently healthy Penaeus Thillai Sekhar, and T. C. Santiago. 2005. Polychaete worms–a vector for monodon postlarvae from India by polymerase chain reaction. Aquaculture white spot syndrome virus (WSSV). Diseases of Aquatic Organisms 63:107– 220:59–67. 111. Phuoc, L. H., M. Corteel, H. J. Nauwynck, M. B. Pensaert, V. Alday-Sanz, W. Zhou, J., W. Fang, X. Yang, S. Zhou, L. Hu, X. Li, X. Qi, H. Su, and L. Xie. 2012. Van den Broeck, P. Sorgeloos, and P. Bossier. 2008. Increased susceptibil- A nonluminescent and highly virulent Vibrio harveyi strain is associated with ity of white spot syndrome virus-infected Litopenaeus vannamei to Vibrio “bacterial white tail disease” of Litopenaeus vannamei shrimp. PLoS One campbellii. Environmental Microbiology 10:2718–2727. [online serial] 7:e2996. Downloaded by [Department Of Fisheries] at 23:53 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:54 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Comparison of the Susceptibility of Four Rainbow Trout Strains to Cold-Water Disease Eric J. Wagnera & Randall W. Oplingera a Utah Division of Wildlife Resources, Fisheries Experiment Station, 1465 West 200 North, Logan, Utah 84321, USA Published online: 05 Aug 2014.

To cite this article: Eric J. Wagner & Randall W. Oplinger (2014) Comparison of the Susceptibility of Four Rainbow Trout Strains to Cold-Water Disease, Journal of Aquatic Animal Health, 26:3, 160-167, DOI: 10.1080/08997659.2014.922514 To link to this article: http://dx.doi.org/10.1080/08997659.2014.922514

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:160–167, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.922514

ARTICLE

Comparison of the Susceptibility of Four Rainbow Trout Strains to Cold-Water Disease

Eric J. Wagner* and Randall W. Oplinger Utah Division of Wildlife Resources, Fisheries Experiment Station, 1465 West 200 North, Logan, Utah 84321, USA

Abstract Susceptibility to cold-water disease was compared among four strains of Rainbow Trout Oncorhynchus mykiss: Arlee strain from Ennis National Fish Hatchery, Montana (AL-EN), the Arlee strain from Jocko River Hatchery, Montana (AL-JR), a cold-water disease-resistant strain (WV), and the Harrison–Hofer strain (HH). Bacterial challenges were either by bath or intraperitoneal injection (50 µL of 0.65 optical density). Each strain was exposed at 75 d after hatch to either the CSF 259-93 (Idaho) or 09–104 isolate (Utah) of Flavobacterium psychrophilum. Injection controls received a phosphate-buffered saline (PBS) solution and bath controls were exposed to uninoculated sterile broth (tryptone yeast extract salts) mixed 1:1 with hatchery well water. For injected fish, the WV had significantly lower mortality (20.0–36.7%) than HH and AL-EN (76.7–96.7%) but did not significantly differ from AL-JR (46.7– 56.7%). Injected fish had significantly higher mortality than bath-exposed fish. For bath-exposed fish, the WV had significantly lower mortality (0%) than the HH (10.0–26.7%), but both Arlee strains had intermediate mortality values (0–13.3%) that did not significantly differ from either the HH or WV strain. There were no significant differences between the two bacterial isolates, indicating similar virulence and similar resistance response of WV to another novel isolate of F. psychrophilum.

Cold-water disease, caused by the bacterium Flavobacterium Center for Cool and Cold Water Aquaculture in West Virginia psychrophilum, can induce necrotic lesions, partial dark skin col- (Silverstein et al. 2009; Leeds et al. 2010). This strain is the oring (black-tail), exophthalmia, anemia, ascites, and vertebral result of three generations of selection for resistance to cold- deformities (Faruk 2002; Nematollahi et al. 2003; Aoki et al. water disease (Leeds et al. 2010). The strain composition of 2005) in salmonids. The bacterium can also induce necrosis this stock, based the pedigree of the four founder strains, is of renal tubular epithelium and haematopoietic tissue (Ekman 16.4% House Creek (College of Southern Idaho), 7.4% Shasta Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 and Norrgren 2003). Cold-water disease has been a significant (Ennis National Fish Hatchery), 16.1% Donaldson (University cause of mortality in a wide variety of cultured fishes world- of Washington), and 60.1% Kamloops/Puget Sound Steelhead wide (Cipriano and Holt 2005). In the United Kingdom, annual cross (Trout Lodge). losses to the disease are estimated to be about 10 million fish There are few controlled studies that compare the suscep- (Faruk 2002). In Utah, we estimated 1.3 million hatchery fish tibility of different fish species or strains to cold-water dis- have been lost to the disease in the last decade. Ekman (2003) ease, also known as Rainbow Trout Fry Syndrome. Nagai noted that 50–60% of the antibiotics used in Swedish aquacul- et al. (2004) compared susceptibility among three stocks of ture are used for treating F. psychrophilum or the closely related Ayu Plecoglossus altivelis; the amphidromous stock had lower F. columnare. mortality than a landlocked or domesticated stock. In a study by One of the recent efforts to control the disease has been Holt et al. (1989) evaluating temperature effects on fish mortality the advent of a disease-resistant strain of Rainbow Trout On- from F.psychrophilum injections, presmolts of Chinook Salmon corhynchus mykiss developed by the team at the National O. tshawytscha and Rainbow Trout had similar mortality rates;

*Corresponding author: [email protected] Received November 5, 2013; accepted April 15, 2014 160 DISEASE SUSCEPTIBILITY OF RAINBOW TROUT STRAINS 161

However, the salmon were twice as heavy as the trout, and ex- the target OD. For the bath exposures, the bath was composed posures were not given at a standard age. Nematollahi et al. of 2 L of TYES broth (0.3 OD) and 2 L of well water and (2003) suggested that salmonids are the most susceptible, and aerated during the exposure. The OD was based on maximum in nonsalmonids the disease is less severe. However, Nagai et al. ODs achievable in our broth cultures in the incubation time (2004) reported that cold-water disease is one of the most serious alloted. For bath exposures, all 30 fish for a treatment were fin- diseases of Ayu in Japan. Also, Lehmann et al. (1991) observed clipped (adipose fin) and exposed together for 30 min. After bath 40–50% mortality among Common Carp Cyprinus carpio,Cru- exposure, the fish were rinsed with freshwater, and 10 fish were cian Carp Carassius carassius, and Tench Tinca tinca that had returned to each treatment aquarium. Thus, 10 bath-exposed fish F. psychrophilum in their spleens and livers. This is similar to and 10 injected fish shared each aquarium. the mortality rates reported for Coho Salmon Oncorhynchus Two isolates of Flavobacterium psychrophilum were tested kisutch (Holt 1987). Rainbow Trout mortality rates around the for each Rainbow Trout strain, the CSF 259-93 isolate from world have ranged from 20% to 90% (Santos et al. 1992; Madsen Idaho, and the 09-104 isolate isolated from sick Rainbow Trout et al. 2005; Barnes and Brown 2011). in Utah. The CSF 259-93 isolate has been used in multiple In an effort to control cold-water disease, the Utah Division studies (Crump et al. 2001; LaFrentz et al. 2003; Crump and of Wildlife’s (UDWR) fish hatchery system has imported the Kay 2008; Castillo et al. 2012). Control fish were injected with Rainbow Trout strain developed by Leeds et al. (2010), hereafter sterile PBS or exposed to a 30-min bath with 2 L sterile TYES referred to as the ‘West Virginia’ strain (WV). The UDWR has broth and 2 L well water. also recently imported other strains of unknown susceptibility For the WV strain, an additional treatment was added for to cold-water disease, including the Arlee strain from Ennis which fish were injected with 50 µL of a 0.4-OD solution of Hatchery, Montana (AL-EN), the Arlee strain from Jocko the CSF 259-93 isolate. This dose has been used in several River Hatchery, Montana (AL-JR), and the Harrison–Hofer previous bacteria challenges related to F.psychrophilum vaccine strain (HH) obtained from the Colorado Division of Wildlife evaluations at the UDWR’s Fisheries Experiment Station. There Resources. The HH strain is a cross between fish from a wild was insufficient tank space for an additional 09-104 treatment population in Harrison Lake, Montan, and the domesticated at 0.4 OD. Additional treatments at 0.4 OD for the other strains Hofer strain originating from Germany (Wagner et al. 2012). was not possible due to tank space limitations. For the WV trial, The source strains and the cross have demonstrated greater the injected fish of the 0.4 OD treatment were cohabitated with resistance to whirling-disease than other Rainbow Trout strains bath-exposed fish as in the other treatments. (Hedrick et al. 2003; Wagner et al. 2006; Wagner et al. 2012). Tank system, fish, and water quality.—The tanks were part However, this strain and the Arlee strains of Rainbow Trout of a recycle system that had a biofilter and sump from which have not been tested for susceptibility to cold-water disease. water was pumped to a headbox and distribution manifold (see The objective of our study was to compare the susceptibility Wagner et al. 2006 for further system details). Each fish strain of these strains with the resistant WV strain. We evaluated was tested on separate dates due to differences in hatching dates, the resistance using two different isolates of Flavobacterium but the same methods were applied in each case. Mean lengths psychrophilum and used both injection and bath exposures. and weights of the fish at the end of the trials were AL-EN = 94.2 mm, 8.4 g; AL-JR = 86.7 mm 7.8 g; HH = 83.1 mm, 6.8 g; and WV = 95.9 mm, 9.5 g. METHODS Fish were fed daily ad libitum with a commercial pellet feed To make our comparisons consistent with other studies, we (Skretting trout diet). Each day, about a fourth (56 L) of the used the same challenge methodology as Leeds et al. (2010). total volume of water was removed, siphoning waste feed and

Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 For the cold-water disease challenges, fish were acclimated feces at the same time. This water was replaced with fresh to aquaria 1 week prior to exposure. Aquaria (110 L) were well water. Water quality was monitored periodically with a each stocked with 20 fish and three replicate tanks were used Hach test kit to ensure temperature, ammonia, and nitrite levels per treatment. Treatments were bacteria isolate (CSF 259-93, were not affecting survival. Temperatures were dependent on air 09-104, or a phosphate-buffered saline injected control) and temperature in the room, but only varied between 13.5◦C and fish strain (WV, AL-JR, AL-EN, HH). Fish were exposed at 16.1◦C during the trials. The well water pH was 7.6, the total 75 d after hatching to Flavobacterium psychrophilum either by hardness was 222 mg/L, and the total alkalinity was 222 mg/L. intraperitoneal injection or bath. Mortalities were removed and recorded daily until the challenge For injections, 50 µL of a 0.65 optical density (OD; 525 nm) ended after 21 d. This duration was the same as in the study by mixture in sterile phosphate-buffered saline (PBS) was injected Leeds et al. (2010), in which most mortality occurred within 14– into 10 fish per tank using a tuberculin syringe. The bacteria used 16 d. Kidney and spleen samples from dead fish were streaked for the mixture were first grown in Petri plates with tryptone on TYES plates and monitored for growth of yellow pigmented yeast extract salt (TYES; Holt et al. 1993) media. After 3–4 d bacteria. of culture, the growth was then transferred with sterile bacteria Broth culture preparation.—The broth cultures used for the loops in sterile test tubes and mixed with sterile water to achieve bath exposures were initiated by adding frozen bacteria stock to 162 WAGNER AND OPLINGER

a flask with 50–100 mL sterile TYES broth. After 2 d, the con- 6.2). For injected fish, the WV had significantly lower mortality tents were transferred to a larger flask with 1–2 L of iron-limited (20.0–36.7%) than HH and AL-EN (76.7–96.7%; P < 0.001) TYES broth and incubated at room temperature (18–20◦C) for but did not significantly differ from AL-JR (46.7–56.7%; P = 3–4 more days. Iron-limited media were used to increase viru- 0.27; Table 1). The percent mortality of injected fish did not lence (LaFrentz et al. 2009; Long et al. 2013). This media was differ between the two bacteria isolates (P = 0.77), but both made by adding 834 µL of 0.03M 2,2 bipyridyl (DPD) per liter differed significantly from controls (P < 0.001). Of all the con- of TYES broth. The DPD was first made in a stock solution trol fish, only one died. This one died the day after injection, (0.3 M: 0.47 g in 10 mL of 70% ethanol). This stock solution indicating the death was likely due to needle damage intraperi- was diluted 1:10 in sterile water, then filter-sterilized (0.2 µm) toneally, rather than disease or water quality issues. Water qual- to make the 0.03-M working solution. In addition to making ity was maintained throughout all the exposure trials, so ammo- injection solutions, the bacteria from the plates were used to nia (<0.9 mg/L total ammonia nitrogen), nitrite (≤0.17 mg/L), augment the number of bacteria in the broth cultures. A portion and temperatures (13.5–16.1◦C) were not a source of mortality. of the broth cultures were also centrifuged in 50 mL tubes. The The comparison of days until death among strains indicated pellets of these tubes were resuspended in a small amount of the that there were significant differences among the strains (Cox supernatant and added to other broth cultures of the same isolate regression: P < 0.001, χ2 = 67.4, df = 3) but not between to concentrate bacteria into 2 L for the bath. Serial 10-fold dilu- bacterial isolates (P = 0.99, χ2 = 0, df = 1; Table 1; Figure 1). tions of both injection and bath solutions were made on TYES Pooling data for both isolates, the mean days until death was plates to estimate the number of bacteria in each solution. significantly higher (P < 0.001, χ2 ≥ 7.1) for the WV strain Statistics.—We used SPSS version 13.0 for all analyses and significance was set at α = 0.05. A fully saturated general linear model was used to compare the percent mortality among the four trout strains, the three challenges of bacteria isolates (CSF 259- 93, 09-104, and PBS control), and between exposure method (bath or injection). Tanks were considered replicates. Given the significant interactions, separate tests were conducted for each exposure method. The percent mortality was transformed prior to analysis using the arcsine-square root procedure (Kirk 1982). Scheffe´ statistic was used for mean separation. Cox regression analysis, using the backward stepwise likelihood ratio option of SPSS, was used to compare time to death (d) among the trout strains and between bacterial isolates. For survivors (censored values), 21 d was inserted as the days-to-death value; mean days to death is the average of all individual days to death and includes survivors in the calculation. The analysis used only the data for injected fish in the 0.65-OD treatments because the mortality rate was so low in the bath-treatment data set that the models using that data would not converge. A Kaplan- Meier test featuring the log-rank (Mantel–Cox) option was used to compare (1) differences, pairwise, among fish strains, and (2)

Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 the 0.40 and 0.65 OD data for the WV strain (pooling the data for the bacterial isolates for both analyses because the isolates were not signiﬁcantly different).

RESULTS The general linear model indicated that effects of Rainbow Trout strain, bacteria isolate, exposure method were all highly significant (P < 0.001), as well as all two-way and three- way interactions (P ≤ 0.007). Injected fish had significantly (P < 0.001) higher mortality than bath-exposed fish, which had low to no mortality. When analyzed separately for each expo- FIGURE 1. Comparison of the cumulative percent mortality among four Rain- bow Trout strains (see Table 1) injected with one of two isolates of Flavobac- sure method, the mortality rates varied significantly among the terium psychrophilum at 0.65 OD: (A) CSF 259-93 from Idaho, and (B) 09-104 strains of Rainbow Trout that were injected (P < 0.001, df = from Utah. Significant differences among strains in the mean days to death are 3, F = 21.6) or exposed in baths (P = 0.003, df = 3, F = noted for each isolate by different lower case letters (a–c). DISEASE SUSCEPTIBILITY OF RAINBOW TROUT STRAINS 163

TABLE 1. Comparison of the mean percent mortality among four strains of Rainbow Trout exposed at 75 d posthatch to two different isolates of Flavobacterium psychrophilum by injection or bath exposure. The optical density (OD; 525 nm wavelength) and plate count dilution estimates are also given. Injected fish received 50 µL (1/20th the amount shown). Significant differences among strains in days to death are noted separately for each bacteria isolate by different letters:xtoz for 09-104 or a to c for 259-93 and the asterisk (*) indicates a significant difference from the 0.65 OD for isolate 259-93. Abbreviations: Arlee strain from Ennis National Fish Hatchery, Montana (AL-EN), the Arlee strain from Jocko River Hatchery, Montana (AL-JR), a cold-water disease-resistant strain (WV), and the Harrison–Hofer strain (HH).

Exposure Rainbow Number of Mean ± SE Mean ± SE method Trout strain Bacteria isolate OD bacteria/mL mortality (%) days to death Injected AL-EN 09-104 0.65 1.44 × 109 76.7 ± 15.3 9.0 ± 1.2 z CSF 259-93 0.65 6.30 × 107 86.7 ± 5.8 7.8 ± 1.0 c PBS control 0.0 ± 0.0 AL-JR 09–104 0.65 No growth 46.7 ± 25.2 14.3 ± 1.3 y CSF 259-93 0.65 No growth 56.7 ± 15.3 14.6 ± 1.2 ab PBS control 0.0 ± 0.0 HH 09–104 0.65 2.2 × 106 96.7 ± 5.8 8.4 ± 0.6 z CSF 259-93 0.65 3.6 × 106 83.3 ± 11.5 11.8 ± 0.9 b PBS control 3.3 ± 5.7 WV 09–104 0.65 3.9 × 106 20.0 ± 10.0 19.3 ± 0.7 x CSF 259-93 0.65 1.4 × 107 36.7 ± 20.8 16.3 ± 1.1 a CSF 259-93 0.4 1.1 × 107 0.0 ± 0.0 20.6 ± 0.4 * PBS control 0.0 ± 0.0 Bath AL-EN 09–104 0.3 1.6 × 107 3.3 ± 5.8 CSF 259-93 0.3 1.9 × 107 6.7 ± 5.8 PBS control 0.0 ± 0.0 AL-JR 09–104 0.3 No growth 0.0 ± 0.0 CSF 259-93 0.3 No growth 13.3 ± 15.3 PBS control 0.0 ± 0.0 HH 09–104 0.3 4.3 × 109 26.7 ± 5.8 CSF 259-93 0.3 8.0 × 109 10.0 ± 10.0 PBS control 0.0 ± 0.0 WV 09–104 0.3 4.6 × 106 0.0 ± 0.0 CSF 259-93 0.3 6.0 × 106 0.0 ± 0.0 CSF 259-93 0.3 4.0 × 106 0.0 ± 0.0 PBS control 0.0 ± 0.0

than the other strains. Each strain was signiﬁcantly different yellow-pigmented bacteria with typical Flavobacterium psy- from each other (P < 0.034, χ2 ≥ 4.5), ranking from highest to chrophilum morphology (long thin rods) were observed in the lowest: WV (17.9 d) > JR (14.5 d) > HH (10.1 d) > AL-EN kidney and spleen samples of dead ﬁsh (Table 2).

Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 (8.4 d). The mean days to death for WV fish exposed to 0.4 OD (20.6 d) was significantly higher than for WV fish exposed to 0.65 OD (17.9 d; P = 0.006, χ2 = 7.4). DISCUSSION For bath-exposed fish, the WV had significantly lower mor- The results of this study indicated that there were significant tality (0%) than the HH (10.0–26.7%), but both Arlee strains differences among the strains in their susceptibility to cold-water had intermediate mortality values (0–13.3%) that did not sig- disease. Not surprisingly, the strain selected for cold-water dis- nificantly differ (P > 0.10) from either the HH or WV strain ease resistance had the lowest mortality. The HH had higher (Table 1). There were no significant differences between the two susceptibility to cold-water disease, despite selection for resis- bacteria isolates, though mortality in both treatments was sig- tance to whirling disease. Selection for one disease does not nificantly higher than the PBS control (P ≤ 0.02). Cumulative necessarily confer resistance to other diseases, and may even mortality after bath exposure for each strain and each isolate is hinder resistance (Fevolden et al. 1992; Allendorf and Spruell showninFigure2. 1999; Hedrick et al. 2001). Nagai et al. (2004) found that a strain Bath-exposed fish, either alive or dead, did not have of Ayu resistant to cold-water disease was just as susceptible to observable lesions, whereas several of the injected fish had vibriosis (caused by Vibrio anguillarum) as two other strains lesions develop around the injection site. Gram-negative, of Ayu that were not resistant to cold-water disease. However, 164 WAGNER AND OPLINGER

TABLE 2. Summary of data (percent positive for each tissue) on reisolation of yellow-pigmented bacteria (YPB, Gram-negative rods) from mortalities after injection of four rainbow trout strains (see Table 1) with one of two different isolates of Flavobacterium psychrophilum.ND= no data.

Strain Isolate YPB In kidney (%) YPB in spleen (%) WV 259–93 16.7 33.3 09–104 25.0 50.0 GH 259–93 100.0 100.0 09–104 81.8 81.8 AL-EN 259–93 100.0 100.0 09–104 ND ND AL-JR 259–93 64.3 64.3 09–104 0.0 20.0

challenges studies with F. psychrophilum in which higher doses of bacteria led to higher mortality (Obach and Laurencin 1991; Madsen and Dalsgaard 1999; Rahman et al. 2000, 2001, 2002; Plant et al. 2011). This effect suggests that lower doses can be controlled by the immune system, but higher bacteria concentrations overcome these defenses. The implication of this dose effect for aquaculture is that cold-water disease can be managed by keeping bacteria growth and numbers below threshold levels. This is a working hypothesis, so further research is needed in an aquaculture setting to determine what these thresholds are. Once determined, regular monitoring and adaptive management (e.g., adjustment of densities, water temperature, fish numbers, water flow rates, rapid removal of mortalities) can keep bacteria loads low and fish healthy. Chemical treatments such as hydrogen peroxide could FIGURE 2. Comparison of the cumulative percent mortality among four Rain- be considered as well (Speare and Arsenault 1997; Rach et al. bow Trout strains (see Table 1) exposed by bath immersion to one of two isolates of Flavobacterium psychrophilum:(A) CSF 259-93 from Idaho, and (B) 09-104 2000; Gimenez-Papiol´ et al. 2009), if doses and durations are from Utah. kept below levels that have been shown to increase the incidence of cold-water disease (Henriksen et al. 2013). Probiotic bacte- Henryon et al. (2005) noted some weak positive correlations ria may also be a tool to keep F. psychrophilum levels low via for Rainbow Trout resistance if fish were previously exposed to competition (Burbank et al. 2011, 2012). The data also suggest either Yersinia ruckeri (cause of enteric redmouth disease), F. that if the bacteria concentrations are sufficient, even resistant psychrophilum (Rainbow Trout Fry Syndrome), or viral haem- strains of Rainbow Trout will succumb to cold-water disease.

Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 orrhagic septicemia (VHS). There was little difference between the two isolates of bac- The bath exposure data indicated that healthy fish can with- teria used for the challenges. Soule et al. (2005) observed that stand a challenge by a substantial dose of bacteria. Several bath a different Utah isolate of F. psychrophilum (03-009) was in exposure studies have had varying success in producing infec- the same lineage (II) as the CSF 259-93 isolate. In this study tion, but wounds or injuries that break the skin and mucus bar- the virulence was comparable between the two bacteria isolates, rier contributed to higher rates of infection by F. psychrophilum whether the fish were exposed by bath or injection. This sug- (Madsen and Dalsgaard 1999) and F. columnare (Bader et al. gests that for future research, one isolate could be used. The 2006). The higher mortality of injected fish corroborates this CSF 259-93 isolate would be preferable because it has been earlier work on the effects of injury. A high percentage of in- used more widely as a reference strain (e.g., Crump et al. 2001; jected fish died in most strains, indicating that if injected bacte- LaFrentz et al. 2003). The results for the WV exposed to the ria numbers are high enough, disease will follow, regardless of 09-104 isolate also indicated that the resistance developed in the trout strain. The lack of mortality in WV injected with a 0.4 OD WV strain using the CSF 259-93 isolate will be applicable to solution of F. psychrophilum, compared with 20–37% in 0.65 Utah and probably other sites beyond Idaho. OD, indicated that a critical concentration of bacteria is needed The relationship between OD and the plate-count numbers to lead to disease. This effect has been observed in numerous is worth discussing as well. It appears that the predictive ability DISEASE SUSCEPTIBILITY OF RAINBOW TROUT STRAINS 165

of OD for bacteria abundance is not as accurate as one would Comparison of these various methods by Paulse et al. (2007) has like. For example, at a 0.65 OD, the plate counts varied from indicated that direct microbial counts are about 43% of that ob- 3.9 × 106 to 1.4 × 109, which is a range of 3 logs. Perhaps vari- tained by flow cytometry; similarly, heterotrophic plate counts ance increases with OD, but the variation in numbers around the are about 4% of the flow cytometry estimates. Wohlsen et al. bath ODs, which were about half that for the injection, was sim- (2006) found that the pour-plate method and the commercial ilarly wide. The scatter in plots by Michel et al. (1999; r2 = 0.61 product Petrifilm produced more consistent and better estimates between OD and F.psychrophilum numbers) also suggested that of coliform bacteria numbers than membrane filtration, the MPN variance was similar across a range of bacteria abundance. An- (most probable number) method, or two commercial coliform other factor to consider is that OD measures both live and dead counting kits. bacteria, whereas plate counts would count only the live colony In summary, the data showed significant differences among forming units. So, it appears that although a general relation- strains of Rainbow Trout tested, the WV strain showing the ship between OD and bacteria numbers exists (e.g., Holt 1987; greatest resistance to cold-water disease. These trials represent Michel et al. 1999), meaningful differences related to OD values the first efforts at comparing susceptibility to cold-water dis- do not correspond well to bacteria numbers. Unfortunately, for ease among Rainbow Trout strains. Because the bacterium can doing research, one cannot wait several days for bacteria counts be vertically transmitted (Brown et al. 1997; Cipriano 2005), on plates to do challenge studies. eggs from infected brood, coupled with inadequate disinfection Plate counts have their pitfalls as well, with variables such as methods (Cipriano 2005; Wagner et al. 2008), lead to transfer of moisture content or air diffusion that may affect colony counts. the bacterial pathogen to rearing hatcheries. By using resistant Another example is our finding that the serial dilutions for the strains and making efforts to develop pathogen-free broodstock, trial with AL-JR had no bacterial growth, yet mortality indicated the disease can be better controlled when the bacterium is not that bacteria were present. We suspect this may have been a represent to cause disease. The data also provide a foundation sult of pouring plates when the media was still too hot, leading to for future studies comparing resistance of other Rainbow Trout potentially toxic polystyrene breakdown products such as ben- strains or among fish species under standard conditions estab- zaldehyde and benzene (Ciucanu et al. 2002) being released, but lished by Leeds et al. (2010). The data also indicate that the we cannot conclude that with certainty. Also, differences among virulence of F. psychrophilum isolates from Utah is similar to serial dilutions of a single sample in this study, which should the Idaho isolate. Bath trials indicated that Rainbow Trout are have been 10-fold, indicated that values were often approxi- able to withstand exposure to high numbers of bacteria without mately but inconsistently 10-fold. Clumping of bacteria could developing the disease, or at least experience much lower mor- also affect colony development, which would underestimate ac- tality levels than fish injected with the bacteria. The exposures tual numbers that would be replicating inside challenged fish. indicated that susceptibility to disease is dependent on bacteria In our experience, auto-agglutination of F. psychrophilum is a numbers, which if controlled, could control or limit mortality common problem when growing broth cultures. The variance and morbidity related to cold-water disease. in bacterial number estimates that results from plate counts is of concern in this study because the resulting numbers suggest that perhaps the fish strains were exposed to different doses that ACKNOWLEDGMENTS affected mortality differences. Ideally all strains would have We thank M.D. Routledge, G. Howe, and R. Mellinthin for been exposed at the same time to the same batch of bacteria, but their help in rearing and providing fish for the study. Thanks given the different spawning times inherent to the strains, this also to two anonymous reviewers and the associate editor whose was not possible. However, every effort was made to minimize comments helped to improve the article. The research was sup-

Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 differences among trials: Fish were exposed to the same OD, the ported by the Utah Division of Wildlife Resources and the Fed- same injection volume, the same age of the broth culture and at eral Aid in Sport Fish Restoration program, Grant F-74-R. the same fish age in the same tank systems. We do not have the data to calculate the variance around a plate count estimate, but given the variance observed for a given OD, the experimental REFERENCES error suggests that fish strains were similarly challenged across Allendorf, F., and P. Spruell. 1999. Genetics, wild trout, and whirling disease: trials. The similarity in results between the two isolates also pro- first, do no harm. Pages 1–4 in J. Bartholomew, editor. Proceedings of the vides support for the reproducibility of the methods for bacterial 5th Annual Whirling Disease Symposium. Whirling Disease Foundation, Bozeman, Montana. challenges. Aoki, M., M. Kondo, K. Kawai, and S. Oshima. 2005. Experimental bath infec- As an alternative to plate counts, bacteria counts on a fil- tion with Flavobacterium psychrophilum, inducing typical signs of Rainbow ter or a slide after dilution could be done, after staining with Trout Oncorhynchus mykiss fry syndrome. Diseases of Aquatic Organisms fluorochromes to indicate viability (APHA et al. 1989; Boulos 67:73–79. et al. 1999). Flow cytometry is another alternative for obtaining APHA (American Public Health Association), American Water Works Associa- tion, and Water Pollution Control Federation. 1989. Standard methods for the bacteria population, cell size, DNA, and biomass-distribution examination of water and wastewater, 17th edition. APHA, American Wa- data (Button and Robertson 1993), though the cost of the inter Works Association, and Water Pollution Control Federation, Washington, strument prohibits its widespread use for aquatic animal health. D.C. 166 WAGNER AND OPLINGER

Bader, J. A., S. A. Moore, and K. E. Nusbaum. 2006. The effect of cutaneous bow Trout (one with suspected resistance to whirling disease) to Myxobolus injury on a reproducible immersion challenge model for Flavobacterium cerebralis infection. Diseases of Aquatic Organisms 55:37–44. columnare infection in Channel Catfish (Ictalurus punctatus). Aquaculture Hedrick, R. P., T. S. McDowell, K. Mukkatira, M. P. Georgiadis, and E. 253:1–9. MacConnell. 2001. Salmonids resistant to Ceratomyxa shasta are susceptible Barnes, M. E., and M. L. Brown. 2011. A review of Flavobacterium psy- to experimentally induced infections with Myxobolus cerebralis. Journal of chrophilum biology, clinical signs, and bacterial cold water disease prevention Aquatic Animal Health 13:35–42. and treatment. Open Fish Science Journal 4:40–48. Henriksen, M. M. M., L. Madsen, and I. Dalsgaard. 2013. Effect of hydrogen Boulos, L, M. Prevost, B. Barbeau, J. Coallier, and R. Desjardins. 1999. peroxide on immersion challenge of Rainbow Trout fry with Flavobacterium LIVE/DEAD BacLight: application of a new rapid staining method for di- psychrophilum. PLoS One [online serial] 8(4):e62590. rect enumeration of viable and total bacteria in drinking water. Journal of Henryon, M., P. Berg, N. J. Olesen, T. E. Kjaer, W. J. Slierendrecht, A. Jokum- Microbiological Methods 37:77–86. sen, and I. Lund. 2005. Selective breeding provides an approach to increase Brown, L. L., W. T. Cox, and R. P. Levine. 1997. Evidence that the causal agent resistance of Rainbow Trout (Oncorhynchus mykiss) to the diseases enteric of bacterial cold-water disease Flavobacterium psychrophilum is transmitted redmouth disease, Rainbow Trout fry syndrome, and viral haemorrhagic sep- within salmonid eggs. Diseases of Aquatic Organisms 29:213–218. ticaemia. Aquaculture 250:621–636. Burbank, D. R., S. E. LaPatra, G. Fornshell, and K. D. Cain. 2012. Isolation Holt, R. A. 1987. Cytophaga psychrophila, the causative agent of bacterial cold- of bacterial probiotic candidates from the gastrointestinal tract of Rainbow water disease in salmonid fish. Doctoral dissertation. Oregon State University, Trout, Oncorhynchus mykiss (Walbaum), and screening for inhibitory activity Corvallis. against Flavobacterium psychrophilum. Journal of Fish Diseases 35:809–816. Holt, R. A., A. Amandi, J. S. Rohovec, and J. L. Fryer. 1989. Relation of Burbank, D. R., D. H. Shah, S. E. LaPatra, G. Fornshell, and K. D. Cain. water temperature to bacterial cold-water disease in Coho Salmon, Chinook 2011. Enhanced resistance to coldwater disease following feeding of probi- Salmon, and Rainbow Trout. Journal of Aquatic Animal Health 1:94–101. otic bacterial strains to Rainbow Trout (Oncorhynchus mykiss). Aquaculture Holt, R. A., J. S. Rohovec, and J. L. Fryer. 1993. Bacterial cold-water disease. 321:185–190. Pages 3–22 in V. Inglis, R. J. Roberts, and N. R. Bromage, editors. Bacterial Button, D. K., and B. R. Robertson. 1993. Use of high-resolution flow cytometry diseases of fish. Blackwell Scientific Publications, Oxford, UK. to determine the activity and distribution of aquatic bacteria. Pages 163–173 Kirk, R. E. 1982. Experimental design. Procedures for the behavioral sciences. in P. F. Kemp, B. F. Sherr, E. B. Sherr, and J. J. Cole, editors. Handbook Brooks/Cole Publishing, Belmont, California. of methods in aquatic microbial ecology. Lewis Publishers, Boca Raton, LaFrentz, B. R., S. E. LaPatra, D. R. Call, G. D. Wiens, and K. D. Cain. 2009. Florida. Proteomic analysis of Flavobacterium psychrophilum cultured in vivo and in Castillo, D., G. Higuera, M. Villa, M. Middelboe, I. Dalsgaard, L. Madsen, iron-limited media. Diseases of Aquatic Organisms 87:171–182. and R. T. Espejo. 2012. Diversity of Flavobacterium psychrophilum and the LaFrentz, B. R., S. E. LaPatra, G. R. Jones, and K. D. Cain. 2003. Passive potential use of its phages for protection against bacterial cold water disease immunization of Rainbow Trout, Oncorhynchus mykiss (Walbaum), against in salmonids. Journal of Fish Diseases 35:193–201. Flavobacterium psychrophilum, the causative agent of bacterial coldwater Cipriano, R. C. 2005. Intraovum infection caused by Flavobacterium psy- disease and Rainbow Trout fry syndrome. Journal of Fish Diseases 26:377– chrophilum among eggs from captive Atlantic Salmon broodfish. Journal 384. of Aquatic Animal Health 17:275–283. Leeds, T. D., J. T. Silverstein, G. M. Weber, R. L. Vallejo, Y.Palti, C. E. Rexroad Cipriano, R. C., and R. A. Holt. 2005. Flavobacterium psychrophilum, cause of III, J. Evenhuis, S. Hadidi, T. J. Welch, and G. D. Wiens. 2010. Response to bacterial cold-water disease and Rainbow Trout fry syndrome. U.S. Fish and selection for bacterial cold water disease resistance in Rainbow Trout. Journal Wildlife Service Fish Disease Leaflet 86. of Animal Science 88:1936–1946. Ciucanu, I., M. Kaykhaii, L. Montero, J. Pawliszyn, and J. Szubra. 2002. Contin- Lehmann, J., D. Mock, F. J. Sturenberg,˝ and J. F. Bernardet. 1991. First isolation uous monitoring of thermooxidative degradation products of polystyrene by of Cytophaga psychrophila from a systemic disease in eel and cyprinids. membrane extraction with sorbent interface and gas chromatography. Journal Diseases of Aquatic Organisms 10:217–220. of Chromatographic Science 40:350–354. Long, A., T. R. Fehringer, M. A. Swain, B. R. LaFrentz, D. R. Call, and Crump, E. M., and W. W. Kay. 2008. Congo red inhibition as a convenient diag- K. D. Cain. 2013. Enhanced efficacy of an attenuated Flavobacterium psy- nostic for Flavobacterium psychrophilum. Journal of Fish Diseases 31:553– chrophilum strain cultured under iron-limited conditions. Fish and Shellfish 557. Immunology 35:1477–1482. Crump, E. M., M. B. Perry, S. C. Clouthier, and W. W. Kay. 2001. Antigenic Madsen, L., and I. Dalsgaard. 1999. Reproducible methods for experimental in- characterization of the fish pathogen Flavobacterium psychrophilum. Applied fection with Flavobacterium psychrophilum in Rainbow Trout Oncorhynchus and Environmental Microbiology 67:750–759. mykiss. Diseases of Aquatic Organisms 36:169–176. Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 Ekman, E. 2003. Natural and experimental infections with Flavobacterium Madsen, L., J. D. Møller, and I. Dalsgaard. 2005. Flavobacterium psychrophilum psychrophilum in salmonid fish. Doctoral dissertation. Swedish University of in Rainbow Trout, Oncorhynchus mykiss (Walbaum), hatcheries: studies on Agricultural Sciences, Uppsala. broodstock, eggs, fry and environment. Journal of Fish Diseases 28:39–47. Ekman, E., and L. Norrgren. 2003. Pathology and immunohistochemistry in Michel, C., D. Antonio, and R. P. Hedrick. 1999. Production of viable cultures three species of salmonids after experimental infection with Flavobacterium of Flavobacterium psychrophilum: approach and control. Research Microbi- psychrophilum. Journal of Fish Diseases 26:529–538. ology 150:351–358. Faruk, M. A. R. 2002. A review on Rainbow Trout fry syndrome (RTFS). Nagai, T., T. Tamura, Y. Iida, and T. Yoneji. 2004. Differences in susceptibility Pakistan Journal of Biological Sciences 5:230–233. to Flavobacterium psychrophilum among three stocks of Ayu Plecoglossus Fevolden, S. E., T. Refstie, and K. H. Roed. 1992. Disease resistance in Rain- altivelis. Fish Pathology 39:159–164. bow Trout (Oncorhynchus mykiss) selected for stress response. Aquaculture Nematollahi, A., A. Decostere, F. Pasmans, and F. Haesebrouck. 2003. 104:19–29. Flavobacterium psychrophilum infections in salmonid fish. Journal of Fish Gimenez-Papiol,´ G., F. Padros,´ A. Roque, A. Estevez,´ and D. Furones. 2009. Diseases 26:563–574. Effects of a peroxide-based commercial product on bacterial load of larval Obach, A., and F. B. Laurencin. 1991. Vaccination of Rainbow Trout On- rearing water and on larval survival of two species of Sparidae under intensive corhynchus mykiss against the visceral form of coldwater disease. Diseases culture: preliminary study. Aquaculture Research 40:504–508. of Aquatic Organisms 12:13–15. Hedrick, R. P., T. S. McDowell, G. D. Marty, G. T. Fosgate, K. Mukkatira, K. Paulse, A. N., V. A. Jackson, and W. Khan. 2007. Comparison of enu- Myklebust, and M. El-Matbouli. 2003. Susceptibility of two strains of Rain- meration techniques for the investigation of bacterial pollution in the DISEASE SUSCEPTIBILITY OF RAINBOW TROUT STRAINS 167

Berg River, Western Cape, South Africa. Water S A (Pretoria) 33:165– to bacterial cold-water disease is moderately heritable and is not adversely 174. correlated with growth. Journal of Animal Science 87:860–867. Plant, K. P., S. E. LaPatra, D. R. Call, and K. D. Cain. 2011. Immunization Soule, M., K. Cain, S. LaFrentz, and D. R. Call. 2005. Combining suppression of Rainbow Trout, Oncorhynchus mykiss (Walbaum), with Flavobacterium subtractive hybridization and microarrays to map the intraspecies phylogeny psychrophilum proteins elongation factor-Tu, SufB FE-S assembly protein of Flavobacterium psychrophilum. Infection and Immunity 73:3799–3802. and ATP synthaseβ. Journal of Fish Diseases 34:247–250. Speare, D. J., and G. J. Arsenault. 1997. Effects of intermittent hydrogen per- Rach, J. J., M. P. Gaikowski, and R. T. Ramsay. 2000. Efficacy of hydrogen oxide exposure on growth and columnaris disease prevention of juvenile peroxide to control mortalities associated with bacterial gill disease infections Rainbow Trout (Oncorhynchus mykiss). Canadian Journal of Fisheries and on hatchery-reared salmonids. Journal of Aquatic Animal Health 12:119– Aquatic Sciences 54:2653–2658. 127. Wagner, E. J., R. E. Arndt, E. J. Billman, A. Forest, and W. Cavender. 2008. Rahman, M. H., A. Kuroda, J. M. Dijkstra, I. Kiryu, T. Nakanishi, and M. Oto- Comparison of the efficacy of iodine, formalin, salt, and hydrogen peroxide take. 2002. The outer membrane fraction of Flavobacterium psychrophilum for control of external bacteria on Rainbow Trout eggs. North American induces protective immunity in Rainbow Trout and Ayu. Fish and Shellfish Journal of Aquaculture 70:118–127. Immunology 12:169–179. Wagner, E. J., M. Bartley, R. Arndt, R. W. Oplinger, and M. D. Routledge. 2012. Rahman, M. H., T. Nakanishi, and M. Ototake. 2001. Development of vac- Comparison of hatchery and field performance between a whirling disease cine against coldwater disease in Ayu, Plecoglossus altivelis. Bulletin of the resistant strain and the Ten Sleep strain of Rainbow Trout. Journal of Aquatic National Research Institute of Aquaculture (Supplement 5):49–53. Animal Health 24:110–120. Rahman, M. H., M. Ototake, Y. Iida, Y. Yokomizo, and T. Nakanishi. 2000. Wagner, E. J., C. Wilson, R. Arndt, P. Goddard, M. Miller, A. Hodgson, R. Efficacy of oil-adjuvanted vaccine for coldwater disease in Ayu Plecoglossus Vincent, and K. Mock. 2006. Evaluation of disease resistance of the Fish altivelis. Fish Pathology 35:199–203. Lake-DeSmet, Wounded Man, and Harrison Lake strains of Rainbow Trout Santos, Y., P. J. Huntly, A. Turnbull, and T. S. Hastings. 1992. Isolation of Cy- exposed to Myxobolus cerebralis. Journal of Aquatic Animal Health 18:128– tophaga psychrophila (Flexibacter psychrophilus) in association with Rain- 135. bow Trout mortality in the United Kingdom. Bulletin of the European Asso- Wohlsen, T., J. Bates, G. Vesey, W. A. Robinson, and M. Katouli. 2006. Eval- ciation of Fish Pathologists 12:209–210. uation of the methods for enumerating coliform bacteria from water samples Silverstein, J. T., R. L. Vallejo, Y. Palti, T. D. Leeds, C. E. Rexroad III, T. using precise reference standards. Letters in Applied Microbiology 42:350– J. Welch, G. D. Wiens, and V. Ducrocq. 2009. Rainbow Trout resistance 356. Downloaded by [Department Of Fisheries] at 23:54 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:55 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Cell Culture Medium Inhibits Antigen Binding Used in an ELISA for Detection of Antibodies against Nervous Necrosis Virus Bomi Choia, Hyun Jung Gyea, Myung-Joo Oha & Toyohiko Nishizawaa a Department of Aqualife Medicine, Chonnam National University, Yeosu 550-749, South Korea Published online: 05 Aug 2014.

To cite this article: Bomi Choi, Hyun Jung Gye, Myung-Joo Oh & Toyohiko Nishizawa (2014) Cell Culture Medium Inhibits Antigen Binding Used in an ELISA for Detection of Antibodies against Nervous Necrosis Virus, Journal of Aquatic Animal Health, 26:3, 168-172, DOI: 10.1080/08997659.2014.922516 To link to this article: http://dx.doi.org/10.1080/08997659.2014.922516

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:168–172, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.922516

COMMUNICATION

Cell Culture Medium Inhibits Antigen Binding Used in an ELISA for Detection of Antibodies against Nervous Necrosis Virus

Bomi Choi, Hyun Jung Gye, Myung-Joo Oh, and Toyohiko Nishizawa* Department of Aqualife Medicine, Chonnam National University, Yeosu 550-749, South Korea

2000), there are still problems at the grow-out stages of fish in Abstract sea cages, due to the horizontal transmission of NNV through We investigated the optimum dilution of nervous necrosis virus the rearing seawater sourced from the environment. Thus, de- (NNV) for use as antigens to detect antibodies by an enzyme- velopment of an effective vaccine against NNV infection is one linked immunosorbent assay (ELISA) in the Sevenband Grouper Epinephelus septemfasciatus. The ELISA values for a standard- of the most important needs in the aquaculture of Sevenband ized suspension of antigens diluted with L-15 medium containing Grouper. Several kinds of NNV vaccines have been developed, 1% fetal bovine serum decreased gradually with the dilution of the including a formalin-inactivated vaccine, a subunit vaccine with antigens, whereas those for the antigens diluted with distilled water a recombinant coat protein, a virus-like particle vaccine, and a (DW) initially increased with the dilution of the antigens, peaked live vaccine (Husgar´ ð et al. 2001; Tanaka et al. 2001; Yuasa at a 320-fold dilution, and then decreased thereafter. Additional studies revealed that binding of NNV antigens to ELISA wells was et al. 2002; Sommerset et al. 2005; Yamashita et al. 2005; Liu inhibited by fetal bovine serum and other substances in the L-15 et al. 2006; Thiery´ et al. 2006; Lin et al. 2007; Kai and Chi 2008; medium. Sera obtained from Sevenband Grouper vaccinated with Nishizawa et al. 2009, 2011a, 2011b, 2012; Oh et al. 2012). live NNV vaccine and survivors from natural NNV-infection were The detection of specific antibodies against a pathogen is subjected to antibody detection by ELISA. All of the sera were important to determine if immunity has been established in vac- positive by ELISA when the standardized suspension was diluted 320-fold, whereas sera from five out of the six survivors and two out cinated fish and is generally carried out using an enzyme-linked of the six vaccinated fish were negative or weakly positive by ELISA immunosorbent assay (ELISA). Several studies using ELISA using NNV antigens diluted 10-fold. We therefore concluded that systems for the detection of specific fish antibodies have been cultured NNV solutions prepared in cell culture media may need reported for aquatic animals (Jorgensen et al. 1991; Yoshimizu to be diluted with distilled water for use in ELISA. et al. 1992; Dixon et al. 1994; Hoglund¨ and Pilstrom¨ 1994, 1995; LaPatra 1996; Nishida et al. 1998; Watanabe et al. 1998; Swain Fish infected with nervous necrosis virus (NNV) of the genus and Nayak 2003; Okuda et al. 2006; Kim et al. 2007, 2008, 2009; Betanodavirus (Nodaviridae), exhibit abnormal swimming be- Takami et al. 2010). However, use of an ELISA for fish anti- Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 havior, vacuolation, and/or necrosis in the central nervous sys- body detection can be problematic because of difficulties such tem (Munday et al. 2002). Serious economic losses due to as low reproducibility, which is partly due to high background NNV infection have occurred in fingerlings and market-sized optical density (OD) caused by nonspecific reactions between Sevenband Grouper Epinephelus septemfasciatus (also known fish antibodies and antigens (Olesen et al. 1991; Hoglund¨ and as Hyporthodus septemfasciatus) in Korea and in other Asian Pilstrom¨ 1994, 1995; Knopf et al. 2000; Kibenge et al. 2002; countries (Sohn et al. 1998; Muroga et al. 2001). Depending on Guo and Woo 2004). It was reported that one of the most impor- the fish species, losses due to NNV infection tend to be very tant causes of those problems was the nonspecific adsorption of high not only at the larval stage but also in the juvenile and fish antibodies with ELISA blocking reagents, but nonspecific grow-out stages. Although it is possible to control NNV infec- adsorption can be suppressed by the pretreatment of fish sera tion in the larval stage by preventing vertical transmission of with a skim milk solution (Kim et al. 2007). To develop an NNV- NNV (Mushiake et al. 1994; Mori et al. 1998; Watanabe et al. specific ELISA, the procedures of Kim et al. (2007) were used

*Corresponding author: [email protected] Received January 28, 2014; accepted April 16, 2014

168 COMMUNICATION 169

to screen sera from Sevenband Grouper infected and/or vacci- The data were shown as relative ELISA values against the av- nated with NNV; however, the detection sensitivity appeared erage of ELISA absorbance in duplicate wells with the cultured low. Thus, in the present study we improved sensitivity by op- NNV solution diluted 10-fold with DW. timizing the dilution of NNV suspension used as the ELISA Inhibition of NNV antigens binding to ELISA plate wells antigen. by L-15 medium components or FBS.—Preparations of L-15 medium with and without 10% FBS (L-1510 and L-150,respectively) were diluted serially twofold with DW from 1:10 to METHODS 1:20,480 and were mixed with an equal volume of cultured NNV Virus.—An isolate of NNV SgNag05 (Kokawa et al. 2008; solution diluted 160-fold with DW (final dilution rate of NNV Nishizawa et al. 2008, 2009) was cultured at 25◦C in Striped antigens: 320-fold). Each solution was placed into duplicate Snakehead Ophicephalus striatus (SSN-1) cells maintained in wells of ELISA plates and fixed by drying at 37◦C overnight. Leibovitz’s L-15 medium (Gibco) containing 10% (v/v) fetal The NNV antigens fixed on the ELISA plate were detected with bovine serum (FBS, Gibco), 150 IU/mL of penicillin G, and the rabbit antiserum against NNV, HRP-conjugated antirabbit 100 µg/mL of streptomycin. Cultured NNV was centrifuged IgG swine serum, and OPD substrate solution, as described (12,000 × g,20min,4◦C) to remove cell debris, and the resul- above. tant supernatant was harvested and stocked at −80◦C until use. Detection of antibodies against NNV in fish sera by ELISA.— The NNV infectivity was titrated using 96-well microplates, After fixing NNV antigens into duplicate wells of ELISA plates which were seeded with SSN-1 cells. After 10 d of culture, at dilutions of 1:10 or 1:320 as described above, the plate wells the appearance of cytopathic effect (CPE) was evaluated to de- were blocked with SM-PBS at 25◦C for 1 h. After being washed termine the 50% tissue culture infectious dose (TCID50). In three times with PBS, 50 µL of fish serum diluted 1:40 in SM- preliminary experiments the infectivity titer of the standardized PBS were placed in the wells and incubated at 25◦Cfor1h.Fish 9.05 NNV suspension was calculated to be 10 TCID50/mL. antibodies reacting with the NNV antigens in the ELISA plates Fish sera.—Anti-NNV fish sera were obtained from Sev- were detected with rabbit antiserum against Sevenband Grouper enband Grouper that had survived a natural NNV infection at Ig (second serum), HRP-conjugated antirabbit IgG swine serum Yeosu, Korea, in 2012, and also from Sevenband Grouper that (third serum), and OPD substrate solution in the same manner had been immunized with live NNV at a low temperature fol- as described above. lowing the methods of Nishizawa et al. (2012) and Oh et al. (2013). Detection of NNV-antigens by ELISA.—The standardized RESULTS AND DISCUSSION NNV stock suspension was used as the antigen for the ELISA Twofold serial dilutions of standard NNV suspensions with after a 10-fold dilution with distilled water (DW). Twofold se- L-151 or DW from 10-fold to 10,240-fold were fixed to ELISA rial dilutions of standard NNV suspensions were prepared with plate wells and detected with anti-NNV rabbit serum and HRP- L-15 medium containing 1% FBS (L-151) or DW from 10-fold conjugated antirabbit Ig serum (Figure 1). The relative ELISA to 10,240-fold. After NNV antigens of each dilution were fixed values of antigens diluted with L-151 decreased gradually with to ELISA plate wells, the ELISA was performed following the the dilution of the antigen and converged to zero at ≥320-fold method described by Kim et al. (2007). Briefly, the diluted dilutions. Conversely, the relative ELISA values of antigens NNV antigen suspension was placed in the wells of ELISA diluted with DW increased with increasing dilution until peak- plates (Greiner bio-one) at 100 µL/well and fixed by drying at ing at a 320-fold dilution (relative ELISA value: 2.75) before 37◦C overnight. The ELISA plate wells were blocked with 5% decreasing drastically and converging to zero at ≥5,120-fold

Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 skim milk in Dulbecco’s phosphate-buffered saline (SM-PBS) dilution (Figure 1). These experiments demonstrated that sufﬁ- at 25◦C for 1 h. After being washed manually three times with cient antigen to give a relative ELISA value of ≥0.1 remained PBS, 50 µL of rabbit antiserum against NNV SgNag05, diluted in the wells when the antigens were diluted ≤160-fold with 1:500 in SM-PBS, was placed in each well and incubated at L-151 and in those diluted ≤2,560-fold with DW, suggesting 25◦C for 1 h. The plate wells were again washed three times that a larger amount of NNV antigens was bound to ELISA with PBS, and 50 µL of antirabbit immunoglobulin G (IgG) wells when diluted with DW rather than with L-151. Because swine Ig conjugated with horseradish peroxidase (HRP; Dako, the ELISA values initially increased in wells containing anti- 1:500 dilution with SM-PBS) was added to the plate wells at gens diluted with DW compared with L-151, we hypothesized 25◦C for 1 h. After washing three times with PBS, 100 µLof that this could be due to the inhibition of NNV antigens bind- o-phenylenediamine (OPD) substrate solution (1 mg/mL OPD, ing to components in the L-151 medium present in the standard 0.03% H2O2, 100 mM Na2HPO4, 50 mM citric acid) was added suspension. ◦ to each well prior to incubation at 25 C for 1 h. The development Thus, we investigated whether L-1510,L-150, and FBS in- reaction was stopped by the addition of 100 µLof2NH2SO4 duced inhibition of the ELISA (Figure 2). When NNV antigens and the absorbance was then read at 492 nm (OD492)usinga were mixed with serial dilutions of L-1510 in DW the ELISA 384 microplate reader (SpectraMax 340PC , Molecular Devices). values increased with the dilution of L-1510 and reached >0.9 170 CHOI ET AL.

in the wells with ≥1,280-fold dilution (Figure 2A). The same tendency was observed when NNV antigens were mixed with serial dilutions of FBS or L-150; the relative ELISA values increased with each dilution and reached >0.9 in the wells with ≥5,120-fold dilution of FBS (corresponding to ≤0.008% of FBS, Figure 2B) and ≥640-fold dilution of L-150 (Figure 2C). Although all of the ELISA wells contained the same amount of NNV antigen, significant declines in ELISA values were observed in the wells that had high concentrations of L-1510, L-150, or FBS. These results confirmed our assumption that the binding of NNV antigens to ELISA plate wells was inhibited by components in the L-1510 medium, such as FBS, amino acids, and/or salts. Next, we subjected the sera obtained from Sevenband Grouper that had survived natural NNV infections or that were vaccinated with live NNV at a low temperature (Nishizawa et al. 2012; Oh et al. 2013) to antibody detection by ELISA FIGURE 1. Change of absorbance values in the ELISA wells fixed with cul- using either 10- or 320-fold dilutions of cultured NNV solution 8.05 tured NNV solution (10 TCID50/100 µL per well) diluted with L-15 medium as antigens (Figure 3). The sera obtained from natural survivors ± containing 1% FBS (L-151) or DW. The data (mean SD) are shown as relative (S1–S6) and in ELISA plate wells coated with the 320-fold values against ELISA absorbance of the wells with the cultured NNV solution diluted NNV antigens exhibited ELISA values from 0.17 to diluted 10-fold with DW. Each data point is an average of ELISA absorbances in duplicate wells. The error bar indicates SD. 0.29, but sera from those plate wells in which the antigens were diluted 10-fold exhibited ELISA values of <0.1. The same tendency was also observed in the sera of fish vaccinated with live NNV,i.e., the sera from vaccinated fish in the ELISA plate wells containing the NNV antigens diluted 320-fold exhibited ELISA values from 0.43 to 0.67, whereas sera in those containing the antigens diluted 10-fold ranged from 0.06 to 0.22. When scoring fish sera by ELISA we have generally concluded that sera with ≥0.10 of ELISA absorbance were positive, those between 0.05 and 0.10 were suspect positive, and those with values ≤0.05 were negative (Kim et al. 2008). According to this criterion, all of the sera from the survivors and vaccinated fish were positive Downloaded by [Department Of Fisheries] at 23:55 16 November 2014

FIGURE 2. Inhibition of binding NNV antigens to ELISA plate wells by FBS and L-15 medium. The NNV antigens were mixed with L-15 media without FBS (L-150), with 10% FBS (L-1510), or with FBS, diluted serially twofold with DW, and ﬁxed to ELISA plate wells. The NNV antigens were detected with rabbit antiserum against NNV and antirabbit IgG swine Ig conjugated with horseradish peroxidase. The data (mean ± SD) are shown as relative values FIGURE 3. Detection of antibodies against NNV from Sevenband Grouper against the ELISA absorbance of the wells with the NNV antigens diluted 320- surviving natural NNV-infection and ﬁsh vaccinated with live NNV by using fold with DW. (A) Dilution of L-1510,(B) diluted with FBS, (C) dilution of ELISA plate wells coated with cultured NNV solution diluted 10- and 320-fold L-150. The error bar indicates SD. with DW. The error bar indicates SD. COMMUNICATION 171

a 320-fold dilution of standard NNV suspension. Furthermore, reaction of the ﬁsh sera was shown to be speciﬁc to the antigens. Based on the present data, we concluded that standard NNV suspensions prepared from cell cultures grown using L-15 medium should be diluted with DW prior to use as ELISA antigens. Further optimization of NNV antigen preparations, as well as the binding of NNV antigens to ELISA plate wells, could be improved with further work, but may be dependent on each virus or cell culture system used. Cultured infectious hematopoietic necrosis virus, viral hemorrhagic septicemia virus, and red seabream iridovirus were generally diluted 10-fold with DW for binding viral antigens to ELISA plate wells (Kim et al. 2007, 2008; Kwon et al. 2010). However, it appears that the optimum dilution of cultured virus solutions for use as ELISA antigens is dependent on the individual virus.

ACKNOWLEDGMENTS This study was supported by grant 2011-0029000 from the midcareer researcher program through the National Research Foundation by the Korean Government, Ministry of Education, Science and Technology.

REFERENCES Dixon, P. F., A. M. Hattenberger-Baudouy, and K. Way. 1994. Detection of carp antibodies to spring viraemia of carp virus by a competitive immunoassay. Diseases of Aquatic Organisms 19:181–186. Guo, F. C., and P. T. K. Woo. 2004. Detection and quantification of Spironucleus barkhanus in experimentally infected Atlantic Salmon Salmo salar. Diseases FIGURE 4. Change of ELISA absorbance values of Sevenband Grouper sera of Aquatic Organisms 61:175–178. considered positive for antibodies against NNV antigens. Specific antibodies Hoglund,¨ J., and L. Pilstrom.¨ 1994. Purification of adult Anguillicola crassus against NNV were detected from sera of the vaccinated fish (V2 and V3) whole-worm extract antigens for detection of specific antibodies in serum and the survivors from a natural NNV-infection (S5) (shown in Figure 3) by from the European Eel (Anguilla anguilla) by ELISA. Fish and Shellfish using ELISA plate wells containing different concentrations of NNV antigens. Immunology 4:311–319. (A) Plate wells coated with cultured NNV solution diluted 320-fold with DW, Hoglund,¨ J., and L. Pilstrom.¨ 1995. Mechanical isolation and characterization (B) plate wells coated with the antigens diluted 10-fold with DW. The error bar of antigens from adult Anguillicola crassus. Fish and Shellfish Immunology indicates SD. 5:51–60. Husgar´ ð, S., S. Grotmol, B. K. Hjeltnes, O. M. Rødseth, and E. Biering. 2001. Immune response to a recombinant capsid protein of striped jack nervous necrosis virus (SJNNV) in Turbot Scophthalumus maximus and Atlantic Hal- by the ELISA using a 320-fold dilution of NNV antigens, ibut Hippoglossus hippoglossus, and evaluation of a vaccine against SJNNV. whereas two out of the six survivors were negative (S2 and S5). Diseases of Aquatic Organisms 45:33–44.

Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 The remaining four fish (S1, S3, S4, and S6) as well as two out Jorgensen, P. E. V., N. J. Olesen, and N. Lorenzen. 1991. Infectious hematopoi- of the six vaccinated fish (V2 and V4) were suspect positive by etic necrosis (IHN) and viral hemorrhagic septicemia (VHS): detection of trout antibodies to the causative viruses by means of plaque neutralization, the ELISA using a 10-fold dilution of NNV antigens (Figure 3). immunofluorescence, and enzyme-linked immunosorbent assay. Journal of The twofold serial dilutions of fish sera (S5, V2, and V3 Aquatic Animal Health 3:100–108. in Figure 3) were subjected to antibody detection by ELISA Kai, Y. H., and S. C. Chi. 2008. Efficacies of inactivated vaccines against be- using 10- and 320-fold dilutions of NNV antigens (Figure 4). tanodavirus in grouper larvae (Epinephelus coioides) by bath immunization. The ELISA absorbance of those sera decreased with the dilu- Vaccine 26:1450–1457. Kibenge, M. T., B. Opazo, A. H. Rojas, and F. S. Kibenge. 2002. Serological ev- tion and converged to 0 in both of the ELISA plates (Figure 4), idence of infectious salmon anaemia virus (ISAV) infection in farmed fishes, confirming that the absorbance was due to the reaction of spe- using an indirect enzyme-linked immunosorbent assay (ELISA). Diseases of cific antibodies against NNV antigens; however, the absorbance Aquatic Organisms 51:1–11. values of sera in the ELISA plate wells containing the antigens Kim, H. J., N. Oseko, T. Nishizawa, and M. Yoshimizu. 2009. Protection of diluted 320-fold were obviously higher than those in the ELISA Rainbow Trout from infectious hematopoietic necrosis (IHN) by injection of infectious pancreatic necrosis virus (IPNV) or Poly(I:C). Diseases of Aquatic plate wells containing the antigens diluted 10-fold. These re- Organisms 83:105–113. sults confirmed that the sensitivity of detection of antibodies Kim, W.-S., M. Mochizuki, T. Nishizawa, and M. Yoshimizu. 2008. Detection against NNV was improved by using the ELISA prepared with of specific antibodies against infectious hematopoietic necrosis virus from 172 CHOI ET AL.

Rainbow Trout sera by ELISA using two novirhabdoviruses. Fish Pathology Oh, M.-J., W.-S. Kim, H. G. Seo, H. J. Gye, and T. Nishizawa. 2013. Change in 43:112–116. infectivity titer of nervous necrosis virus (NNV) in brain tissue of Sevenband Kim, W.-S., T. Nishizawa, and M. Yoshimizu. 2007. Non-specific adsorption Grouper with Poly(I:C) administration. Journal of Fish Diseases 36:159–162. of fish immunoglobulin M (IgM) to blocking reagents on ELISA plate wells. Oh, M.-J., I. Takami, T. Nishizawa, W.-S. Kim, C.-S. Kim, S.-R. Kim, and M. Diseases of Aquatic Organisms 78:55–59. A. Park. 2012. Field tests of Poly(I:C) immunization with nervous necrosis Knopf, K., K. Naser, M. H. van der Heijden, and H. Taraschewski. 2000. Eval- virus (NNV) in Sevenband Grouper Epinephelus septemfasciatus (Thunberg). uation of an ELISA and immunoblotting for studying the humoral immune Journal of Fish Diseases 35:187–191. response in Anguillicola crassus infected European Eel Anguilla anguilla. Okuda, R., T. Nishizawa, and M. Yoshimizu. 2006. Infection-cycles of Renibac- Diseases of Aquatic Organisms 43:39–48. terium salmoninarum in hatcheries of Masu Salmon Oncorhynchus masou Kokawa, Y., I. Takami, T. Nishizawa, and M. Yoshimizu. 2008. A mixed infec- by monitoring specific antibodies. Fish Pathology 41:175–178. tion in Sevenband Grouper Epinephelus septemfasciatus affected with viral Olesen, N. J., N. Lorenzen, and P. E. V. Jørgensen. 1991 Detection of Rain- nervous necrosis (VNN). Aquaculture 284:41–45. bow Trout antibody to Egtved virus by enzyme-linked immunosorbent assay Kwon, S. R., T. Nishizawa, I. Takami, and M. Yoshimizu. 2010. Antibody (ELISA), immunofluorescence (IF), and plaque neutralization tests (50% detection against Red Sea Bream iridovirus (RSIV) in Yellowtail Seriola PNT). Diseases of Aquatic Organisms 10:31–38. quinqueradiata using ELISA. Fish Pathology 45:73–76. Sohn, S.-G., M.-A. Park, M.-J. Oh, and S.-K.Chun. 1998. A fish nodavirus iso- LaPatra, S. E. 1996. The use of serological techniques for virus surveillance and lated from cultured Sevenband Grouper, Epinephelus septemfasciatus. Jour- certification of finfish. Annual Review of Fish Diseases 6:15–28. nal of Fish Diseases 11:97–104. Lin, C. C., J. H. Y. Lin, M. S. Chen, and H. L. Yang. 2007. An oral nervous Sommerset, I., R. Skern, E. Biering, H. Bleie, I. U. Fiksdal, S. Grove, and A. necrosis virus vaccine that induces protective immunity in larvae of grouper H. Nerland. 2005. Protection against Atlantic Halibut nodavirus in turbot is (Epinephelus coioides). Aquaculture 268:265–273. induced by recombinant capsid protein vaccination but not following DNA Liu, W., C. H. Hsu, C. Y. Chang, H. H. Chen, and C. S. Lin. 2006. Immune vaccination. Fish and Shellfish Immunology 18:13–29. response against grouper nervous necrosis virus by vaccination of virus-like Swain, P., and S. K. Nayak. 2003. Comparative sensitivity of different serologi- particles. Vaccine 24:6282–6287. cal tests for seromonitoring and surveillance of Edwardsiella tarda infection Mori, K., K. Mushiake, and M. Arimoto. 1998. Control measures for viral of Indian major carps. Fish and Shellfish Immunology 15:333–340. nervous necrosis in striped jack. Fish Pathology 33:443–444. Takami, I., S. R. Kwon, T. Nishizawa, and M. Yoshimizu. 2010. Protection of Munday, B. L., J. Kwang, and N. Moody. 2002. Betanodavirus infections of Japanese flounder Paralichthys olivaceus from viral hemorrhagic septicemia teleost fish: a review. Journal of Fish Diseases 25:127–142. (VHS) by Poly(I:C) immunization. Diseases of Aquatic Organisms 89:109– Muroga, K. 2001. Viral, and bacterial diseases of marine fish and shellfish in 115. Japanese hatcheries. Aquaculture 202:23–44. Tanaka, S., K. Mori, M. Arimoto, T. Iwamoto, and T. Nakai. 2001. Protective Mushiake, K., T. Nishizawa, T. Nakai, I. Furusawa, and K. Muroga. 1994. immunity of Sevenband Grouper, Epinephelus septemfasciatus Thunberg, Control of VNN in Striped Jack: selection of spawners based on the detection against experimental viral nervous necrosis. Journal of Fish Diseases 24:15– of SJNNV gene by polymerase chain reaction (PCR). Fish Pathology 29:177– 22. 182. Thiery,´ R., J. Cozien, J. Cabon, F. Lamour, M. Baud, and A. Schneemann. 2006. Nishida, H., M. Yoshimizu, and Y. Ezura. 1998. Detection of antibody against Induction of a protective immune response against viral nervous necrosis in lymphocystis disease virus in Japanese flounder by enzyme linked im- the European Sea Bass Dicentrarchus labrax by using betanodavirus virus- munosorbent assay. Fish Pathology 33:207–211. like particles. Journal of Virology 80:10201–10207. Nishizawa, T., H. J. Gye, I. Takami, and M.-J. Oh. 2012. Potentiality of a Watanabe, K., T. Nishizawa, and M. Yoshimizu. 2000. Selection of brood stock live vaccine with nervous necrosis virus (NNV) for Sevenband Grouper candidates of Barfin Flounder using an ELISA system with recombinant pro- Epinephelus septemfasciatus at a low rearing temperature. Vaccine 30:1056– tein of Barfin Flounder nervous necrosis virus. Diseases of Aquatic Organisms 1063. 41:219–223. Nishizawa, T., Y.Kokawa, T. Wakayama, S. Kinoshita, and M. Yoshimizu.2008. Watanabe, K., S. Suzuki, T. Nishizawa, K. Suzuki, M. Yoshimizu, and Y. Ezura. Enhanced propagation of fish nodaviruses in BF-2 cells persistently infected 1998. Control strategy for viral nervous necrosis of Barfin Flounder. Fish with snakehead retrovirus (SnRV). Diseases of Aquatic Organisms 79:19–25. Pathology 33:445–446. Nishizawa, T., I. Takami, Y. Kokawa, and M. Yoshimizu. 2009. Fish immuniza- Yamashita, H., Y. Fujita, H. Kawakami, and T. Nakai. 2005. The efficacy of in- tion using a synthetic double-stranded RNA Poly(I:C), an interferon inducer, activated virus vaccine against viral nervous necrosis (VNN). Fish Pathology offers protection against RGNNV, a fish nodavirus. Diseases of Aquatic Or- 40:15–21. ganisms 83:115–122. Yoshimizu, M., S. Direkbusarakom, T. Nomura, Y. Ezura, and T. Kimura. Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 Nishizawa, T., I. Takami, M. Yang, and M.-J. Oh. 2011a. Live vaccine of 1992. Detection of antibody against Aeromonas salmonicida in the serum viral hemorrhagic septicemia virus (VHSV) for Japanese flounder using a of salmonid fish by the enzyme linked immunosorbent assay. Fish Pathology fish rearing temperature at around 21◦C, instead of Poly(I:C) administration. 27:73–82. Vaccine 29:8397–8404. Yuasa, K., I. Koesharyani, D. Roza, K. Mori, M. Katata, and T. Nakai. 2002. Im- Nishizawa, T., I. Takami, M. Yoshimizu, and M.-J. Oh. 2011b. Required dose of mune response of Humpback Grouper, Cromileptes altivelis (Valenciennes) fish nervous necrosis virus (VNN) for Poly(I:C) immunization of Sevenband injected with the recombinant coat protein of betanodavirus. Journal of Fish Grouper Epinephelus septemfasciatus. Aquaculture 331:100–104. Diseases 25:53–56. This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:55 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Development of a Multiplex Assay to Measure the Effects of Shipping and Storage Conditions on the Quality of RNA Used in Molecular Assays for Detection of Viral Haemorrhagic Septicemia Virus A. Siaha, H. Duesundb, K. Frischc, A. Nylundd, P. McKenziec & S. Saksidaa a British Columbia Centre for Aquatic Health Sciences, 871A Island Highway, Campbell River, British Columbia V9W 2C2, Canada b Cermaq ASA, Grev Wedelsplass 5, N-0105 Oslo, Norway c Cermaq, #203-919 Island Highway Campbell River, British Columbia V9W 2C2, Canada d Department of Biology, University of Bergen Thormohlensgt 55, N-5020 Bergen, Norway Published online: 05 Aug 2014.

To cite this article: A. Siah, H. Duesund, K. Frisch, A. Nylund, P. McKenzie & S. Saksida (2014) Development of a Multiplex Assay to Measure the Effects of Shipping and Storage Conditions on the Quality of RNA Used in Molecular Assays for Detection of Viral Haemorrhagic Septicemia Virus, Journal of Aquatic Animal Health, 26:3, 173-180, DOI: 10.1080/08997659.2014.902874 To link to this article: http://dx.doi.org/10.1080/08997659.2014.902874

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:173–180, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.902874

ARTICLE

Development of a Multiplex Assay to Measure the Effects of Shipping and Storage Conditions on the Quality of RNA Used in Molecular Assays for Detection of Viral Haemorrhagic Septicemia Virus

A. Siah* British Columbia Centre for Aquatic Health Sciences, 871A Island Highway, Campbell River, British Columbia V9W 2C2, Canada H. Duesund Cermaq ASA, Grev Wedelsplass 5, N-0105 Oslo, Norway K. Frisch Cermaq, #203-919 Island Highway Campbell River, British Columbia V9W 2C2, Canada A. Nylund Department of Biology, University of Bergen Thormohlensgt 55, N-5020 Bergen, Norway P. McKenzie Cermaq, #203-919 Island Highway Campbell River, British Columbia V9W 2C2, Canada S. Saksida British Columbia Centre for Aquatic Health Sciences, 871A Island Highway, Campbell River, British Columbia V9W 2C2, Canada

Abstract In routine diagnostics, real-time reverse transcriptase quantitative PCR (RT-qPCR) has become a powerful method for ﬁsh health screening. Collection, transportation, and storage conditions of specimens could dramatically affect their integrity and could consequently affect RT-qPCR test results. In this study, to assess the expression proﬁle of α

Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 elongation factor 1 alpha (ELF-1 ) gene, head kidney (HK) tissues from Atlantic Salmon Salmo salar were exposed at room temperature, 4◦C, −20◦C, and −80◦C as well as in 70% ethanol for 6, 12, 24, 48, and 72 h. Data showed a signiﬁcant increase of RT-qPCR cycle threshold (Ct) values for ELF-1α ranging from 14.7 to 26.5 cycles for tissues exposed to room temperature. In order to mimic the sample transportation conditions, different temperatures of storage were used and tissue quality was evaluated using ELF-1α gene expression. Data showed that Ct values for ELF-1α increased signiﬁcantly when the tissues were transported on ice for 2 h, stored at −20◦C, thawed on ice for 6 h, and stored again at −80◦C. The HK tissues collected from Atlantic Salmon challenged with viral hemorrhagic septicemia virus (VHSV) through intraperitoneal injection were exposed at room temperature for 0, 6, 12, 24, 48, 72, and 96 h. Data showed a good correlation of values for ELF-1α and VHSV Ct although the ELF-1α mRNA of the host degraded faster than the RNA of VHSV. Based on these data, HK tissues could be transported on ice or ice packs without the quality of the tissue being affected when stored at −80◦C upon arrival at the laboratory. In addition, 70%

*Corresponding author: [email protected] Received November 25, 2013; accepted February 14, 2014 Published online August 5, 2014 173 174 SIAH ET AL.

ethanol could be used as a preservative for long-distance transportation. For an efﬁcient diagnostic test, a duplex VHSV–ELF-1α was developed and optimized. Data showed that the sensitivity of the duplex assay for VHSV was similar to the singleplex.

Viral hemorrhagic septicemia virus (VHSV), which belongs laboratories from 27 European countries was implemented for to the family of Rhabdoviridae and the genus Novirhabdovirus, the collection and transport of human blood samples (Pazzagli is an enveloped negative-strand RNA virus. With 11,158 nu- et al. 2013). This large-scale quality assessment provided guide- cleotides, the VHSV genome encodes for the nucleoprotein (N), lines for the preanalytical procedures of RNA analysis (Pazzagli the phosphoprotein (P), the matrix protein (M), the glycoprotein et al. 2013). Such a study has not yet been initiated for fish health (G), the nonstructural viral protein (Nv), and the polymerase (L) screening. The implementation of sample quality controls for an (Schutze et al. 1999). Based on phylogenetic analysis of VHSV RNA-based diagnostic would help to determine acceptance or sequences, four genotypes have been found with the genotype rejection criteria. IV that are predominantly located in North America and eastern During the last decade, the importance of reference genes Asia (Skall et al. 2005) and have two subtypes divided as IVa in RT-qPCR has been highlighted (Radonic´ et al. 2004), and and IVb (Elsayed et al. 2006). endogenous controls were suggested as key components of di- Viral hemorrhagic septicemia virus causes serious diseases agnostic assays using RT-qPCR (Snow et al. 2006). Indeed, in marine and freshwater fishes (Skall et al. 2005) and is desig- in fish health screening, several studies have shown the im- nated as a reportable aquatic pathogen by the U.S. Department portance of using endogenous genes as quality controls of the of Agriculture, the Canadian Food Inspection Agency, and the samples (Snow et al. 2006; Olsvik et al. 2008; Christiansen et al. World Organization for Animal Health (OIE). Virus cell culture 2011; Bland et al. 2012). These endogenous controls are widely has been traditionally used as the gold standard diagnostic test used as reference genes, assuming that the level of their expres- for VHSV (Lorenzen et al. 1999). However, virus propagation sion is unaffected under suitable conditions of sample storage by cell culture can take more than 2 weeks and confirmation (Snow et al. 2006). Among the identified reference genes, elon- of the pathogen requires additional tests. As an alternative to gation factor 1 alpha (ELF-1α) was selected as the most suit- cell culture, PCR-based techniques are used as confirmatory able endogenous control gene in studies with Atlantic Salmon diagnostic tests for VHSV and are recognized by the OIE. Salmo salar (Ingerslev et al. 2006; Jorgensen et al. 2006; Snow Techniques based on real-time quantitative PCR are consid- et al. 2006; Christiansen et al. 2011; Bland et al. 2012). In ad- ered to be powerful and accurate methods used to screen for dition, reference genes will ensure that the RT-qPCR testing nucleic acids of pathogens in fish health monitoring programs was processed properly. However, the main detractors for us- (Purcell et al. 2011). These techniques detect nucleic acids spe- ing endogenous genes for quality control in RT-qPCR screening cific to the targeted pathogen, thus allowing a high specificity are the cost of the analysis and the turnaround time of the as- and high sensitivity compared with other standard diagnostic say because this doubles the number of tests being performed methods. Several assays using real-time reverse transcriptase when compared with only testing for the targeted pathogen quantitative PCR (RT-qPCR) have been successfully developed RNA. to detect VHSV in fish (Duesund et al. 2010; Matejusova et al. Screening for nucleic acids of pathogens and endogenous 2010; Garver et al. 2011; Jonstrup et al. 2013). However, con- gene expression in the same and single RT-qPCR analysis cerns have been raised regarding the possible effects that RNA may represent a more efficient way, both in terms of cost and Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 integrity may have on efficacy of the assays and thus conse- turnaround time, to identify pathogen nucleic acids and at the quently influence the analysis. For instance, nucleic acids are same time evaluate the quality of the samples. Several tech- sensitive to nucleases such as RNases, which are activated over niques based on multiplex PCR have been successfully per- a wide temperature range (Fersht 1977:325–329). Indeed, the formed to detect simultaneously the nucleic acids of several fish analytical methods in molecular diagnostic assays are closely pathogens (del Cerro et al. 2002; Altinok et al. 2008; Liu et al. related to the conditions of storage and handling of the samples 2008). An important challenge of multiplex assays is the loss during transportation from the sampling locations to the labora- of sensitivity to the targeted sequences, which may lead to false tory after collection. In clinical diagnostics, several studies have negative results. reported the loss of viral RNA when stored at room temperature The objectives of this study were to evaluate the level of over a few days (Jose´ et al. 2003, 2005; Orpetveit et al. 2010). ELF-1α mRNA in different shipping, handling, and short-term For example, Halfon et al. (1996) showed 100% loss of hep- storage conditions (up to 3 d) and develop a duplex assay com- atitis C virus RNA after 5 d at room temperature. To standard- bining ELF-1α and VHSV. Our goal was to design an assay ize preanalytical procedures for molecular diagnostic assays, a capable of simultaneously assessing the quality of the samples pan-European external quality assessment (EQA) involving 102 and detecting VHSV. MULTIPLEX ASSAY TO MEASURE QUALITY OF RNA AND DETECT VHSV 175

METHODS maintained at 20◦C using MEM-10 medium (minimal essential Sampling and RNA extraction.— Head kidney (HK) samples medium supplemented with 10% fetal calf serum, 200 mM were dissected from Atlantic Salmon. Approximately 30 mg Glutamax-1, and 5 mL sodium bicarbonate [7.5% w/v] were of each sample were collected and RNA was extracted using used according to the procedure described in Afonso et al. Qiagen RNeasy Mini Kit according to the manufacturer’s pro- 2012). For North American VHSV (genogroup IV) propagation, ◦ tocol (Qiagen). Total RNA was eluted with 50 µLofRNase- EPC cells were inoculated with the virus and incubated at 15 C free water and stored at −80◦C for further analysis. Agarose for 45–60 min, and then fresh medium was added. Cytopathic gel electrophoresis was performed on samples stored at room effects (CPE) were assessed every day and virus was harvested temperature and at −80◦C for RNA quality assessment. after complete cell lysis. The suspension was centrifuged ◦ Complementary DNA synthesis and RT-qPCR.—A The RT- at 2,500 × g for 15 min at 4 C after which supernatant was ◦ qPCR was performed using the AgPath-ID One-Step RT- aliquoted into cryotubes and stored at −80 C. Virus stock titer PCR kit according to the manufacturer’s recommendations was determined using a plaque assay as described previously (Life Technologies). Singleplex analysis was performed us- (Burke and Mulcahy 1980). For optimization, RNA was ing a total reaction final volume of 12.5 µL and final con- extracted from the supernatant of the cell culture infected centrations of 300 and 600 nM of ELF-1α forward (5-CCCT with VHSV IV, and 2 µL of the extracted RNA was added to CCAGGACGTTTACAAA-3) and reverse primers (5-CACA 10.5 µL of the reaction final volume. Different concentrations CGGCCCACAGGTACA-3), respectively, and a final con- of the primer forward and reverse sets, respectively, were used: centration of 225 nM of the FAM-probe (5-VIC-ATCG 200 and 200 nM, 300 and 300 nM, 400 and 400 nM, 600 and GTGGTATTGGAAC-MGB-3) (Olsvik et al. 2005), 400 nM 600 nM, 900 and 900 nM, 300 and 600 nM, 600 and 300 nM, of each VHS forward and reverse primers, and 125 nM of the 600 and 900 nM, and 900 and 600 nM. Probe concentration was FAM-probe. Two microliters of RNA were added to each re- kept at 120 nM. Different concentrations of the probe ranging action. The assay for ELF-1α was designed and developed by from 100 to 250 nM (in 25-nM increments) were tested using Olsvik et al. (2005). the optimal concentrations of primers. The lowest Ct values For duplex analysis, the optimal conditions for the primers and highest final fluorescence value were used as criteria for and probe were 40 nM final concentration of ELF-1α forward the selection of optimum primers–probe concentrations. and reverse primers and 100 nM final concentration of the probe, The efficiency of the assay was performed using a 10-fold 400 nM of each VHS forward and reverse primers, and 250 nM serial dilution of RNA extracted from the virus cell culture of the probe. supernatant at 106 PFU per reaction. The assay was performed Analysis by RT-qPCR was performed in QuantStudio 12K using five serially diluted samples ranging from 105 to 101 PFU system (Life Technologies). Synthesis of complementary DNA per reaction in duplicates. (cDNA) was performed at 50◦C for 30 min. The thermocycling Analytical specificity and sensitivity.—The optimized assay parameters consisted of a 15-min AmpliTaq Gold polymerase was evaluated for specificity against infectious hematopoietic activation and denaturation at 95◦C followed by 45 cycles of necrosis virus (IHNV) North American strain, a member of 95◦C for 15 s and 60◦Cfor1min. the Rhabdoviridae family, and RNA from infectious pancreatic VHSV assay design.—A new TaqMan RT-qPCR assay necrosis virus (IPNV), a member of the Birnaviridae family. targeting the North American VHSV N-gene genogroup Also, RNA was extracted from 106 PFU using supernatant from IV was designed using PrimerExpress (Life Technologies). the cell culture propagation of each virus. The forward and reverse primers (5-CAATGACAACTCCAA The assay sensitivity was performed using an artificial single- GATCTCCTACA-3 and 5-CAAGGAGTCCACTGCGTACT strand DNA amplicon provided by Integrated DNA Technolo- Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 TC-3 , respectively) and probe sequences (5 -FAM-CTG gies (Coralville, Iowa). A 10-fold serial dilution of the artificial TCTCAGATCAGTGGG-MGB-3) target the position at 1,000– amplicon ranging from 2.9 × 106 to 0.29 copies was performed. 1,066 bp of the VHSV N-gene. In silico validation has Comparison between singleplex (1 target) and duplex (2 tar- been performed by blasting the primers and probe sequences gets).—For the duplex ELF-VHS assay, different concentrations against the GenBank database, and data showed 100% ho- of primers, probes, 25 × RT-PCR Enzyme Mix, dNTP, and mology with VHSV sequences genogroup IV isolated in MgCl2 were tested and optimized using kidney tissue spiked North America (accession numbers: DQ427105, HQ415763, with VHSV. To compare the singleplex VHSV and duplex HQ453210, JQ396250, HQ453211, HQ214133, GQ385941, ELF–VHSV, a 10-fold serial dilution of VHSV-spiked Atlantic GQ255380, HQ415762, X59241, HQ214135, JQ396249, Salmon kidney tissue was analyzed using both single and du- HQ168407, HQ214134, JQ396247, EF079895, HQ168406) and plex assays. Atlantic Salmon HK tissue was spiked with VHSV Asia (AB6726, AB179621, AB490792, KF477302, JF792424, serially diluted from 106 to 102 PFU in triplicate. JQ651390, KC685626, JQ651385). Sample quality evaluation.—Head kidney samples (30 mg) Assay optimization and efficiency.—Different combined were exposed to different conditions of storage (n = 6 for each ◦ ◦ ◦ primers and probe concentrations were evaluated for assay condition): room temperature, 4 C, −20 C, and −80 C and in ◦ optimization. Epithelioma papulosum cyprini (EPC) cells ethanol at −20 C for 6, 12, 24, 48, and 72 h. Extraction of 176 SIAH ET AL.

RNA and RT-qPCR analysis were performed on each sample as sis was performed on samples stored at room temperature and described above. at −80◦C for RNA quality assessment. Data showed a grad- To mimic the conditions of sample transportation, samples ual degradation of RNA, and the RNA smear completely disap- of HK tissue were stored under five different conditions: group peared after 48 h of exposure at room temperature but no change 1: stored in the freezer (−80◦C); group 2: transported on ice for was observed at −80◦C (Figure 1). After 12 h of exposure to 2 h and then stored in the freezer (−80◦C); group 3: transported room temperature, Ct values increased from 14.7 to 19.4 cycles. on ice for 2 h, stored in the freezer (−20◦C) for 12 h, and Under the other conditions, ELF-1α Ct values ranged between then stored in the freezer (−80◦C); group 4: transported on 14 and 16 (Figure 2). The Ct values corroborate the agarose gel ice for 2 h, stored in the freezer (−20◦C) for 12 h, thawed electrophoresis profiles (Figure 1). on ice for 2 h, and then stored in the freezer (−80◦C); group 5: The shipping and handling conditions were simulated using transported on ice, stored in the freezer (−20◦C) for 12 h, thawed different conditions of storage as described above. The data on ice for 6 h, and then stored in the freezer (−80◦C). The same showed that ELF-1α Ct values increased significantly when experiment was performed simultaneously with samples stored the tissues were transported under the conditions represented in 70% of molecular-grade absolute ethanol in order to evaluate in group 5 (on ice, stored at −20◦C, thawed on ice for 6 h, the ethanol’s ability to protect the sample against degradation. and stored at −80◦C; Figure 3). If the tissue was stored in Extraction of RNA and RT-qPCR analysis were performed on 70% ethanol, no differences were observed between any of the each sample as described above. different conditions of storage at that temperature (Figure 3). Viral RNA degradation experiment.—Two grams of head kidney tissue were collected from Atlantic Salmon challenged VHSV RNA Degradation Experiment with VHSV IVa through intraperitoneal injection. Subsamples Head kidney tissues collected from Atlantic Salmon chal- (30 mg) were cut and incubated at room temperature for 0 h as lenged with VHSV IVa (Isolate BC 99–292) through intraperi- a control (and then kept at −80◦C) and for 6, 12, 24, 48, 72, and toneal injection were exposed at room temperature for 6, 12, 24, 96 h. Extraction of RNA and duplex RT-qPCR analysis were 48, 72, and 96 h. The ELF-1α Ct values increased from 14.3 for performed on each sample as described above. the control to 29.7 after 96 h of exposure. The ELF-1α Ct values Statistical analysis.—Sigma Stat for Windows (Jandel) was were significantly higher after 6, 12, 24, 48, 72, and 96 h than the used for statistical analysis. To assess multiple comparisons, a controls (Figure 4A). Similarly, VHSV Ct values were highly 2 parametric one-way ANOVA was performed on data. When the correlated (r = 0.995) with ELF-1α Ct values (Figure 4B). The distribution was not normal, a Kruskal–Wallis one-way ANOVA VHSV Ct values increased from 19.6 for the control to 25.6 on ranks was used. For more statistical stringency, significance after 96 h of exposure (Figure 4A). was defined at P < 0.01. Duplex Analysis An optimized ELF-1α–VHS duplex assay was compared RESULTS with the singleplex assay. The Ct values ranged from 24.1 to 33 for the singleplex assay and from 24.2 to 33.4 for the duplex Analytical Optimization assay (Figure 5). At the lowest concentration (102 PFU), only a A new assay was designed for the detection of VHSV 0.4 Ct value difference was recorded between the two assays. genogroup IV using optimized sets of primers and probe at the concentrations of 300 and 600 nM for forward and reverse primers, respectively, and 125 nM of the probe concentration. DISCUSSION The efficiency of the assay using a serial dilution of RNA ex- Due to its high sensitivity and specificity, RT-qPCR is be-

Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 tracted from EPC cell culture supernatant was evaluated at 98% coming one of the main diagnostic assays used to monitor ﬁsh (R2 = 0.998; data not shown). The sensitivity was assessed us- health (Purcell et al. 2011). Several studies have shown that RT- ing a 10-fold serial dilution of the single-strand DNA amplicon qPCR is more sensitive than cell culture and other molecular ranging from 2.9 × 106 to 0.29 copies. The data showed that the tests (Hope et al. 2010; Pierce et al. 2013). As a diagnostic test, sensitivity of the assay was estimated at three copies. No ampli- RT-qPCR should be performed because a high throughput test ﬁcation was recorded when IPNV and IHNV RNA were used as for rapid turnaround time results at a lower cost. To meet these templates, which demonstrated that there was no cross-reaction requirements, the integrity of the samples cannot be tested using with these viruses. traditional methods such as gel electrophoresis. Good quality of the samples is a necessity for accurate diagnostic tests required Sample Quality Control by accreditation processes such as ISO 17025 as issued by the Samples of HK tissues from Atlantic Salmon were exposed at International Organization for Standardization. Consequently, room temperature, 4◦C, −20◦C, −80◦C and in ethanol at −20◦C several studies have shown the importance of using an endoge- for 6, 12, 24, 48, and 72 h, respectively. Data showed an increase nous control such as ELF-1α mRNA in parallel with the targeted of Ct values for ELF-1α ranging from 14.7 to 26.5 cycles for pathogen nucleic acid as a quality control for the samples (In- tissues exposed to room temperature. Agarose gel electrophore- gerslev et al. 2006; Jorgensen et al. 2006; Snow et al. 2006; MULTIPLEX ASSAY TO MEASURE QUALITY OF RNA AND DETECT VHSV 177

FIGURE 1. Agarose gel electrophoresis of total RNA extracted from Atlantic Salmon HK stored at room temperature and at −80◦C for 0, 6, 12, 24, 48, and 72 h.

Christiansen et al. 2011; Bland et al. 2012). Recently, a new To ensure that ELF-1α could be used as an indicator of the StaRT-PCR approach, which incorporates several endogenous quality of Atlantic Salmon RNA in an RT-qPCR diagnostic test, controls (actin β1, 18S rRNA, and elongation factor 1 α), was HK tissue was exposed to different temperatures (room temper- published for accurate VHSV screening (Pierce et al. 2013). ature, 4◦C, −20◦C, −80◦C and in ethanol stored at −20◦C) for 6, 12, 24, 48, and 72 h). Data showed that ELF-1α Ct values increased signiﬁcantly from 14.7 to 19.4 after 12 h exposure at 30 room temperature (Figure 2). No differences of Ct values were -80C ◦

28 4C * recorded for the other storage temperatures although at 4 C, the * Room Temperature Ct value started to signiﬁcantly increase after 72 h of exposure. 26 -20C Ethanol (-20C) To mimic the real situation of storage conditions during 24 the shipping of the samples from the aquaculture farms to the

22 * laboratory, HK samples were stored under ﬁve different condi- Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 * α 20 tions. Data showed that ELF-1 Ct values increased when the HK samples were transported on ice for 2 h, frozen at −20◦C, Ct Values 18 thawed on ice for 2 or 6 h, and then stored at −80◦C (Figure 3). *

* ◦ 16 Based on these results, constant cold conditions (4 C or less) using ice for a short time or ice packs for long-term transporta- 14 tion could be considered to be adequate storage conditions 12 that would not affect the quality of the HK sample for RNA ◦ 10 analyses. In the laboratory, samples should be stored at −80 C, 0 6 12 24 48 72 and freeze–thaw cycles should be avoided. It is noteworthy that Exposure time (Hours) Ct values of ELF-1α remained the same at all the conditions when 70% ethanol was used as a preservative for HK samples α FIGURE 2. Kinetic profile of ELF-1 Ct values of Atlantic Salmon HK tissue (Figure 3). A commercially available preservative such as stored at −80◦C (solid circles), 4◦C (open circles), room temperature (solid triangles), −20◦C (open triangles), and ethanol (−20◦C) (solid squares) for 0, RNAlater was not used in this study due to its high cost and 6, 12, 24, 48, and 72 h. Markers represent mean values of six samples ± SE. inaccessibility to fish farmers. Alternatively, as it was shown Asterisk (*) corresponds to the significant differences (ANOVA: P < 0.01). in this study, 70% ethanol could be used as a preservative to 178 SIAH ET AL.

20 A A 35 * ELF 19 VHS 30 * * * 18 * * * * 25 * * * * *

Ct Values 17

20 Ct Values Ct 16

15 15 Group 1 Group 2 Group 3 Group 4 Group 5 0H 6H 12H 24H 48H 72H 96H 20 B Exposure time (Hours) B 19 27 2 R = 0.995 18 26

Ct Values 17 24

23 16 22 VHS Ct values VHS Ct 15 21 Group 1 Group 2 Group 3 Group 4 Group 5 20 FIGURE 3. Amplification data (Ct values) of ELF-1α from Atlantic Salmon 19 HK tissue stored in five different conditions with or without 70% ethanol: HK samples stored at −80◦C (group 1); transported in ice and then stored at −80◦C 12 14 16 18 20 22 24 26 28 30 32 (group 2); transported in ice, stored at −20◦C for 12 h, and then stored at −80◦C ELF 1 Ct values (group 3); transported in ice, stored at −20◦C for 12 h, thawed in ice for 2 h, andthenstoredat−80◦C (group 4); transported on ice, stored at −20◦Cfor FIGURE 4. (A) Kinetic pattern of ELF-1α and VHSV Ct values of HK tissues 12 h, thawed in ice for 6 h, and then stored at −80◦C (group 5) without (A) collected from Atlantic Salmon challenged with VHSV IVa (Isolate BC 99–292) preservative and (B) stored in 70% ethanol. Asterisk (*) corresponds to the through intraperitoneal injection and exposed at room temperature for 0, 6, 12, significant differences (ANOVA: P < 0.01). 24, 48, 72, and 96 h. (B) Correlation between ELF-α and VHSV Ct values. Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 protect tissue degradation during transportation, at least for short-term storage (up to 3 d). Head kidney tissues collected from Atlantic Salmon handling and storage conditions on viral RNA stability. While challenged with VHSV IVa (Isolate BC 99–292) through studies have reported an unstable viral RNA at different storage intraperitoneal injection were exposed at room temperature conditions (Cuypers et al. 1992), others showed a resistance for up to 96 h. Data showed that ELF-1α Ct values increased of viral RNA to degradation after 4 d (de Gerbehaye et al. significantly from 14.4 to 23.8 after 6 h of exposure at room 2002) and up to 30 d (Hasan et al. 2012) of storage at room temperature (Figure 4A). This drastic increase could be ex- temperature. Weesendorp et al. (2010) documented that viral plained by the presence of nucleases in the original sample. For RNA in tissues has a half-life time ranging from 0.95 to 2.55 VHSV, the Ct values increased significantly from 19.6 to 23.1. d. How the handling and storage can affect the integrity of viral Although ELF-1α and VHSV patterns were similar and highly RNA remains a subject for debate in the scope of diagnostic correlated (Figure 4B), degradation of ELF-1α mRNA was analyses. Consequently, guidelines have been implemented quicker than that for VHSV RNA (Figure 4A). In human viral by the Clinical and Laboratory Standards Institute (2005) for RNA diagnostic studies, conflicting data have been reported on handling and storage of samples for molecular testing. MULTIPLEX ASSAY TO MEASURE QUALITY OF RNA AND DETECT VHSV 179

40 use of ELF-1α mRNA as an endogenous quality control for At- lantic Salmon HK samples has increased during the last decade. 38 Our data showed that storage at 4◦Corlessusingiceorice 36 packs or placing tissues in 70% ethanol could constitute ade-

34 quate transportation conditions for short or long periods of time, respectively. Once at the laboratory, samples should be stored at 32 −80◦C for further RNA analyses. Cycles of freeze–thaw should

30 be avoided to limit tissue degradation. The duplex assay will

Ct Values Ct allow simultaneous evaluation of the quality of the tissue and 28 the screening of VHSV without affecting the sensitivity, the

26 cost of the analysis, and the turnaround time of the screening. Also, ELF-1α mRNA could represent a good indicator of RNA 24 quality; however, further studies should investigate the accept- 22 able range of ELF-1α mRNA among ﬁsh in a population and 1.0 2.0 3.0 4.0 5.0 6.0 7.0 at different life stages (juvenile, adult, freshwater, marine) and Log quantity of spiked VHSv (pfu) physiological states (dead, healthy, stressed).

FIGURE 5. Comparison of both duplex ELF-1α–VHSV (solid circles) and singleplex VHSV (open circles) Ct values performed on RNA extracted from ACKNOWLEDGMENTS HK tissue of Atlantic Salmon spiked with serially diluted VHSV supernatant The authors acknowledge the valuable technical support pro- 6 2 (10 –10 PFU). vided by Zina Richmond and Robert Johns from the British Columbia Centre for Aquatic Health Sciences. Both VHSV and The mechanisms by which RNA from aquatic infectious HK collected from Atlantic Salmon challenged with VHSV IVa viruses such as VHSV is degraded during the handling and (Isolate BC 99–292) through intraperitoneal injection were pro- storage conditions remain unclear. Due to this uncertainty, the vided by Kyle Garver from Fisheries and Oceans Canada at the possibility of false negatives if the quality of the tissue is poor Pacific Biological Station, Nanaimo, British Columbia. remains likely. Hence, the use of an endogenous control would provide a rapid analysis tool for RNA quality. This is supported by a recent study conducted by Bland et al. (2012) using ELF- REFERENCES 1α mRNA within the range of 17.76–25.97 as samples with Afonso, L. O. B., Z. Richmond, A. A. Eaves, J. Richard, L. M. Hawley, and K. A. Garver. 2012. Use of ultraviolet C (UVC) radiation to inactivate infec- acceptable quality for testing. When considering the use of en- tious hematopoietic necrosis virus (IHNV) and viral hemorrhagic septicemia dogenous genes as a quality control for tissue samples, the cost virus (VHSV) in fish processing plant effluent. Journal of Aquaculture Re- and the turnaround time of the analysis will increase dramati- search and Development [online serial] 3(1):1000120. cally, thus limiting their use. Consequently, combining both the Altinok, I., E. Capkin, and S. Kayis. 2008. Development of multiplex PCR endogenous gene and the pathogen nucleic acids in the same assay for simultaneous detection of five bacterial fish pathogens. Veterinary Microbiology 131:332–338. assay would provide an important development. Bland, F., R. McIntosh, N. Bain, and M. Snow. 2012. Development and vali- During the last decade, multiplex PCR-based assays were de- dation of a range of endogenous controls to support the implementation of veloped to detect multiple nucleic acids of aquatic fish pathogens practical taqman real-time PCR based surveillance for fish diseases within simultaneously (del Cerro et al. 2002; Altinok et al. 2008; Liu aquaculture. Journal of Fish Diseases 35:447–454. et al. 2008). The main challenge of these assays is to detect Burke, J. A., and D. Mulcahy. 1980. Plaquing procedure for infectious

Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 hematopoietic necrosis virus. Applied Environmental and Microbiology multiple RNA sequences without affecting the sensitivity of the 39:872–876. assay in comparison with the assays performed alone. In this Christiansen, D. H., P. S. Ostergaard, M. Snow, O. B. Dale, and K. Falk. 2011. study, a duplex RT-qPCR assay combining ELF-1α and VHSV A low-pathogenic variant of infectious salmon anemia virus (ISAV-HPR0) was developed for routine screening of VHSV in HK tissue is highly prevalent and causes a non-clinical transient infection in farmed from Atlantic Salmon. The duplex assay was optimized to get a Atlantic Salmon (Salmo salar L.) in the Faroe Islands. Journal of General Virology 92:909–918. similar sensitivity as the singleplex assay. Comparing both the Clinical and Laboratory Standards Institute. 2005. Collection, transport, prepa- single and duplex assays on serially diluted HK tissue spiked ration, and storage of specimens for molecular methods; approved guideline. with VHSV, data showed that both assays were similar and Clinical and Laboratory Standards Institute, MM13-A, Wayne, Pennsylvania. provided the same Ct values for all the samples (Figure 5). Cuypers, H. T., D. Bresters, I. N. Winkel, H. W. Reesink, A. J. Weiner, M. Houghton, C. L. van der Poel, and P. N. Lelie. 1992. Storage conditions of blood samples and primer selection affect the yield of cDNA polymerase CONCLUSION chain reaction products of hepatitis C virus. Journal of Clinical Microbiology 30:3220–3224. Assessing the quality of the tissue is a key component for De Gerbehaye, A. I. M., M. Bodeus, A. Robert, Y. Horsmans, and P. Goubau. the interpretation of RT-qPCR data, and consequently quality 2002. Stable hepatitis C virus RNA detection by RT-PCR during four days control features should be implemented into the procedure. The storage. BMC Infectious Diseases 2:1–6. 180 SIAH ET AL.

del Cerro, A., I. Marquez, and J. A. Guijarro. 2002. Simultaneous detection of Liu, Z., Y. Teng, H. Liu, Y. X. Jiang, H. Xie, Li, J. Lv, L. Gao, J. He, X. Aeromonas salmonicida, Flavobacterium psychrophilum,andYersinia ruck- Shi, F. Tian, J. Yang, and C. Xie. 2008. Simultaneous detection of three fish eri, three major fish pathogens, by multiplex PCR. Applied and Environmental rhabdoviruses using multiplex real-time quantitative RT-PC assay. Journal of Microbiology 68:5177–5180. Virological Methods 149:103–109. Duesund, H., S. Nylund, K. Watanabe, K. F. Ottem, and A. Nylund. 2010. Matejusova, I., P. McKay, F. Bland, and M. Snow. 2010. Application of a Characterization of a VHS virus genotype III isolated from Rainbow Trout sensitive, specific and controlled real-time PCR assay to surveillance indicates (Oncorhychus mykiss) at a marine site on the west coast of Norway. Virology a low prevalence of viral haemorrhagic septicaemia virus (VHSV) in wild Journal [online serial] 7:article (19). herring, Clupea harengus L., in Scottish waters. Journal of Fish Diseases Elsayed, E., M. Faisal, M. Thomas, G. Whelan, W. Batts, and J. Winton. 2006. 33:841–847. Isolation of viral haemorrhagic septicaemia virus from Muskellunge, Esox Olsvik, P. A., K. K. Lie, A. E. Jordal, T. O. Nilsen, and I. Hordvik. 2005. masquinongy (Mitchill), in Lake St Clair, Michigan, USA reveals a new Evaluation of potential reference genes in real-time rt-PCR studies of Atlantic sublineage of the North American genotype. Journal of Fish Diseases 29:611– Salmon. BMC Molecular Biology [online serial] 6:21. 619. Olsvik, P. A., L. Softeland, and K. K. Lie. 2008. Selection of reference genes for Fersht, A. R. 1977. Enzyme structure and mechanism. Freeman, Reading, qRT-PCR examination of wild populations of Atlantic Cod Gadus morhua. Pennsylvania. BMC Research Notes [online serial] 1:47. Garver, K. A., L. M. Hawley, C. A. McClure, T. Schroeder, S. Aldous, F. Orpetveit, I., A. B. Mikalsen, H. Sindre, O. Evensen, B. H. Dannevig, and Doig, M. Snow, S. Edes, C. Baynes, and J. Richard. 2011. Development and P. J. Midtlyng. 2010. Detection of infectious pancreatic necrosis virus in validation of a reverse transcription quantitative pcr for universal detection of subclinically infected Atlantic Salmon by virus isolation in cell culture or viral hemorrhagic septicemia virus. Diseases of Aquatic Organisms 95:97– real-time reverse transcription polymerase chain reaction: influence of sam- 112. ple preservation and storage. Journal of Veterinary Diagnostic Investigation Halfon, P., H. Khiri, V. Gerolami, M. Bourliere, J. M. Feryn, P. Reynier, A. 22:886–895. Gauthier, and G. Cartouzou. 1996. Impact of various handling and storage Pazzagli, M., F. Malentacchi, L. Simi, C. Orlando, R. Wyrich, K. Gunther,¨ C. C. conditions on quantitative detection of hepatitis C virus RNA. Journal of Hartmann, P. Verderio, S. Pizzamiglio, C. M. Ciniselli, A. Tichopad, M. Ku- Hepatology 25:307–311. bista, and S. Gelmini. 2013. SPIDIA-RNA: first external quality assessment Hasan, M. R., R. Tan, G. N. Al-Rawahi, E. Thomas, and P. Tilley. 2012. Short- for the pre-analytical phase of blood samples used for RNA based analyses. term stability of pathogen-specific nucleic acid targets in clinical samples. Methods 59:20–31. Journal of Clinical Microbiology 50:4147–4150. Pierce, L. R., J. C. Willey, E. L. Crawford, V. V. Palsule, D. W. Leaman, M. Hope, K. M., N. R. N. Casey, G. H. Groocock, R. G. Getchell, P. R. Faisal, R. K. Kim, B. S. Shepherd, L. M. Stanoszek, and C. A. Stepien. 2013. Bowser, and J. W. Casey. 2010. Comparison of quantitative RT-PCR with A new StaRT-PCR approach to detect and quantify fish viral hemorrhagic cell culture to detect viral hemorrhagic septicemia virus (VHSV) IVb in- septicemia virus (VHSV): enhanced quality control with internal standards. fections in the Great Lakes. Journal of Aquatic Animal Health 22:50– Joural of Virological Methods 189:129–142. 61. Purcell, M. K., R. G. Getchell, C. A. McClure, and K. A. Garver. 2011. Ingerslev, H. C., E. F. Pettersen, R. A. Jakobsen, C. B. Petersen, and H. I. Werge- Quantitative polymerase chain reaction (PCR) for detection of aquatic an- land. 2006. Expression profiling and validation of reference gene candidates imal pathogens in a diagnostic laboratory setting. Journal of Aquatic Animal in immune relevant tissues and cells from Atlantic Salmon (Salmo salar L.). Health 23:148–161. Molecular Immunology 43:1194–1201. Radonic,´ A., S. Thulke, I. M. Mackay, O. Landt, W. Siegert, and A. Nitsche. Jonstrup, S. P., S. Kahns, H. F. Skall, T. S. Boutrup, and N. J. Olesen. 2013. 2004. Guideline to reference gene selection for quantitative real-time PCR. Development and validation of a novel taqman-based real-time rt-pcr as- Biochemical and Biophysical Research Communications 313:856–862. say suitable for demonstrating freedom from viral haemorrhagic septicaemia Schutze, H., E. Mundt, and T. Mettenletter. 1999. Complete genomic sequence virus. Journal of Fish Diseases 36:9–23. of viral hemorrhagic septicemia virus, a fish rhabdovirus. Virus Genes 19:59– Jorgensen, S. M., E. J. Kleveland, U. Grimholt, and T. Gjoen. 2006. Validation 65. of reference genes for real-time polymerase chain reaction studies in Atlantic Skall, H. F., N. J. Olesen, and S. Mellergaard. 2005. Viral haemorrhagic septi- Salmon. Marine Biotechnology 8:398–408. caemia virus in marine fish and its implications for fish farming—a review. Jose,´ M., S. Curtu, R. Gajardo, and J. I. Jorquera. 2003. The effect of storage Journal of Fish Diseases 28:509–529. at different temperatures on the stability of Hepatitis C virus RNA in plasma Snow, M., P. McKay, A. J. McBeath, J. Black, F. Doig, R. Kerr, C. O. Cun- samples. Biologicals 31:1–8. ningham, A. Nylund, and M. Devoid. 2006. Development, application and Jose,´ M., S. Curtu, R. Gajardo, and J. I. Jorquera. 2005. Stability of HCV, validation of a taqman real-time RT-PCR assay for the detection of infectious Downloaded by [Department Of Fisheries] at 23:55 16 November 2014 HIV-1 and HBV nucleic acids in plasma samples under long-term storage. salmon anaemia virus (ISAV) in Atlantic Salmon (Salmo salar). Develop- Biologicals 33:9–16. ments in Biologicals 126:133–145. Lorenzen, E., B. Carstensen, and N. Olse. 1999. Inter-laboratory comparison of Weesendorp, E., E. M. Willems, and W. L. Loeffen. 2010. The effect of tis- cell lines for susceptibility to three viruses: VHSV,IHNV and IPNV. Diseases sue degradation on detection of infectious virus and viral RNA to diagnose of Aquatic Organisms 37:81–88. classical swine fever virus. Veterinary Microbiology 141:275–281. This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:56 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Histopathologic Changes in Disease-Resistant-Line and Disease-Susceptible-Line Juvenile Rainbow Trout Experimentally Infected with Flavobacterium psychrophilum David P. Marancika, Timothy D. Leedsa & Gregory D. Wiensa a National Center for Cool and Cold Water Aquaculture, 11861 Leetown Road, Kearneysville, West Virginia 25430, USA Published online: 07 Aug 2014.

To cite this article: David P. Marancik, Timothy D. Leeds & Gregory D. Wiens (2014) Histopathologic Changes in Disease- Resistant-Line and Disease-Susceptible-Line Juvenile Rainbow Trout Experimentally Infected with Flavobacterium psychrophilum, Journal of Aquatic Animal Health, 26:3, 181-189, DOI: 10.1080/08997659.2014.920735 To link to this article: http://dx.doi.org/10.1080/08997659.2014.920735

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:181–189, 2014 American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.920735

ARTICLE

Histopathologic Changes in Disease-Resistant-Line and Disease-Susceptible-Line Juvenile Rainbow Trout Experimentally Infected with Flavobacterium psychrophilum

David P. Marancik,* Timothy D. Leeds, and Gregory D. Wiens* National Center for Cool and Cold Water Aquaculture, 11861 Leetown Road, Kearneysville, West Virginia 25430, USA

Abstract A challenge to selectively breeding fish populations for improved disease resistance in aquaculture is an incomplete understanding of how artificial selection affects innate immunity at the host–pathogen level. The objective of this study was to determine whether Rainbow Trout Oncorhynchus mykiss bred for differential susceptibility to bacterial cold- water disease exhibited altered tissue damage and cellular inflammatory response following experimental challenge with Flavobacterium psychrophilum. Fish from disease-resistant (ARS-Fp-R) and disease-susceptible (ARS-Fp-S) lines were experimentally challenged as juveniles, and mortalities, as well as survivors, were sampled for histopathology during the acute phase of the disease. Microscopic lesions were quantified or semiquantified and statistically compared for changes over time and between genetic lines. Significant progression in the degree of perisplenitis, splenic necrosis, splenic inflammatory infiltrates, average splenic ellipsoid area, total splenic ellipsoid area, and peritonitis was present over time in both genetic lines on at least one postinfection time point. No differences were found between renal inflammatory infiltrates and renal hematopoietic cell depletion over time. Perisplenitis was significantly lower in fish from the ARS-Fp-R line on day 9 postinfection than in fish from the ARS-Fp-S line. The ARS-Fp-R line demonstrated a trend towards reduced splenic necrosis compared with the ARS-Fp-S line that approached significance, and fish from the ARS-Fp-S line were 3.6 times more likely than fish from the ARS-Fp-R line to have a higher splenic necrosis lesion score after day 3 postinfection. These findings support the hypothesis that differential survival is a result of divergence in disease magnitude and not altered disease course between genetic lines. Characterization of histopathologic changes between genetic lines and over time helps elucidate mechanisms of disease resistance and contributes to our understanding of disease pathogenesis in fish infected with F. psychrophilum.

Infectious disease can substantially limit on-farm production Khoo et al. 2004; Baerwald et al. 2011; Verrier et al. 2013). Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 of Rainbow Trout Oncorhynchus mykiss due to direct mortal- These studies highlight that fish disease resistance has a strong ity and reduced fitness of fish that survive infection (National genetic component and that there is potential for exploiting the Agricultural Statistics Service 2012). An approach to reducing naturally occurring variation in selective-breeding programs. fish loss is to select and propagate disease-resistant fish through A current challenge to producing disease-resistant fish lines is family-based breeding programs (Kjøglum et al. 2008; Gjedrem an incomplete understanding of how artificial selection affects 2010). Previous studies in Rainbow Trout have demonstrated ge- innate immunity and host–pathogen interactions at the mecha- netic variation to a number of different disease etiologies based nistic level. on controlled laboratory challenge of full-sib families (Dorson A pedigreed Rainbow Trout line, designated ARS-Fp-R, et al. 1995; Hedrick et al. 2003; Henryon et al. 2005; Over- has been selectively bred at the National Center for Cool and turf et al. 2010). Additionally, numerous quantitative trait loci Cold Water Aquaculture (NCCCWA; Kearneysville, West Vir- have been associated with disease resistance (Ozaki et al. 2001; ginia) that demonstrates relative increased survival following

*Corresponding authors: [email protected]; [email protected] Received January 7, 2014; accepted March 10, 2014 181 182 MARANCIK ET AL.

experimental (Hadidi et al. 2008; Silverstein et al. 2009; Leeds pool consisted of 43 full-sib families (mean = 1,489 eggs per et al. 2010) and natural (Wiens et al. 2013a) challenge with family) and the ARS-Fp-S pool consisted of 11 full-sib families Flavobacterium psychrophilum, the causative agent of Rainbow (mean = 393 eggs per family). All broodstock and fish used in Trout fry syndrome (RTFS) and bacterial cold-water disease this study were certified to be free of common salmonid bacterial (BCWD). It is hypothesized that artificial selection has aug- and viral pathogens by two independent diagnostic laboratories mented innate resistance to F. psychrophilum in the ARS-Fp-R as described previously (Leeds et al. 2010; Wiens et al. 2013a). line. Through comparison to a reference susceptible line, des- Bacterial isolate.—Flavobacterium psychrophilum CSF ignated the ARS-Fp-S line, intrinsic factors that differentially 259–93 was previously collected from a BCWD field case and affect survival of the ARS-Fp-R line can be explored. maintained at −80◦C in TYES (trypticase yeast extract agar + One limitation in describing host response to F. psy- skim milk) media supplemented with 10% (volume per volume) chrophilum is an incomplete understanding of disease patho- glycerol. Frozen stock was cultivated on TYES media for 5 days genesis in infected fish. Lesions described in moribund and at 15◦C prior to experimental challenge. dead juvenile Rainbow Trout include exophthalmia, peritonitis, Experimental design.—A total of 239 fish from the ARS- perisplenitis, and nephritis (Nematollahi et al. 2003; Ferguson Fp-R line and 240 fish from the ARS-Fp-S line were randomly 2006; Nilsen et al. 2011), suggesting a septicemic-type disease. assigned to tanks of 40 fish/tank supplied with 2.4 L/min of 13◦C Prevention and treatment of RTFS and BCWD would benefit flow-through spring water, hand fed daily with a commercial from a more thorough characterization of the earlier stages of pellet feed, and allowed a 7-day acclimation period prior to disease and the precursory events leading to mortality. experimental challenge. The mean ± SD starting weights of fish We hypothesized that higher survival in the ARS-Fp-R line from the ARS-Fp-R and ARS-Fp-S lines were 7.6 ± 4.3 g and may be due to differences in tissue damage or cellular inflamma- 7.3 ± 3.8 g, respectively. As expected, all fish were confirmed tory response secondary to bacterial infection. To test this hy- to be female at necropsy. pothesis, microscopic changes were described for each genetic Fish were anesthetized with 100 mg/L tricaine methanesul- line and quantified over time following experimental challenge. fonate (Tricaine-S; Western Chemical, Ferndale, Washington) These results propose mechanisms responsible for differential and challenged intraperitoneally with 1.97 × 107 CFU/fish survival between genetic lines and elucidate host–pathogen in- of F. psychrophilum (strain CSF 259–93) in 25 µl of chilled teractions occurring during the progression of disease from sub- phosphate-buffered saline (PBS). Eighty fish from each genetic clinical infection to mortality. line were injected with 25 µl of chilled PBS to be used as noninfected controls. Injections were completed using a re- peater pipette (Eppendorf, Hauppauge, New York) and 26- METHODS gauge, 12.7-mm needles. Mortality was observed daily in all Experimental animals.—All fish were maintained at the NC- tanks. One adjacent tank of 40 challenged fish and one tank CCWA following Standard Operating Procedures for the Care of 40 PBS-injected fish from each genetic line were reserved and Use of Research Animals (Rainbow Trout). The NCC- for histopathology sampling and not included in the mortality CWA Institutional Animal Care and Use Committee approved assessment. this study (Protocol #76). The ARS-Fp-R and ARS-Fp-S ge- In the histopathology sampling tanks, on day 0, five fish from netic lines were derived from the same founder population that the ARS-Fp-R line and five from the ARS-Fp-S line were eu- was developed in 2005 and thus differed only as a result of thanized with 250 mg/L Tricaine-S prior to experimental chal- artificial selection for BCWD postchallenge survival (Wiens lenge to serve as baseline controls. On days 1, 3, 6, and 9 et al. 2013a). Rainbow Trout broodstock from each line were postchallenge, five infected fish and two PBS-injected fish from

Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 spawned during a 5-week period in February and March 2012 at each genetic line were sampled (Figure 1). Four fish from the the NCCCWA. Single-sire × single-dam matings were made ARS-Fp-R line and four fish from the ARS-Fp-S line were within each genetic line between 3-year-old females and 1-year- found dead between days 6 and 18 postchallenge with min- old neomales, and eggs were incubated separately by full-sib imal postmortem autolysis and were additionally sampled for family in upwelling jars. The neomales had been previously histopathology. Sampled fish were fixed in 10% neutral buffered given an intraperitoneal (IP) injection three times per week formalin (Fisher-Scientific) for 72 h prior to being placed in de- with 1.5 mg salmonid pituitary extract (Argent Laboratories, calcification solution (Cal-Rite; Thermo Scientific, Waltham, Redmond, Washington) per kilogram of body weight beginning Massachusetts) for 8 h. Each fish was serially cross-sectioned approximately 3 months before spawning to induce sexual mat- using anatomical locations as a guide to minimize variability in uration at 1 year of age. Water temperature in the incubation tissue sectioning between fish. Cross sections were made at the jars was manipulated so that all families would hatch within caudal aspect of the eye, the dorsocranial insertion of the op- a 1-week period (Leeds et al. 2010). Eggs were pooled within erculum, the insertion of the pectoral fin, the midsection of the line at the eyed stage and reared in flow-through spring water liver, the caudal end of the liver, the spleen, the cranial aspect (approximately 12.5◦C), and the resulting fish were fed a com- of dorsal fin, the vent, and the caudal peduncle. Tissues were mercial diet (Zeigler Brothers, Gardners, Pennsylvania) from processed routinely, embedded in paraffin, sectioned at 5 µm, swim-up until the beginning of the experiments. The ARS-Fp-R and stained with hematoxylin and eosin. HISTOPATHOLOGIC CHANGES IN TROUT INFECTED WITH FLAVOBACTERIUM PSYCHROPHILUM 183

Perisplenitis, average splenic ellipsoid area, total splenic ellipsoid area, and renal hematopoietic cell depletion were quantified using Nikon NIS-Elements imaging software (Tokyo, Japan). Perisplenitis was defined as deposition and disruption of the splenic capsule with fibrin and inflammatory cells and calculated by averaging the widest pixel diameter from all four splenic borders at 20 × magnification. The average splenic ellipsoid area was calculated for all ellipsoids present in five randomly selected 20 × fields (2 million pixels each). Total splenic ellipsoid area was quantified by measuring the sum of the ellipsoid area present in five randomly selected 20 × fields (2 million pixels each). All quantification pertaining to renal pathology was performed in the cranial kidney at the aspect of the pectoral fin. Quantification of renal hematopoietic cell depletion was performed based on color threshold and average area measurement of five randomly chosen 40 × fields. Each field was manually examined to ensure exclusion of nonhematopoietic cells, such FIGURE 1. Assessment of differential survival between Rainbow Trout from as red blood cells and mature leukocytes. = = the ARS-Fp-R (n 198) and ARS-Fp-S (n 200) lines experimentally chal- Splenic necrosis, splenic inflammatory infiltrates, renal in- lenged with Flavobacterium psychrophilum. Control fish were injected with PBS alone (n = 40 per genetic line), and no mortality was observed. Arrows flammatory infiltrates, and peritonitis were measured in each denote days that fish were sampled for histopathology. [Color figure available fish using semiquantitative methods under blinded conditions. online] Grading was performed by examining a single histologic tissue section under 4 × and 100 × power. Tissue alterations were Fish from the ARS-Fp-R line and ARS-Fp-S line were ex- scored as 0 (no identifiable lesions), 1 (minimal change), 2 amined under light microscopy. Notable changes were observed (moderate change), and 3 (severe change) (Table 1). in the spleen, kidney, and peritoneal cavity of infected fish from Statistical analysis.—GraphPad Prism version 5.0 (Graph- both genetic lines; thus, disease parameters in these organs were Pad Software, La Jolla, California) was used to analyze survival chosen for lesion quantification. data (time to event) and mean day to death. Survival curve

TABLE 1. Semiquantitative grading methods for splenic necrosis, splenic inflammatory infiltrates, renal inflammatory infiltrates, and peritonitis.

Lesion score Histopathologic change 0 = absent 1 = minimal change 2 = moderate change 3 = severe change Splenic necrosis No or questionable Scattered to multifocal Multifocal to coalescing areas of Multifocal, coalescing, or lesions areas of pyknosis pyknosis, karyorrhexis, and diffuse areas of pyknosis, and karyorrhexis deposition of cellular debris karyorrhexis, and cellular affecting <10% of affecting 10–50% of the tissue debris with loss of tissue

Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 the tissue parenchyma affecting >50% of the tissue Splenic and No or questionable Inflammatory cell Inflammatory cells demonstrate Inflammatory cells are renal lesions infiltrates are rare or multifocal, perivascular present within most inflammatory have scattered, distribution affecting 10–50% perivascular and infiltrates perivascular of vessels with infiltration into interstitial spaces distribution affecting <10% of the interstitium affecting >50% of vessels <10% of tissue or interstitial spaces Peritonitis No or questionable Vascular congestion, Vascular congestion and edema, Vascular congestion, edema, lesions edema, and rare areas of hemorrhage, and frequent areas of inflammatory cell perivascular distribution of hemorrhage, and fibrin infiltrates within fibrin and inflammatory cells and inflammatory cell adipose tissue and affecting 10–50% of vessels infiltration affecting free peritoneal space and inflammatory cells with >50% of perivascular and affecting <10% of infiltration into <10% of the interstitial spaces tissue or space interstitium 184 MARANCIK ET AL.

TABLE 2. Number of ﬁsh from the ARS-Fp-R (R) and ARS-Fp-S (S) lines in which histopathologic changes were observed during the acute phase of bacterial cold-water disease.

Day postinfection Day 0 Day 1 Day 3 Day 6 Day 9 Mortalities Disease sign R, Sa R, S R, S R, S R, S R, Sb Quantified data Increase in splenic ellipsoid prominence (total 0, 0 0, 1 5, 5 5, 5 5, 5 3, 4 and average width) Perisplenitis 0, 0 0, 0 0, 0 1, 3 2, 3 4, 4 Renal hematopoietic cell depletion 0, 0 0, 0 0, 0 0, 0 2, 1 1, 2 Semiquantified data Peritonitis 0, 0 1, 1 3, 3 3, 4 4, 2 4, 4 Splenic necrosis 0, 0 0, 0 0, 0 1, 3 4, 5 4, 4 Splenic inflammatory infiltrates 0, 0 1, 0 0, 1 3, 1 5, 4 4, 4 Renal inflammatory infiltrates 0, 0 0, 0 0, 0 2, 2 3, 3 3, 3 Quantification not performed Increased prominence of endothelial 0, 1 0, 0 3, 4 4, 4 5, 5 4, 4 macrophages Myocarditis 0, 0 0, 0 0, 0 0, 1 0, 0 0, 0 Splenic congestion 2, 1 3, 2 3, 4 3, 4 5, 5 3, 2 Injection site inflammation 0, 0 0, 0 0, 2 3, 3 3, 4 4, 3

aFive ﬁsh from each genetic line were sampled on days 0, 1, 3, 6, and 9 postinfection. Table numbers indicate the number of affected ﬁsh per line. bA total of four mortalities from each genetic line were examined.

comparison between the ARS-Fp-R and ARS-Fp-S lines was ARS-Fp-S lines and varied only in lesion severity and preva- performed using a log-rank test. For quantified histopathologic lence. Lesion descriptions were qualitatively described for each data, the significance of sample day and genetic line was an- sample day, and the observed frequency of quantified and semi- alyzed using a one-way analysis of variance (ANOVA). Semi- quantified lesions and lesions that could not be adequately quan- quantified histopathologic data were analyzed using PROC LO- tified are summarized in Table 2. GISTIC in SAS (SAS 9.3; SAS Institute, Cary, North Carolina) Perisplenitis, average ellipsoid area, and total ellipsoid area to test for effects of the day postchallenge, the genetic line, and (Figure 2), but not renal hematopoietic cell depletion (data not the 2-way interaction of these effects on lesion score severity. shown), demonstrated a significant difference between control When applicable, an odds ratio estimate was used to demon- fish and infected fish on at least one sample day (P < 0.05). strate the likelihood that a lesion score would change between Perisplenitis was significantly greater in the ARS-Fp-S line then sample days within each genetic line. All statistical analyses the ARS-Fp-R line on day 9 postinfection (P < 0.05) (Fig- were run with a significance level of P < 0.05. ure 2A). No significant difference in average or total ellipsoid area was present between genetic lines (Figure 2B–C). For semi-

Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 quantified lesions, the day postchallenge significantly affected RESULTS lesion score severity for splenic necrosis, splenic inflammatory Mortality Study infiltrates, and peritonitis (P < 0.05; Figure 3) but not renal The mean ± SD percent survival of the ARS-Fp-R line at inflammatory infiltrates (data not shown). The effect of genetic 82.9 ± 4.1% (n = 198) was significantly greater than that of the line on lesion score was not significant for any semiquantified ARS-Fp-S line at 60.8 ± 6.5% (n = 200) (P < 0.01; Figure 1). parameter, although splenic necrosis was approaching signifi- The mean ± SD day to death was 6.9 ± 3.7 d for the ARS-Fp-R cance (P = 0.10). Based on the odds ratio that describes relative line and 10.4 ± 4.3 d for the ARS-Fp-S line (P < 0.0001). risk, the ARS-Fp-S line was 3.6 times more likely than the ARS-Fp-R line to have a more severe splenic necrosis lesion Histopathology Study score after day 3 postinfection (95% confidence interval = 0.8– The PBS-injected fish and the fish sampled on day 0 demon- 16.5). strated no significant differences in quantitative or semiquan- Histopathologic changes in challenged fish on day 1.— titative lesion measurements and thus were pooled together to Microscopic changes in fish at 24 h postchallenge were con- serve as a single control population. fined to the free peritoneal spaces and were consistent with IP General histopathologic characteristics were similar between injection of bacteria and PBS. In all fish examined, the free F. psychrophilum-challenged fish from both the ARS-Fp-R and space between visceral organs and the visceral adipose tissue HISTOPATHOLOGIC CHANGES IN TROUT INFECTED WITH FLAVOBACTERIUM PSYCHROPHILUM 185 Downloaded by [Department Of Fisheries] at 23:56 16 November 2014

FIGURE 2. Histopathologic lesions quantified using image-analysis software through the acute phase of disease in Rainbow Trout from ARS-Fp-R (R) and FIGURE 3. Semiquantified histopathologic lesions through the acute phase ARS-Fp-S (S) lines challenged with Flavobacterium psychrophilum. Software of disease in Rainbow Trout from ARS-Fp-R (R) and ARS-Fp-S (S) lines chal- demonstrated statistical differences between sample days for (A) perisplenitis, lenged with Flavobacterium psychrophilum. Analysis of (A) splenic necrosis, < (B) average ellipsoid area, and (C) total ellipsoid area (P 0.05). Sample days (B) splenic inflammatory infiltrates, and (C) peritonitis demonstrated a signifi- with significant differences in lesion measurement are labeled with differing cant effect of day (P < 0.05) on lesion score. The effect of genetic line was not letters. Perisplenitis was significantly greater in the ARS-Fp-S line than the significant, although it approached significance for splenic necrosis (P = 0.10). < ARS-Fp-R line on day 9 postinfection (P 0.05). Error bars represent standard Error bars represent standard deviation. [Color figure available online] deviation. [Color figure available online] 186 MARANCIK ET AL.

was filled with variable amounts of homogenous basophilic material admixed with scores of filamentous bacterial rods. In one fish from each genetic line, bacteria and basophilic material were associated with scattered to moderate numbers of degen- erate and nondegenerate neutrophils and indistinct mononuclear cells (Table 2). However, inflammatory infiltrates were largely absent in most fish examined on day 1 postinfection and quantified changes showed no significant differences between infected fish within each genetic line and control samples. Histopathologic changes in challenged fish on day 3.—There was evidence of reticuloendothelial system stimulation characterized by increased prominence of endothelial lining histiocytes and splenic ellipsoids. Compared to control fish (Fig- ure 4A–B), there was increased prominence of splenic ellipsoids (Figure 4C–D) with significant increase in average and total splenic ellipsoid area in fish from both ARS-Fp-R and ARS- Fp-S lines, but no significant differences were detected between lines (Figure 2B–C). Mild splenic inflammation was present in one ARS-Fp-S fish with scattered epithelioid macrophages and lymphocytes within vascular channels and splenic parenchyma (Table 2). Scores of slender, filamentous bacteria were evident within the peritoneal cavity, but basophilic material was no longer present. Bacteria often adhered to serosal surfaces of organs and the peritoneum, but no infiltration of tissue was observed. Mild levels of peritonitis were identified in 6 out of 10 fish (Ta- ble 2). Bacteria were associated with small rafts of macrophages, lymphocytes, and scattered neutrophils. Histopathologic changes in challenged fish on day 6.— Ellipsoids remained prominent within the spleen, and there was diffuse vascular congestion. Both genetic lines exhibited a significant increase in total splenic ellipsoid area, but not average FIGURE 4. Photomicrograph of spleen samples from (A–B) control fish and ellipsoid area, compared to day 3 samples, indicating an increase (C–F) fish experimentally challenged with Flavobacteriump sychrophilum.Nor- mal splenic architecture is shown (A) with intermixed white pulp (†)andred in the number of visible reticuloendothelial centers (Figure 2B– pulp (*; 4×) and (B) with an unstimulated splenic ellipsoid. Experimentally C). Epithelioid macrophages and few neutrophils were present challenged fish demonstrate prominent splenic ellipsoids at (C) low magnifica- within the splenic parenchyma and vascular channels of three tion and (D) high magnification (4× and 20×, respectively), (E) perisplenitis fish from the ARS-Fp-R line and one fish from the ARS-Fp-S with abundant eosinophilic fibrin and mononuclear cells (20×), and (F) splenic × line (Table 2), and there was mild perisplenitis characterized by necrosis with loss of red and white pulp and expansive clear spaces (4 ). The insert shows infiltration of splenic vasculature spaces and tissue by epithelioid expansion and disruption of the splenic capsule with fibrin and macrophages (arrowheads; 100×). Quantification in (B), (D) and (E) was per- Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 low numbers of macrophages (Figure 4E). There were signs of formed using Nikon image-analysis software. [Color figure available online] mild to moderate splenic necrosis (Figure 3A) characterized by pyknosis and karyorrhexis of hematopoietic tissue and deposition of cellular debris in one fish from the ARS-Fp-R line and hemorrhage obscuring visceral adipose tissue (Figure 3C). In- three from the ARS-Fp-S line (Table 2). flammation rarely extended into pancreatic acinar cells. Bacteria Two fish from each genetic line also had renal inflamma- were visible within phagocytes and within visceral fat. tory infiltrates characterized by low numbers of epithelioid Histopathologic changes in challenged fish on day 9.—There macrophages and lymphocytes within vascular channels and was substantial progression of disease in the spleen. Perispleni- hematopoietic tissue of the head kidney (Table 2). Additional tis was mild to severe with deposition of fibrin and low num- findings included sustained prominence of histiocytes lining the bers of macrophages and lymphocytes. Perisplenitis was signif- cardiac endothelium, aggregates of lymphocytes within the epi- icantly increased in the fish from the ARS-Fp-S line compared cardium of one ARS-Fp-S fish, and granulomatous inflamma- to those from the ARS-Fp-R line (Figure 2A). Splenic necrosis tion in the dermis, subdermis, and skeletal muscle at the injec- was mild to severe with variable loss of white pulp, frequent py- tion site of six fish. Peritonitis ranged from absent to severe with knosis and karyorrhexis of hematopoietic cells, and deposition neutrophils, macrophages, lymphocytes, and scattered areas of of basophilic cellular debris. Inflammatory infiltrates included HISTOPATHOLOGIC CHANGES IN TROUT INFECTED WITH FLAVOBACTERIUM PSYCHROPHILUM 187

numerous large epithelioid macrophages, neutrophils, and small phenotype has been demonstrated to be relatively stable between lymphocytes filling the lumen of arterioles and veins and infil- laboratory and field challenges and across a variety of body trating into the splenic parenchyma. weights, ranging from 0.2 g to over 800 g (Hadidi et al. 2008; Renal hematopoietic tissue inflammatory infiltrates were ab- Wiens et al. 2013a). Although the ARS-Fp-R line had a more sent to moderate in both ARS-Fp-R and ARS-Fp-S fish, with acute mean day to death than the ARS-Fp-S line, this did not scattered macrophages, neutrophils, and small lymphocytes. appear to result in earlier manifestation of microscopic lesions. Cords of renal hematopoietic tissue were sparsely populated This observed difference in mean day to death was not repeatable in two ARS-Fp-R fish and one ARS-Fp-S fish (Table 2) and dis- in subsequent laboratory challenges in the same year-class when played scattered cellular pyknosis and karyorrhexis, but quantifi- fish were 40 g and 300 g (D. P. Marancik and G. D. Wiens, cation of hematopoietic tissue for each population was not sig- unpublished) or in prior year-class studies, including fish of nificantly different than that of control fish (results not shown). similar weight (Hadidi et al. 2008; Leeds et al. 2010; Wiens et al. Peritonitis ranged from absent to severe with characteristics 2013a). The reason behind this change in mortality kinetics is similar to those described for day 6 (Figure 3C). unknown but appears to be related to this particular experiment Histopathologic changes in mortalities.—Moderate to severe and is not a fundamental feature of the survival phenotype. splenic necrosis and perisplenitis was present in all fish mortali- The most apparent and consistent histopathologic changes ties that were examined (Table 2). Additionally, there was exten- occurred in the spleen of challenged fish. The ARS-Fp-R line sive congestion, deposition of fibrin, and variable hemorrhage, had a lower degree of perisplenitis and an observed trend to- and the parenchyma was expanded by large, clear spaces inter- wards reduced splenic necrosis that approached significance. preted to be dilated vascular channels (Figure 4F). Inflammatory Perisplenitis has been described in both experimentally and nat- infiltrates of lymphocytes and macrophages were quantified as urally infected moribund and dead Rainbow Trout and has been moderate in all fish examined. Quantified total ellipsoid area proposed as being pathognomonic for F. psychrophilum infec- was decreased compared to day 9 samples, but there was no tion (Rangdale et al. 1999; Nilsen et al. 2011). The reason behind change in average ellipsoid area (Figure 2B–C). Splenic necro- the relatively high prevalence of perisplenitis in juvenile Rain- sis and total ellipsoid area were negatively correlated (Pearson’s bow Trout is not well understood. In this study, the early stages r =−0.57). of perisplenitis began with the deposition of fibrin and few in- Fish had absent to moderate infiltration of the renal inter- flammatory cells within the splenic capsule. Over time, there stitium and vascular spaces with epithelioid macrophages and was progressive capsular expansion leading to the disruption of small lymphocytes. Cords of hematopoietic tissue within the the capsule and the infiltration of fibrin and inflammatory cells cranial kidney of three ARS-Fp-R and three ARS-Fp-S fish into the outer splenic parenchyma. There was no correlation be- appeared depleted and displayed necrosis, but quantification of tween the degree of perisplenitis and the splenic inflammatory the complete dataset revealed no significant differences between infiltrate lesion score, suggesting these disease processes are either line or compared with control fish. occurring independently. Further investigation into why F. psy- Mild to severe peritonitis was present in all fish mortalities chrophilum infection appears to have an affinity for the splenic that were examined (Table 2). Inflammatory cell type was con- capsule is warranted. sistent with that previously described for days 6 and 9. Inflam- The association of splenic changes with survival phenotype mation often infiltrated and obscured pancreatic tissue and was is consistent with evidence that pathologic and immunologic associated with scattered degeneration and necrosis of exocrine processes occurring in the spleen influence the outcome of dis- gland cells. ease in RTFS and BCWD (Decostere et al. 2001; Hadidi et al. 2008; Orieux et al. 2013). It is not currently known how splenic

Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 inflammation and necrosis influence survival in the ARS-Fp-R DISCUSSION and ARS-Fp-S lines on a pathophysiologic level. Physiologic The quantification of histopathologic lesions over time changes resulting from splenic disease have been associated proved valuable in elucidating host–pathogen interactions and with anemia, thrombocytopenia, and leucopenia, leading to mor- exploring differential disease response in selectively bred lines tality in mammals (Robbins and Cotran 2005), and this is an area of fish. Similar disease progression between the genetic lines of interest for future studies. and lower antemortem lesion severity in the spleen of fish from The cause of reduced splenic lesions in the ARS-Fp-R line the ARS-Fp-R line support the hypothesis that differential sur- is unknown at this time but may be associated with bacterial vival is a result of divergence in disease magnitude. Comparable loads, bacterial toxins, or inflammatory cell products. Previ- histomorphologic features in ARS-Fp-R and ARS-Fp-S fish on ously quantified F. psychrophilum colony forming units in in- day 0 suggest that the observed changes were induced by infec- fected fish from the ARS-Fp-R and ARS-Fp-S lines have shown tion and not a preformed trait. significantly lower splenic bacterial loads in the ARS-Fp-R line The relative survival difference of the ARS-Fp-R and ARS- on day 5 postinfection (Hadidi et al. 2008). However, in the Fp-S lines was similar to that observed for prior year-classes low number of fish from the ARS-Fp-R line that succumb to (Hadidi et al. 2008; Wiens et al. 2013a, 2013b). This survival experimental challenge, splenic bacterial loads are not different 188 MARANCIK ET AL.

from those in mortalities from the ARS-Fp-S line as measured itself. Likewise, a sharp decrease in total ellipsoid area, but by quantitative PCR (Marancik and Wiens 2013). Additional not average ellipsoid area, in mortalities indicates a loss in the studies examining multiple tissues through the acute phase of number of ellipsoid centers. Total ellipsoid size was negatively disease are needed to associate phenotypic response with the correlated with the severity of splenic necrosis, suggesting loss kinetics of infection. of ellipsoids due to necrosis or apoptosis. Reticuloendothelial Splenic and renal inflammatory infiltrates and renal response to F. psychrophilum infection is not well described in hematopoietic cell depletion did not demonstrate significant the literature. Ellipsoid and macrophage centers represent an changes over time or between genetic lines. Inflammation of area of interest for future studies to elucidate their immunologic the spleen and kidney and hematopoietic cell necrosis in the function and to validate their role as an early histopathologic kidney have been described in naturally infected moribund and indicator of bacterial infection in fish. dead Rainbow Trout (Rangdale et al. 1999; Nematollahi et al. Disease parameters quantified in this study were chosen 2003; Nilsen et al. 2011). The lack of observed significance based on their conspicuous alteration during the course of dis- in this study may be representative of a sample size that in- ease and their ability to be measured using quantifiable and sufficiently accounted for variation in individual host response. semiquantifiable methods. This approach inherently excludes Although the effect of age, sex, and environmental conditions certain disease parameters, such as cardiac reticuloendothelial were controlled for, to the limits of the experimental model, system activation, that could not be adequately measured but interfamily and individual fish genetics represent inherent di- that may serve an important role in the disease response. Fu- versity in host response that is difficult to account for at the ture exploration using molecular and biochemical strategies may population level (Roed et al. 1993). Mapping quantitative trait elucidate cellular and subcellular mechanisms of resistance not loci is a relatively newly employed method that may increase perceived in this study. Results from this study suggest that our understanding of how genetic diversity affects phenotypic immunologic and physiologic mechanisms associated with the response. Major trait loci have been identified that affect BCWD spleen are a priority for further study. resistance (Vallejo et al. 2010, 2014; Wiens et al. 2013b), and further work to validate and fine-map relevant quantitative trait loci are underway. ACKNOWLEDGMENTS Peritonitis was evident in this study but has not been com- We acknowledge technical contributions from Travis More- monly described in naturally F. psychrophilum-infected fish. land and Joel Caren. The authors would like to thank Susan Inflammatory changes observed within visceral adipose tissue Knowles for critical review of the manuscript and Vicki Blazer’s are likely specific to the IP challenge route. Lesion severity histology laboratory at the National Fish Health Laboratory for was progressive through the course of infection, suggesting that their technical support. This work was supported by Agricul- inflammation of the peritoneal cavity may have contributed to tural Research Service CRIS Project 1930-32000-005 “Host- clinical disease in this challenge model. The induction of peri- Pathogen and Environmental Interactions in Cool and Cold Wa- tonitis does not appear to negatively impact the ability of this ter Aquaculture” and Agriculture and Food Research Initiative experimental model to select fish resistant to natural F. psy- competitive grant number 2012-67015-30217 from the U.S. De- chrophilum infections, in which peritonitis presumably does not partment of Agriculture (USDA) National Institute of Food and occur, as the survival phenotype of the ARS-Fp-R line has been Agriculture. Mention of trade names or commercial products in demonstrated to be stable in field trials (Wiens et al. 2013a). this publication is solely for the purpose of providing specific in- Although IP injection of bacteria would seemingly result in formation and does not imply recommendation or endorsement immediate host immunostimulation, cellular changes in the peri- by the USDA. The USDA is an equal opportunity employer.

Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 toneal cavity were not detectable in tissues by histopathology on day 1 postinfection and cellular reactions were fairly subtle on day 3. This is consistent with previous observations that in- REFERENCES flammatory cellular responses are often absent in fish during the Baerwald, M. R., J. L. Petersen, R. P. Hedrick, G. J. Schisler, and B. May. peracute stage of disease (Ferguson 2006). 2011. A major effect quantitative trait locus for whirling disease resistance The increased prominence of splenic ellipsoids and endothe- identified in Rainbow Trout (Oncorhynchus mykiss). Heredity 106:920–926. lial lining macrophages in the heart were the earliest signs of sys- Dalmo, R. A., K. Ingebrigtsen, and J. Bøgwald. 1997. Non-specific defense mechanisms in fish, with particular reference to the reticuloendothelial system temic cellular change. As part of the reticuloendothelial system, (RES). Journal of Fish Diseases 20:241–273. these cellular components are believed to be important for endo- Decostere, A., E. D’Haese, M. Lammens, H. Nelis, and F. Haesebrouck. 2001. cytosis and degradation of foreign particulates and may serve as In vivo study of phagocytosis, intracellular survival and multiplication of antigen-presenting cells during infection (Dalmo et al. 1997). Flavobacterium psychrophilum in Rainbow Trout, Oncorhynchus mykiss An observed peak in average ellipsoid area on day 3 postin- (Walbaum), spleen phagocytes. Journal of Fish Diseases 24:481–487. Dorson, M., E. Quillet, M. G. Hollebecq, C. Torhy, and B. Chevassus. 1995. fection and continued increase in total ellipsoid area through Selection of Rainbow Trout resistance to viral haemorrhagic septicaemia day 9 indicate development of new or previously unobserved virus and transmission of resistance by gynogenesis. Veterinary Research ellipsoid centers, rather than sustained increase in ellipsoid size 26:361–368. HISTOPATHOLOGIC CHANGES IN TROUT INFECTED WITH FLAVOBACTERIUM PSYCHROPHILUM 189

Ferguson, H. W. 2006. Systemic pathology of fish: a text and atlas of normal Orieux, N., D. G. Douet, M. Le Henaff,´ and J. P. Bourdineaud. 2013. Prevalence tissues in teleosts and their responses in disease, 2nd edition. Scotian Press, of Flavobacterium psychrophilum bacterial cells in farmed Rainbow Trout: London. characterization of metallothionein A and interleukin1-β genes as markers Gjedrem, T. 2010. The first family-based breeding program in aquaculture. overexpressed in spleen and kidney of diseased fish. Veterinary Microbiology Reviews in Aquaculture 2:2–15. 162:127–135. Hadidi, S., G. W. Glenney, T. J. Welch, J. T. Silverstein, and G. D. Wiens. Overturf, K., S. LaPatra, R. Towner, N. Campbell, and S. Narum. 2010. Relation- 2008. Spleen size predicts resistance of Rainbow Trout to Flavobacterium ships between growth and disease resistance in Rainbow Trout, Oncorhynchus psychrophilum challenge. Journal of Immunology 180:4156–4165. mykiss (Walbaum). Journal of Fish Diseases 33:321–329. Hedrick, R. P., T. S. McDowell, G. D. Marty, G. T. Fosgate, K. Mukkatira, K. Ozaki, A., T. Sakamoto, S. Khoo, K. Nakamura, N. R. M. Coimbra, T. Akutsu, Myklebust, and M. El-Matbouli. 2003. Susceptibility of two strains of Rain- and N. Okamoto. 2001. Quantitative trait loci (QTL) associated with resis- bow Trout (one with suspected resistance to whirling disease) to Myxobolus tance/susceptibility to infectious pancreatic necrosis virus (IPNV) in Rainbow cerebralis infection. Diseases of Aquatic Organisms 55:37–44. Trout (Oncorhynchus mykiss). Molecular Genetics and Genomics 265:23–31. Henryon, M., P. Berg, N. J. Olesen, T. E. Kjaer, W. J. Slierendrecht, A. Jokum- Rangdale, R. E., R. H. Richards, and D. J. Alderman. 1999. Histopathological sen, and I. Lund. 2005. Selective breeding provides an approach to increase and electron microscopical observations on Rainbow Trout fry syndrome. resistance of Rainbow Trout (Oncorhynchus mykiss) to the diseases, enteric Veterinary Record 144:251–254. red mouth disease, Rainbow Trout fry syndrome, and viral haemorrhagic Robbins, S., and R. Cotran. 2005. Pathologic basis of disease, 7th edition. septicaemia. Aquaculture 250:621–636. Elsevier-Saunders, Philadelphia. Khoo, S. K., A. Ozaki, F. Nakamura, T. Arakawa, R. Nickolov, T. Sakamoto, T. Roed, K. H., K. T. Fjalested, and A. Stromsheim. 1993. Genetic variation in non- Akutsu, M. Mochizuki, I. Denda, and N. Okamoto. 2004. Identification of a specific immune parameters in Atlantic Salmon (Salmo salar). Aquaculture novel chromosomal region associated with infectious hematopoietic necrosis 111:292–294. (IHN) resistance in Rainbow Trout Oncorhynchus mykiss. Fish Pathology Silverstein, J. T., R. L. Vallejo, Y. Palti, T. D. Leeds, C. E. Rexroad III, T. 39:95–101. J. Welch, G. D. Wiens, and V. Docrocq. 2009. Rainbow Trout resistance Kjøglum, S., M. Henryon, T. Aasmundstad, and I. Korsgaard. 2008. Selective to bacterial cold-water disease is moderately heritable and is not adversely breeding can increase resistance of Atlantic Salmon to furunculosis, infectious correlated with growth. Journal of Animal Science 87:860–867. salmon anaemia and infectious pancreatic necrosis. Aquaculture Research Vallejo, R. L., Y, Palti, S. Liu, T. D. Leeds, J. P. Evenhuis, G. Gao C. E. Rexroad 39:498–505. III, and G. D. Wiens. 2014. Detection of QTL in Rainbow Trout affect- Leeds, T. D., J. T. Silverstein, G. M. Weber, R. L. Vallejo, Y.Palti, C. E. Rexroad ing survival when challenged with Flavobacterium psychrophilum. Marine III, J. Evenhuis, S. Hadidi, T. J. Welch, and G. D. Wiens. 2010. Response to Biotechnology 16:349–360. selection for bacterial cold water disease resistance in Rainbow Trout. Journal Vallejo, R. L., G. D. Wiens, C. E. Rexroad III, T. J. Welch, J. P. Evenhuis, T. of Animal Science 88:1936–1946. D. Leeds, L. L. Janns, and Y. Palti. 2010. Evidence of major genes affecting Marancik, D. P., and G. D. Wiens. 2013. A real-time polymerase chain re- resistance to bacterial cold water disease in Rainbow Trout using Bayesian action assay for identification and quantification of Flavobacterium psy- methods of segregation analysis. Journal of Animal Science 88:3814–3832. chrophilum and application to disease resistance studies in selectively bred Verrier, E. R., M. Dorson, S. Mauger, C. Torhy, C. Ciobotaru, C. Hervet, Rainbow Trout Oncorhynchus mykiss. FEMS Microbiology Letters 339:122– N. DeChamp, C. Genet, P. Boudinot, and E. Quillet. 2013. Resistance to a 129. rhabdovirus (VHSV) in Rainbow Trout: identification of a major QTL related National Agricultural Statistics Service. 2012. Trout production. Available: to innate mechanisms. PLoS ONE [online serial] 8(2):e55302. www.nass.usda.gov. (June 2014). Wiens, G. D., S. E. LaPatra, T. J. Welch, J. P. Evenhuis, C. E. Rexroad III, and Nematollahi, A., A. Decostere, F. Pasmans, and F. Haesebrouck. 2003. T. D. Leeds. 2013a. On-farm performance of Rainbow Trout (Oncorhynchus Flavobacterium psychrophilum infections in salmonid fish. Journal of Fish mykiss) selectively bred for resistance to bacterial cold water disease: effect of Diseases 26:563–574. rearing environment on survival phenotype. Aquaculture 388–391:128–136. Nilsen, H., A. B. Olsen, Ø. Vaagnes, H. Hellberg, K. Bottolfsen, H. Skjel- Wiens, G. D., R. L. Vallejo, T. D. Leeds, Y. Palti, S. Hadidi, S. Liu, J. P. stad, and D. J. Colquhoun. 2011. Systemic Flavobacterium psychrophilum Evenhuis, T. J. Welch, and C. E. Rexroad III. 2013b. Assessment of genetic infection in Rainbow Trout, Oncorhynchus mykiss (Walbaum), farmed in correlation between bacterial cold water disease resistance and spleen index fresh and brackish water in Norway. Journal of Fish Diseases 34:403– in a domesticated population of Rainbow Trout: identification of QTL on 408. chromosome Omy19. PLoS ONE [online serial] 8(10):e75749. Downloaded by [Department Of Fisheries] at 23:56 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:58 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Survival of Fish-Pathogenic Strains of Aeromonas hydrophila under Starvation Xiaojun Zhangab, Wenlong Caia, Zhen Taoac & Cova R. Ariasa a Aquatic Microbiology Laboratory, School of Fisheries, Aquaculture, and Aquatic Sciences, Auburn University, Auburn, Alabama 36849, USA b Present address: Huaihai Institute of Technology of China, 2220055, Liamuimgang , China c Present address: School of Marine Sciences, Ningbo University, 315211, Ningbo, China Published online: 21 Aug 2014.

To cite this article: Xiaojun Zhang, Wenlong Cai, Zhen Tao & Cova R. Arias (2014) Survival of Fish-Pathogenic Strains of Aeromonas hydrophila under Starvation, Journal of Aquatic Animal Health, 26:3, 190-193, DOI: 10.1080/08997659.2014.922515 To link to this article: http://dx.doi.org/10.1080/08997659.2014.922515

PLEASE SCROLL DOWN FOR ARTICLE

This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:190–193, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.922515

COMMUNICATION

Survival of Fish-Pathogenic Strains of Aeromonas hydrophila under Starvation

Xiaojun Zhang,1 Wenlong Cai, Zhen Tao,2 and Cova R. Arias* Aquatic Microbiology Laboratory, School of Fisheries, Aquaculture, and Aquatic Sciences, Auburn University, Auburn, Alabama 36849, USA

Aeromonas hydrophila is ubiquitous in natural freshwater Abstract environments, including lakes, rivers, and ponds (Rippey and The survival of Aeromonas hydrophila under low-nutrient con- Cabelli 1979), and it has been reported as part of the natu- ditions was investigated in this study. The behavior of three strains ral microbiota of fishes (Trust and Sparrow 1974). The vast isolated from Common Carp Cyprinus carpio (China) and Channel Catfish Ictalurus punctatus (USA) was compared when cells were majority of freshwater ecosystems are oligotrophic, and it is starved at different temperatures (4, 15, 25, and 35◦C) over a 4- accepted that aquatic bacteria must adjust to rapid changes in week period. Temperature played a major role in cell survival, and nutrient availability in order to survive (Poindexter 1981). Sev- cell viability decreased rapidly at 4◦C. Conversely, cells stored at ◦ ◦ eral studies have characterized the behavior of A. hydrophila 15 C did not lose viability over time. Warmer temperatures (25 C in low-nutrient water (Kersters et al. 1996; Brandi et al. 1999; and 35◦C) decreased cell numbers by approximately one order of magnitude. Ultrastructural changes in cell morphology were ob- Croci et al. 2001) and indicate that this bacterium can survive served in starved cells. Our data confirm that A. hydrophila can for extended periods even in chlorinated tap water. Moreover, persist in the aquatic environment for extended periods, but sur- A. hydrophila cells appear to enter the viable but nonculturable vival is strongly influenced by temperature. (VBNC) state under starvation at low temperatures (Mary et al. 2002). These studies confirm that A. hydrophila can quickly Aeromonas hydrophila is a gram-negative bacterium and one adapt to low-nutrient environments after being cultured in high- of the causative agents of motile Aeromonas septicemia (MAS) nutrient media. Temperature also plays a key role in how this in fish. This pathogen affects a variety of freshwater fish species bacterium responds to starvation. However, the strains used in around the world (Austin and Austin 1999), including Nile these studies were isolated from water or from human patients. Tilapia Oreochromis niloticus (Ruangpan et al. 1986), Chan- Little information is available regarding the survival of fish- nel Catfish Ictalurus punctatus (Defigueiredo and Plumb 1977), pathogenic strains during periods of starvation. The objective of Goldfish Carassius auratus (Brenden and Huizinga 1986), this study was to determine the culturability of three strains of Common Carp Cyprinus carpio (Lamers and Vanmuiswinkel A. hydrophila isolated from epizootic events in China and the 1986), and European Eel Anguilla anguilla (Esteve et al.

Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 USA under conditions of low-nutrient availability at different 1994). Recently, catfish producers in Alabama have experienced temperatures. severe losses due to an atypical A. hydrophila strain of unusual virulence (Hemstreet 2010). Since 2009, more than 7.5 million pounds of market-size catfish have been lost due to an epidemic METHODS of MAS that started in the heart of the catfish-producing region in Bacterial strains.—Three previously characterized A. western Alabama (Pridgeon and Klesius 2011). Similarly, carp hydrophila strains were used in this study. Strain ML09- production in China has been seriously affected by outbreaks 119, considered an atypical and highly virulent strain of A. of A. hydrophila which seemingly intensify over time (Nielsen hydrophila, was isolated from diseased Channel Catfish during et al. 2001). Currently, no preventive or palliative measures are a large-scale disease outbreak in Alabama in 2009 (Tekedar effective against this disease (Austin and Austin 1999). et al. 2013). Strain AL01 was isolated from Channel Catfish

*Corresponding author: [email protected] 1Present address: Huaihai Institute of Technology of China, Liamuimgang 2220055, China. 2Present address: School of Marine Sciences, Ningbo University, Ningbo 315211, China. Received December 13, 2013; accepted March 22, 2014 190 COMMUNICATION 191

in 2001 during a typical MAS episode, and it displayed low TABLE 1. Total number per milliliter (mean ± SE) of colony-forming units of three strains of Aeromonas hydrophila obtained when cells were maintained virulence toward Channel Catfish in artificial challenges (Arias, ◦ in sterile distilled water at 4 C. Data were log10 transformed to ensure nor- personal observation). Strain HC060718-1 was isolated from a mality. Within columns, significantly different means (P < 0.05) are denoted Common Carp during a severe MAS outbreak in that species by different lowercase letters; within rows, significantly different means are ◦ in Jiangsu Province, China. Bacteria were stored at −80 C denoted by different uppercase letters. as glycerol stocks and routinely cultured on Luria Bertani (LB) agar or broth with shaking (125 rpm) at 28 ± 2◦Cfor Day ML09-119 AL01 HC060718-1 24 h. 0 6.602 ± 0.008 zZ 6.230 ± 0.005 zZ 6.556 ± 0.002 zZ Microcosm inoculation.—Individual colonies from each 1 6.648 ± 0.044 zZ 6.318 ± 0.073 zY 6.603 ± 0.055 zZ strain were inoculated into 5 mL of LB broth and incubated 7 4.958 ± 0.194 yZ 4.763 ± 0.348 yZ 5.382 ± 0.207 yZ ◦ at 28 ± 2 C overnight with shaking. An aliquot (50 µL) of 14 4.679 ± 0.211 yZ 4.600 ± 0.372 yZ 4.816 ± 0.188 xZ a 24-h culture was inoculated into 15 mL of LB broth and 21 4.726 ± 0.352 yZ 4.034 ± 0.523 yZ 4.784 ± 0.158 xZ incubated overnight as before. Cultures were then centrifuged at 28 3.131 ± 0.082 xZ 3.071 ± 0.103 xZ 3.558 ± 0.361 wZ 5,000 × g for 10 min, and pellets were washed twice with 0.85% (weight : volume) NaCl sterile solution to avoid any carryover of nutrients. Cells were finally resuspended in sterile distilled RESULTS AND DISCUSSION 6 water to an approximate concentration of 10 colony-forming The behavior of A. hydrophila under conditions of starvation units (CFU)/mL. Four aliquots of 10 mL each were transferred ◦ at different temperatures was determined over a 4-week period. to sterile tubes and stored in the dark at 4, 15, 25, and 35 C, Table 1 summarizes cell survival at 4◦C. Initial cells counts respectively. Temperatures were selected based on the upper (approximately 106 CFU/mL) were statistically identical in all (35◦C) and lower limits (15◦C) of MAS occurrence in aquacul- ◦ strains. The first significant reduction was observed at day 7 in ture ponds. A lower temperature (4 C) was added to test whether all strains. Cell numbers remained constant for 2 weeks in strains A. hydrophila can survive winter temperatures. For statistical ML09-119 and AL01 but dropped significantly by day 28. analysis, there were three biological replicates per temperature. Significant reductions in cell counts in strain HC060718-1 were Cell enumeration.—Cell counts were determined immedi- observed on days 7, 14, and 28. By the end of the experiment ately after cells were resuspended in sterile distilled water and (day 28), all strains had experienced a reduction of three orders at 1, 7, 14, 21, and 28 d postinoculation. Cell suspensions were of magnitude in culturability. Our results are in agreement with µ 10-fold serially diluted in sterile distilled water and 100 L those of Mary et al. (2002), in which a sharp decrease in cell cul- of each dilution plated on LB agar in triplicate. Colonies were ◦ turability of A. hydrophila type strain ATCC 7966 was observed enumerated after 24 h of incubation at 28 C. within the first few days of incubation at 4◦C. According to the Statistical analysis.—Cell counts were converted to base- Mary et al. (2002) study, A. hydrophila entered the VBNC state 10 logarithms to meet the model assumption of a normal dis- at low temperatures because cells appeared to remain viable (as ± tribution, and mean log10 CFUs/mL SEs were calculated. per the LIVE/DEAD BacLight Bacterial Viability kit), but they One-way analysis of variance (ANOVA) was used to determine could no longer be cultured on media. However, raising the the differences in A. hydrophila CFU/mL from the short-term temperature to 25◦C did not reverse the VBNC stage, bringing survival study and Welch’s ANOVA (allowing for unequal vari- into question the true nature of those VBNC forms. Whether A. ance) in different strains. If the results of either the ANOVA hydrophila can survive as VBNC cells at low temperatures < or Welch’s ANOVA were statistically significant (P 0.05), remains uncertain. Tukey’s and Scheffe’s´ method were applied to perform post Cell viability at 15◦C slightly varied between strains through-

Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 α = hoc, pairwise comparisons at 0.05 for the means of the out the 4-week study (Table 2). Overall, cells maintained via- log A. hydrophila counts or Dunnett’s T3 test (allowing un- bility over time, and no significant decrease in cell counts was equal variance) for post hoc, pairwise comparisons for counts α = in different strains at 0.05. TABLE 2. Total number per milliliter (mean ± SE) of colony-forming units Ultrastructural analysis.—Changes in morphology were of three strains of Aeromonas hydrophila obtained when cells were maintained monitored under starvation conditions using scanning electron in sterile distilled water at 15◦C. See Table 1 for additional information. microscopy as previously described (Arias et al. 2012). Briefly, cells (5 µL of culture) were fixed in 2.5% glutaraldehyde (vol- Day ML09-119 AL01 HC060718-1 ◦ ume : volume) at 4 C overnight. Samples were filtered through 0 6.602 ± 0.007 zZ 6.230 ± 0.003 xw,Z 6.556 ± 0.009 yx,Z Isopore membrane (0.2 µm GTBP 02500; Millipore) dehydrated 1 6.777 ± 0.034 zY 6.598 ± 0.006 zy,X 6.899 ± 0.006 zZ in a graded ethanol series (50, 70, 90, and 100%), critical-point 7 6.685 ± 0.103 zZ 6.734 ± 0.091 zZ 6.733 ± 0.097 zy,Z ± ± ± driedinCO2 in an EMS 850 (Electron Microscopy Science) 14 6.476 0.114 zZ 6.144 0.056 wY 6.469 0.130 xZ and coated with gold palladium alloy in an EMS 550X (Elec- 21 6.492 ± 0.077 zZ 6.479 ± 0.189 zyx,Z 6.129 ± 0.156 wZ ± ± ± tron Microscopy Science). The coated samples were examined 28 6.160 0.213 yZ 6.408 0.210 yxw,Z 6.416 0.082 xZ using a Zeiss EVO 50. 192 ZHANG ET AL.

TABLE 3. Total number per milliliter (mean ± SE) of colony-forming units of three strains of Aeromonas hydrophila obtained when cells were maintained in ultrapure water at 25◦C. See Table 1 for more information.

Day ML09-119 AL01 HC060718-1 0 7.146 ± 0.004 zZ 6.845 ± 0.007 zZ 6.663 ± 0.003 zZ 1 6.500 ± 0.209 yZ 6.682 ± 0.102 zZ 6.627 ± 0.111 zZ 7 6.219 ± 0.194 yZ 6.260 ± 0.170 zy,Z 6.287 ± 0.178 zy,Z 14 5.476 ± 0.044 xY 5.925 ± 0.134 yZ 6.114 ± 0.163 zy,Z 21 4.964 ± 0.236 wZ 5.373 ± 0.378 xZ 5.514 ± 0.487 yx,Z 28 5.030 ± 0.333 xw,Z 5.211 ± 0.376 xZ 5.076 ± 0.655 xZ

FIGURE 1. Morphological changes to Aeromonas hydrophila cells during starvation in sterilized distilled water as determined by scanning electronic ◦ observed during the experiment. Conversely, at 25 C there was microscopy. Panel (A) shows 24-h-old cells immediately prior to inoculation ◦ a stronger effect on cell survival, with cell counts decreasing into microcosms, panel (B) cells maintained under starvation for 28 d at 25 C. = µ over time to approximately 105 CFU/mL (Table 3). The initial Bars 1 m. and final counts did not differ between strains, but ML09-119 maintained under low-nutrient conditions (Arias et al. 2012). presented fewer culturable cells than the other two strains at Such a drastic change in their ultrastructure is likely to affect the day 14. Previous studies compared the survival of A. hydrophila attachment and colonization capabilities of this fish pathogen, at 10◦C and 20◦C (Croci et al. 2001) and room temperature but further studies on its infectivity and pathogenicity are (Kersters et al. 1996), with similar results. Lower temperatures needed to test this hypothesis. (10–15◦C) favored the survival of A. hydrophila cells regard- In summary, our data show that all three strains of A. less of the bacterial strain or water chemistry used. By contrast, hydrophila behaved very similarly under starvation and that higher temperatures (≥20◦C) reduced culturability over time. At all can survive for extended periods at moderate and warm 35◦C, all strains showed a significant decrease in cell numbers at temperatures. Based on our study and others (Croci et al. day 7, after which they remained constant for the duration of the 2001), the threshold between survival and loss of culturability study (Table 4). To the best of our knowledge, these are the first ◦ ◦ ◦ occurs between 4 C and 10 C; above 15 C, cells lose some data on A. hydrophila survival at 35◦C. Although 35◦C might viability over time but can survive in water without nutrients seem excessively high for natural aquatic environments, cat- for several weeks. Our results support the notion that water can fish ponds typically reach (and could exceed) this temperature act as a reservoir and serve as a dispersal mechanism for this during the summer months. pathogen. Future studies should assess the virulence of starved The morphology of starved cells at day 21 was compared A. hydrophila cells to better define the life cycle of this species with that of fresh (24-h) cultures. As can be seen in Figure 1, in the aquatic environment. the morphology of A. hydrophila cells dramatically changed during starvation, with cells becoming shorter. Cell length decreased significantly, from 2.0 ± 0.2 µm in fresh cultures ACKNOWLEDGMENTS to 1.0 ± 0.2 µm in starved cells. The observed changes We thank Michael Miller (Advanced Microscopy and Imag- were similar in all three strains examined. “Rounding up” is ing Laboratory, Auburn University) for his help with scan- a common phenomenon in cells subjected to starvation and ning microscopy. This research was funded by the USDA–

Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 has previously been described in several gram-negative species ARS/Auburn University Speciﬁc Cooperative Agreement Pre- (Wai et al. 1999). Starved cells were covered by a “veil of vention of Diseases of Farmed Raised Fish and USDA–ARS secreted slime” that has also been observed in bacterial cells CRIS project 6420-32000-022-00D. Xiaojun Zhang is the recipient of a fellowship from the Jiangsu Overseas Research and Training Program for University Prominent Young and Middle- TABLE 4. Total number per milliliter (mean ± SE) of colony-forming units of three strains of Aeromonas hydrophila obtained when cells were maintained Aged Teachers and Presidents. Wenlong Cai thanks Shanghai in ultrapure water at 35◦C. See Table 1 for additional information. Ocean University for partially supporting his research fellowship. Zhen Tao is the recipient of a graduate research fellowship Day ML09-119 AL01 HC060718-1 funded by the Chinese Scholarship Council and Qingdao Ocean 0 7.146 ± 0.006 zZ 6.845 ± 0.009 zZ 6.662 ± 0.005 zZ University. 1 6.264 ± 0.093 yZ 6.169 ± 0.148 zZ 5.759 ± 0.432 zy,Z 7 4.457 ± 0.523 xZ 4.243 ± 0.656 yZ 4.248 ± 0.652 yx,Z 14 4.839 ± 0.329 xZ 4.629 ± 0.456 yZ 4.631 ± 0.416 yx,Z REFERENCES 21 4.659 ± 0.475 xZ 4.019 ± 0.541 yZ 3.892 ± 0.398 xZ Arias, C. R., S. LaFrentz, W. Cai, and O. Olivares-Fuster. 2012. Adaptive response to starvation in the ﬁsh pathogen Flavobacterium columnare: cell vi- 28 5.075 ± 0.112 xZ 5.108 ± 0.247 yZ 4.793 ± 0.122 yx,Z ability and ultrastructural changes. BMC Microbiology [online serial] 12:266. COMMUNICATION 193

Austin, B., and D. A. Austin 1999. Bacterial fish pathogens: diseases of farmed Nielsen, M. E., L. Hoi, A. A. Schmidt, D. Qian, T. Shimada, J. Y. and wild fish. Springer, New York. Shen, and J. L. Larsen. 2001. Is Aeromonas hydrophila the dominant Brandi, G., M. Sisti, F. Giardini, G. F. Schiavano, and A. Albano. 1999. Survival motile Aeromonas species that causes disease outbreaks in aquaculture ability of cytotoxic strains of motile Aeromonas spp. in different types of production in the Zhejiang Province of China? Diseases of Aquatic water. Letters in Applied Microbiology 29:211–215. Organisms 46:23–29. Brenden, R. A., and H. W. Huizinga. 1986. Pathophysiology of experimental Poindexter, J. S. 1981. Oligotrophy: fast and famine existence. Advances in Aeromonas hydrophila infection in Goldfish, Carassius auratus. Journal of Microbial Ecology 5:63–89. Fish Diseases 9:163–167. Pridgeon, J. W., and P. Klesius. 2011. Molecular identification and virulence of Croci L., S. De Pasquale, L. Cozzi, and L. Toti. 2001. Behavior of Aeromonas three Aeromonas hydrophila isolates cultured from infected Channel Catfish hydrophila in bottled mineral water. Journal of Food Protection 64:1836– during a disease outbreak in west Alabama (USA) in 2009. Diseases of 1840. Aquatic Organisms 94:249–253. Defigueiredo, J., and J. A. Plumb. 1977. Virulence of different isolates of Rippey, S. R., and V. J. Cabelli. 1979. Membrane filter procedure for enumer- Aeromonas hydrophila in Channel Catfish. Aquaculture 11:349–354. ation of Aeromonas hydrophila in freshwaters. Applied and Environmental Esteve, C., C. Amaro, and A. E. Toranzo. 1994. O-serogrouping and surface Microbiology 38:108–113. composition of Aeromonas hydrophila and Aeromonas jandaei pathogenic Ruangpan, L., T. Kitao, and T. Yoshida. 1986. Protective efficacy of for eels. FEMS Microbiology Letters 117:85–90. Aeromonas hydrophila vaccines in Nile Tilapia. Veterinary Immunology and Hemstreet, B. 2010. An update on Aeromonas hydrophila from a fish health Immunopathology 12:345–350. specialist for summer 2010. Catfish Journal 24:4. Tekedar, H. C., G. C. Waldbieser, A. Karsi, M. R. Liles, M. J. Griffin, S. Kersters, I., G. Huys, H. Van Duffel, M. Vancanneyt, K. Kersters, and W. Vamenta, T. Sonstegrad, M. Hossain, S. C. Schroeder, L. Khoo, and M. L. Verstaete. 1996. Survival potential of Aeromonas hydrophila in freshwaters Lawrence. 2013. Complete genome sequence of Channel Catfish epidemic and nutrient-poor waters in comparison with other bacteria. Journal of Ap- isolate, Aeromonas hydrophila strain Ml09-119. Genome Announcements plied Bacteriology 80:266–276. [online serial] 1(5):e00755-13. Lamers, C. H. J., and W. B. Vanmuiswinkel. 1986. Natural and acquired agglu- Trust, T. J., and A. H. Sparrow. 1974. The bacterial flora in the alimentary tract tinins to Aeromonas hydrophila in carp (Cyprinus carpio). Canadian Journal of freshwater salmonid fishes. Canadian Journal of Microbiology 20:1219– of Fisheries and Aquatic Sciences 43:619–624. 1228. Mary, P.,N. E. Chihib, O. Charafeddine, C. Defives, and J. P.Hornez. 2002. Star- Wai, S. N., Y. Mizunoe, and S. Yoshida. 1999. How Vibrio vation survival and viable but nonculturable states in Aeromonas hydrophila. cholerae survive during starvation. FEMS Microbiology Letters 180: Microbial Ecology 43:250–258. 123–131. Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 This article was downloaded by: [Department Of Fisheries] On: 16 November 2014, At: 23:58 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Journal of Aquatic Animal Health Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/uahh20 Development and Characterization of a Largemouth Bass Cell Line Rodman G. Getchella, Geoffrey H. Groococka, Emily R. Cornwella, Vanessa L. Schumacherb, Lindsay I. Glasnera, Barry J. Bakera, Stephen A. Frattinic, Gregory A. Woostera & Paul R. Bowsera a Aquatic Animal Health Program, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, 930 Campus Road, Ithaca, New York 14853, USA b Roche Diagnostics GmbH, Nonnenwald 2, 82377 Penzberg, Germany c The Feinstein Institute for Medical Research, 350 Community Drive, Manhasset, New York 11030, USA Published online: 21 Aug 2014.

To cite this article: Rodman G. Getchell, Geoffrey H. Groocock, Emily R. Cornwell, Vanessa L. Schumacher, Lindsay I. Glasner, Barry J. Baker, Stephen A. Frattini, Gregory A. Wooster & Paul R. Bowser (2014) Development and Characterization of a Largemouth Bass Cell Line, Journal of Aquatic Animal Health, 26:3, 194-201, DOI: 10.1080/08997659.2014.922517 To link to this article: http://dx.doi.org/10.1080/08997659.2014.922517

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions Journal of Aquatic Animal Health 26:194–201, 2014 C American Fisheries Society 2014 ISSN: 0899-7659 print / 1548-8667 online DOI: 10.1080/08997659.2014.922517

ARTICLE

Development and Characterization of a Largemouth Bass Cell Line

Rodman G. Getchell,* Geoffrey H. Groocock, and Emily R. Cornwell Aquatic Animal Health Program, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, 930 Campus Road, Ithaca, New York 14853, USA Vanessa L. Schumacher Roche Diagnostics GmbH, Nonnenwald 2, 82377 Penzberg, Germany Lindsay I. Glasner and Barry J. Baker Aquatic Animal Health Program, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, 930 Campus Road, Ithaca, New York 14853, USA Stephen A. Frattini The Feinstein Institute for Medical Research, 350 Community Drive, Manhasset, New York 11030, USA Gregory A. Wooster and Paul R. Bowser Aquatic Animal Health Program, Department of Microbiology and Immunology, College of Veterinary Medicine, Cornell University, 930 Campus Road, Ithaca, New York 14853, USA

Abstract The development and characterization of a new cell line, derived from the ovary of Largemouth Bass Micropterus salmoides, is described. Gonad tissue was collected from Largemouth Bass that were electrofished from Oneida Lake, New York. The tissue was processed and grown in culture flasks at approximately 22◦C for more than 118 passages during an 8-year period from 2004 to 2011. The identity of these cells as Largemouth Bass origin was confirmed by sequencing a portion of the cytochrome b gene. Growth rate at three different temperatures was documented. The cell line was susceptible to Largemouth Bass virus (LMBV) and its replication was compared with that of Bluegill Lepomis macrochirus fry (BF-2), one of the cell lines recommended for LMBV isolation by the American Fisheries Society Fish Health Section Blue Book. Quantitative PCR results from the replication trial showed the BF-2 cell line produced approximately 10-fold more LMBV copies per cell than the new Largemouth Bass cell line after 6 d, while Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 the titration assay showed similar quantities in each cell line after 1 week.

Cell lines provide an important diagnostic tool even though more than 80% of world aquaculture products originate (Lakra the availability of molecular assays is increasing. Close to et al. 2011). 300 cell lines from finfish have been established since the As a result of its popularity as a game fish, Largemouth Bass first gonadal cell line was developed from Rainbow Trout On- Micropterus salmoides may now be the most widely introduced corhynchus mykiss (RTG-2; Wolf and Quimby 1962; Fryer and fish in North America (USDA 2006). Largemouth Bass are cul- Lannan 1994; Lakra et al. 2011). During the last 20 years, 60% tured at private, state, and federal facilities for recreational fish- of the new cell lines have come from Asian investigators, where ing and stocking programs. According to the 2005 Census of

*Corresponding author: [email protected] Received February 18, 2014; accepted April 16, 2014

194 NEW LARGEMOUTH BASS CELL LINE 195

Aquaculture (USDA 2006), 192 U.S. fish farms reared Large- tion flask and an additional 15 mL of 0.25% trypsin added. Cells mouth Bass, posting total sales of US$10.6 million that year. were later harvested in aliquots after 8 h of digestion with con- Largemouth Bass were introduced in China in 1983 (Deng et al. tinuous mixing. Harvested cells were washed and resuspended 2011) and have become widely cultured with annual produc- in phosphate-buffered saline (PBS) three times. Finally, cells tion levels over 100,000 tons (Bai et al. 2008). Prior to 2011, were suspended in Hank’s minimal essential media (HMEM) the only significant viral pathogen of Largemouth Bass was the with 20% fetal bovine serum (FBS) and 100 U/mL penicillin, iridovirus, Largemouth Bass virus (LMBV), which had been 100 µg/mL streptomycin, and 2 mM glutamine (PSG), and then isolated from Largemouth Bass in the eastern USA since 1991 aliquoted into primary 25-cm2 culture flasks. Flasks were in- (Plumb et al. 1996; Grizzle et al. 2002). But recently, three new cubated at room temperature, at approximately 22◦C (hereafter viruses have been isolated from cultured Largemouth Bass in 22◦C). The cell lines were weaned onto HMEM with 10% FBS China, a ranavirus (Deng et al. 2011), a megalocytivirus (Ma by passage 20. The flasks containing cells that adhered and et al. 2011), and a rhabdovirus (Ma et al. 2013). In Italy, a be- grew were propagated. None of the flasks with testicular cells tanodavirus has been isolated from cultured Largemouth Bass remained viable from the few males that were processed. For (Bovo et al. 2011). The emergence of viral pathogens remains the initial 4–5 years, the LMBG cells were very slow growing. a significant constraint for aquaculture production and for the Complete monolayer formation with 1:2 subcultivations took al- sustainability of biodiversity in the natural environment (Sahoo most a month to occur. Morphologic changes in the cells began and Goodwin 2012). As viruses cannot reproduce on their own, to appear over time, initially becoming fibroblastic (not shown), the development of new fish cell lines aids in the detection and then later becoming more polygonal in appearance from approx- characterization of these emerging pathogens. imately passage 35 onwards (Figure 1A, B). The final polygonal Largemouth Bass virus has been isolated from 16 species morphology was stable at passage 94 (Figure 1C). of fish in the USA (Iwanowicz et al. 2013) and will replicate At the time of manuscript submission, the LMBG cells had in several cell lines, and optimal cell culture methods for the been passaged 118 times. Complete monolayers had formed detection of this virus have been determined (Piaskoski et al. with 1:3 subcultivations in <24 h. The cells were generally 1999; McClenahan et al. 2005). Five cell lines previously tested polygonal and had variably distinct margins, marked anisocyto- for their LMBV susceptibility are: Bluegill fry (BF-2; ATCC sis (variable cell size), marked anisokaryosis (variable nucleus certified cell line [CCL]-91), Fathead Minnow (ATCC CCL- size), and an irregularly round nucleus. A small portion of the 42), epithelioma papulosum cyprini (ATCC cell repository line cells had moderate perinuclear cytoplasmic vacuolation. The [CRL]-2872), Channel Catfish ovary (ATCC CRL-2772), and morphology was epithelial-like, suggesting that the cells were Chinook Salmon embryo (CHSE-214; ATCC CRL-1681). All likely derived from gonadal epithelial tissue. five of these lines have similar susceptibility to LMBV infection Plaque assay.—To determine the viral susceptibility of (McClenahan et al. 2005). LMBG and the titer of LMBV stock frozen at −80◦C, LMBG The objective of this study was to establish and characterize cells were inoculated with serial 10-fold dilutions of LMBV a new cell line from the gonad of Largemouth Bass (LMBG) stock solution and incubated at 30◦C. The isolate was the sixth and evaluate the replication of LMBV in this cell line versus passage of LMBV from the Santee–Cooper Reservoir (Plumb BF-2 cells. et al. 1996). The assay was stopped when plaques were clearly visible, and the cells were then fixed with 10% neutral buffered formalin (NBF) and stained with crystal violet, and plaques METHODS were counted under a light microscope. Fish.—Largemouth Bass of various sizes were captured in Briefly, the LMBG cells were first removed from the surface 2 Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 June 2004 from Oneida Lake, New York, by electrofishing and of 75-cm flasks with 0.25% trypsin and centrifuged for 3 min transported to the Cornell University Aquatic Animal Health at 1,200 × g. The supernatant was discarded and the pellet Program. Twenty Largemouth Bass were euthanized and the of cells at the bottom of the tube was resuspended in 6 mL of gonads were removed for processing and culture. This work HMEM with 5% FBS, penicillin, streptomycin, glutamine, and was conducted under an animal protocol approved by the Insti- HEPES (4-[2-hydroxyethyl]-1-piperazineethanesulfonic acid) tutional Animal Care and Use Committee of Cornell University. buffer (HMEM-5/PSG + H) added. Then 1 mL of the re- Cell line initiation.—Gonads from each of 20 adult Large- suspended cells in solution followed by 1 mL of HMEM- mouth Bass were collected aseptically and processed separately. 5/PSG + H media were added to each well of the six-well tissue Following a procedure Bowser and Plumb (1980) developed to culture plates. The plates were incubated at 22◦C in a sealed con- create an ovary cell line from Channel Catfish Ictalurus punc- tainer until they were confluent. Next, a serial 10-fold dilution of tatus, Largemouth Bass tissues were minced and suspended in stock LMBV was performed as follows. The existing media in 20 mL of Hank’s balanced salt solution (HBSS), and the suspen- all of the wells was removed and discarded. Then each well was sion was centrifuged at 1,500 × g for 5 min. The HBSS was then inoculated with 0.5 mL of the appropriate concentration from decanted and tissues were resuspended in 5 mL of 0.25% trypsin. most dilute (negative control) to least dilute (positive control). The tissue suspension was immediately transferred to a diges- Stock virus was directly added to the positive control wells, and 196 GETCHELL ET AL.

FIGURE 1. (A) Early stage LMBG cells (approximately passage 35) with polygonal morphology. (B) Intermediate LMBG cells (passage 70) with polygonal morphology and perinuclear vacuolation. Photo taken on Leica SP5 confocal microscope at 63 × magniﬁcation. (C) LMBG cells in monolayer with polygonal morphology (passage 94). (D) LMBG monolayer (passage 94) ﬁxed with a 10% solution of NBF and stained with crystal violet during plaque-forming assay with a10−4 dilution of LMBV. Two plaques (shown by arrows) are visible.

the same amount of HMEM-5/PSG + H media was added to Identity confirmation.—To confirm the new LMBG cell line the negative control wells. The plates were then incubated at was truly comprised of Largemouth Bass cells, DNA was iso- 30◦C for 1 h, the optimal temperature for LMBV replication lated from a 25-mg section of frozen Largemouth Bass skele- in Largemouth Bass (McClenahan et al. 2005). After 1 h, a tal muscle and two LMBG cell cultures at passage number Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 2% methylcellulose overlay (Methocel; Fluka Chemie, Buchs, 80, using a DNeasy Tissue kit (Qiagen, Valencia, California), Switzerland) was added to each well to preserve the plaques according to the manufacturer’s instructions for animal cells. by preventing the virus from spreading. The plates were then Primers were designed to amplify a fragment of the cytochrome incubated at 30◦C and observed daily for presence of viral cy- b gene using the default parameters in Primer Express 3.0 (Ap- topathic effects (CPE) for 8 d. When plaques were observed the plied Biosystems) and the Micropterus salmoides mitochon- media was removed from the wells and 1 mL of 10% NBF was drion complete genome sequence from GenBank accession added to each well and left for 1 h. Lastly, NBF was removed number NC 008106. The primers chosen were expected to am- and cells were stained with crystal violet dye. Once the dye plify a 401-bp amplicon in Largemouth Bass (forward primer, had been removed and the cell monolayer allowed to dry for 5-GGCAACCCAGATCCTCACAG-3; reverse primer, 5- 1–3 h, the total number of plaques in each well was counted GGGTGGCATTGTCGACTGA-3). Isolated DNA, including three times: from left to right, from top to bottom, and from a negative and a no-template control, was amplified using PCR right to left. The mean number of plaques was calculated from with an AccuPrime DNA Polymerase Kit (Invitrogen, Carlsbad, the three tabulated counts from each of the three replicate wells California). Agarose gel electrophoresis followed by ethidum and used to determine the titer of the virus. bromide staining was used to visualize the PCR product. The NEW LARGEMOUTH BASS CELL LINE 197

bands were sequenced in both directions and the resulting se- plates were incubated at 30◦C and observed daily for CPE. Af- quences were analyzed with the National Center for Biotechnol- ter 1 week, wells with positive CPE were recorded. Viral titers ogy Information (NCBI) Basic Alignment Search Tool (Altschul were expressed as tissue culture infectious dose (TCID50/mL; et al. 1990) and aligned with the same Largemouth Bass se- the dose that produced a cytopathic effect in 50% of the inocu- quence used to design the primers using Clustal W 2.0 (Larkin lated cultures) according to Reed and Muench (1938). All wells et al. 2007). This alignment was used to generate a maximum from the 10−6 dilutions were harvested and stored in 1.5-mL likelihood tree using PHYML (Guindon and Gascuel 2003), Eppendorf microcentrifuge tubes and frozen for later analysis with 10,000 bootstraps. by quantitative PCR (qPCR). The cells in the negative control Growth curve.—The trypan blue exclusion method was used wells were harvested individually and counted to estimate cell to determine the number of viable LMBG cells/cm2 using a density. hemocytometer. Growth of LMBG cells from passage number LMBV replication in LMBG and BF-2 cell lines.—One 48- 80 was documented at 15, 22, and 30◦C, at 3–4-d intervals for well tissue culture plate was prepared for each cell line (LMBG 42 d. and BF-2) and 500 µL of resuspended cells were added to each Briefly, media was carefully removed from three 75-cm2 tis- well. The plates were incubated at 22◦C in a sealed container, sue culture flasks with a complete monolayer of LMBG cells. allowed to grow for 1 d, and checked to confirm confluence. Using 3 mL of 0.25% trypsin, cells were completely detached A known stock of LMBV at a concentration of 4.9 × 107 from each flask, then neutralized with 9 mL of minimal essen- PFU/mL was diluted to achieve a multiplicity of infection tial media containing 10% FBS, with penicillin, streptomycin, (MOI) of 1.0 in each well. Due to the size difference between glutamine, and Hanks’ salts (HMEM-10/PSG + H) added. Sus- LMBG (passage 98) and BF-2 cells, different inoculation pended cells from each flask were combined to give a total concentrations were used for each cell line (1.0 × 104 PFU/mL volume of 36 mL. A 1:10 dilution of these cells using HMEM- for LMBG cells and 3.0 × 104 PFU/mL for BF-2 cells). Each 10/PSG + H resulted in a total volume of 360 mL. Of these well had 250 µL of media removed before they were inoculated day-0 cells, 10 µL were added to 40 µL of trypan blue stain with 250 µL of diluted LMBV, except for three wells that in three separate Eppendorf microcentrifuge tubes. The total received 250 µL HMEM-5/PSG + H as negative controls. The number of cells from each sample was counted using a hemo- plates were incubated at 30◦C and three wells from each plate cytometer following standard procedures (Strober 2001). From were harvested every 8 h for 4 d, and then a final sample was the 360 mL of pooled cells a volume of 5 mL was aliquoted taken at 6 d. The contents of each well, with the exception of into each of seventy-two 25-cm2 tissue culture flasks. Twenty- the negative control wells that were counted to estimate cell four flasks were incubated at each of three different tempera- density, were harvested into 1.5-mL microcentrifuge tubes and tures (15, 22, and 30◦C). Cells from two flasks were counted frozen at −80◦C for later analysis by qPCR. at each of 12 time points (days 1, 3, 5, 7, 9, 11, 14, 17, 21, Quantitative PCR.—The qPCR method used in this study 28, 35, and 42). To count cells, media was carefully removed for the LMBV titration and LMBV replication trials was as de- and 0.5 mL of 0.25% trypsin was added. Once all the cells scribed by Getchell et al. (2007). The method, in brief, is as detached from the bottom of the flask, 1.5 mL of HMEM- follows. The qPCR mixture was made of 1,250 µL of TaqMan 10/PSG + H was added to neutralize the trypsin, and the re- 2 × Master-Mix (Life Technologies, Carlsbad, California), 5 µL suspended cells were transferred to a 14-mL tube. Following of 0.3 µM LMBV398F forward primer (5-TGATTGGCAA- the procedure described above, 10 µL of cells were mixed with CACTAGCGATCT-3), 5 µLof0.3µM LMBV459R re- 40 µL of trypan blue in a 1.5-mL microcentrifuge tube and then verse primer (5-CCTAGCTCCTGCTTGATCGG-3), 5 µLof counted. 0.2 µM LMBV422FAM probe (5-[6- carboxyfluorescein]- Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 Titration of LMBV in LMBG and BF-2.—Two 96-well cell TCAATCCCGCCCCCGCC-[tetramethyl rhodamine]-3 ), and culture plates were made from both LMBG and BF-2 cells, re- 820 µL of QIAGEN RNase-Free H2O for a full 96-well plate. spectively. The LMBG and BF-2 cells were counted and then di- The PCR mixture was added to each well in the amount of luted to the same concentration, approximately 12,750 cells/mL. 20 µL. Samples, standards, and positive and negative controls Of these resuspended cells 200 µL were added to each well. The were added to each well in the amount of 5 µL. Every sample, plates were incubated at 22◦C in a sealed container and allowed including controls, was run in triplicate. The qPCR plates were to grow to confluence for 1 d. Tenfold serial dilutions of LMBV run on an ABI 7500 detector, a linear regression was calculated stock were made at concentrations ranging from 10−1 to 10−6. from five DNA standards ranging from 4.0 × 102 to 4.0 × 106, On one of the plates 24 wells were used for the positive control and all plates were run for 42 cycles. and inoculated directly from LMBV stock, and the second 24 Cell density.—For LMBV titration and replication studies wells were used for the negative control and contained HMEM- conducted with LMBG and BF-2 cell lines, negative control 5/PSG + H media only. An additional 24 wells were used for wells were used to estimate cell density per well. Media was the serial dilutions of 10−1 to 10−6, so there were two rows (24 removed from the negative control wells and 50 µL of trypsin wells) per concentration. Media was removed and each well was (0.25% for LMBG, 0.05% for BF-2) was added to each well. inoculated with 100 µL of the appropriate virus dilution. The Once it had been confirmed that the cells had lifted from the 198 GETCHELL ET AL.

surface, 200 µL of HBSS was added and pipetted up and down RESULTS to further separate the cells. Then, all 250 µL were removed and transferred to a 1.5-mL Eppendorf microcentrifuge tube. Plaque Assay Contents of the tube were mixed with a vortex mixer for 10 s. After 5 d incubation CPE was observed, and the plaque assay The same trypan blue exclusion method and counting procedure was stopped. No plaques were present in the control wells or −7 −6 −5 described above was used to determine the total number of cells the 10 dilution. For both 10 and 10 dilutions, very few per well. plaques were formed. On average 41 plaques were observed at −4 Statistical analysis.—This study presents quantitative data the 10 dilution (Figure 1D). The titer for this dilution was × 4 = × 3 as means and SD. Differences were analyzed with the Student’s calculated to be 6.0 10 PFU/mL (SD 2.0 10 ). Plaques −3 t-test. The mean viral loads at each time point in the LMBV coalesced at the 10 dilution and were too numerous to count. replication trial were compared using a two-sample t-test on Infection with LMBV was confirmed by positive qPCR results log-transformed, standardized qRT-PCR results. Values were from those flasks showing CPE (data not shown). considered significant at α < 0.05. Identity Confirmation Agarose gel electrophoresis followed by ethidium bromide staining of the PCR products revealed single bands at approximately 400 bp from DNA isolated from both LMBG samples, matching the size of the band in the Largemouth Bass skeletal muscle positive control (Figure 2). The sequence of the Large- mouth Bass skeletal muscle was deposited in NCBI GenBank under accession number KJ174324. The nucleotide alignment of the PCR amplicon isolated from both LMBG samples was identical to M. salmoides (data not shown). The phylogenetic tree Downloaded by [Department Of Fisheries] at 23:58 16 November 2014

FIGURE 3. Phylogenetic relationships of cytochrome b sequences from LMBG, representative ﬁsh, and other vertebrate species. This alignment was made using PHYML (Guindon and Gascuel 2003) with 10,000 bootstraps. The FIGURE 2. Electrophoresis gel showing ampliﬁed cytochrome b products resulting maximum-likelihood tree was visualized using NJplot (Perriere` and from Largemouth Bass tissue (T), LMBG cells (F), LMBG cells (C), and the Gouy 1996). The branch lengths are based on the number of inferred substitu- ladder (L). tions as indicated by the bar and the bootstrap values are percentage relatedness. NEW LARGEMOUTH BASS CELL LINE 199

FIGURE 4. Growth curve of LMBG cells at 15, 22, and 30◦C, at 3–4-d intervals for 42 d. The mean number of viable cells/cm2 ( ± SD) was determined using the trypan blue exclusion method and a hemocytometer. Bars represent SD.

confirmed that the LMBG and PCR amplicons of M. salmoides was performed to compare results between cell lines at each were most closely related to the Florida Largemouth Bass M. time point and found that LMBV production from the BF-2 salmoides floridanus (Figure 3). cells was significantly higher from the LMBG cells at all time points except 8 h (Figure 5). Growth Curves The initial concentration of LMBG cells at day 0 was 3.2 × Quantitative PCR 102 cells/cm2. Cells incubated at 15◦C never formed a confluent Quantitative PCR data from the LMBV titration assay that monolayer during the duration of the 42 d experiment, but there were collected at saturation showed similar relative quantities was a slight increase to 7.0 × 102 cells/cm2 by day 14, then a steady decrease to 4.6 × 10 cells/cm2 by day 42 (Figure 4). For the LMBG cells incubated at 30◦C a monolayer formed quickly but only increased to 1.5 × 103 cells/cm2 by day 14 before decreasing to 6.1 × 102 cells/cm2 by day 35. For those LMBG cells incubated at 22◦C a monolayer was visible by day 3 Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 5, and by day 17 the LMBG cell number peaked at 3.0 × 10 cells/cm2 followed by a slow decline to 1.2 × 103 cells/cm2 by day 42. Based on these growth curves LMBG cells appeared to have adapted to 22◦C.

Titration and Replication of LMBV in LMBG and BF-2 Cell Lines Both LMBG and BF-2 cell lines produced the same amount of virus at saturation, which was 108.47 TCID50/mL (data not shown). The LMBV replication by time showed that BF-2 cells produced more detectable viral genome equivalents at an initial FIGURE 5. Comparison of LMBV replication in LMBG and BF-2 cell lines faster rate than did LMBG cells. The ﬁnal concentration of (mean ± SD) by time. Bars represent SD. The mean viral load in BF-2 cells LMBV measured at day 6 in BF-2 cells was approximately 10- was signiﬁcantly higher than in LMBG cells at each time point except 8 h (t-test: fold higher than in LMBG cells (Figure 5). A Student’s t-test P ≤ 0.05). 200 GETCHELL ET AL.

of viral genomic DNA per cell from each cell line (LMBG Bellmund, Po Ting Wong, and Sara Huckabone for technical mean = 915, SD = 393) and (BF-2 mean = 862, SD = 542). A assistance. We also thank the staff at the Cornell Biological Student’s t-test was performed to compare results between cell Field Station for their assistance with electrofishing. This report lines and found that LMBV production from the BF-2 cells was is a resulting product from several supporting programs includ- not different from that of the LMBG cells (df = 46, P = 0.69). ing the AQUAVET Summer Research Fellowship Program, the Cornell University College of Veterinary Medicine Veterinary Investigator Program, the TST BOCES New Visions Life Sci- DISCUSSION ences Program, and the NIH Research Experience Fellowship. Cell lines are populations of cells capable of continuous or The statements, findings, conclusions, views, and recommenda- prolonged growth and division in vitro and the lines are often tions are those of the authors and do not necessarily reflect the comprised of immortalized cells. This LMBG cell line grows views of any of those organizations. continuously in cell culture, its identity has been confirmed, and it should be a valuable addition to the research and diagnostic community. The ideal growth temperature for this new REFERENCES cell line appears to be 22◦C, though these cells also can be ◦ Altschul, S. F., W. Gish, W. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic grown at 30 C, which is the optimum replication temperature local alignment search tool. Journal of Molecular Biology 215:403–410. ◦ for LMBV. A visible monolayer is formed within 24 h at 22 C Bai, J. J., D. J. Lutz-Carrillo, Y. C. Quan, and S. X. Liang. 2008. Taxonomic sta- and 30◦C. When the cell line was first developed it was main- tus and genetic diversity of cultured Largemouth Bass Micropterus salmoides tained at room temperature, 22◦C, and over time was kept only in China. Aquaculture 278:27–30. Bovo, G., A. Gustinelli, F. Quaglio, F. Gobbo, V. Panzarin, A. Fusaro, F. Mu- at that temperature. So, one might expect that the LMBG cells ◦ tinelli, M. Caffara, and M. L. Fioravanti. 2011. Viral encephalopathy and have adapted to grow at 22 C. Further testing of growth tem- retinopathy outbreak in freshwater fish farmed in Italy. Diseases of Aquatic ◦ ◦ peratures between 22 C and 30 C is needed to establish the Organisms 96:45–54. optimum temperature for this new cell line. The LMBG cells Bowser, P. R., and J. A. Plumb. 1980. Fish cell lines: establishment of a line were susceptible to LMBV, and distinct plaques appeared in from ovaries of Channel Catfish. In Vitro (Rockville) 16:365–368. Deng, G. C., S. J. Li, J. Xie, J. J. Bai, K. C. Chen, D. M. Ma, X. Y. Jiang, H. the monolayer in 4–5 d. Early plaque assays of frozen LMBV H. Lao, and L. Y. Yu. 2011. Characterization of a ranavirus isolated from 4 stock on LMBG cells showed quantities of 6.0 × 10 PFU/mL, cultured Largemouth Bass (Micropterus salmoides) in China. Aquaculture while later end-point titrations showed 108.47 TCID50/mL after 312:198–204. the LMBV isolate had been passaged in culture several times. Fryer, J. L., and C. N. Lannan. 1994. Three decades of fish cell culture: a current Caution also should be exercised when comparing the qPCR listing of cell lines derived from fishes. Journal of Tissue Culture Methods 16:87–94. results from the LMBV titration assay with the LMBV replica- Getchell, R. G., G. H. Groocock, V. L. Schumacher, S. G. Grimmett, G. A. tion trial; relative quantities of viral genomic DNA per cell are Wooster, and P. R. Bowser. 2007. Quantitative polymerase chain reaction reported in the titration assay, whereas the LMBV replication assay for Largemouth Bass virus. Journal of Aquatic Animal Health 19:226– trial plotted relative quantities of viral genomic DNA per 50 ng 233. RNA. At the lower MOI in the replication trial the BF-2 cells Grizzle, J. M., I. Altinok, W. A. Fraser, and R. Francis-Floyd. 2002. First isolation of Largemouth Bass virus. Diseases of Aquatic Organisms 50:233– produced about 10-fold more viral copies than did the LMBG 235. cells. On the other hand, the qPCR results from the end-point Guindon, S., and O. Gascuel. 2003. A simple, fast, and accurate algorithm titration assay showed that the LMBG and BF-2 cell lines pro- to estimate large phylogenies by maximum likelihood. Systematic Biology duced similar quantities of LMBV. A number of other new cell 52:696–704. lines are susceptible to piscine iridoviruses, although they are Iwanowicz, L., C. Densmore, C. Hahn, and P. McAllister. 2013. Identification of Largemouth Bass virus in the introduced Northern Snakehead inhabit- mostly derived from marine species of aquaculture significance

Downloaded by [Department Of Fisheries] at 23:58 16 November 2014 ing the Chesapeake Bay watershed. Journal of Aquatic Animal Health 25: in Asia and Europe (Wen et al. 2013). This LMBG cell line may 191–196. be a good substrate to use to test for the presence of viruses, Kim, R., and M. Faisal. 2010. Comparative susceptibility of representative Great particularly iridoviruses, in freshwater species. Additional work Lakes ﬁsh species to the North American viral hemorrhagic septicemia virus on this cell line that could be performed includes karyotyping sublineage IVb. Diseases of Aquatic Organisms 91:23–34. Lakra, W. S., T. R. Swaminathan, and K. P. Joy. 2011. Development, character- the cells and testing the susceptibility of LMBG cells to some of ization, conservation and storage of ﬁsh cell lines: a review. Fish Physiology the newer viral isolates from Largemouth Bass, including viral and Biochemistry 37:1–20. hemorrhagic septicemia virus (Kim and Faisal 2010). Finally, Larkin, M. A., G. Blackshields, N. P. Brown, R. Chenna, P. A. McGettigan, H. increased availability of gonad cells may prove to be a valuable McWilliam, F. Valentin, I. M. Wallace, A. Wilm, R. Lopez, J. D. Thompson, resource for diagnostic testing; therefore, this cell line will be T. J. Gibson, and D. G. Higgins. 2007. Clustal W and clustal X version 2.0. Bioinformatics 23:2947–2948. deposited with the ATCC. Ma, D. M., G. C. Deng, J. J. Bai, S. J. Li, X. Y. Jiang, and X. J. Yang. 2011. Pathogeny of disease characterized by swollen liver and spleen in Largemouth Micropterus salmoides ACKNOWLEDGMENTS Bass ( ). Journal of Fishery Sciences of China 18:654– 659. John Plumb, Auburn University, Auburn, Alabama, kindly Ma,D.M.,G.C.Deng,J.J.Bai,S.J.Li,L.Y.Yu,Y.Quan,X.J.Yang,X. provided the LMBV strain used in this study. We thank Chelsea Y. Jiang, Z. Zhu, and X. Ye. 2013. A strain of Siniperca chuatsi rhabdovirus NEW LARGEMOUTH BASS CELL LINE 201

causes high mortality among cultured Largemouth Bass in South China. Sahoo, P. K., and A. E. Goodwin. 2012. Viruses of freshwater fin- Journal of Aquatic Animal Health 25:197–204. fish in the Asian–Pacific region. Indian Journal of Virology 23:99– McClenahan, S. D., B. H. Beck, and J. M. Grizzle. 2005. Evaluation of cell 105. culture methods for detection of Largemouth Bass virus. Journal of Aquatic Strober, W. 2001. Monitoring cell growth. Current Protocols in Immunology Animal Health 17:365–372. Supplement 21:A.3A.1–A.3A.2. Perriere,` G., and M. Gouy. 1996. WWW-Query: an on-line retrieval system for USDA (U.S. Department of Agriculture) National Agriculture Statis- biological sequence banks. Biochimie 78:364–369. tics Service. 2006. 2002 Census of agriculture: census of aquaculture Piaskoski, T. O., J. A. Plumb, and S. R. Roberts. 1999. Characterization of the (2005). USDA, Volume 3, Special Studies Part 2, AC-02-SP-2, Beltsville Largemouth Bass virus in cell culture. Journal of Aquatic Animal Health Maryland. 11:45–51. Wen, C. M., C. C. Ku, and C. S. Wang. 2013. Viral susceptibility, transfection and Plumb, J. A., J. M. Grizzle, H. E. Young, A. D. Noyes, and S. Lamprecht. growth of SPB–a fish neural progenitor cell line from the brain of Snubnose 1996. An iridovirus isolated from wild Largemouth Bass. Journal of Aquatic Pompano, Trachinotus blochii (Lacepede). Journal of Fish Diseases 36:657– Animal Health 8:265–270. 667. Reed, L. J., and H. Muench. 1938. A simple method of estimating fifty per cent Wolf, K., and M. C. Quimby. 1962. Established eurythermic line of fish cells in endpoints. American Journal of Hygiene 27:493–497. vitro. Science 135:1065–1066. Downloaded by [Department Of Fisheries] at 23:58 16 November 2014