North American Journal of Aquaculture Development Of

This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:12 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Development of Captive Breeding and Seed Production Techniques for Giant River Catfish Sperata seenghala M. Aminur Rahmana, A. Arshadb, Fatimah Md. Yusoffb, S. M. N. Aminc, K. Marimuthud & R. Arae a Laboratory of Marine Biotechnology, Institute of Bioscience, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia b Laboratory of Marine Biotechnology, Institute of Bioscience and Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia c Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia d Department of Biotechnology, Faculty of Applied Sciences, AIMST University, Kedah Darul Aman, Malaysia e Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia Published online: 15 Apr 2014.

To cite this article: M. Aminur Rahman, A. Arshad, Fatimah Md. Yusoff, S. M. N. Amin, K. Marimuthu & R. Ara (2014) Development of Captive Breeding and Seed Production Techniques for Giant River Catfish Sperata seenghala, North American Journal of Aquaculture, 76:2, 97-103, DOI: 10.1080/15222055.2013.855282 To link to this article: http://dx.doi.org/10.1080/15222055.2013.855282

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 North American Journal of Aquaculture 76:97–103, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.855282

ARTICLE

Development of Captive Breeding and Seed Production Techniques for Giant River Catﬁsh Sperata seenghala

M. Aminur Rahman* Laboratory of Marine Biotechnology, Institute of Bioscience, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia A. Arshad and Fatimah Md. Yusoff Laboratory of Marine Biotechnology, Institute of Bioscience and Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia S. M. N. Amin Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia K. Marimuthu Department of Biotechnology, Faculty of Applied Sciences, AIMST University, Kedah Darul Aman, Malaysia R. Ara Department of Aquaculture, Faculty of Agriculture, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia

Abstract The Giant River Catfish Sperata seenghala, locally known as “Guizza Air” (and hereafter referred to as just Guizza), has significant aquacultural and conservational values, but closer monitoring of the fish is now badly needed. To protect the species from extinction, an experiment was undertaken to breed and produce seed of Guizza in a captive-rearing system. In our study, Guizza were successfully bred and the naturally produced fry were reared at different stocking densities in nine earthen nursery ponds having an area of 0.012 ha each with an average depth of

Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 0.8 m. Three stocking densities were evaluated in three replicates each. Fry produced from natural propagation that were stocked at 100,000/ha were defined as treatment 1 (T1), those stocked at 150,000/ha as treatment 2 (T2), and those stocked at 200,000/ha as treatment 3 (T3). At stocking, all reared fry were of the same age-group with a mean ± SD length and weight of 1.74 ± 0.27 cm and 0.20 ± 0.03 g, respectively. Fry in all the experimental ponds were reared with commercial Saudi Bangla (SABINCO) nursery feed (32.06% crude protein) for the first 14 d and starter-I (31.53% crude protein) for days 15–56. Growth (final weight, final length, weight gain, length gain, and specific growth rate) and survival of the fingerlings were significantly higher in T1 than in T2 and T3. Feed conversion rate was significantly lower in T1 than in T2 and T3. Moreover, the physicochemical measurements and plankton population of the pond water were within the optimal level for fish culture, with plankton abundance being higher in T1 than in T2 and T3. Consistently higher net benefits were obtained from T1 than from T2 and T3. Hence, of the three stocking densities, 100,000 fry/ha was the most suitable density for rearing Guizza fingerlings in nursery ponds.

*Corresponding author: [email protected] Received April 17, 2013; accepted October 7, 2013 97 98 RAHMAN ET AL.

The Giant River Catfish Sperata seenghala, locally referred of the greater Mymensingh region during January–February, to as “Guizza Air” (and hereafter referred to as just Guizza), is 2009. The collected brood fish were transported to the Fresh- an important bagrid catfish and was once available in rivers, water Station at the Bangladesh Fisheries Research Institute, floodplains, inundated fields, swamps, ditches, canals, and Mymensingh, and then reared in earthen ponds. Fish were other freshwater areas throughout Afghanistan, Pakistan, India, fed twice daily with supplemental feed (34.27% crude pro- Nepal, Malaysia, and Bangladesh (Jayaram 1977, 2002; tein, 11.55% crude lipid, 10.86% crude fiber, 15.48% ash, and Jhingran 1991; Talwar and Jhingran 1991; Muchlisin et al. 2004; 27.84% nitrogen-free extract on dry matter basis) composed of Rahman 2005). It is a popular species of catfish and a favorite rice bran (30%), mustard oilcake (29%), fish meal (40%), and of consumers due to its delicious taste and therefore is in great vitamin-premix (1%) at the rate of 5–6% of the estimated fish demand, fetching higher prices in the market than major Indian biomass. The pond was fertilized at biweekly intervals with carps (Catla Catla catla [also known as Gibelion catla], Rohu cow dung (1,000 kg/ha) and urea and triple super phosphate Labeo rohita, and Mrigal Cirrhinus mrigala; Tripathi 1996). (50 kg/ha), following Rahman et al. (2005b). The fish is carnivorous in nature and consumes a variety Natural propagation in artificial nests.—For stimulating the of living creatures including fish, frogs, snakes, insects, earth- natural propagation of Guizza, artificial nests (holes) were con- worms, tadpoles, crustaceans, and debris (Rahman 2005). It structed on the pond bottom. Each nest was 0.7 m in diameter fights well and provides good sport fishing. In its natural habitat and 0.3 m in depth. A total of 30 nests were made in two ponds it spawns twice a year from May to July and from September having an area of 0.04 ha each with 15 nests per pond. Three to November (Talwar and Jhingran 1991; Rahman et al. 2005b, months before the onset of the breeding season, 30 pairs (30 2011). Absolute fecundity of this fish varies from 13,005 to males and 30 females) of Guizza breeders (650–1200 g) were 119,943 eggs, with an estimated egg diameter of 0.80–1.40 mm; stocked in the ponds and fed with supplemental feed as above. the size range of spawners is 32.50–48.70 cm in length (M. A. Fresh groundwater was supplied everyday to maintain a natu- Rahman, unpublished). There are several reasons for the lack of ral condition (e.g., temperature adjustment, water circulation, aquaculture of the tropical bagrid catfishes: these fishes are diffi- mate selection, sex stimulation, and courtship induction) for cult to spawn artificially, are sensitive to water quality changes, breeding. During mid-May to early June 2009, a total of 18,287 and are easily stressed (Rahman et al. 2005b, 2011). Hatcheries Guizza fry (762 ± 192 fry/nest [mean ± SD] with lengths of have difficulties in synchronizing the maturity between male and 1.74 ± 0.27 cm and weights of 0.20 ± 0.03 g) were collected female broodstock, which makes it hard to artificially produce from 24 holes by completely draining the pond and were then seeds for stocking in grow-out ponds (Muchlisin et al. 2004; transferred to earthen nursery ponds for 8 weeks. Rahman et al. 2005b). Rearing of fry and fingerlings in nursery ponds.—The Although the Giant River Catfish is categorized as “Least collected Guizza fry were reared at the Freshwater Station, Concern” for now because current field surveys indicate that it Bangladesh Fisheries Research Institute, Mymensingh, for 8 is still relatively abundant, closer monitoring of its fisheries is weeks from 8 June to 3 August in nine earthen nursery ponds badly needed (IUCN 2013). However, in Bangladesh the natural with a surface area of 0.012 ha and an average depth of 0.8 m. habitats and breeding grounds of this bagrid catfish have been All the ponds had similar rectangular size, depth, basin confor- severely degraded in recent years due to poor management, in- mation, contour, and bottom type. Three treatments of differ- creasing water pollution, destruction of breeding grounds, and ing stocking densities of fry were employed with three repli- overfishing. As a result, the natural stock of this important fish- cates each. Stocking densities were as follows: 100,000 fry/ha ery is declining day by day (Rahman et al. 2005b, 2011). Limited (treatment1; T1), 150,000 fry/ha (treatment 2; T2), and 200,000

Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 attempts have been made to develop an aquaculture of Guizza fry/ha (treatment 3; T3). The ponds were drained, cleared of with major Indian carps in earthen ponds (Rahman et al. 2011), aquatic vegetation, and exposed to full sunlight and had a well- but no systematic attempts have been made to develop breed- designed system of inlet and outlet. After the ponds were dry, ing protocols, except for a few on induced breeding without quicklime (CaCO; 250 kg/ha) was spread over the pond bot- success (Rahman et al. 2005b). In the present study, fish were tom to kill harmful animals and pathogens. The ponds were successfully bred for the first time under natural propagation then filled with groundwater and fertilized with organic manure techniques. The naturally produced fry were then reared at dif- (cow dung; 2,500 kg/ha). Seven days after manure was applied, ferent stocking densities to determine the optimal density, based the pond water was sprayed with dipterex (2, 22-Trichloro-1- on the highest growth and production, of Guizza fingerlings in Hydroxy Ethyl Phosphate; manufactured by Ciba) at the rate of a controlled nursery rearing system. 1.0 ppm to eradicate harmful insects and predatory zooplankton. Fry having a mean ± SD length of 1.74 ± 0.27 cm and weight of 0.20 ± 0.03 g, taken from 30 randomly sampled individuals, METHODS were then stocked on the same day in all the experimental ponds. Collection and rearing of broodstock.—A total of 100 indi- Fry in all the experimental ponds were fed with commercial viduals of adult Guizza weighing between 650 and 1300 g were Saudi Bangla (SABINCO) fish feed, i.e., nursery (32.06% crude collected from the Brahmaputra River basin and the floodplains protein) for the first 14 d and starter-I (31.53% crude protein) BREEDING AND SEED PRODUCTION OF GIANT RIVER CATFISH 99

from days 15–56. The rate of feeding was 14% of the estimated The FCR was calculated according to the formula proposed body weight of fry for the first 2 weeks, 12% for the second 2 by Castell and Tiews (1980) as follows: weeks, 10% for the third 2 weeks, and 8% for the fourth 2 weeks. After the fry were stocked, the ponds were fertilized with cow feed fed (dry weight) FCR = . dung (1,000 kg/ha) at weekly intervals to accelerate the primary live weight gain productivity of the ponds throughout the entire experiment. The physicochemical characteristics of the pond water were After 8 weeks of rearing, the fingerlings were harvested by monitored weekly between 0900 and 1000 hours. Temperature repeated netting, followed by draining the ponds. The live fin- (◦C) and dissolved oxygen (mg/L) were determined directly by gerlings were counted and weighed individually. Survival (%) a digital water quality analyzer (Model 58; YSI, USA), pH by a and production (number/ha) of fingerlings were then calculated digital pH-meter (Model 3020; Jenway, UK), water transparency directly and compared among the treatments. The prices of fin- (cm) by a Secchi disk, and ammonia nitrogen by a HACH water gerlings (based on the size categories) fixed by the Bangladesh analysis kit (DR 2000; USA). Total alkalinity was measured Fisheries Research Institute, where the experiment was con- following the standard method (Stirling 1985; APHA 1992). ducted, were as follows (in takas [Tk]; US$1 = Tk 77.40 as of Quantitative and qualitative estimates of plankton in the nurs- December 2013): Tk 6.00/piece for 9.0–10.0 cm, Tk 4.00/piece ery ponds were taken weekly. Ten liters of water collected from for 8.0–9.0 cm, and Tk 3.00/piece for 7.0–8.0 cm. different locations and depths in each pond were filtered through Data analysis.—The data on growth, survival, production, fine-meshed plankton net (25 µm) to obtain a 50-mL sample. water quality, and plankton abundance of different treatments The samples were preserved immediately with 5% buffered for- were tested through one-way analysis of variance using Stat malin in plastic bottles. Plankton density was estimated by using View version 4.0 (Abacus Concepts, Berkeley, California) fol- a subsampling technique. A Sedgwick–Rafter cell was used un- lowed by Tukey’s multiple comparison test. Standard deviation der a calibrated compound microscope for plankton counting. in each variable and treatment was calculated and expressed as Plankton count was performed using the formula proposed by mean ± SD. The level for statistical significance was set at P ≤ Rahman (1992) and Stirling (1985). 0.05. A simple cost-benefit analysis was done to estimate the Twenty individuals from each pond were sampled weekly un- net benefits from different treatments. til they attained the fingerling stage. Growth in terms of length and weight, specific growth rate (SGR), and food conservation ratio (FCR) was estimated. Length and weight gain were cal- RESULTS culated simply by deducting the mean initial length and weight Water Quality values from that of the final. The SGR, which is the instanta- Temperature, pH, alkalinity, and ammonia nitrogen did not neous change in weight of a fish, was calculated as the percent show any significant (P > 0.05) differences among the treat- increase in body weight per day over any given time interval ments, while Secchi disk transparency significantly (P < 0.05) (Brown, 1957) as follows: increased with greater fish density (from T1 to T3), but dissolved oxygen decreased as fish density increased(Table 1).

ln W2 − ln W1 SGR (%/day) = × 100, Plankton Enumeration T − T 2 1 The recorded phytoplankton population comprised four broad groups, i.e., Bacillariophyceae (5 genera), Chlorophyceae Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 where W1 = the initial live body weight (g) at time T1 (day) and (14 genera), Cyanophyceae (7 genera), and Euglenophyceae (3 W2 = the ﬁnal live body weight at time T2. genera), which are summarized in Table 2. The abundances

TABLE 1. Mean ± SD values of water quality characteristics from weekly samples over the 8-week experiment. Numbers in the same row with the same lowercase letter are not signiﬁcantly different (P > 0.05).

Treatments

Water quality characteristic T1 (100,000 fry/ha) T2 (150,000 fry/ha) T3 (200,000 fry/ha) Water temperature (◦C) 30.96 ± 1.28 z 30.88 ± 1.32 z 30.98 ± 1.38 z Transparency (cm) 34.68 ± 3.55 x 45.32 ± 3.88 y 53.75 ± 3.52 z Dissolved oxygen (mg/L) 4.96 ± 0.82 z 4.38 ± 0.73 y 3.90 ± 0.69 x pH 7.65 ± 0.28 z 7.55 ± 0.25 z 7.52 ± 0.24 z Total alkalinity (mg/L) 134.15 ± 26.38 z 132.85 ± 25.78 z 130.68 ± 28.28 z Ammonia nitrogen (mg/L) 0.30 ± 0.18 z 0.35 ± 0.19 z 0.38 ± 0.26 z 100 RAHMAN ET AL.

TABLE 2. Mean ± SD values of plankton abundance from weekly samples of pond water over the 8-week experiment. Numbers in the same row with the same lowercase letter are not signiﬁcantly different (P > 0.05).

Treatments

Plankton groups and totals T1 (100,000 fry/ha) T2 (150,000 fry/ha) T3 (200,000 fry/ha) Phytoplankton (cells/L) Bacillariophyceae 3,793 ± 219 z 3,322 ± 209 y 2,775 ± 250 x Chlorophyceae 4,578 ± 207 z 4,019 ± 252 y 3,281 ± 279 x Cyanophyceae 3,025 ± 222 z 2,650 ± 273 y 2,178 ± 283 x Euglenophyceae 2,578 ± 254 z 2,284 ± 237 y 2,081 ± 259 y Total 13,974 ± 880 z 12,275 ± 766 y 10,315 ± 560 x Zooplankton (individuals/L) Crustacea 7,834 ± 464 z 6,525 ± 581 y 5,222 ± 652 x Rotifera 9,625 ± 501 z 8,509 ± 523 y 6,884 ± 562 x Total 17,459 ± 1,266 z 15,034 ± 1,403 y 12,106 ± 1,175 x

of Chlorophyceae, Bacillariophyceae, and Cyanophyceae were followedbyT2 (Tk 220,463/ha) and then T3 (Tk 120,965/ha) significantly higher (P < 0.05) in T1 than in T2 and in T2 (Table 4). than in T3, but the values of Euglenophyceae in T2 and T3 did not differ significantly (P > 0.05). Among the phytoplank- DISCUSSION ton groups, Chlorophyceae was the most dominant in all the Growth (final length, length gain, final weight, weight gain, treatments followed by Bacillariophyceae, Cyanophyceae, and and SGR) and survival of Guizza fingerlings was significantly then Euglenophyceae. The mean total phytoplankton abundance higher in T than in T and T although the same food was was significantly higher (P < 0.05) in T than in T and T .The 1 2 3, 1 2 3 applied at an equal ratio in all the treatments. The causes might recorded zooplankton population consisted of two major groups, include competition for food and space due to higher numbers i.e., Crustacea (5 genera) and Rotifera (7 genera). Rotifera were of fish in T and T . The results of the present experiment agree dominant over Crustacea during the entire experimental dura- 2 3 well with the findings obtained by Houde (1975), Haque et al. tion in all the treatments. However, the abundance of Rotifera (1994), Kohinoor et al. (1994), Mollah and Hossain (1998), Is- and Crustacea was significantly higher (P < 0.05) in T than 1 lam (2002), Rahman and Rahman (2003), and Rahman et al. in T and in T than in T . The abundance of total zooplankton 2 2 3 (2005a) during fry and fingerling rearing experiments with Sea was also significantly higher (P < 0.05) in T1 than in T2 and in T2 than in T3. 10.0 Treatment-1 9.0 Treatment-2 Growth and Production of Fingerlings Treatment-3 8.0 The greatest length and weight increase at weekly intervals

Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 was obtained in T1 followedbyT2 and then T3 (Figures 1, 2). 7.0

The initial length and weight of fry in all the experimental ponds 6.0 at stocking were the same (Table 3). The mean final length and weight of the fingerlings were significantly higher (P < 0.05) 5.0 in T1 than in T2 and in T2 than in T3. Weight and length gains 4.0

followed the same trends as final weight and length. The SGR Mean length (cm) 3.0 in T1 was significantly higher (P < 0.05) than in T2 and T3. Significantly lower FCR was obtained in T1 than in T2 and in 2.0 T than in T . The highest survival (%) was also observed in T 2 3 1, 1.0 followedbyT2 and then T3 (P < 0.05). Significantly higher numbers of fingerlings were found in 0.0 012345678 T3 (90,333/ha) than in T2 (86,415/ha) and in T2 than in T1 (69,250/ha) (Table 3). Conversely, the total cost of produc- Weeks tion was the lowest in T1 (Tk 98,499/ha) followed by T2 (Tk FIGURE 1. Mean length increase of Guizza fingerlings at different stocking 125,197/ha) and then the highest in T3 (Tk 150,034/ha). Despite densities, i.e., treatment 1 (100,000 fry/ha), treatment 2 (150,000 fry/ha), and this, the highest net benefit was obtained in T1 (Tk 318,751/ha) treatment 3 (200,000 fry/ha), over the nursing period of 8 weeks. BREEDING AND SEED PRODUCTION OF GIANT RIVER CATFISH 101

4.5 and Rahman et al. (2005a). Transparency was consistently Treatment-1 higher in T3, possibly due to the reduction of the plankton popu- 4.0 Treatment-2 lation by the higher density of ﬁsh in the ponds (Rahman 1992; Treatment-3 Haque et al. 1994; Rahman et al. 2005a). The concentration of 3.5 dissolved oxygen was lower in ponds stocked with a higher den-

3.0 sity of fish than in ponds where stocking density was low. This might be due to the higher consumption rate of oxygen by the 2.5 higher density of fish and other aquatic organisms (Boyd 1982). However, the dissolved oxygen level was within the suitable 2.0 range for fish culture (Mollah and Hossain 1998; Rahman et al. 2005a). The pH values agree well with the findings of Mollah

Mean live weight (g) weight live Mean 1.5 and Hossain (1998), Rahman and Rahman (2003), and Rahman

1.0 et al. (2005a) and are within the range of good water quality for rearing of fry and fingerlings in nursery ponds. Higher total alka- 0.5 linity values might be due to higher amounts of lime doses during the pond preparation and subsequent liming during the experi- 0.0 mental period (Boyd 1982; Jhingran 1991; Rahman et al. 2005a). 012345678 The level of ammonia nitrogen recorded from the experimental Weeks ponds was lower than reported by Dewan et al. (1991). However, FIGURE 2. Mean weight increase of Guizza fingerlings at different stocking the present level of ammonia nitrogen content in the experimen- densities, i.e., treatment 1 (100,000 fry/ha), treatment 2 (150,000 fry/ha), and tal ponds was not lethal to the fish (Kohinoor et al. 1998, 2001). treatment 3 (200,000 fry/ha), over the nursing period of 8 weeks. The quantity of both phytoplankton and zooplankton in the present study was inversely related with the stocking density of Bream Archosargus rhomboidalis, Common Carp Cyprinus car- fry. The quantity of plankton was higher in T1, where stocking pio, Putitor Mahseer Tor putitora, Calbasu Labeo calbasu, Sil- density of fry was low, than in T2 and T3. The zooplankton abun- ver Barb Puntius gonionotus, and Sharptooth Catfish Clarias dances were consistently higher than those of phytoplankton. Similar results have also been recorded in various grow-out fish gariepinus. The FCR values in T1 are significantly lower than and fry or fingerling rearing ponds (Wahab et al. 1994; Haque those in T2 and T3. The FCRs of the present study are lower than the FCR values reported by many workers (Reddy and Katre et al. 1998; Kohinoor et al. 1999; Chakraborty et al. 2003; 1979; Das and Ray 1989; De Silva and Davy 1992; Islam 2002; Rahman et al. 2008, 2009). Higher plankton concentrations Islam et al. 2002; Rahman et al. 2005a). However, the lower FCR in water normally indicate higher productivity of the pond. values in this study indicate better food utilization efficiency, de- The higher abundance of zooplankton than phytoplankton spite the values increasing with higher stocking densities. might be due to regular manuring with organic fertilizer (cow The temperature of the experimental ponds was within the dung) (Islam 2002; Keshavanath et al. 2002; Rahman et al. suitable range for nursery ponds and agrees well with the find- 2005a; Rahman et al. 2008). Nevertheless, the highest plankton ings of Haylor and Mollah (1995), Mollah and Hossain (1998), abundances in T1 might lead to the productivity of pond water

TABLE 3. Growth performance measures, survival, feed utilization, and production of Guizza ﬁngerlings after 8 weeks of rearing; values are means ± SDs. Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 Numbers in the same row with the same lowercase letter are not signiﬁcantly different (P > 0.05).

Treatments

Measurements T1 (100,000 fry/ha) T2 (150,000 fry/ha) T3 (200,000 fry/ha) Initial length (cm) 1.74 ± 0.27 z 1.74 ± 0.27 z 1.74 ± 0.27 z Final length (cm) 9.15 ± 0.12 z 8.09 ± 0.10 y 7.20 ± 0.13 x Initial weight (g) 0.20 ± 0.03 z 0.20 ± 0.03 z 0.20 ± 0.03 z Final weight (g) 4.18 ± 0.06 z 2.87 ± 0.14 y 2.23 ± 0.06 x Weight gain (g) 3.98 ± 0.06 z 2.68 ± 0.11 y 2.02 ± 0.08 x Length gain (cm) 7.41 ± 0.12 z 6.35 ± 0.07 y 5.46 ± 0.13 x Speciﬁc growth rate (%/day) 5.43 ± 0.03 z 4.78 ± 0.07 y 4.31 ± 0.05 x Feed conversion ratio 1.68 ± 0.08 x 2.25 ± 0.11 y 2.65 ± 0.09 z Survival (%) 69.25 ± 3.93 z 57.61 ± 3.09 y 45.17 ± 3.30 x Production (number/ha) 69,250 ± 3,929 x 86,415 ± 4,634 y 90,333 ± 6,602 z 102 RAHMAN ET AL.

TABLE 4. Costs and beneﬁts (in takas [Tk]; US$1 = Tk 77.40 as of December 2013) from the rearing of Guizza ﬁngerlings in 1-ha earthen ponds for a period of 8 weeks.

Treatments

Items T1 (100,000 fry/ha) T2 (150,000 fry/ha) T3 (200,000 fry/ha) Costs Pond lease (Tk 30,000.00/ha/yr) 4,615 4,615 4,615 Lime (Tk 6.00/kg) 1,500 1,500 1,500 Cow dung (Tk 0.35/kg) 2,975 2,975 2,975 Dipterex (Tk 800.00/kg) 6,848 6,848 6,848 Fry (Tk 0.50/piece) 50,000 75,000 100,000 Feeds: Nursery (Tk 25.00/kg) 2,058 2,389 2,695 Starter-1 (Tk 23.00/kg) 20,663 22,030 21,561 Labor (Tk 70.00/day) 7,840 7,840 7,840 Miscellaneous 2,000 2,000 2,000 Total costs 98,499 125,197 150,034 Benefits Gross benefits (fingerlings)a 417,001 345,660 270,999 Net benefits (gross benefits minus total costs) 318,751 220,463 120,965

a The prices for ﬁngerlings ﬁxed by the Bangladesh Fisheries Research Institute, where the experiment was conducted, were as follows: Tk 6.00/piece forT1, Tk 4.00/piece for T2, and Tk 3.00/piece for T3.

thereby enhancing the growth performance, feed efficiency, and with those reported by Kohinoor et al. (1994), Rahman and Rah- survival of Guizza fry to a greater extent (see references above). man (2003), Rahman et al. (2005a), and Chakraborty and Mirza In our study, significantly higher numbers of fingerlings were (2007) in various carp and barb nursery ponds. produced in ponds stocked with 200,000 fry/ha (T3) than in the Owing to the environmental and man-made interventions in ponds with 150,000 fry/ha (T2) and 100,000 fry/ha (T1). Despite the aquatic ecosystem, natural spawning and feeding grounds this, consistently higher net benefits were obtained from T1 than of this important indigenous catfish species have been drasti- from T2 and T3. The higher market price of the larger fingerlings cally degraded. Under the prevailing situation, production of produced in ponds with 100,000 fry/ha substantially increased adequate healthy and quality seeds through the application of the net benefits compared with the net benfits obtained from the our present captive natural propagation and nursery rearing tech- smaller fingerlings produced at higher stocking densities. Simi- niques might have great implications for the protection of Guizza lar results were obtained by Rahman et al. (2005a), Chakraborty from extinction as well as for their stock enhancement and bio- and Mirza (2007), and Rahman et al. (2013) in their fingerling diversity conservation. On the other hand, this study represents rearing experiments with critically endangered Putitor Mahseer, the first successful attempt to produce fingerlings of Guizza Bata Labeo bata, and Gulsha Mystus cavasius, respectively, in through natural propagation. It showed that the growth, sur-

Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 earthen nursery ponds. Overall, highest production and bene- vival, production, and net benefits of Guizza fingerlings were fits were obtained in ponds stocked with 100,000 fry/ha, where inversely related to the stocking densities of fry. In all respects, growth and survival rates of fingerlings were higher than in the fish at a stocking density of 100,000 fry/ha performed better ponds stocked with higher densities. During the entire exper- than those at the higher stocking densities. Therefore, in the iment, the physicochemical characteristics of the pond water light of the present study, a stocking density of 100,000 fry/ha were within the suitable range for nursery management; the should be recommended for the rearing of Guizza fingerlings growth of fingerlings was dependent on the quality and quan- for 8 weeks in earthen nursery ponds. tity of food available. However, availability of plankton varied among the ponds, being more abundant in ponds stocked with lower densities. In the present experiment, the amount of feed ACKNOWLEDGMENTS supplied in different ponds was based on the number of fry We would like to thank the Director General and Chief Sci- stocked and the amount provided was kept at the same level. entific Officer, Bangladesh Fisheries Research Institute, My- Hence, the observed low growth at higher stocking densities mensingh, for providing experimental facilities at the pond could be due to less availability of natural food (plankton) and complex of the Freshwater Station, Mymensingh, as well as some changes in environmental characteristics (Kohinoor et al. the research assistants and field workers for their technical help 1997). The findings of the present study are in close agreement in successfully conducting the experiment. BREEDING AND SEED PRODUCTION OF GIANT RIVER CATFISH 103

REFERENCES gonionotus)andMirrorCarp(Cyprinus carpio var. specularis) in mini ponds. APHA (American Public Health Association). 1992. Standard methods for the Bangladesh Journal of Zoology 25:129–133. examination of water and wastewater. APHA, Washington, D.C. Kohinoor, A. H. M., M. S. Islam, N. Begum, and M. G. Hussain. 1999. Produc- Boyd, C. E. 1982. Water quality management for pond fish culture. Elsevier, tion of Thai Sharpunti (Puntius gonionotus Bleeker) in polyculture with carps Amsterdam. using low-cost feed. Bangladesh Journal of Fisheries Research 3:157–164. Brown, M. E. 1957. Experimental studies on growth. Pages 361–400 in M. E. Kohinoor, A. H. M., M. L. Islam, M. A. Wahab, and S. H. Thilsted. 1998. Effect Brown, editor. The physiology of fishes, volume 1. Academic Press, New of Mola (Amblypharyngodon mola Ham.) on the growth and production York. of carp in polyculture. Bangladesh Journal of Fisheries Research 2:111– Castell, J. D., and K. Tiews, editors. 1980. Report of the EIFAC, IUNS and ICES 126. working group on standardization of methodology in fish nutrition research. Kohinoor, A. H. M., M. A. Wahab, M. L. Islam, and S. H. Thilsted. 2001. EIFAC (European Inland Fisheries Advisory Commission) Technical Paper Culture potentials of Mola (Amblypharyngodon mola), Chela (Chela cachius) 36. and Punti (Puntius sophore) under monoculture system. Bangladesh Journal Chakraborty, B. K., M. I. Miah, M. J. A. Mirza, and M. A. B. Habib. 2003. of Fisheries Research 5:123–134. Rearing and nursing of local Sarpunti, Puntius sarana (Hamilton), at different Mollah, M. F. A., and Z. Hossain. 1998. Stocking density effects on growth and stocking densities. Pakistan Journal of Biological Sciences 6:797–800. survival of African Catfish (Clarias gariepinus) fry in net cages. Bangladesh Chakraborty, B. K., and M. J. A. Mirza. 2007. Effect of stocking density on Journal of Agricultural Sciences 25:5–9. survival and growth of endangered Bata, Labeo bata (Hamilton–Buchanan), Muchlisin, Z. A., R. Hashim, and A. S. Chong. 2004. Preliminary study on the in nursery ponds. Aquaculture 265:156–162. cryopreservation of tropical bagrid catfish (Mystus nemurus) spermatozoa; Das, I., and A. K. Ray. 1989. Growth performance of Indian major carps Labeo the effect of extender and cryoprotectant on the motility after short-term rohita (Ham.) on duckweed incorporated pelleted feed: a preliminary study. storage. Theriogenology 62:25–34. Journal of the Inland Fisheries Society of India 21:1–6. Rahman, A. K. A. 2005. Freshwater fishes of Bangladesh, 2nd edition. Zoolog- De Silva, S. S., and F. B. Davy. 1992. Fish nutrition research for semi-intensive ical Society of Bangladesh, University of Dhaka, Dhaka. culture systems in Asia. Asian Fisheries Science 5:129–144. Rahman, M. A., A. Arshad, and S. M. N. Amin. 2011. Evaluation of growth and Dewan, D., M. A. Wahab, M. C. M. Beveridge, M. H. Rahman, and B. K. production of the threatened Giant River Catfish, Sperata seenghala (Sykes) Sarkar. 1991. Food selection, electivity and dietary overlap among planktiv- in polyculture with indigenous major carps. African Journal of Biotechnology orous Chinese and Indian major carp fry and fingerlings grown in extensively 10:2999–3008. managed, rain-fed ponds in Bangladesh. Aquaculture Research 22:277–294. Rahman, M. A., M. A. Mazid, M. R. Rahman, M. N. Khan, M. A. Hossain, Haque, M. S., M. A. Wahab, M. I. Wahid, and M. S. Haq. 1998. Impacts of and M. G. Hussain. 2005a. Effect of stocking density on survival and growth Thai Silver Barb (Puntius gonionotus Bleeker) inclusion in the polyculture of critically endangered Mahseer, Tor putitora (Hamilton), in nursery ponds. of carps. Bangladesh Journal of Fisheries Research 2:15–22. Aquaculture 249:275–284. Haque, M. Z., M. A. Rahman, M. M. Hossain, M. A. Rahaman, and M. Aminur. Rahman, M. A., K. M. A. Uddin, and M. Zaher. 2005b. Development of artificial 1994. Effect of stocking densities on the growth and survival of Mirror Carp breeding techniques for Long-whiskered Catfish, Sperata aor and Giant River (Cyprinus carpio var. specularis) fry in rearing ponds. Bangladesh Journal of Catfish, Sperata seenghala of Bangladesh. Bangladesh Journal of Fisheries Zoology 22:109–116. Research 9:11–12. Haylor, G. S., and M. F. A. Mollah. 1995. Controlled hatchery production of Rahman, M. A., M. Zaher, and K. M. Azimuddin. 2008. Evaluation of growth, African Catfish, Clarias gariepinus: the influence of temperature on early survival, and production of an endangered fish, Labeo gonius (Hamilton) development. Aquatic Living Resources 8:431–438. fingerlings in earthen nursery ponds. Journal of Applied Aquaculture 20:62– Houde, E. D. 1975. Effects of stocking density and food density on survival, 78. growth and yield of laboratory-reared larvae of Sea Bream Archosargus rhom- Rahman, M. A., M. Zaher, and K. M. Azimuddin. 2009. Development of fin- boidalis (L.) (Sparidae). Journal of Fish Biology 7:115–127. gerling production techniques in nursery ponds for the critically endangered Islam, M. S. 2002. Evaluation of supplementary feeds for semi-intensive pond Reba Carp, Cirrhinus ariza (Hamilton, 1807). Turkish Journal of Fisheries culture of mahseer, Tor putitora (Hamilton). Aquaculture 212:263–276. and Aquatic Sciences 9:165–172. Islam, M. S., S. Dewan, M. G. Hussain, M. A. Hossain, and M. A. Mazid. 2002. Rahman, M. A., M. Zaher, K. M. Azimuddin, S. Yeasmine, M. M. Khan, and Feed utilization and wastage in semi-intensive pond culture of mahseer, Tor A. Arshad. 2013. Stocking density effects on growth and production of the putitora (Ham.). Bangladesh Journal of Fisheries Research 6:1–9. threatened silurid catfish, Mystus cavasius (Hamilton) fingerlings in nursery IUCN (International Union for Conservation of Nature). 2013. IUCN red list ponds. Aquaculture Research 44:1132–1139. Downloaded by [Department Of Fisheries] at 20:12 15 June 2014 of threatened species, version 2013.1. IUCN Global Species Programme Red Rahman, M. R., and M. A. Rahman. 2003. Studies on the growth, survival List Unit, Cambridge, UK. and production of Calbasu (Labeo calbasu Ham.) fry at different stocking Jayaram, K. C. 1977. Aid to the identification of the siluroid fishes of India, densities in primary nursing. Bulletin of the Faculty of Science, University Burma, Sri Lanka, Pakistan, and Bangladesh: Bagridae. Records of the Zoo- of the Ryukyus, Japan 76:245–255. logical Survey of India, Miscellaneous Publication 8, Kolkata. Rahman, M. S. 1992. Water quality management in aquaculture. BRAC Jayaram, K. C. 2002. The freshwater fishes of the Indian region. Narendra Prokashana, Dhaka, Bangladesh. Publishing House, Delhi, India. Reddy, S. R., and S. Katre. 1979. Growth rate and conversion efficiency of Jhingran, V. G. 1991. Fish and fisheries of India, 3rd edition. Hindustan Pub- the air-breathing catfish, Heteropneustes fossilis, in relation to ration size. lishing, Delhi, India. Aquaculture 18:35–40. Keshavanath, P., B. Gangadhar, T. J. Ramesh, A. A. van Dam, M. C. M. Stirling, H. P., editor. 1985. Chemical and biological methods of water analysis Beveridge, and M. C. J. Verdegem. 2002. The effect of periphyton and sup- for aquaculturists. Institute of Aquaculture, University of Stirling, Stirling, plemental feeding on the production of the indigenous carps Tor khudree and UK. Labeo fimbriatus. Aquaculture 213:207–218. Talwar, P. K., and A. G. Jhingran. 1991. Inland fishes of India and adjacent Kohinoor, A. H. M., M. Z. Haque, M. G. Hussain, and M. V. Gupta. 1994. countries, volume 2. Oxford and IBH Publishing, New Delhi. Growth and survival rate of Thai Punti, Puntius gonionotus (Bleeker) spawn Tripathi, S. D. 1996. Present status of breeding and culture of catfishes in South in nursery ponds at different stocking densities. Journal of the Asiatic Society Asia. Aquatic Living Resources 9(Supplement 1):219–228. of Bangladesh Science 20:65–72. Wahab, M. A., Z. F. Ahmed, M. S. Haq, and M. Begum. 1994. Compatibility Kohinoor, A. H. M., M. A. Hossain, and M. G. Hussain. 1997. Semi-intensive of Silver Carp in the polyculture of cyprinid fishes. Progressive Agriculture culture and production cost of Pabda (Ompok pabda) with Rajpunti (Puntius 5:221–227. This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:13 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Critical Thermal Maxima of Two Geographic Strains of Channel and Hybrid Catfish Heather A. Stewarta & Peter J. Allena a Department of Wildlife, Fisheries and Aquaculture, Mississippi State University, Box 9690, Mississippi State, Mississippi 39762-9690, USA Published online: 15 Apr 2014.

To cite this article: Heather A. Stewart & Peter J. Allen (2014) Critical Thermal Maxima of Two Geographic Strains of Channel and Hybrid Catfish, North American Journal of Aquaculture, 76:2, 104-111, DOI: 10.1080/15222055.2013.856827 To link to this article: http://dx.doi.org/10.1080/15222055.2013.856827

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 North American Journal of Aquaculture 76:104–111, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.856827

ARTICLE

Critical Thermal Maxima of Two Geographic Strains of Channel and Hybrid Catﬁsh

Heather A. Stewart and Peter J. Allen* Department of Wildlife, Fisheries and Aquaculture, Mississippi State University, Box 9690, Mississippi State, Mississippi 39762-9690, USA

Abstract Critical thermal maxima have been used extensively to provide physiologically and ecologically valuable reference points that identify early signs of thermal stress. In catfish pond culture, daily temperature maxima up to 36◦Cand daily fluctuations of as much as 6◦C are observed. These extreme conditions will probably be exacerbated by the effects of global climate change. Channel Catfish Ictalurus punctatus have a broad natural distribution from southern Canada to northern Mexico. If regional genetic differences could cause strains with a southern distribution to have greater thermal tolerance than strains with a northern distribution, and consequently a greater critical thermal maximum, then hybrid catfish (Channel Catfish × Blue Catfish I. furcatus) strains might be expected to have greater critical thermal maxima than their respective Channel Catfish strains because Blue Catfish have a more southern range of distribution. To examine this, we quantified differences of acute thermal tolerance in two geographically distinct strains of Channel Catfish and their hybrid crosses with an industry standard strain of Blue Catfish. The catfish were subjected to water temperature increases of 2.0 ± 0.1◦C · h−1 until they lost equilibrium. Standard length, which ranged from 162 to 320 mm, had a significant effect on survival, survival being greater in larger fish. The critical thermal maximum ranged from 38.6◦C to 40.3◦C. Southern Channel Catfish tolerated higher temperatures than northern Channel Catfish did, and both strains of Channel Catfish tolerated higher temperatures than their hybrid catfish strains did. This study indicates that geographically distinct catfish strains differ in acute thermal tolerance and suggests heritability for this trait, as evidenced by similar responses in Channel Catfish and their corresponding hybrid crosses with Blue Catfish.

Environmental conditions such as temperature affect its introduction, critical thermal methodology has been modified aquaculture productivity by altering fish growth, reproductive and standardized to be defined as the mean maximum thermal Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 capacity, physiology, behavior, immune system function, and point a fish can reach over a slow and constant exposure to mortality (Brett 1956, 1979; McCauley and Beitinger 1992; increasing temperature until locomotive activity becomes dis- Brandt 1993; Le Morvan et al. 1998; Lang et al. 2003; Brander organized and the fish loses its ability to maintain dorsoventral 2007; Ficke et al. 2007). Since the 1920s, research has been orientation (Cowles and Bogert 1944; Cox 1974; Becker and ongoing to determine the temperature tolerance of various fish Genoway 1979; Currie et al. 1998). In natural environments, species (Hathaway 1927) to cold and heat. Determinations of fish typically encounter temperatures outside of their tolerance heat tolerance, primarily use two methods: incipient upper lethal range for brief periods, known as acute fluctuations. Examin- temperature (IULT; Fry 1947; Allen and Strawn 1968) and crit- ing thermal sensitivity to acute fluctuations using CTmax can ical thermal maximum (CTmax; Becker and Genoway 1979; provide guidelines for best culture management practices (Brett D´ıaz and Buckle¨ 1999). CTmax provides a more ecologically 1956; Hutchison 1976). relevant point of reference, identifying earlier signs of stress Pond aquaculture systems are extremely prevalent in the U.S. than IULT (Paladino et al. 1980; D´ıaz and Buckle¨ 1999). Since catfish industry, more than 95% of Channel Catfish Ictalurus

*Corresponding author: [email protected] Received August 5, 2013; accepted October 14, 2013

104 CRITICAL THERMAL MAXIMA OF CHANNEL AND HYBRID CATFISH 105

punctatus being grown in ponds (Brune et al. 2004; Jackson METHODS 2004). Water temperatures in ponds cannot be controlled, and Fish source and acclimation.—Two strains of Channel shallow depths (<1.5 m) may have large daily fluctuations (i.e., Catfish were used: Delta Select (from the Mississippi Delta, ◦ ◦ 3–6 C), reaching high daily maxima up to 36–40 C(Waxetal. Mississippi) and Red River (from the Red River, North Dakota). 1987; Arnold et al. 2013; Liu et al. 2013). These strains were used because of their disparate geographic Catfish aquaculture utilizes Channel Catfish and hybrids be- distributions. The Delta Select is a commercial strain developed tween female Channel Catfish and male Blue Catfish I. fur- at the Thad Cochran National Warmwater Aquaculture Cen- catus, both species that occupy broad geographic ranges. The ter (NWAC) in Stoneville, Mississippi. It is a first-generation natural geographic distribution of Channel Catfish ranges from composite of families collected from 80 to 100 random spawns southern Canada to northern Mexico, encompassing a natu- from 10 commercial catfish farms in the Mississippi Delta. The ◦ ral thermal range of 5–35 C (McCauley and Beitinger 1992; Red River strain was originally collected in 1988 from the Red Bennett et al. 1998; Tavares-Dias and Moraes 2007). Female River, North Dakota (Hudson Bay drainage), where it does not Channel Catfish are frequently hybridized with male Blue Cat- regularly experience the same high temperatures as the Delta fish to produce offspring with faster growth, better feed con- Select strain and is not a typical strain utilized in the catfish version efficiency, more uniform size at harvest, greater toler- industry (Li et al. 1998). In addition, two hybrid strains, crosses ance of low oxygen levels and crowding in pond systems, and between D&B Blue Catfish and each of the previously men- greater resistance to enteric septicemia (Andrews and Stickney tioned pure Channel Catfish strains, were used. The D&B Blue 1972; Wolters and Johnson 1994; Bosworth et al. 1998; Li et al. Catfish strain is commonly used in aquaculture to produce 2001). The natural distribution of Blue Catfish extends further hybrids for the catfish industry (Xu et al. 2012). The D&B south than that of Channel Catfish, ranging from the Missis- broodstock catfish used in this experiment were from the U.S. sippi River basin and the Gulf of Mexico coast through Mexico Department of Agriculture Catfish Genetics Research Unit and and into Guatemala and Belize (Graham 1999). Because fish were originally obtained from Dycus Farms, Arkansas (Xu et al. are poikilotherms, an understanding of their thermal capacity is 2012). Similarly, the Delta Select × D&B blue (Delta Select hy- crucial for research on habitat selection, metabolism, growth brid) is representative of commonly used strains in southeastern rates, and reproductive migration (Crawshaw and Hammel U.S. commercial catfish production. 1974). Broodfish were strip-spawned at NWAC in early May 2012 Previous studies have failed to find any acute thermal toler- following protocols described by Bosworth et al. (2011). The ance differences between Channel Catfish strains (Hart 1952; same six Delta Select and five Red River females were used Allen and Strawn 1968; Cheetham et al. 1976; Reutter and for both channel and hybrid strains to minimize individual dif- Herdendorf 1976; Bennett et al. 1998). However, these studies ferences. After fingerlings were eating a formulated diet, they either did not directly compare strains, used different heating were transported in late June to the Mississippi State University rates, or varied in methodology or were limited in inference South Farm Aquaculture Facility (South Farm), approximately by small sample sizes. Therefore, to address geographic influ- 30 d after hatching. ences on CTmax in catfish, research is needed that allows for At the South Farm, the fingerlings underwent eight months of direct comparison, that is, utilizes strains that have clearly de- acclimation at 30 ± 1◦C in four circular tanks (430 L) divided fined breeding lines, are raised in an aquaculture setting, and are by strain. The water supplying the tanks was nonchlorinated, subjected to uniform conditions (e.g., acclimation temperature) aerated well water maintained in a recirculating system. A sam- to. Although it would be highly beneficial to the aquaculture ple of 10 fingerlings from each strain was measured and weighed

Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 industry, little is known about CTmax values of hybrid catfish to obtain a baseline wet weight to calculate the 2% body weight and whether thermal tolerance traits are heritable. feeding rate. Food was withheld 24 h before trials to ensure a The objective of this study was to quantify acute thermal postabsorptive state and reduce stress during handling (Barton tolerance differences in two geographically distinct strains of et al. 1988; Mørkøre et al. 2008). Channel Catfish and their hybrid crosses with Blue Catfish, using rates of temperature increase that mimic pond condi- System design.—The experimental set-up of the acute thermal tions. It was hypothesized that regional genetic differences trial had three primary components: a sump tank, a header tank, would cause strains with an originally southern natural range and an insulated fiberglass tank holding the experimental aquaria (Delta Select Channel Catfish and Delta Select × D&B Blue (Figure 1). The sump was a 149 × 100 × 60 cm oval tank Catfish) to have greater thermal tolerance than strains with a with two heaters inside: one 4000 W, 200–250 V heater and natural northern range (Red River Channel Catfish and Red one 1700 W, 100–140 V heater (SmartOne; Process Technol- River × D&B Blue Catfish) and consequently a greater CT- ogy, Mentor, Ohio). Two inline pumps (model 3 utility pump; max. The hybrid catfish strains were also expected to have a Aquatic Eco-systems, Inc., Apopka, Florida) were used to pump higher CTmax than their respective Channel Catfish strains since heated water to the header tank and a third pump to circulate Blue Catfish have a more southern distribution than Channel water around the heaters for consistent thermal regulation. The Catfish do. sump tank was connected to the header tank by inflow polyvinyl 106 STEWART AND ALLEN

FIGURE 1. Acute thermal trial experimental set-up: top view diagram of the setup used to test critical thermal maxima.

chloride (PVC) tubing with an additional overflow outlet that Acute thermal trials were conducted on 30 fish per strain over returned to the sump tank via PVC tubing. All junctions were fit- 14 trials, for a total of 120 fish. Only three strains were tested ted with uniseals to prevent leaks. The header tank was a 35.5 × per day, with three replicates of each to minimize daily fluctua- Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 84 × 61 cm rectangular tank (PT-564; Polytank, Inc., Litchfield, tions, and tested strains were alternated to ensure equal trials for Minnesota) covered in thick insulation. Inside the header tank, all four strains. Fish were placed in the trial set-up at 30◦C ± three airstones aerated the well water prior to its entering the 0.03◦C for 30–40 min to adjust to the new setting before acute experimental aquaria. Tubing covered with insulation was used thermal trials began according to the acclimation methodol- to supply the heated, oxygenated water from the header tank to ogy of D´ıaz and Buckle¨ (1999). During the summer months the experimental aquaria. Water flow was regulated with needle of 2009, data from six different catfish aquaculture ponds valves. Within the insulated fiberglass tank (212 × 61 × 56 cm; found water temperatures increased by 1–2◦C each hour (E. L. model MT-700 min-o-cool; Frigid Units, Inc., Toledo, Ohio) Torrans, U.S. Department of Agriculture Agricultural Research were nine 9.5-L aquaria (30.5 × 15.25 × 20 cm), each cov- Service, unpublished data from 2010). Based on these data, a ered with reflective bubble insulation, and each with an external heating rate of 2◦C per hour was determined to give an accu- standpipe that maintained a water depth of 17 cm. The external rate estimation of CTmax. Water temperature was increased at standpipe allowed the aquaria to be submerged in a water bath to arateof2.0◦C ± 0.1◦Ch−1 until loss of equilibrium (LOE) further insulate internal temperatures. Water from the insulated occurred, to achieve an environmentally realistic rate of temper- fiberglass tank drained into the sump tank by a PVC pipe, where ature increase (Perez-Casanova´ et al. 2008). LOE was defined it was reheated and oxygenated. as the failure of a fish to retain dorsoventral orientation for 1 min CRITICAL THERMAL MAXIMA OF CHANNEL AND HYBRID CATFISH 107

TABLE 1. Catﬁsh strain body size differences depicted by minimum, maximum, and mean ± SE weight and length. Each strain used in the experiment had a sample size of 30 individuals.

Weight (g) Length (mm) Catﬁsh strain Min. Max. Mean ± SE Min. Max. Mean ± SE Southern Channel 60 241.9 162.8 ± 7.7 200 320 275.2 ± 4.9 Southern Hybrid 34.1 235.6 120.3 ± 8.9 162 290 235.8 ± 6.1 Northern Channel 59.6 258.2 137.3 ± 10.9 198 315 251.5 ± 6.8 Northern Hybrid 50 246.8 105.3 ± 7.5 167 275 224.0 ± 4.8

(Bennett and Beitinger 1997). CTmax for each individual fish RESULTS was recorded as the water temperature at which LOE was ob- Significant differences in length of fish used in the exper- served. Behavioral responses to increasing temperature were iment between geographic range (P < 0.01) and catfish type monitored and recorded throughout the trial. Typical observa- (P < 0.01) were observed; however, no significant differences tions were erratic swimming, surfacing, aquatic surface respi- were observed in geographic range × catfish type interaction ration as evidenced by bubbles, rapid ventilation rate, irregular (P = 0.30). Catfish from the Red River (northern) were shorter opercular movement, splashing, and muscular spasms. (295 mm) than the Delta Select (southern; 305 mm). Hybrid Dissolved oxygen levels were monitored every 30 min and catfish were shorter (282.5 mm) than their respective Channel remained close to saturation (mean ± SE, 100% ± 0.96%, 7.14 Catfish types (317.5 mm). Notably, length had no significant ± 0.09 mg/L; always ≥93%) in all tanks throughout the trial. effect on CTmax (P = 0.36). Significant effects of geographic Temperature was recorded with digital handheld thermometers range (P < 0.01) and catfish type (P < 0.01) on CTmax were ◦ to the nearest 0.1 C every 30 min throughout the trial. After found but the geographic range × catfish type interaction was CTmax was reached, fish were removed, measured (standard not significant (P = 0.36). Other nonsignificant interactions length to the nearest 0.1 mm), and weighed (to the nearest were length × catfish type, length × range, and length × catfish 0.5 g), and then directly returned to their original acclimation type × range (P > 0.34). Northern catfish had lower CTmax ◦ temperature (30 ± 1 C) to allow recovery (Table 1). Fish were values (39.08◦C) than southern catfish (39.25◦C; Figure 2A). placed in different recovery tanks to distinguish strains and Hybrid catfish had lower CTmax values (39.07◦C) than the individuals and then monitored the next 24 h for survival (Currie respective Channel Catfish with which they were crossed et al. 1998). Mortalities were recorded, dead fish were removed, (39.26◦C; Figure 2B). and mortality rates were determined. Length had a significant effect on survival (P = 0.02), survival increasing as the fish increased in size (Figure 3). An apparent regional effect on survival was observed—survival was Statistical analysis.—An analysis of covariance (ANCOVA)us- higher in southern catfish—but this was not statistically sig- ing the proc glm procedure in SAS (version 9.2; SAS Institute, nificant (P = 0.08). During acute trials, fish exhibited erratic Inc., Cary, North Carolina) was used to examine effects of geo- swimming and irregular opercular movement, as had been ob- graphic range (north or south), catfish type (channel or hybrid), served in the study by Hlohowskyj and Wissing (1985). Some

Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 and the interaction between all main effects on CTmax with fish never lost equilibrium but instead stopped breathing; that a covariate of standard length. Because weight and length are is, death was the endpoint, similar to the report of Bettoli et al. highly correlated, only length was used to measure size. A gen- (1985). These latter fish (n = 11) were not attributed to a spe- eral linear model using the proc logistic procedure in SAS was cific treatment group and had a slightly higher thermal endpoint; used to determine significance of range, catfish type, and length therefore, they were removed from statistical analyses. variables on survival after the acute thermal trials. The forward selection technique was used to remove nonsignificant variables DISCUSSION from the model. A logistic regression predictive model was then Regional differences in CTmax were observed, supporting generated from the significant factors. A two-way analysis of the hypothesis that geographic distribution would affect thermal variance (ANOVA) was used to determine effects of geographic tolerance. As predicted, catfish with a southern distribution had range, catfish type, and the interaction of these variables on a greater CTmax than did catfish with a northern distribution. length. The Shapiro–Wilk and Kolmogorov–Smirnov tests were These geographic differences in thermal tolerance were also implemented to confirm normality. A Tukey’s honestly signifi- observed in the hybrid catfish, suggesting a genetic component cant difference (HSD) post hoc test was used to compare means. for this trait. Statistical significance was determined against an α of 0.05. All CTmax values are often characterized by low variability and data are presented as means ± SEs unless otherwise noted. small changes between environmental conditions (reviewed by 108 STEWART AND ALLEN

FIGURE 3. Probability of survival relative to fish length of (A) geographically distinct catfish strains (northern and southern) and (B) all catfish 24 h after acute trials. A significant effect of length on survival (P = 0.02) was observed, the FIGURE 2. Mean (± SE) critical thermal maximum (CTmax) by (A) geo- larger fish having increased survival. A trend (P = 0.08) of southern catfish graphic range (northern and southern) and (B) type (Channel Catfish and hybrid having increased survival over northern catfish was also observed. A general catfish Channel Catfish × Blue Catfish) of catfish. Different letters indicate linear model using the proc logistic procedure in SAS was run to determine the significant differences between strains or types (ANCOVA, Tukey’s HSD post effect of length and geographic range on survival. A logistic regression predictive hoc test, P < 0.05). model was then generated for probability of survival using the equations pN =

Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 1/(1 + exp(logitN)), where logitN = 11.5804 – 0.0335length + 0.9609, and Beitinger et al. 2000), suggesting tight control of thermal toler- pS = 1/(1 + exp(logitS)), where logitS = 11.5804 – 0.0335length. (Tukey’s < ance. The low variability in CTmax values suggests tight control HSD post hoc test, P 0.05). over upper lethal temperature tolerance. For example, standard deviations (SD) in CTmax values for a group of fish subjected to 2013) and are likely to increase under the influence of climate a similar acclimation temperature are often small (Currie et al. change. 1998; Beitinger et al. 2000), such as in this study (SD: 0.2– The only known previous study on geographic differences in 0.4◦C). However, even small changes in environmental temper- acute thermal tolerance in catfish by Hart (1952) found little to ature, such as 2.5–6◦C as predicted by global climate models no geographic variation in IULT of Channel Catfish from Florida (Karl et al. 2009), can have large consequences for suscepti- and Ohio. However, the study was limited by small sample sizes ble freshwater fish species (Magnuson et al. 1979; Regier et al. (ranging from four to nine fish per site), the use of wild-caught 1990; Morgan et al. 2001). Thus, in this study, although the CT- fish held for brief periods in the laboratory (i.e., 4–5 d), and max differences between Channel Catfish strains and between greater variations in body size (51.9–436.5 g) and age. Wild- channel and hybrid catfish were small (0.4◦C), the biological caught fish may be stressed both by transfer and the environ- importance may be large, considering that current pond temper- ment of tank settings (Clearwater and Pankhurst 1997), which atures reach nearly lethal limits (Arnold et al. 2013; Liu et al. can make interpretation of the resulting data more difficult. In CRITICAL THERMAL MAXIMA OF CHANNEL AND HYBRID CATFISH 109

comparison, the present study used two strains maintained sepa- (62–264 mm). Barrionuevo and Fernandes (1995) found that rately in aquaculture settings such that their genetically distinct body size will affect critical thermal minimum (CTmin) but not heritages were retained and similar rearing environments were CTmax in Curimbata´ Prochilodus scrofa, whereas Cox (1974) ensured, presumably minimizing stress and allowing for clearer found that body size has an effect on CTmax in Bluegill Lepomis application of results to aquaculture conditions. macrochirus and Cook et al. (2006) found an inverse relation- Few studies have examined strain or population differ- ship between CTmax and body size in Striped Bass Morone ences in CTmax. In three strains of Brook Trout Salvelinus saxatiliis. In terms of the effect of body size on survival or fontinalis originating from different geographic locations but recovery, data are limited. Similar to the present study, Meffe held for multiple generations in a common hatchery environ- et al. (1995) observed improved thermal recovery with increas- ment, thermal tolerance reflected historic geographic origins ing body size in Eastern Mosquitofish. A possible explanation (McDermid et al. 2012). Specifically, strains with a more south- could be related to energy storage or a length:weight body ratio, ern origin had higher acute thermal tolerance than did those with as suggested by Ospina and Mora (2004). Therefore, recovery a more northern origin. Similarly, Barramundi Lates calcar- following high temperature exposure may rely more on the im- ifer, originating from different geographic locations and subse- mune system and body reserves; thus, larger fish may have an quently held in hatchery environments, have been found to have advantage. different extents of acute thermal tolerance, the tropical popu- A heating rate of 2◦Ch−1, comparable with that reported by lations having higher tolerance than the subtropical populations Perez-Casanova´ et al. (2008), was used in this study because (Newton et al. 2010). Further, Meffe et al. (1995) found that in it was representative of the rates observed in Channel Cat- Eastern Mosquitofish Gambusia holbrooki traits for improved fish aquaculture ponds. In a previous study on Channel Catfish, thermal tolerance were genetically heritable. In this study, the Currie et al. (1998) used a heating rate of 0.3◦Cmin−1, which differences in CTmax values observed between the Channel is more standard for the literature but does not replicate envi- Catfish strains (i.e., the southern strain having a greater CTmax ronmental conditions. CTmax may be affected by slower rates than the northern one) were similarly observed in the hybrid of temperature change, possibility allowing low levels of ac- cross for both strains, suggesting heritability of acute thermal climation. Interestingly, the CTmax range for both Channel tolerance. The genetic underpinnings and the strength for selec- Catfish strains in this study overlapped with those found for tion for this trait are unknown, but suggest that further research Channel Catfish (40.3◦C) by Currie et al. (1998). These results on genetic control and expression are needed. are consistent with the literature suggesting that CTmax for Contrary to the original hypothesis, hybrid catfish had lower the acclimation temperature of 29.5 ± 0.5◦C (mean ± SD) CTmax values than Channel Catfish. Because Blue Catfish have would be around 40◦C (Cheetham et al. 1976; Bennett et al. a natural distribution that extends further south than that of 1998). Although Becker and Genoway (1979) recommended Channel Catfish, presumably exposing the former to higher standardization criteria, no universally adopted heating rate for temperatures, it was hypothesized that they would demonstrate determining CTmax has yet been established. For example, the a greater thermal tolerance or CTmax than Channel Catfish. rates used by Cheetham et al. (1976), Watenpaugh et al. (1985), However, verified knowledge about the species’ thermal toler- Kita et al. (1996), and Bennett et al. (1998) were 1◦Cmin−1, ance is limited. Possible explanations for the results include the 0.3◦Cmin−1,5◦Ch−1, and 0.15◦Cmin−1, respectively. Beitinger relative exposure of these species to thermal variability, reduced et al. (2000) stated that a correct heating rate needs to be slow temperature tolerance in Blue Catfish, and epistasis (Burke and enough that the core temperature of the fish being studied does Arnold 2001). In terms of thermal variability, Blue Catfish oc- not lag behind the water temperature, yet not so fast that the

Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 cur in subtropical to tropical areas, where water temperatures fish do not have time to reacclimate thermally during the trial. are less variable in seasonal fluctuation than those in temperate According to Jobling (1981), using different heating rates can areas, where Channel Catfish occur (Ficke et al. 2007). This ther- lead to substantial variations in values of CTmax. However, the moplasticity in Channel Catfish would be beneficial for tolerat- combined results of this study and that of Currie et al. (1998) do ing predicted seasonal temperature fluctuation increases (Regier not support this generalization for channel and hybrid catfish. et al. 1990). It is also possible that Blue Catfish have a lower Rapid changes in water temperatures can lead to thermal CTMax than Channel Catfish, although this has not been tested. shock, which may affect thermoregulatory physiology and be- A third possibility is epistatic effects, which may occur in hy- havior. As ectotherms, catfish rely on environmental tempera- bridized organisms and can lead to asymmetrical hybrid fitness tures for metabolic rate and physiological needs (Morgan et al. (Rhode and Cruzan 2005). 2001). Preferred and avoided temperature ranges may be mod- Although length did not have an effect on CTmax, it did ified due to temperature-induced damage to the central periph- affect survival after thermal trials. Similarly, Bennett et al. eral receptors or preoptic region (Crawshaw and Hammel 1974; (1998) found no relationship between length and CTmax in Prosser and Nelson 1981). If the thermal shock is below lethal Channel Catfish of the same age range. In contrast, Recsetar levels, fish may be able to alter their physiology to a point et al. (2012) found no effect of length on survival of Chan- at which alternative habitats can be utilized (Bevelhimer and nel Catfish, possibly due to the smaller size of fish examined Bennett 2000; Browse and Xin 2001; Portner¨ 2002). CTmax 110 STEWART AND ALLEN

is biologically important, because at this temperature fish are Bennett, W. A., and T. L. Beitinger. 1997. Temperature tolerance of the unable to escape conditions that will quickly lead to their death Sheepshead Minnow, Cyprinodon variegatus. Copeia 1997:77–87. (Beitinger et al. 2000). Climate change is a concern for biota Bennett, W. A., R. W. Mccauley, and T. L. Beitinger. 1998. Rates of gain and loss of heat tolerance in Channel Catfish. Transactions of the American Fisheries of all ecosystems; however, freshwater fish are more vulnerable Society 127:1051–1058. than some organisms to increasing temperatures due to limited Bettoli, P. W., W. H. Neill, and S. W. Kelsch. 1985. Temperature preference and thermal refuge (Morgan et al. 2001). Catfish production ponds heat resistance of Grass Carp, Ctenopharyngodon idella (Valenciennes), Big- provide no alternative habitat for fish to utilize during times of head Carp, Hypophthalmichthys nobilis (Gray), and their F1 hybrid. Journal thermal stress; therefore, the fish must acclimate or die. of Fish Biology 27:239–247. Bevelhimer, M., and W. Bennett. 2000. Assessing cumulative thermal stress in In conclusion, this study indicates that catfish geographic fish during chronic intermittent exposure to high temperatures. Environmental strains (i.e., northern versus southern origin) and types (i.e., Science and Policy 3(Supplement 1):211–S216. Channel Catfish versus hybrid catfish) differ in short-term ther- Bosworth, B. G., N. Chatakondi, and J. Avery. 2011. Producing hybrid cat- mal tolerance. Moreover, thermal recovery improved with in- fish fry: workshop manual. U.S. Department of Agriculture, Agricultural creasing body size. To obtain a clearer and broader representa- Research Service, and Mississippi State University, National Warmwater Aquaculture Center, Mississippi State. Available: http://tcnwac.msstate.edu/ tion of long-term temperature effects on survival and growth in 2011%20Hybrid%20Catfish%20Manual.pdf. (December 2013.) and among various strains of channel and hybrid catfish, chronic Bosworth, B. G., W. R. Wolters, D. J. Wise, and M. H. Li. 1998. Growth, elevated thermal effects should be examined (Jobling 1981). Fi- feed conversion, fillet proximate composition and resistance to Edwardsiella nally, given the results of this study that imply genetic control of ictaluri of Channel Catfish, Ictalurus punctatus (Rafinesque), Blue Catfish, temperature tolerance, genetic evaluation of thermal resistance Ictalurus furcatus (Lesueur), and their reciprocal F1 hybrids fed 25% and 45% protein diets. Aquaculture Research 29:251–257. in channel and hybrid catfish strains should be further explored. Brander, K. M. 2007. Global fish production and climate change. Proceedings of the National Academy of Sciences of the USA 104:19709–19714. ACKNOWLEDGMENTS Brandt, S. B. 1993. The effect of thermal fronts on fish growth: a bioenergetics evaluation of food and temperature. Estuaries and Coasts 16:142– We thank Brian Bosworth and the staff at the Thad Cochran 159. National Warmwater Aquaculture Center for their assistance in Brett, J. R. 1956. Some principles in the thermal requirements of fishes. Quar- spawning and rearing the fry used in this experiment; Kurt Gam- terly Review of Biology 31:75–86. perl for his helpful comments on experimental design; Mack Brett, J. R. 1979. Environmental factors and growth. Pages 599–675 in W. S. Fondren and South Farm Aquaculture Facility staff; research Hoar, D. J. Randall, and J. R. Brett, editors. Fish physiology, volume 8. Academic Press, New York. assistant Chris Rego and students at Mississippi State Univer- Browse, J., and Z. Xin. 2001. Temperature sensing and cold acclimation. Current sity, who helped monitor, weigh, and measure fish; and two Opinion in Plant Biology 4:241–246. anonymous reviewers for providing helpful comments on this Brune, D. E., G. Schwartz, A. G. Eversole, J. A. Collier, T. E. Schwedler. manuscript. We would also like to thank Natural Resource En- 2004. Partitioned aquaculture systems Pages 561–584 in C. Tucker and J. terprises, the Mississippi Agricultural and Forestry Experiment Hargreaves, editors. Biology and culture of Channel Catfish: developments in aquaculture and fisheries science. Elsevier, San Diego, California. Station, the U.S. Department of Agriculture Agricultural Re- Burke, J. M., and M. L. Arnold. 2001. Genetics and the fitness of hybrids. search Service, and the National Institute of Food and Agricul- Annual Review of Genetics 35:31–52. ture for funding. Cheetham, J. L., C. T. Garten, C. L. King, and M. H. Smith. 1976. Temperature tolerance and preference of immature Channel Catfish (Ictalurus punctatus). Copeia 1976:609–612. REFERENCES Clearwater, S. J., and N. W. Pankhurst. 1997. The response to capture and Allen, K. O., and K. Strawn. 1968. Heat tolerance of Channel Catfish Ictalurus confinement stress of plasma cortisol, plasma sex steroids and vitellogenic punctatus. Proceedings of the Annual Conference Southeastern Association oocytes in the marine teleost, red gurnard. Journal of Fish Biology 50:429– Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 of Game and Fish Commissioners 21(1967):399–411. 441. Andrews, J. W., and R. R. Stickney. 1972. Interactions of feeding rates and envi- Cook, A. M., J. Duston, and R. G. Bradford. 2006. Thermal tolerance of a ronmental temperature on growth, food conversion, and body composition of northern population of Striped Bass Morone saxatilis. Journal of Fish Biology Channel Catfish. Transactions of the American Fisheries Society 101:94–99. 69:1482–1490. Arnold, M., E. Torrans, and P. Allen. 2013. Influences of cyclic, high tem- Cowles, R. B., and C. M. Bogert. 1944. A preliminary study of the thermal peratures on juvenile Channel Catfish growth and feeding. North American requirements of desert reptiles. Bulletin of the American Museum of Natural Journal of Aquaculture 75:77–84. History 83:265–296. Barrionuevo, W. R., and M. N. Fernandes. 1995. Critical thermal maxima and Cox, D. K. 1974. Effects of three heating rates on the critical thermal maximum minima for curimbata´, Prochilodus scrofa Steindachneri, of two different of Bluegill. Pages 158–163 in J. W. Gibbons and R. R. Sharitz, editors. sizes. Aquaculture Research 26:447–450. Thermal ecology: proceedings. Springfield. Barton, B. A., C. B. Schreck, and L. G. Fowler. 1988. Fasting and diet con- Crawshaw, L. I., and H. T. Hammel. 1974. Behavioral regulation of internal tent affect stress induced changes in plasma glucose and cortisol in juvenile temperature in the Brown Bullhead, Ictalurus nebulosus. Comparative Bio- Chinook Salmon. Progressive Fish-Culturist 50:16–22. chemistry and Physiology Part A Physiology 74:51–60. Becker, C. D., and R. G. Genoway. 1979. Evaluation of the critical thermal Currie, R. J., W. A. Bennett, and T. L. Beitinger. 1998. Critical thermal minima maximum for determining thermal tolerance of freshwater fish. Environmen- and maxima of three freshwater game-fish species acclimated to constant tal Biology of Fishes 4:245–256. temperatures. Environmental Biology of Fishes 51:187–200. Beitinger, T. L., W. A. Bennett, and R. W. McCauley. 2000. Temperature tol- D´ıaz, F., and L. F. Buckle.¨ 1999. Effect of the critical thermal maximum on erances of North American freshwater fishes exposed to dynamic changes in the preferred temperatures of Ictalurus punctatus exposed to constant and temperature. Environmental Biology of Fishes 58:237–275. fluctuating temperatures. Journal of Thermal Biology 24:155–160. CRITICAL THERMAL MAXIMA OF CHANNEL AND HYBRID CATFISH 111

Ficke, A. D., C. A. Myrick, and L. J. Hansen. 2007. Potential impacts of global McDermid, J. L., F. A. Fischer, M. Al-Shamlih, W. N. Sloan, N. E. Jones, and climate change on freshwater fisheries. Reviews in Fish Biology and Fisheries C. C. Wilson. 2012. Variation in acute thermal tolerance within and among 17:581–613. hatchery strains of Brook Trout. Transactions of the American Fisheries Fry, F. E. J. 1947. Effects of the environments on animal activity. University of Society 141:1230–1235. Toronto Press, Toronto. Meffe, G. K., S. C. Weeks, M. Mulvey, and K. L. Kandl. 1995. Genetic dif- Graham, K. 1999. A review of the biology and management of Blue Catfish. ferences in thermal tolerance of Eastern Mosquitofish (Gambusia holbrooki; Pages 37–49 in E. R. Erwin, W. A. Hubert, C. F. Rabeni, H. L. Schramm Jr., Poeciliidae) from ambient and thermal ponds. Canadian Journal of Fisheries and T. Coon, editors. Proceedings of the International Ictalurid symposium. and Aquatic Sciences 52:2704–2711. American Fisheries Society, Symposium 24, Bethesda, Maryland. Morgan, I. J., D. G. Mcdonald, and C. M. Wood. 2001. The cost of living for Hart, J. S. 1952. Geographic variations of some physiological and morphological freshwater fish in a warmer, more polluted world. Global Change Biology characters in certain freshwater fish. Publications of the Ontario Fisheries 7:345–355. Research Laboratory 72:1–79. Mørkøre, T., I. Pablo, T. Mazo, V. Tahirovic, and O. Einen. 2008. Impact of Hathaway, E. S. 1927. Quantitative study of the changes produced by acclima- starvation and handling stress on rigor development and quality of Atlantic tization in the tolerance of high temperatures by fishes and amphibians. U. S. Salmon (Salmon salar L). Aquaculture 277:231–238. Bureau of Fisheries Bulletin 43:169–192. Newton, J. R., C. Smith-Keune, and D. R. Jerry. 2010. Thermal tolerance Hlohowskyj, I., and T. E. Wissing. 1985. Seasonal changes in the critical varies in tropical and sub-tropical populations of Barramundi (Lates calcar- thermal maxima of Fantail (Etheostoma flabellare), Greenside (Etheostoma ifer) consistent with local adaptation. Aquaculture 308(Supplement 1):S128– blennioides), and Rainbow (Etheostoma caeruleum) darters. Canadian Jour- S132. nal of Zoology 63:1629–1633. Ospina, A. F., and C. Mora. 2004. Effect of body size on reef fish tolerance to Hutchison, V. H. 1976. Factors influencing thermal tolerance of individ- extreme low and high. Environmental Biology of Fishes 70:339–343. ual organisms. Pages 10–26 in W. G. Esch and W. R. McFarlane, ed- Paladino, V. K., J. R. Spotila, J. P. Schubauer, and K. T. Kowalski. 1980. The itors. Thermal ecology II. U.S. National Technical Information Service, critical thermal maximum: a technique used to elucidate physiological stress Springfield, Virginia. and adaptation in fish. Reviews of Canadian Biology 39:115–122. Jackson, D. C. 2004. Natural history and fisheries Pages 15–35 in C. Tucker and Perez-Casanova,´ J. C., L. O. B. Afonso, S. C. Johnson, S. Currie, and A. K. J. Hargreaves, editors. Biology and culture of Channel Catfish: developments Gamperl. 2008. The stress and metabolic responses of juvenile Atlantic Cod in aquaculture and fisheries science. Elsevier, San Diego, California. Gadus morhua L. to an acute thermal challenge. Journal of Fish Biology Jobling, M. 1981. Temperature tolerance and the final preferendum – rapid 72:899–916. methods for the assessment of optimum growth temperatures. Journal of Fish Portner,¨ H. O. 2002. Climate variations and the physiological basis of tem- Biology 19:439–455. perature dependent bigeography: systemic to molecular hierarchy of ther- Karl, T. R., J. M. Melillo, and T. C. Peterson, editors. 2009. Global climate mal tolerance in animals. Comparative Biochemistry and Physiology Part A change impacts in the United States. Cambridge University Press, New York. Molecular and Integrative Physiology 132:739–761. Kita, J., S. Tsuchida, and T. Setoguma. 1996. Temperature preference and Prosser, C. L., and D. O. Nelson. 1981. The role of nervous systems in tempera- tolerance, and oxygen consumption of the Marbled Rockfish, Sebastiscus ture adaptation of poikilotherms. Annual Review of Physiology 43:281–300. marmoratus. Marine Biology 125:467–471. Recsetar, M. S., M. P.Zeigler, D. L. Ward, S. A. Bonar, and C. A. Caldwell. 2012. Lang, R. P., R. P. Romaire, and T. R. Tiersch. 2003. Induction of early spawning Relationship between fish size and upper thermal tolerance. Transactions of of Channel Catfish in heated earthen ponds. North American Journal of the American Fisheries Society 141:1433–1438. Aquaculture 65:73–81. Regier, H. A., J. A. Holmes, and D. Pauly. 1990. Influence of temperature Le Morvan, C., D. Troutaud, and P. Deschaux. 1998. Differential effects of changes on aquatic ecosystems: an interpretation of empirical data. Transac- temperature on specific and nonspecific immune defenses in fish. Journal of tions of the American Fisheries Society 119:374–389. Experimental Biology 201:165–168. Reutter, J. M., and C. E. Herdendorf. 1976. Thermal discharge from a nuclear Li, M. H., B. B. Manning, E. H. Robinson, B. G. Bosworth, and W. R. Wolters. power plant: predicted effects on Lake Erie fish. Ohio Journal of Science 2001. Comparison of growth, processing yield, and body composition of 76:39–45. USDA103 and Mississippi “normal” strains of Channel Catfish fed diets Rhode, J. M., and M. B. Cruzan. 2005. Contributions of heterosis and epistasis containing three concentrations of protein. Journal of the World Aquaculture of hybrid fitness. American Naturalist 166(5):E124–E139. Society 32:402–408. Tavares-Dias, M., and F. R. Moraes. 2007. Haematological and biochemical ref- Li, M. H., E. H. Robinson, and W. R. Wolters. 1998. Evaluation of three strains of erence intervals for farmed Channel Catfish. Journal of Fish Biology 71:383– Downloaded by [Department Of Fisheries] at 20:13 15 June 2014 Channel Catfish Ictalurus punctatus fed diets containing three concentrations 388. of protein and digestible energy. Journal of the World Aquaculture Society Watenpaugh, D. E., T. L. Beitinger, and D. W. Huey. 1985. Temperature tol- 29:155–160. erance of nitrite-exposed Channel Catfish. Transactions of the American Liu, S., X. Wang, F. Sun, J. Zhang, J. Feng, H. Liu, K. V. Rajendran, L. Sun, Fisheries Society 114:274–278. Y. Zhang, Y. Jiang, E. Peatman, L. Kaltenboeck, H. Kucuktas, and Z. Liu. Wax, C. L., J. W. Pote, and N. C. Dellman. 1987. A climatology of pond 2013. RNA-Seq reveals expression signatures of genes involved in oxygen temperatures for aquaculture in Mississippi. Mississippi Agricultural and transport, protein synthesis, folding and degradation in response to heat stress Forestry Experiment Station Technical Bulletin 149. in catfish. Physiological Genomics 334:1–45. Wolters, W. R., and M. R. Johnson. 1994. Enteric septicemia resistance in Blue Magnuson, J. J., L. B. Crowder, and P. A. Medvick. 1979. Temperature as an Catfish and three Channel Catfish strains. Journal of Aquatic Animal Health ecological resource. American Zoologist 19:331–343. 6:329–334. McCauley, R. W., and T. Beitinger. 1992. Predicted effects of climate warm- Xu, D.-H., P. H. Klesius, B. G. Bosworth, and N. Chatakondi. 2012. Suscep- ing on the commercial culture of the Channel Catfish, Ictalurus punctatus. tibility of three strains of Blue Catfish, Ictalurus furcatus (Valenciennes), to GeoJournal 28:61–66. Ichthyophthirius multifiliis. Journal of Fish Diseases 35:887–95. This article was downloaded by: [Department Of Fisheries] On: 16 June 2014, At: 01:11 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Investigating the Utility of Measuring 11α- Ketotestosterone and Vitellogenin in Surface Mucus as an Alternative to Plasma Samples in Assessments of the Reproductive Axis of White Bass Nicholas A. Barkowskia & Alf H. Haukenesa a Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Stop 4912, Pine Bluff, Arkansas 71601, USA Published online: 15 Apr 2014.

To cite this article: Nicholas A. Barkowski & Alf H. Haukenes (2014) Investigating the Utility of Measuring 11α- Ketotestosterone and Vitellogenin in Surface Mucus as an Alternative to Plasma Samples in Assessments of the Reproductive Axis of White Bass, North American Journal of Aquaculture, 76:2, 112-118, DOI: 10.1080/15222055.2013.860066 To link to this article: http://dx.doi.org/10.1080/15222055.2013.860066

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 North American Journal of Aquaculture 76:112–118, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.860066

ARTICLE

Investigating the Utility of Measuring 11α-Ketotestosterone and Vitellogenin in Surface Mucus as an Alternative to Plasma Samples in Assessments of the Reproductive Axis of White Bass

Nicholas A. Barkowski1 and Alf H. Haukenes* Aquaculture/Fisheries Center, University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Stop 4912, Pine Bluff, Arkansas 71601, USA

Abstract Sex steroids and vitellogenin measured in plasma have been used in a variety of fish species to investigate the reproductive system. Surface mucus has been shown to be an effective substitute for plasma in several species of fish. The goal of this study was to investigate the utility of mucus sampling as an approach to investigating the hypothalamic–pituitary–gonadal (HPG) axis in White Bass Morone chrysops.Wemeasured11α-ketotestosterone and vitellogenin in both surface mucus and plasma samples from White Bass collected from a wild population during their annual spawning migration. Mucus and plasma samples were collected from male and female White Bass that were treated to stimulate the HPG axis (carp pituitary extract, 10 mg/kg; estradiol, 10 mg/kg) or treated with saline injections. Mucus and plasma concentrations of both vitellogenin and 11α-ketotestosterone were positively correlated with one another. Treatments with estradiol led to higher amounts of vitellogenin in both plasma and mucus in both male and female fish. Assessments of 11α-ketotestosterone in mucus correctly identified the sex of 90% of spawning White Bass, as did vitellogenin in 94%. These data demonstrate the potential utility of mucus sampling as a tool to less invasively assess the annual reproductive cycle, identify stimulation of the HPG axis, and identify fish sex in hatchery populations of White Bass.

Traditional methods for investigating reproductive markers and marine species (Schultz et al. 2005) have been assayed suc- such as sex steroids and vitellogenin (VTG) include the assay cessfully for 11α-ketotestosterone (11-KT) and VTG. An im- Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 of plasma samples. Blood sampling and plasma collection is a portant feature when characterizing the HPG axis of hatchery stressful procedure and physiological responses to acute stress fish is that the sampling methodology should employ a mini- are known to negatively impact fish performance (Barton and mum amount of handling stress. Ultimately, sampling surface Iwama 1991). Blood collection from small fish often requires mucus may offer an alternative to blood collection that is less sacrificing the fish to collect samples (Houston 1990), and in stressful to the organism under study. many instances whole-body extractions of constituents from the Understanding the HPG axis in White Bass Morone chysops fish are required for assay (Sink et al. 2007). Surface mucus is important due to their use in commercial aquaculture. In is a potential alternative sampling method for measuring many general, principle hormone signals involved with reproduc- steroids and VTG and would allow for sampling a wide range tive development in teleosts are the result of a cascade of of fish sizes when making assessments of the hypothalamic– events occurring along the HPG axis. In males, gonadotropins pituitary–gonadal (HPG) axis of fishes. Surface mucus of nu- stimulate the production of testosterone, which is synthesized merous freshwater (Moncaut et al. 2003; Maltais and Roy 2009) into 11-KT by Leydig cells located in the testis and induces

*Corresponding author: [email protected] 1Present address: U.S. Army Corps of Engineers, 231 South LaSalle Street, Chicago, Illinois 60604, USA. Received July 31, 2013; accepted October 18, 2013

112 SURFACE MUCUS ASSESSMENTS OF THE WHITE BASS REPRODUCTIVE AXIS 113

spermatogenesis by acting on Sertoli cells (Evans and Claiborne gion. Gravid females have a larger abdominal girth than males. 2006). In females, gonadotropins released into the bloodstream The fish were held in 100-L tanks within a water recirculating initiate the production of testosterone in the ovaries. Testos- system supplied with 20◦C dechlorinated city water formulated terone is then converted into 17β-estradiol (E2) in theca cells to a salinity measurement of 1–2 ‰. Fish were acclimated to also found in the ovaries; E2 is released into circulation and acts these conditions for 24–72 h before any sampling procedure and on receptors in the liver, leading to the production of VTG by given opportunity to forage on Fathead Minnow Pimephales hepatocytes (Takemura and Kim 2001; Moncaut et al. 2003). promelas. Vitellogenin is a precursor to egg yolk, and thus leads to ovar- Mucus and plasma sampling.—Mucus and plasma samples ian maturation in female fish (Heppell et al. 1999), but it is were collected from nontreated fish to obtain baseline levels also a diagnostic feature of endocrine disruption in male fishes of VTG and 11-KT in fish during their spawning migrations. (Mylchreest et al. 2003). Additional mucus and plasma samples were collected from White Bass are used as broodstock for the production of fish after an injection of E2 or carp pituitary extract (CPE). hybrid Striped Bass Morone saxatilis (McGinty and Hodson These treatments were selected to stimulate the HPG axis. In 2008). Sunshine bass (female White Bass × male Striped Bass) instances when multiple fish were held in the same tank, they have become the prevailing hybrid produced by aquaculturists. were tagged by applying a uniquely coded Floy T-bar tag (Floy A limited availability of female Striped Bass broodstock, the Tags, Seattle, Washington) for later identification. Each fish was fact that White Bass mature at younger ages than Striped Bass anesthetized in a bath of tricaine methanesulfonate (Tricaine-S, (making them easier and less expensive to maintain), and the Western Chemical, Inc., Ferndale, Washington) at 100 mg/L advanced growth of this hybrid cross during phase II–III has of water. An area (∼5cm2) posterior to the operculum and led to the industry preference for sunshine bass (Morris et al. dorsal to the pectoral fin was then swabbed using a 2-cm2 1999; Rudacille and Kohler 2000). Domesticated lines of White section of polyurethane sponge (Buf-Puf gentle facial sponge, Bass have been established along with successful breeding pro- 3M, St. Paul, MN). The sponge was placed into a uniquely grams (Kohler et al. 1994; Smith et al. 1996; Woods 2001). The labeled vial (fish identification code, treatment, and date) and development of less-invasive sampling methods to understand placed on ice. The samples were transported to freezers where the annual reproductive cycle can better characterize the events they were stored at −80◦C until assayed. Blood samples were associated with successful reproduction. White Bass reproduc- also collected from the caudal vasculature of each fish using a tive events occur over an annual cycle (Newton and Kilambi 3-mL syringe with an attached 3.8-cm-long, 21-gauge needle 1969). Preparation for spawning occurs several months in ad- (Houston 1990). Approximately 1.0 mL of blood was collected vance of ovulation in response to changes in temperature and and transferred into a labeled microcentrifuge tube coated with photoperiod regimes (Pope et al. 1997; Guy et al. 2002). Current heparin and stored on ice (about 4◦C) until centrifuged (Ac- research programs in hybrid Striped Bass aquaculture are focus- cuspin microR; Fisher Scientific, Pittsburgh, Pennsylvania) at ing on better control of the reproductive cycle through control of 2,750 × gravity for 5 min at 4◦C. Plasma samples were trans- temperature and photoperiod conditions. Mucus sampling may ferred into clean microcentrifuge tubes, labeled, and stored at provide an option for collecting fish response data to these en- −80◦C until assayed. vironmental manipulations. The objectives of this study were Hormone and gonadotropin treatments.—Estradiol (Cayman to (1) characterize the relationship between plasma and mucus Chemical, Ann Arbor, Michigan) was dissolved in a peanut oil concentrations of 11-KT and VTG in White Bass, (2) evalu- carrier to create an injectable solution containing 10 mg E2/mL. ate the sensitivity of surface mucus to detect stimulation of the Intraperitoneal injections of either an E2 dose (10 mg E2/kg of

Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 HPG axis, and (3) test the efficacy of using 11-KT and VTG body weight) or a sham injection containing only the peanut measurements in surface mucus to identify the sex of White oil carrier were administered to groups of fish. Estradiol was Bass. injected into four male and eight female fish. Four additional males were injected with only the peanut oil carrier. At 24 h after injection, mucus and plasma samples were collected as de- METHODS scribed previously. Methods of sampling fish treated with CPE Test fish.—Female and male White Bass (means: 29.8 cm were similar to those described for E2 treatment. Plasma and SL [SD, 3.2], 651.8 g [SD, 210.7]) were collected by angling mucus samples from 10 male and 5 female fish served to rep- and electrofishing from Caney Bayou during March and April resent nontreated fish. Seven male and seven female fish were 2010. Caney Bayou is a tributary of the lower Arkansas River injected with CPE at 10 mg/kg body weight (Stoller Fisheries, adjacent to the campus of University of Arkansas at Pine Bluff Spirit Lake, Iowa), and 24 h after injection, samples of blood and and supports a significant White Bass fishery during their an- mucus were collected. Additionally, six male fish were injected nual spawning migration (Baker 2012). Fish were transported with phosphate-buffered saline (PBS), which served as the car- live to wet laboratory facilities on the campus of University of rier for CPE. All plasma and mucus samples were collected and Arkansas at Pine Bluff. Fish were initially identified as male or stored using methods described previously. Following sampling, female by examining the relative girth of their abdominal re- all fish were euthanized in a lethal MS-222 bath (250 mg/L), and 114 BARKOWSKI AND HAUKENES

a gross inspection of gonads was performed to confirm earlier 0.00 pg/mL. Interplate (n = 4) and intraplate (n = 4) variation observations of fish sex. for both plasma and mucus was also determined using methods Assays of VTG OD and 11-KT.—We extracted VTG and 11- similar to those described for VTG OD. When assaying samples KT from the sponges using methods similar to those described collected during our experiments, all samples for both VTG and by Schultz et al. (2007). Each sponge sample was thawed and 11-KT assays were chosen at random to minimize the influence placed into the barrel of a 3-mL syringe. For each sample, two of interplate variation on any inferences made from the data. 0.5-mL aliquots of ice-cold tris-buffered saline solution (1× Statistics.—A linear regression was performed to character- solution, pH 7.5; Fischer Scientific, Pittsburgh, Pennsylvania) ize the strength of the relationship between plasma and mu- were forced through the syringe with the plunger to extract the cus measures for both VTG OD and 11-KT concentrations. An sponge contents and transfer the fluid into a microcentrifuge ANCOVA was performed to test for parallelism of the samples tube. The extracted sample was centrifuged at 13,000 × gravity spiked with 11-KT. A significant interaction between sample for 15 min at 4◦C, and the supernatant was transferred into a and the covariate (spiked amount of 11-KT added to the sample) labeled vial. Two drops from a Pasteur pipette of a protease in- was used as an indication of a lack of parallelism. Wilcoxon’s hibitor (16 µg/mL benzamidine HCl, 10 µg/mL phenanthroline, rank-sum test was performed to compare nontreated fish with 10 µg/mL aprotinin, 10 µg/mL leupeptin, 10 µg/mL pepstatin those injected with the peanut oil carrier to determine if the car- A, and 1mM phenylmethyl sulfonyl fluoride; BD PharMingen, rier elicited a response independent of E2 or CPE. A two-way San Diego, California) were added to each VTG surface mucus ANOVA was used to assess differences in ranked VTG OD and sample. Before VTG and 11-KT ELISA analyses, the mucus and 11-KT between nontreated and treated, as well as between male plasma samples were assayed for protein concentrations using and female fish, for both plasma and mucus samples. Before the a Bradford assay (Pierce Biotechnology, Rockford, Illinois). two-way ANOVA was performed, data were rank-transformed A semiquantitative ELISA kit for Striped Bass VTG to satisfy the assumptions of parametric testing. If a signifi- (Cayman Chemical, Ann Arbor, Micigan) was used to assay cant interaction term was found, pairwise contrasts were used samples. Standard amounts of protein in the sample (0.4 µgfor to illustrate the source of the interaction. All analyses were con- mucus and 0.5 µg for plasma) were added to each microplate ducted using Statistical Analysis Software (SAS Institute, Cary, well to improve upon the quantitative aspects of the assay. The North Carolina) and a threshold of significance was set at α ≤ manufacturer’s instructions for the assay were followed, and a 0.05. VTG optical density (VTG OD) at a wavelength of 492 nm for Sex identification.—An evaluation of a technique to identify each sample was recorded via a microplate reader (Powerwave fish sex using mucus VTG OD and 11-KT concentrations was XS; Biotek, Winooski, Vermont). All samples were analyzed performed. The methods to identify males and females followed in duplicate. Three plasma and mucus samples were diluted to procedures adapted from Meyers et al. (1993), which was based protein concentrations of 40, 4, 0.4, and 0.004 µg/mL of buffer on the variation surrounding a negative control sample to iden- supplied with the ELISA kit, and the VTG OD determined to tify a threshold value for a VTG-OD or 11-KT concentration. illustrate the relationship between VTG OD and relative amount The threshold used for classification purposes was the value of of VTG in each sample. Interplate variation was determined by 2 SDs greater than the mean of the negative controls. In this in- analyzing the same plasma and mucus sample on three (plasma) stance nontreated male White Bass served as negative controls or four (mucus) plates. An estimate of intraplate assay variation for sex identification when using the VTG assay, and nontreated was determined by producing multiple values for one sample females served as negative controls for the 11-KT identifica- for plasma (n = 5) and mucus (n = 6) on the same plate. The tions. For example, when a VTG OD value for a fish was >2

Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 standard deviation surrounding the mean VTG OD recorded for SDs from the mean of VTG OD of nontreated males, the fish intraplate and interplate variation sample was used to calculate was classified as a female; if the VTG values were below this a coefficient of variation for each type of sample. threshold they were classified as a male. The percent of fish cor- Plasma and mucus concentrations of 11-KT were measured rectly classified was then determined. A Fisher’s exact test was using a commercially available ELISA kit (Cayman Chemi- used to determine if there were differences in the proportions cal). Samples were measured in duplicate; when duplicate val- of fish correctly classified using only surface mucus 11-KT or ues yielded a CV greater than 20% they were assayed again, VTG OD. and the second value was used for statistical analyses. Samples spiked with known amounts of 11-KT were assayed to illus- RESULTS trate parallelism, a test to determine the sensitivity of assay for measuring known increases in 11-KT in plasma and mucus Vitellogenin samples. Specifically, 25 µL of four plasma samples were each An increase in VTG OD was observed as the concentration of spiked with equal volumes of standard concentrations of 11-KT sample added to microplate wells increased for both mucus and prepared at 12.50, 6.25, 3.13, 1.57, and 0.00 pg/mL of buffer plasma samples (data not shown). The CV for plasma samples supplied with the ELISA kit. Four mucus samples were similarly calculated was 74% (n = 3) for interplate and 11% (n = 5) for spiked with 11-KT standards at 6.25, 3.13, and 1.57, 0.78, and intraplate variation for VTG OD. The CV for mucus samples SURFACE MUCUS ASSESSMENTS OF THE WHITE BASS REPRODUCTIVE AXIS 115

FIGURE 1. Linear regression describing the relationship between vitellogenin optical densities (VTG OD) for plasma and mucus samples collected from White Bass.

calculated was 59% (n = 4) for interplate and 16% (n = 6) for intraplate variation. A significant linear relationship between plasma and mucus VTG OD of White Bass was observed for all the samples collected (r2 = 0.32, n = 28, P = 0.001; Fig- ure 1). The mean VTG OD in males injected with peanut oil and nontreated males did not significantly differ from each other for either mucus or plasma samples (Wilcoxon’s rank-sum test: = = P 0.27 for mucus, P 0.60 for plasma; data not shown). FIGURE 2. Mean (+1 SD) vitellogenin optical densities (VTG OD) for (a) The two-way ANOVA of VTG OD for plasma samples re- plasma and (b) mucus samples from nontreated and 17β-estradiol-injected male vealed significant main effect differences (treatment, sex) with and female White Bass. A significant main effect difference in the overall mean of VTG OD for treatment and fish sex was observed in plasma. Plasma no significant interaction between these effects (F3, 23 = 47.71, < P < 0.0001). Fish injected with E (F = 118.86, P < 0.0001) samples from females had higher VTG OD values than males (P 0.0001) 2 1 and treated fish had a significantly higher VTG OD values nontreated fish had larger plasma VTG OD measurements than nontreated fish, (P = 0.0002). In surface mucus, a significant interaction between treatment and larger plasma VTG OD measurements were observed in and fish sex and treatment was observed for VTG OD. The results of pairwise females than in males (F1 = 21.34, P = 0.0002; Figure 2a). contrasts of surface mucus to distinguish specific differences among the means Analysis of VTG OD in mucus samples provided a slightly are shown. Significant differences among individual means are illustrated by different outcome; the overall model was significant (two-way different letters (w, x). ANOVA: F3, 23 = 18.55, P < 0.0001). However, a significant interaction between treatment and fish sex illustrated a different Concentrations of 11-KT ranged from 0.11 to 40.67 pg/mg response to treatment between male and female fish (F1 = 5.77, of protein in plasma and from 0.14 to 1.12 pg/mg of protein in P = 0.030). Nontreated females had a significantly higher (P = mucus samples. The concentrations of 11-KT found in plasma 0.0002) mean VTG OD in mucus than nontreated males, while and mucus from nontreated and PBS-injected fish did not differ no significant difference was observed for VTG OD between significantly (Wilcoxon’s rank-sum test: P = 0.55 for mucus,

Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 = E2-treated male and female ﬁsh (Figure 2b). P 0.093 for plasma; data not shown). There were signiﬁcant

11␣-Ketotestosterone Plasma and mucus samples that were spiked with 11-KT demonstrated a parallel relationship and positive slope as the spiked amount of 11-KT was increased (data not shown). No interactions between the sample and the covariate (11-KT added to the spiked samples) were found using the ANCOVA for plasma (F3 = 0.020, P = 0.99) or mucus (F3 = 2.29, P = 0.13). The interplate variation for concentrations of 11-KT in plasma was 2–10% (n = 4) and for intraplate was 7–12% (n = 4). Interplate variation for mucus samples ranged from 5% to 7% (n = 4) and intraplate ranged from 10% to 26% (n = 4). A signiﬁcant positive linear relationship between 11-KT concentrations in plasma 2 = and mucus samples was detected (linear regression: r 0.22, FIGURE 3. Linear regression describing the relationship between 11α- n = 28, P = 0.005; Figure 3). ketotestosterone (11-KT) concentrations in plasma and mucus samples. 116 BARKOWSKI AND HAUKENES

of White Bass were also found (F3, 28 = 15.72, P < 0.0001). Samples from male fish contained greater concentrations of 11- KT than females (F1 = 46.57, P < 0.0001), and no differences were found between nontreated and CPE-injected fish (F1 = 0.50, P = 0.49). No interaction between sex of fish and treatment was observed (F1 = 0.08, P = 0.77; Figure 4b). Sex Identification Overall, fish sex was correctly classified for 90% of the sex- ually mature White Bass using 11-KT and for 94% using VTG OD. These proportions of fish correctly classified did not differ when using either VTG OD or 11-KT (Fisher’s exact test: P = 0.3687; Table 1). Samples of VTG OD in mucus correctly identified the sex of 100% of the nontreated females and E2-treated females. All E2-treated males were classified as females via the method described. This technique correctly classified 75% (n = 4) of peanut oil injected males. The concentration of 11-KT in mucus correctly identified the sex in 90% of the nontreated males (n = 10), 83% of the CPE-treated males (n = 6), and 83% of the saline-injected males (n = 6). All female fish were correctly classified based on 11-KT.

+ α FIGURE 4. Mean ( 1 SD) concentrations of 11 -ketotestosterone (11-KT) DISCUSSION for (a) plasma and (b) mucus collected from male and female White Bass. No significant main effect differences were attributed to treatment with carp This study provides insight into the utility of mucus sam- pituitary extract (CPE). A significant main effect difference for fish sex was pling as a means of investigating the annual reproductive cycle observed for both mucus and plasma values of 11-KT. Male White Bass had of White Bass. The less-invasive manner of sampling mucus significantly higher amounts of 11-KT than females (P < 0.0001 for both plasma relative to plasma sampling enhances the potential of this tech- and mucus samples). nique to be more widely adopted for use in natural and captive populations of fish. Significant relationships between mucus differences in 11-KT concentrations in plasma among groups of and plasma samples for both VTG OD and 11-KT are also sampled fish (two-way ANOVA: F3, 28 = 21.23, P < 0.0001). reported for other species. Mucus and plasma concentrations Male fish had significantly higher concentrations of 11-KT in of VTG have been measured in Shorthead Redhorse Moxos- plasma than females (F1 = 63.16, P < 0.0001). However, 11-KT toma macrolepidotum, Copper Redhorse M. hubbsi (Maltais and concentrations were similar in plasma for both nontreated and Roy 2009), Atlantic Salmon Salmo salar (Meucci and Arukwe CPE-injected fish (F = 0.00, P = 0.98). No interaction between 2005), Cichlasoma dimerus (Moncaut et al. 2003), and Striped sex and treatment was observed (F1 = 0.53, P = 0.47; Figure 4a). Bass (Kishida et al. 1992) through Western-blot and gel elec- Significant differences in concentrations of 11-KT in mucus trophoresis. Our use of commercially available VTG ELISA

Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 TABLE 1. Mean vitellogenin optical densities (VTG OD) and percentages of White Bass correctly classiﬁed as male or female using the results of vitellogenin (VTG)or11α-ketotestosterone (11-KT) assays of mucus samples. The categorization method was adapted from an approach described by Meyers et al. (1993) for the interpretation of enzyme-linked immunosorbant assay optical densities. Asterisks denote negative controls. Nontreated males and females used to generate the negative control statistics were not classiﬁed (n/c). Numbers in parentheses indicate the sample sizes.

Assay Treatment (n) Mean VTG OD SD Percent correctly classiﬁed VTG Nontreated males* (6) −0.008 0.004 n/c Nontreated females (6) 0.092 0.043 100 E2 females (8) 0.335 0.262 100 E2 males (4) 0.173 0.110 0 Peanut oil males (4) −0.003 0.007 75 11-KT Nontreated females* (5) 0.204 0.028 n/c Nontreated males (10) 0.396 0.104 90 CPE females (7) 0.190 0.036 100 CPE males (6) 0.506 0.349 83 Saline males (6) 0.388 0.169 83 SURFACE MUCUS ASSESSMENTS OF THE WHITE BASS REPRODUCTIVE AXIS 117

kits adds to the overall efficiency of generating these data. Con- of endocrine disruption. For example, male Atlantic Sturgeon centrations of 11-KT have also been measured in fish species Acipenser oxyrinchus were reported (Heise et al. 2009) to have (Schultz et al. 2005). The correlation coefficient for the rela- measurable concentrations of VTG in their plasma that were tionship between plasma and mucus 11-KT reported here was attributed to a potential endocrine disrupting chemical (EDC). less than reported by Schultz et al. (2005) who found a stronger Our observation that E2-treated male White Bass had a measure- 2 relationship (r = 0.83) in koi (an ornamental variant of Com- able increase in surface mucus vitellogenin also led to E2-treated mon Carp Cyprinus carpio) between plasma and mucus 11-KT. males being classified as females. Thus, in a severe instance of The lower correlation coefficient in our animals may indicate a endocrine disruption our surface mucus sampling was able to need to further optimize the assay or that there is a less direct detect an anticipated response along the HPG axis (Pollack et al. relationship between surface mucus and plasma concentrations 2003). Ultimately, measurement of VTG in mucus, rather than of 11-KT. It is clear, however, that surface mucus measurements plasma, might be employed to more easily monitor populations of 11-KT were sensitive enough to detect broad differences in in areas with unknown amounts of EDCs or occupying regions hormone levels in male and female fish. known to have higher levels of EDCs because it is a more easily Interplate variation for VTG OD was high in both mucus acquired sample. and plasma samples. This observation is probably the result Mucus samples are inherently easier to collect than the blood of the repeated freeze and thaw of the VTG sample used to samples typically used to characterize reproductive cycles in assess interplate variation. The sample used for interassay vari- fish. In the case of White Bass it will be necessary to evaluate ation was frozen at −80◦C and thawed for as many as five larger sample sizes, a variety of fish ages, and samples collected analyses. Known to be an unstable compound (Goodwin et al. throughout the year to further establish the utility of this assay. 1992), the repeated freeze–thaw cycle of VTG may have led to However, the ease of sampling does offer a less-invasive method its degradation. Brodeur et al. (2006) found that freezing sam- to monitor endocrine responses, and this assay approach may be ples increased the measured concentrations of VTG in Fathead sensitive enough to detect response data to environmental ma- Minnow; an increase thought to be the result of antibody binding nipulations designed to control the reproductive cycle of White to degradation products of VTG. Differences in the characteris- Bass broodstock held for hybrid Striped Bass production. tics of the antibody used in our study may explain the different response reported here. The monoclonal antibody used in this ACKNOWLEDGMENTS study was prepared to bind to intact VTG molecules. Any in- This project was financially supported by funding supplied ferences made from data presented here should not have been by the Southern Regional Aquaculture Center (U.S. Department impacted by the high interplate variation because the process of Agriculture’s Cooperative State Research, Education, and Ex- of randomizing sample selection would have controlled for any tension Service). We also acknowledge Steve Lochmann, Mike bias. In contrast to VTG OD, interplate and intraplate variation Eggleton, S. Adam Fuller, and Anita Kelly for their guidance for 11-KT were lower. Intraplate variation for 11-KT plasma and review of this manuscript. Finally, we thank Clint Peacock, samples satisfied the criteria of the ELISA kit used (<20%), Brandon Baker, Dan Grigas, Mini Jose, Brett Timmons, Jimmy but the intraplate variation had a range that was slightly higher Faulkner, Dave Heikes, and Alichia Wilson for their assistance (CV = 10–26%). In comparison, these values for intraplate vari- with data collection, preparation, and materials to ensure a suc- ation were greater than those reported for 11-KT assays from cessful study. other studies (Mandich et al. 2004; Schultz et al. 2005). Our use of a smaller sample size to obtain these values probably

Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 contributes to the higher CV reported here. REFERENCES Sex identification using steroid hormones and VTG in plasma Baker, B. W., and S. E. Lochmann. 2012. A population assessment and minimum is a relatively common practice with fish. Plasma VTG and 11- length limit evaluation for White Bass in the Arkansas River, Arkansas. Proceedings of the Annual Conference of the Southeastern Association of KT have been successfully used to determine sex of Arapaima Fish and Wildlife Agencies 66:6–11. Arapaima gigas (Chu-Koo et al. 2009), California Sheephead Barton, B. A., and G. K. Iwama. 1991. Physiological changes in fish from stress Semicossyphus pulcher (Loke-Smith et al. 2010), and Bluefin in aquaculture with emphasis on the response and effects of corticosteroids. Tuna Thunnus thynnus (Micera et al. 2010). Plasma reproduc- Annual Review of Fish Diseases 1:3–26. tive markers in sturgeon species have also been studied intensely Brodeur, J. C., K. B. Woodburn, F. Zhang, M. J. Bartels, and G. M. Klecka. 2006. Plasma sampling and freezing procedures influence vitellogenin measure- (Petochi et al. 2011; Keyvanshokook and Gharaei 2010; Feist ments by enzyme-linked immunoassay in the Fathead Minnow (Pimephales et al. 2004). In these reports fish sex identification is complicated promelas). Environmental Toxicology and Chemistry 25:337–348. by the more cryptic nature of secondary sexual characteristics. Chu-Koo, F., R. Dugue,´ M. A. Aquilar, A. C. Daza, F. A. Bocanegra, C. C. In our study, a low error rate in identification of fish sex was Veintemilla, F. Duponchelle, J.-F. Rnno, S. Tello, and J. Nunez. 2009. Gen- observed for both male and female White Bass. Our classifica- der determination in the Paiche or Pirarucu (Arapaima gigas) using plasma vitellogenin, 17β-estradiol, and 11-ketotestosterone levels. Fish Physiology tion methods using only mucus correctly classified all female and Biochemistry 35:125–136. fish and high percentages of male fish. Mucus sampling may Evans, D. H., and J. B. Claiborne. 2006. The physiology of fishes, 3rd edition. also offer utility in studies screening populations for the impact CRC Press, Boca Raton, Florida. 118 BARKOWSKI AND HAUKENES

Feist, G., J. P. Van Eenennaam, S. I. Doroshov, C. B. Schreck, R. P. Schneider, the enzyme-linked immunosorbent assay (ELISA) to detect soluble antigen of and M. S. Fitzpatrick. 2004. Early identification of sex in cultured White Renibacterium salmoninarum in Alaskan Pacific salmon. Diseases of Aquatic Sturgeon, Acipenser trasmontanus, using plasma steroid levels. Aquaculture Organisms 16:191–197. 232:581–590. Micera, E., R. Zupa, A. Zarrilli, A. Camarda, A. M. Moramarco, F. Acone, G. Goodwin, A. E., J. M. Grizzle, J. T. Bradley, and B. H. Estridge. 1992. Mon- DeMatrio, and A. Corriero. 2010. A rapid latex agglutination test for gen- oclonal antibody-based immunoassay of vitellogenin in the blood of male der identification in the Atlantic Bluefin Tuna, Thunnus thynnus (Linnaeus). Channel Catfish (Ictalurus punctatus). Comparative Biochemistry and Phys- Aquaculture Research 41:1396–1401. iology 101:441–446. Moncaut, N., F. L. Nostro, and M. C. Maggese. 2003. Vitellogenin detection Guy, C. S., R. D. Schultz, and C. A. Cox. 2002. Variation of gonad development, in surface mucus of the South American cichlid fish Cichlasoma dimerus growth, and condition of White Bass in Fall River Reservoir, Kansas. North (Heckel, 1840) induced by estradiol-17β. Effects on liver and gonads. Aquatic American Journal of Fisheries Management 22:643–651. Toxicology 63:127–137. Heise, R. J., R. B. Bringolf, R. Patterson, W. G. Cope, and S. T. Ross. 2009. Morris, J. E., C. C. Kohler, and C. C. Mischke. 1999. Pond culture of hybrid Plasma vitellogenin and estradiol concentrations in adult Gulf sturgeon from Striped Bass in the north central region. North Central Regional Aquaculture the Pascagoula River drainage, Mississippi. Transactions of the American Center, Fact Sheet 107, Iowa State University, Ames. Fisheries Society 138:1028–1035. Mylchreest E., S. Snajdr, J. J. Korte, and G. T. Ankley. 2003. Comparison Heppell, S. A., L. F. Jackson, G. M. Weber, and C. V. Sullivan. 1999. Enzyme- of ELISAs for detecting vitellogenin in the Fathead Minnow (Pimephales linked immunosorbent assay (ELISA) of vitellogenin in temperate basses promelas). Comparative Biochemistry and Physiology Part C 134:251–257. (genus Morone): plasma and in vitro analyses. Transactions of the American Newton, S. H., and R. V. Kilambi. 1969. Determination of sexual maturity of Fisheries Society 128:532–541. White Bass from ovum diameters. Southwestern Naturalist 14:213–220. Houston, A. H. 1990. Blood and circulation. Pages 273–334 in C. B. Schreck and Petochi, T., P. Di Marco, V. Donadelli, A. Longobardi, I. Corsalini, D. Bertotto, P. B. Moyle, editors. Methods for fish biology. American Fisheries Society, M. G. Finoia, and G. Marino. 2011. Sex and reproductive stage identification Bethesda, Maryland. of sturgeon hybrids (Acipener naccarii × Acipenser baerii) using different Keyvanshokooh, S., and A. Gharaei. 2010. A review of sex determination tools: ultrasounds, histology, and sex steroids. Journal of Applied Ichthyology and searches for sex-specific markers in sturgeon. Aquaculture Research 27:637–642. 41:e1–e7. Pollack, S. J., M. A. Ottinger, C. V. Sullivan, and L. C. Woods III. 2003. The Kishida, M., T. R. Anderson, and J. L. Specker. 1992. Induction by β-estradiol effects of the soy isoflavone genistein on the reproductive development of of vitellogenin in Striped Bass (Morone saxatilis): characterization and quan- Striped Bass. North American Journal of Aquaculture 65:226–234. tification in plasma and mucus. General and Comparative Endocrinology Pope, K. L., D. W. Willis, and D. O. Lucchesi. 1997. Influence of temperature 88:29–39. and precipitations on age-0 White Bass abundance in tow South Dakota Kohler, C. C., R. J. Sheehan, C. Habicht, J. A. Malison, and T. B. Kayes. 1994. natural lakes. Journal of Freshwater Ecology 12:599–605. Habituation to captivity and controlled spawning of White Bass. Transactions Rudacille, J. B., and C. C. Kohler. 2000. Aquaculture performance comparison of the American Fisheries Society 12:964–974. of sunshine bass, palmetto bass, and White Bass. North American Journal of Loke-Smith, K. A., M. A. Sundberg, K. A. Young, and C. G. Lowe. 2010. Use Aquaculture 62:114–124. of morphology and endocrinology to predict sex in California Sheephead: Schultz, D. R., N. Perez, A. J. Mendez, D. Snodgrass, J. E. Serafy, E. D. Prince, evidence of altered timing of sex change at Santa Catalina Island, California. W. A. Crow Jr., and T. R. Capo. 2007. Tracking gender factors in fish surface Transactions of the American Fisheries Society 139:1742–1750. mucus: temporal patterns in individual koi (Cyprinus carpio). Journal of Mandich, A., A. Massari, S. Bottero, P. Pizzicori, H. Goos, and G. Marino. Applied Ichthyology 23:187–188. 2004. Plasma sex steroid and vitellogenin profiles during gonad development Schultz, D. R., N. Perez, C. K. Tan, A. J. Mendez, T. R. Capo, D. Snodgrass, E. in Mediterranean Amber Jack (Seriola dumerilii). Marine Biology 144:127– D. Prince, and J. D. Serafy. 2005. Concurrent levels of 11-ketotestosterone in 138. fish surface mucus, muscle tissue and blood. Journal of Applied Ichthyology Maltais, D., and R. L. Roy. 2009. Purification and partial characterization of 21:394–398. vitellogenin from Shorthead Redhorse (Moxostoma macrolepidotum)and Sink, T. D., S. Kumaran, and R. T. Lochmann. 2007. Development of a whole- Copper Redhorse (Moxostoma hubbsi) and detection in plasma and mucus body cortisol extraction procedure for determination of stress in golden, with heterologous antibody. Fish Physiology and Biochemistry 35:241–254. Notemigonus crysoleucas. Fish Physiology and Biochemistry 33:189– McGinty, A. S., and R. G. Hodson. 2008. Hybrid Striped Bass: hatchery 193. phase. Southern Regional Aquaculture Center, Publication 301, Stoneville, Smith, T. I. J., W. E. Jenkins, and L. D. Heyward. 1996. Production and Downloaded by [Department Of Fisheries] at 01:11 16 June 2014 Mississippi. extended spawning of cultured White Bass broodstock. Progressive Fish- Meucci, V., and A. Arukwe. 2005. Detection of vitellogenin and zona radi- Culturist 58:85–91. ata protein expressions in the surface mucus of immature juvenile Atlantic Takemura, A., and B. H. Kim. 2001. Effects of estradiol- 17β treatment on in Salmon (Salmo salar) exposed to waterborne nonylphenol. Aquatic Toxicol- vitro and in vivo sythesis of two distinct vitellogenins in tilapia. Comparative ogy 73:1–10. Biochemistry and Physiology Part A 129:641–651. Meyers, T. R., S. Short, C. Farrington, K. Lipson, H. J. Geiger, and R. Gates. Woods, L. C. III. 2001. Domestication and strain evaluation of Striped Bass 1993. Establishment of a negative-positive threshold optical density value for (Morone saxatilis). Aquaculture 202:343–350. This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:15 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 The Influence of Dietary Lysine on Yellow Perch Maturation and the Quality of Sperm Karolina Kwaseka, Konrad Dabrowskib, Joanna Nyncac, Michal Wojnod & Macdonald Wicke a School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA; and Department of Animal Sciences, Ohio State University, 2027 Coffey Road, Columbus, Ohio 43210, USA b School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA c School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA; and Polish Academy of Sciences, Tuwina 10, 10-748 Olsztyn, Poland, d School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA e Department of Animal Sciences, Ohio State University, 2027 Coffey Road, Columbus, Ohio 43210, USA Published online: 15 Apr 2014.

To cite this article: Karolina Kwasek, Konrad Dabrowski, Joanna Nynca, Michal Wojno & Macdonald Wick (2014) The Influence of Dietary Lysine on Yellow Perch Maturation and the Quality of Sperm, North American Journal of Aquaculture, 76:2, 119-126, DOI: 10.1080/15222055.2013.856826 To link to this article: http://dx.doi.org/10.1080/15222055.2013.856826

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 North American Journal of Aquaculture 76:119–126, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.856826

ARTICLE

The Inﬂuence of Dietary Lysine on Yellow Perch Maturation and the Quality of Sperm

Karolina Kwasek School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA; and Department of Animal Sciences, Ohio State University, 2027 Coffey Road, Columbus, Ohio 43210, USA Konrad Dabrowski* School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA Joanna Nynca School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA; and Polish Academy of Sciences, Tuwina 10, 10-748 Olsztyn, Poland Michal Wojno School of Environment and Natural Resources, Ohio State University, 2021 Coffey Road, Columbus, Ohio 43210, USA Macdonald Wick Department of Animal Sciences, Ohio State University, 2027 Coffey Road, Columbus, Ohio 43210, USA

Abstract The objective of the present study was to determine whether lysine (Lys) supplementation influences the maturation and sperm quality of Yellow Perch Perca flavescens. Juveniles raised on a formulated commercial diet and weighing approximately 75 g each were randomly distributed into six 400-L tanks. This experiment included two wheat-gluten- based diets in triplicate: (−) Lys (Lys-deficient) and (+) Lys (Lys-supplemented; 2.23% in dry feed) diets. In addition, 16 control fish were kept under identical conditions and fed a commercial diet. The weight of males was larger in the control group than in the (+) Lys and (−) Lys groups. The sperm concentration was significantly higher in the Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 control and (+) Lys groups than in the (−) Lys group. Sperm motility was lower in the (−) Lys group than in the control and (+) Lys groups. The control group had significantly higher protein concentration in its seminal plasma than did the (+) Lys and (−) Lys groups. The seminal plasma trypsin inhibitor activity showed the same trend. The seminal plasma free amino acid concentrations of arginine, methionine, threonine, glutamine, alanine, and glycine differed significantly among treatments. This is the first report demonstrating the negative effect of dietary lysine level in plant-protein-based diets on reproduction in fish.

The unpredictable and variable reproductive performance of is largely dependent on the nutritional status of the broodstock many ﬁsh species is a limiting factor in the successful produc- in which nutrient deﬁciency can reduce semen quality (Vassallo- tion of juveniles (Coward et al. 2002). Reproductive well-being Agius et al. 2001) and thus fertilization ability. Successful

*Corresponding author: [email protected] Received June 19, 2013; accepted October 13, 2013 119 120 KWASEK ET AL.

fertilization in fish depends greatly on the quality of the sperm cluded two diets: (−) Lys (Lys deficient) and (+) Lys (Lys sup- (Ciereszko and Dabrowski 1994). Hence, improved knowledge plemented; 2.23% free Lys; Bachem, New York) wheat-gluten- of sperm characteristics, such as motility, density, and seminal based (MP Biomedicals, Solon, Ohio) and fish-meal-based diets plasma proteins, is necessary to understand the reproductive in triplicate as previously described in Kwasek et al. 2012. The responses of fish kept in captivity. total amount of Lys in the (−) Lys diet was approximately 1% The role of seminal plasma is to create an optimal environ- (1.42% Lys in wheat gluten; 4.51% Lys in fish meal; NRC 1994). ment for spermatozoa that consists of stable osmolality lev- In addition, due to limited capacity and fish number, a total of els, ion concentrations, and pH. Seminal plasma also facilitates 16 control fish were kept under identical conditions in a sep- the synthesis of proteins involved in many of the physiological arate tank and fed a commercial diet (BioOregon, Westbrook, functions of semen, such as spermatozoa protection during their Maine). Fish were fed near to satiation at readjusted feeding storage in the reproductive system. There are different sperm rates that were equal across all treatments based on each day’s parameters that contribute to sperm quality, such as the concen- projected change in weight. Fish size was checked periodically tration of spermatozoa, sperm volume, motility, pH, and sem- in order to monitor individual fish growth. The water tempera- inal plasma protein concentration (Lanes et al. 2010). Seminal ture during the experiment followed seasonal cycles and varied plasma of teleost fish is characterized by a low protein concen- between 7.8◦C and 24.0◦C. The city water that was used was filtration (1–3 g/L) when compared with protein levels in blood. tered through activated charcoal filters and additionally treated For that reason, it is possible that antiproteinases may be one with sodium thiosulfate to keep chlorine levels < 0.1 mg/L. The of the main proteins in teleost seminal plasma. This suggestion photoperiod was 13 h light: 11 h dark. is supported by the relationship between the inhibitory activ- In March 2008 Yellow Perch broodstock were evaluated ity and the mean protein concentration in fish seminal plasma for the advancement of maturation and sex based on size, ro- (Dabrowski and Ciereszko 1994; Wojtczak et al. 2003). bustness, and sperm release. Males and females were injected The free amino acid (FAA) composition of seminal plasma with human chorionic gonadotropin (300 IU/kg; Sigma-Aldrich, varies among fish species (Billard and Menezo 1984; Lahn- Saint Louis, Missouri) (Dabrowski et al. 1994). Fish were anaes- steiner 2009). These FAAs are assumed to originate from the thetized with tricaine methanesulfonate (50 mg/L), and sperm epithelium of the spermatic cord or from the proteolysis of sem- from the males from the control (n = 9), the (+) Lys group (n = inal plasma proteins (Lahnsteiner et al. 1994). The actual role of 11), and the (−)Lys(n = 11) group was sampled by stripping FAAs in the seminal plasma of teleost fish is unclear. In mam- for fertility and biochemical analysis. mals, seminal FAAs are involved in the enhanced metabolic Sperm samples were centrifuged twice: at 3,000 × g for activity of spermatozoa and in detoxification and they affect 3 min and 10,000 × g for 10 min. The resulting seminal plasma sperm viability. It is possible that FAAs in fish seminal plasma was stored at −80◦C. Sperm motility was evaluated after adding contribute to osmolality and have a positive effect on sperm hatchery water (dechlorinated Columbus, Ohio, city water) and fertilizing ability (He and Woods 2003). recorded as the relative percent of motile sperm. Approximately We have previously shown that lysine (Lys) directly affects 50 µL of this water was placed on a glass slide, and approx- the growth and blood plasma FAA pool in Yellow Perch Perca imately 10 µL of semen was added, mixed, and covered with flavescens and that a wheat-gluten-based diet supplemented with a cover slip. The percentage of motile spermatozoa was im- Lys improves fish performance on a diet containing plant protein mediately estimated using light microscopy at an initial mag- (Kwasek et al. 2012). In the present study, which is a continua- nification of 20× and subsequently at 40×. Prior to fertilization of that previous study, we analyzed dietary effects on male tion, the sperm from individuals with the highest motility (80–

Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 maturation. 90%) was diluted 20 fold in ice-cold Moore extender (Rinchard Sperm production in Yellow Perch can be affected by many et al. 2005). The sperm concentration was diluted for the fer- factors, such as photoperiod, temperature (Dabrowski et al. tilization to 50,000 spermatozoa per egg. Sperm concentration 1996), and broodstock diet. Understanding the effects of these was measured according to Ciereszko and Dabrowski (1993). factors on sperm quantity and sperm viability (motility, fertiliz- Protein concentration was determined by the method of Brad- ing ability) is an important step for optimizing fertilization pro- ford (1976) using bovine serum albumin as a standard. Antit- tocols and methods of broodstock evaluation and management. rypsin activity of seminal plasma was evaluated by inhibition Therefore, the objective of the present study was to determine of Atlantic Cod Gadus morhua trypsin (Sigma Chemical, Saint whether dietary Lys influences the maturation and sperm quality Louis, Missouri). Amidase activity was determined according of Yellow Perch. to Geiger and Fritz (1983) with modification (Dabrowski and Ciereszko 1994; Ciereszko et al. 1998). Trypsin inhibitor ac- METHODS tivity (APA) was expressed as U/L, where one inhibitory unit Two-year-old Yellow Perch weighing approximately 75 g (U) corresponds to the apparent amount of inhibitor able to each were randomly distributed into six 400-L tanks at 32 ± 1 block one unit of trypsin activity (defined as the hydrolysis of fish per tank. All fish were individually marked with PIT tags 1 µM of N-Benzoyl-DL-arginine-4-nitroanilide hydrochloride (Biomark, Boise, Idaho; Baras et al. 2000). This experiment in- per minute at room temperature ∼25◦C). INFLUENCE OF DIETARY LYSINE 121

Fish seminal plasma samples (control, n = 9; both Lys groups, n = 11) were extracted for FAA analysis with 0.1 M HCl in 1:1 (v/v) containing 200 µmol/L norleucine internal standard according to Cohen et al. (1989). The norleucine recovery was approximately 104%. Samples were then spun at 12,000 × g for 15 min at 4◦C. The supernatants were filtered (Millipore, 10 kDa cutoff at 2,000 × g for 90 min at 4◦C). Blanks (0.1 M HCl + 200 µmol/L norleucine) (Terjesen et al. 2004) and external standards (Sigma acid–neutral and basic amino acids) were prepared along with sample preparation. Samples, blanks, and external standards were stored at −80◦C until analysis. The same concentration of glutamine in 0.1 M HCl as external standard was prepared on the day of analysis and added to the basic amino acids standard. Amino acids were precolumn derivatized with phenylisothiocyanate (Cohen et al. 1989). Sample precip- itates were removed by 10 min of centrifugation at 10,000 × g (Terjesen et al. 2004). Free amino acids were quantified by a Waters Pico Tag reversed-phase HPLC (Waters Corporation, FIGURE 1. The mean weight (whiskers indicate SD) of Yellow Perch males from control (n = 9), (+)Lys(n = 12), and (−)Lys(n = 15) groups (from Milford, Massachusetts) equipped with an application-specific < × Kwasek et al. 2012). Different letters indicate a statistical difference at P column (3.9 mm 30 cm), a Waters 717 autosampler, two Wa- 0.05. ters 501 pumps, a Waters 441 absorbance detector at 254 nm, ◦ and a column heater set at 46 C. Eluent 1 and Eluent 2, pur- concentration and APA activity in seminal plasma than did the chased from Waters Corporation, were used throughout the in- (+) Lys and (−) Lys groups (P < 0.05) (Figure 2). vestigation as mobile phases. Each amino acid was identified by The concentration of Lys in the seminal plasma of the (−) spiking with known amino acid standards and retention times of Lys group tended to be lower than in the control and (+)Lys external standards. Free amino acid concentrations (expressed groups, but differences between treatments were not significant µ as mol/L seminal plasma) were calculated using internal and (P > 0.05) (Figure 3). The arginine (Arg) concentration was external standards (Cohen et al. 1989). significantly lower in the (−) Lys group than in the control (P ± Data are presented as mean standard deviation (SD). The < 0.05), while there was no difference between the (−)Lys data were analyzed by using a nonparametric two-sample test and (+) Lys treatments (P > 0.05). The methionine (Met) con- or a one-way analysis of variance (ANOVA) (SPSS version centration was different between (+) Lys and (−) Lys groups < 17.0; SPSS, Chicago, Illinois). A P 0.05 was considered (P < 0.05) but not different from the control group (P > 0.05) significantly different. (Figure 3). The tyrosine (Tyr) level was the highest in the (+) Lys group and differed from the level in the (−) Lys group (P < 0.05) but not from the control (P > 0.05). The threonine (Thr) RESULTS concentration showed the same trend. Among the dispensable The growth results (Kwasek et al. 2012) showed that the amino acids, the (+) Lys group had higher glutamine (Gln) lev- ± mean weight of males was larger in the control group (131 els that were significantly different from the (−) Lys group (P

Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 < + ± − 26 g) (P 0.05) than in the ( )Lys(66 10 g) and ( )Lys < 0.05) but not from the control groups (P > 0.05) (Figure 3). ± groups (54 10 g) (Figure 1). The number of individuals with The control and (+) Lys groups had higher glycine (Gly) con- − scoliosis and lordosis tended to be higher in the ( ) Lys group centrations than did the (−) Lys group (P < 0.05). There was no ± + ± ± (15 10%) than in the ( ) Lys group (2 3%; mean difference in phenylalanine (Phe), valine (Val), aspartate (Asp), SD for three replicate tanks). However, these numbers were not glutamate (Glu), asparagine (Asn), hydroxyproline (Hpro), ser- = significantly different (P 0.0809). There were no deformities ine (Ser), taurine (Tau), histidine (His), proline (Pro), isoleucine observed in the control group. Although not significant, the (Ile), leucine (Leu), tryptophan (Trp), and ornithine (Orn) con- percentage of identified (spermiating) males among all fish fed centrations among all three groups (P > 0.05) (Table 1). There + − the ( ) Lys diet was 12%, whereas among those fed ( )Lys were also no differences in the level of total FAA (P > 0.05). diet it was 18% (P = 0.3827). The survival of the fish at the end of the spawning season (July 2008) reached approximately 86% for both (−) Lys and (+) Lys groups. The mortality was mainly DISCUSSION due to the sampling and handling of the fish. Broodstock nutrition greatly affects the quality of sperm, and Sperm concentration and motility were significantly higher in a verification of sperm quality is an important step that allows control and (+) Lys groups than in the (−) Lys group (Figure 2) for the selection of superior males for increased fertilization (P < 0.05). The control group had significantly higher protein and hatching rates. Dietary ascorbic acid was shown to improve 122 KWASEK ET AL.

TABLE 1. The concentration of free amino acids in Yellow Perch seminal plasma (µmol/L). The table shows average concentrations from control, (+) Lys, and (−) Lys groups with SDs. See the text for the deﬁnitions of the amino acid abbreviations.

Control (−)Lys (+)Lys Free amino acids Average SD Average SD Average SD Asp 163 185 66 130 37 20 Glu 219 249 105 125 90 42 Hpro 48 53 48 42 51 48 Ser 2,135 479 1,586 662 2,034 820 Asn 475 147 514 215 635 149 Tau 528 274 487 166 667 239 His 394 111 468 289 519 159 Pro 457 1,198 2,441 3,639 1,181 1,602 Val 609 161 622 289 716 236 Ile 186 52 222 156 228 83 Leu 432 141 416 215 541 232 Phe 252 40 228 133 276 71 Trp 47 9 51 23 51 12 Orn524733215425 Downloaded by [Department Of Fisheries] at 20:15 15 June 2014

FIGURE 2. The quality of Yellow Perch sperm (sperm concentration, sperm motility, seminal plasma protein concentration, and seminal plasma trypsin inhibitor activity) from control, (+) Lys, and (−) Lys groups. Bars are means and whiskers indicate SD. Different letters indicate a statistical difference at P < 0.05. INFLUENCE OF DIETARY LYSINE 123 Downloaded by [Department Of Fisheries] at 20:15 15 June 2014

FIGURE 3. Mean free amino acid concentration (whiskers indicate SD) in Yellow Perch seminal plasma from control, (+) Lys, and (−) Lys groups. See text for the deﬁnitions of the amino acid abbreviations. Different letters above the whiskers indicate a statistical difference at P < 0.05. There was no statistical difference in the concentrations of Lys, Val, and indispensable amino acids (IDAA). 124 KWASEK ET AL.

sperm concentration, motility, and fertilizing ability in Rain- had plasma testosterone and 11-ketotestosterone levels lower bow Trout Oncorhynchus mykiss (Canyurt and Akhan 2008). than high-stress-responding males. Consequently, high-stress- Ciereszko and Dabrowski (1995) showed that the nutrient con- sensitive fish began spermiating earlier and maintained a sper- centration, in this case ascorbic acid, in the seminal fluid closely miation response longer than low-stress-responding fish. Our reflected the concentration of this vitamin in the broodstock diet. results in Yellow Perch suggest that males that were under po- However, it did not affect semen quality at the beginning of the tential “nutritional stress” accelerated spermiation. This specu- spawning season. A deficiency of ascorbic acid reduced sperm lation, however, needs to be investigated further. concentration and motility later during the spawning period. In Lahnsteiner (2009) indicated that in vitro supplementation of addition, Lee and Dabrowski (2004) showed that dietary sup- Met, Leu, and Ile (at 2,500 µmol/L) had a positive effect on the plementation of vitamin C can increase sperm quality in Yellow sperm viability of Rainbow Trout and Common Carp Cyprinus Perch. carpio. He suggested that Met could play an important role in Motility has been shown to be of predictive value in esti- the antioxidant defense, whereas Leu and Ile were assumed to mating reproductive success in fish (Ciereszko and Dabrowski, be sources of additional energy for spermatozoa during the pe- 1994). The present study showed that two sperm quality parame- riod of testicular storage. The results of the present experiment ters, sperm concentration and sperm motility, were significantly with Yellow Perch showed that the FAA composition of sem- higher in the control and (+) Lys groups than in the (−)Lys inal plasma is tightly regulated and easily impacted by dietary group, suggesting that Lys deficiency has a significant effect on deficiencies of indispensable amino acids. For instance, the con- the quality of sperm. The control and (+) Lys groups had signif- centration of Lys in the seminal plasma of the (−) Lys group was icantly higher protein concentration and APA in seminal plasma not different from that in the control and (+) Lys groups (P > than did the (−) Lys group. These findings are the first indica- 0.05). The Arg level, however, was significantly lower in the (−) tion, in fish, that reproductive performance can be impacted by Lys group than in the control (P < 0.05), but there was no dif- dietary indispensable amino acid levels. ference between the (−) Lys and (+) Lys treatments (P < 0.05). The highest protein concentration in seminal plasma was The Met concentrations were not different between treatment found in Yellow Perch when compared with Rainbow Trout groups (P > 0.05). When our data are compared with results and Lake Whitefish Coregonus clupeaformis (Dabrowski and obtained by Billard and Menezo (1984) (Table 2), the (+)Lys Ciereszko 1994). Since the protein concentration in fish semi- Yellow Perch seminal plasma showed higher concentrations of nal plasma is very low (1–2 mg/mL) in comparison with blood, FAAs than did the Rainbow Trout (all FAA) and Common Carp it is possible that a species-specific system of proteins with an- (except Asp, Glu, Tau, Ala, Pro, and Gly) seminal plasmas. The titrypsin activity would constitute the main system of proteins seminal plasma level of free Lys in the (+) Lys Yellow Perch in this fluid (Dabrowski and Ciereszko 1994). Seminal plasma was 20-fold higher than in Rainbow Trout and 5-fold higher inhibitors are related to blood plasma inhibitors and differ in than in Common Carp (Arg levels in Yellow Perch were 35- their affinity toward serine proteinases, but their exact role still and 37-fold higher than in Rainbow Trout and Common Carp, remains unclear. It has been shown, however, that some seminal respectively, and Met levels were 22-fold higher than in both plasma protease inhibitors that originate from blood and skin Rainbow Trout and Common Carp). Furthermore, the level of mucus might indicate contamination of milt by proteases and Lys in the seminal plasma of European Perch Perca fluviatilis their inhibitors (Kowalski et al. 2003). Mommens et al. (2008) (also known as Eurasian Perch) was shown to be 20 µmol/L suggested that a very high antiproteinase activity of Atlantic (Lahnsteiner 2010), which is a 22-fold lower concentration than Halibut Hippoglossus hippoglossus seminal plasma might be in Yellow Perch seminal plasma.

Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 related to serine proteinase inhibitors participating in the an- Yellow Perch blood plasma levels of FAAs (Kwasek et al. tibacterial defense of sperm during storage. 2012) are compared with the seminal plasma levels of FAAs The addition of 0.1% Trp to the diets of Ayu Plecoglossus in Table 2. The Lys level in the (+) Lys fed fish was approxi- altivelis resulted in the rise of the serum testosterone levels and mately two times higher in the seminal plasma than in the blood caused earlier spermiation of males and induced maturation of plasma. However, the level of Lys in the (−) Lys group was five females (Akiyama et al. 1996). Kawabata et al. (1992) indicated times higher in the seminal plasma than in the blood plasma. that the sexual behavior and sperm release in the male Rose Similarly, the blood plasma levels of Arg in the (+) Lys and (−) Bitterling Rhodeus ocellatus could be induced by FAAs, such Lys groups (134 and 94 µmol/L, respectively) were almost four as cysteine (Cys), Ser, alanine (Ala), Gly, and Lys, placed in times lower than the seminal plasma levels. An opposite pattern dialysis tubes suspended in tanks. Our study showed numeri- was observed with the amino acids Asp, Glu, Hpro, Tau, and cally more advanced maturation (spermiation) of males in the Prointhe(+) Lys group and Gly in the (−) Lys group. It is clear (−) Lys group than in the (+) Lys group. that there is a barrier between blood and seminal plasma com- It is thus possible that dietary Lys deficiency induced stress partments and amino acid metabolism and transport in these in male Yellow Perch, which could be explained by results body fluids is regulated independently. Nutritional changes reported by Castranova et al. (2005). These authors indicated are indicated far more in the FAA levels in the blood; how- that low-stress-responding male Striped Bass Morone saxatilis ever, the male reproductive system appears to be less impacted INFLUENCE OF DIETARY LYSINE 125

TABLE 2. Comparison of seminal plasma free amino acid concentrations (µmol/L) among Yellow Perch ([ + ] Lys treatment), Rainbow Trout, Common Carp (Billard and Menezo 1984), and European Perch (Lahnsteiner 2010). The table also shows blood plasma free amino acids of Yellow Perch from the (+)Lys treatment (as averages from small and large ﬁsh; please refer to Kwasek et al. [2012] for details). The Lys level in blood plasma in the (−) Lys group was on average 59 µmol/L. See the text for the deﬁnitions of the amino acid abbreviations. The abbreviation “nd” indicates the amino acid was not detected.

Yellow Perch Rainbow Trout Common Carp European Perch Free amino acids Seminal plasma Blood plasma Asp 37 347 20a 948a Glu 90 158 23b 7,110b 960 Hpro 51 87 nd Ser 2,034 312 10 320 nd Tau 667 4,553 82c 3,127c His 519 323 9 276 nd Ala 797 516 19 822 2,090 Pro 1,181 3,955 5,413 nd Tyr 778 163 355 nd Ile 228 196 173 1,120 Leu 541 397 408 nd Trp 51 29 23 70 Orn 54 28 10 21 Asn 635 75 1,120 Lys 451 281 23 93 20 Arg 485 134 14 13 7,790 Gly 1,641 1,646 14 2,118 nd Met 371 226 17 17 1,600 Thr 751 227 15 623 nd Val 716 387 13 336 530 Phe 276 162 195 280 Gln 1,415 901 nd

a Including Asp NH2. b Including Glu NH2. cIncluding phosphoethanoloamine and phosphoserine.

by diets and may also reflect the specific metabolic needs of Bradford, M. 1976. A rapid and sensitive method for quantitation of microgram sperm. quantities of protein utilizing the principle of protein-dye-binding. Analytical In conclusion, this is the first report demonstrating the neg- Biochemistry 72:248–254. Canyurt, M. A., and S. Akhan. 2008. Effect of ascorbic acid supplementation on ative effect of dietary Lys level in plant-protein-based diets on sperm quality of Rainbow Trout (S. Oncorhynchus mykiss). Turkish Journal Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 reproduction in fish. of Fishery and Aquatic Sciences 8:171–175. Castranova, D. A., V. W. King, and L. C. Woods III. 2005. The effects of stress on androgen production, spermiation response and sperm quality in high ACKNOWLEDGMENTS and low cortisol responsive domesticated male striped bass. Aquaculture This project was supported by the U.S. Department of Agri- 246:413–422. culture Special Grant 600006883. Ciereszko, A., and K. Dabrowski. 1993. Estimation of sperm concentration of Rainbow Trout, Whitefish and Yellow Perch by spectrophotometric technique. Aquaculture 109:367–373. Ciereszko, A., and K. Dabrowski. 1994. Relationship between biochemical REFERENCES constituents of fish semen and fertility: the effect of short-term storage. Fish Akiyama, T., M. Shiraaishi, T. Yamamoto, and T. Unuma 1996. Effect of dietary Physiology and Biochemistry 12:357–367. tryptophan on maturation of Ayu Plecoglossus altivelis. Fisheries Sciences Ciereszko, A., and K. Dabrowski. 1995. Sperm quality and ascorbic acid con- 62:776–782. centration in Rainbow Trout semen are affetcted by dietary vitamin C: an Baras E., C. Malbrouck, M. Houbart, P. Kestemnot, and C. Melard.´ 2000. The across-season study. Biology of Reproduction 52:982–988. effect of PIT tags on growth and physiology of age-0 cultured Eurasian Perch Ciereszko, A., B. Piros, K. Dabrowski, D. Kucharczyk, M. J. Luczynski, S. Perca fluviatilis of variable size. Aquaculture 185:159–173. Dobosz, and J. Glogowski. 1998. Serine proteinase inhibitors of seminal Billard, R., and Y. Menezo. 1984. The amino acid composition of Rainbow plasma of teleost fish: distribution of activity, electrophoretic profiles and Trout (Salmo gairdneri) seminal fluid and blood plasma: a comparison with relation to proteinase inhibitors of blood. Journal of Fish Biology 53:1292– carp (Cyprinus carpio). Aquaculture 41:255–258. 1305. 126 KWASEK ET AL.

Cohen, S., M. Meys, and T. Tarvin 1989. The pico-tag method: a manual of Lahnsteiner, F. 2010. A comparative study on the composition and impor- advanced techniques for amino acid analysis. Millipore Corporation, Milford, tance of free amino acids in semen of Gilthead Sea Bream, Sparus aurata, Massachusetts. and perch, Perca fluviatilis. Fish Physiology and Biochemistry 36:1297– Coward, K., N. R. Bromage, O. Hibbit, and J. Parrington. 2002. Gamete physi- 1305. ology, fertilization and egg activation in teleost fish. Reviews in Fish Biology Lahnsteiner, F., R. A. Patzner, and T. Weisman 1994. The testicular main ducts and Fisheries 12:33–58. and the spermatic ducts in some cyprinid fishes – II. Composition of the Dabrowski, K., and A. Ciereszko. 1994. Proteinase inhibitor(s) in seminal seminal fluid. Journal of Fish Biology 44:459–467. plasma of teleost fish. Journal of Fish Biology 45:801–809. Lanes,C.F.C.,M.H.Okamoto,A.Bianchini,L.F.Marins,andL.A.Sam- Dabrowski, K., A. Ciereszko, L. Ramseyer, D. Culver, and P. Kestemont 1994. paio. 2010. Sperm quality of Brazilian flounder Paralichthys orbignyanus Effects of hormonal treatment on induced spermiation and ovulation in the throughout the reproductive season. Aquaculture Research 41:199– Yellow Perch (Perca flavescens). Aquaculture 120:171–180. 207. Dabrowski, K., R. E. Ciereszko, A. Ciereszko, G. P. Toth, S. A. Christ, D. Lee, K.-J., and K. Dabrowski. 2004. Long-term effects and interaction of di- El-Saidy, and J. S. Ottobre. 1996. Reproductive physiology of Yellow Perch etary vitamins C and E on growth and reproduction of Yellow Perch, Perca (Perca flavescens) environmental and endocrinological cues. Journal of Ap- flavescens. Aquaculture 230:377–389. plied Ichthyology 12:139–148. Mommens, M., M. Wojtczak, A. Ciereszko, and I. Babiak. 2008. Seminal plasma Geiger, R., and H. Fritz. 1983. Trypsin. Pages 121–124 in H. U. Bergmeyer, proteins of Atlantic Halibut (Hippoglossus hippoglossus L.). Fish Physiology editor. Methods of enzymatic analysis, volume 5. Verlag Chemie, Weinheim, and Biochemistry 34:349–355. Germany. NRC (National Research Council). 1994. Nutrient requirements of poultry, 9th He, S., and L. C. Woods III. 2003. Effects of glycine and alanine on short-term edition. National Academy Press, Washington, D.C. storage and cryopreservation of Striped Bass (Morone saxatilis) spermatozoa. Rinchard, J., K. Dabrowski, J. J. Van Tessel, and R. A. Stein. 2005. Optimization Cryobiology 46:17–25. of fertilization success in Sander vitreus is influenced by the sperm: egg ratio Kawabata, K., K. Tsubaki, T. Tazaki, and S. Ikeda. 1992. Sexual behavior and ova storage. Journal of Fish Biology 67:1157–1161. induced by amino acids in the Rose Bitterling Rhodeus ocellatus ocellatus. Terjesen, B. F., K. Park, M. Tesser, M. Portella, Y. Zhang, and K. Dabrowski Nippon Suisan Gakkaishi 58:839–844. 2004. Lipoic acid and ascorbic acid affect plasma free amino acids selec- Kowalski, R., M. Wojtczak, J. Glogowski, and A. Ciereszko 2003. Gelatinolytic tively in the teleost fish Pacu (Piaractus mesopotamicus). Journal of Nutrition and anti-trypsin activities in seminal plasma of Common Carp: relationship to 134:2930–2934. blood, skin mucus and spermatozoa. Aquatic Living Resources 16:438–444. Vassallo-Agius, R., T. Watanabe, G. Yoshizaki, S. Satoh, and Y. Takeuchi. 2001. Kwasek, K., K. Dabrowski, K. Ware, J. M. Reddish, and M. Wick. 2012. The Quality of eggs and spermatozoa of Rainbow Trout fed an n-3 EFA deficient effect of lysine-supplemented wheat gluten-based diet on Yellow Perch Perca diet and its effects on the lipid and fatty acid components of eggs, semen and flavescens (Mitchill) performance. Aquaculture Research 43:1384–1391. livers. Fisheries Science 67:818–827. Lahnsteiner, F. 2009. The role of free amino acids in semen of Rainbow Trout Wojtczak, M., J. Glogowski, M. Koldras, D. Kucharczyk, and A. Ciereszko. Oncorhynchus mykiss and carp Cyprinus carpio. Journal of Fish Biology 2003. Characterization of protease inhibitors of seminal plasma of cyprinids. 75:816–833. Aquatic Living Resources 16:461–465. Downloaded by [Department Of Fisheries] at 20:15 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:16 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 A Simple Cultivation Method for Chesapeake Bay Ulva intestinalis for Algal Seed Stock Ji Lia, Patrick Kangasa & Daniel E. Terlizzib a Department of Environmental Science and TechnologyUniversity of Maryland, 0538 Animal Sciences Building, College Park, Maryland 20742, USA b Sea Grant Extension Program, Institute for Marine and Environmental TechnologyUniversity of Maryland, Columbus Center, 701 East Pratt Street, Baltimore, Maryland 21202, USA Published online: 15 Apr 2014.

To cite this article: Ji Li, Patrick Kangas & Daniel E. Terlizzi (2014) A Simple Cultivation Method for Chesapeake Bay Ulva intestinalis for Algal Seed Stock, North American Journal of Aquaculture, 76:2, 127-129, DOI: 10.1080/15222055.2013.869287 To link to this article: http://dx.doi.org/10.1080/15222055.2013.869287

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 North American Journal of Aquaculture 76:127–129, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.869287

TECHNICAL NOTE

A Simple Cultivation Method for Chesapeake Bay Ulva intestinalis for Algal Seed Stock

Ji Li* and Patrick Kangas Department of Environmental Science and Technology, University of Maryland, 0538 Animal Sciences Building, College Park, Maryland 20742, USA Daniel E. Terlizzi Sea Grant Extension Program, Institute for Marine and Environmental Technology, University of Maryland, Columbus Center, 701 East Pratt Street, Baltimore, Maryland 21202, USA

Mao et al. 2009). Research has demonstrated macroalgae to be Abstract a sustainable waste management strategy that can be integrated Ulva spp. are valuable economically as edible green algae, and with animal aquaculture for nutrient remediation. Additionally, they also have environmental value in biofiltration. Cultivation of macroalgae can be a valuable aquaculture crop in their own right. Ulva spp., in fact macroalgal cultivation generally, is limited in the USA. As part of a larger effort to develop macroalgal culti- Therefore, developing macroalgae cultivation will not only vation in the Chesapeake Bay for nutrient remediation, we de- bring economic benefits as food, animal feeds, and fertilizer, but veloped a “seeding” method for Chesapeake Bay isolates of U. also ecological benefits (Neori et al. 2004). This global trend intestinalis. Ulva intestinalis thalli were induced to produce and towards sustainability using macroalgae aquaculture can also release free-swimming zoospores by desiccation and rehydration. contribute to the solution of eutrophication in the Chesapeake The zoospores then were induced to settle on the nylon nets by dark incubation. Rhizoids from the zoospores formed on the nets, Bay. and the nets became covered with green filaments which devel- Marine tubular species of the green algal genus Ulva are oped tubular thalli. Nets with thalli can be incubated in indoors widely distributed in the Chesapeake Bay region and dom- or deployed in natural water to generate production-scale cultiva- inate the rocky intertidal and subtidal habitats. These algae tion of U. intestinalis in the Chesapeake Bay. Such cultivation can grow in the Chesapeake Bay estuaries in a wide range of salin- be integrated with the growing oyster aquaculture industry in the Chesapeake Bay area, U. intestinalis serving as biofilters to remove ity. The rapid growth of Ulva spp. is attributed to their high oyster wastes and increase economic benefit through water quality photosynthetic rates and high carbon and nitrogen uptake ca- enhancement and nutrient credits, and as an aquaculture product pacity. The wide salinity tolerance (Cohen and Fong 2004; in their own right. Ouisse et al. 2011; Imchen 2012), rapid growth rate (Middel- Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 boe and Sand-Jensen 2000; Taylor et al. 2001; Cohen and Fong Eutrophication has serious consequences in the Chesapeake 2004), and potential high economic value as a food source have Bay, USA, contributing to harmful algal blooms and hypoxia made Ulva spp. a favorable candidate for bioremediation ap- events that threaten ecosystem and human health (Boynton et al. plications and for large-scale cultivation in Chesapeake Bay 1995; Glibert et al. 2001; Glibert and Magnien 2004; Kemp et al. area. 2005). Macroalgae are considered to be important not only in Ulva spp. are important edible macroalgae in East Asia coastal food webs but also in nutrient cycling. Interest in the use (e.g., Japan), but they are not being cultivated in the Chesa- of macroalgae as biofilters has spanned several decades of re- peake Bay region. Moreover, current North American produc- search in North America and South America (Chopin et al. tion of macroalgae is minuscule compared with Asian pro- 2001; Halling et al. 2005; Buschmann et al. 2008), Europe duction which represents ∼99.7% of total global production (Halling et al. 2005; Matos et al. 2006; Abreu et al. 2011), (FAO 2009). There are only very limited commercial macroal- Africa (Msuya and Neori 2008), and Asia (Kang et al. 2008; gae activities in the USA. One of the important reasons is that

*Corresponding author: [email protected] Received June 10, 2013; accepted November 15, 2013 127 128 LI ET AL.

macroalgae are not traditional in American diets; other limiting factors include lack of permitting and of investment funding (Mumford 1990). As a result, there is limited cultivation experience in the USA. Commercial-scale macroalgae aquaculture is in early developmental stages in the USA within the Chesapeake Bay region. This study introduces a simple method to prepare the Chesapeake Bay native species U. intestinalis as algal seed stock that can lead to large-scale cultivation.

METHODS Healthy U. intestinalis thalli (∼2-kg wet weight) were collected at lower Choptank River, a tributary of Chesapeake Bay, Maryland, in the spring (water temperature ∼ 15◦C, 15 psu); they were then wrapped in absorbent paper towels, moistened with seawater, and transported in a cooler to the laboratory. Species identification was based on morphological characters. The thalli were carefully washed in seawater at ambient salinity at time of collection and cleaned with autoclaved seawater to remove attached microalgae, epiphytes, and small debris. Cleaned thalli were covered with paper towels and placed in dark to desiccate overnight at 15◦C. The thalli were then re- hydrated in a 20-gal tank with filtered seawater in the bottom. FIGURE 1. Ulva intestinalis rhizoids formed in the nylon screen and devel- These processes simulate part of a tidal cycle in order to stim- oped filaments after (a) 4 d of settlement, (b) 1 week, (c) 2 weeks, and (d) 3 ulate and facilitate the production and release of zoospores. weeks. Screens with U. intestinalis thalli can be (e, f) attached to oyster float Released free-swimming zoospores could be observed under and deployed in the natural water. a dissecting microscope from ∼ 10 min after the rehydration. Following zoospore release, the zoospore suspension was dis- DISCUSSION pensed into a shallow plate (10-cm diameter) with nylon nets Macroalgae aquaculture and development of cultivation = × (mesh size 0.5 0.5 cm) in the bottom. The plates were kept methods for species of interest contributes to “demand-side” ap- under dark conditions for at least 2 h to ensure the attachment of proaches to manage the nutrients in the water column, but also to zoospores on the screen. The “seeded” screens were incubated ◦ the development of sustainable aquaculture in the Chesapeake in Guillards’ f/2 medium at 15 psu at 15 C in a 12 light : 12 Bay. By 2050, aquaculture will be the predominant source of µ 2· dark photoperiod, light intensity of 100 mol photons/(m s). seafood (Diana et al. 2013). It is important to create more effi- The medium was replenished every 5 d. cient and diverse aquaculture systems at every production level. The cultivation of U. intestinalis can be integrated with the growing oyster aquaculture industry in the Chesapeake Bay area, the Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 macroalgae serving as biofilters to remove oyster wastes and in- RESULTS crease economic benefit through water quality enhancement and Rhizoids were observed forming and growing on the nets nutrient credits, and as an aquaculture product in their own right. under an inverted microscope within 5 d of incubation (Figure Floating culture of macroalgae around the oyster farm will not 1a). The thalli regenerated and developed a filamentous mass. impede water flows, and the algae will help oxygenate the water. The screens then were covered with green filaments and de- Oysters may even stimulate algal growth (Cognie and Barille veloped tubular thalli (Figure 1b, c). The U. intestinalis thalli 1999), adding to the sustainability of an integrated multitrophic ∼ on the screens reached over 2 cm in 3 weeks (Figure 1d). aquaculture system. Successful integration of macroalgae with The U. intestinalis thalli can be incubated indoors or grown in oyster aquaculture in the Chesapeake Bay may also increase the the field by deploying the nets to algal rafts. Seeded nets were social acceptance of macroalgae production in the USA. attached to the side of oyster floats for in situ performance evaluation. Ulva intestinalis tubular thalli turned into leafy thalli under high-intensity natural light. The thalli grew to over 35 cm FUNDING in 15–20 d in the spring (Figure 1e, f). Summer growth rates This study was supported by the Maryland Industrial Part- of greater than 20 g dry weight/(m2·d2) were found in on-land nerships program (number 4728), the University of Maryland algal raceways at Baltimore Harbor, Maryland. Agricultural Experiment Station, and National Oceanic and TECHNICAL NOTE 129

Atmospheric Administration Aquaculture Sea Grant (018383- ods of Gracilaria chilensis in an integrated seaweed and fish cage culture. 001) 2012. We thank Kevin McClarren and the Marinetics Oys- Aquaculture International 13:409–422. ter Company for their cooperation and assistance. Imchen, T. 2012. Effect of temperature, salinity and biofilm on the zoospores settlement of Enteromorpha flexuosa (Wulfen) J.Agardh. Indian Journal of Geo-Marine Sciences 41:355–358. REFERENCES Kang, Y. H., J. A. Shin, M. S. Kim, and I. K. Chung. 2008. A preliminary Abreu, M. H., R. Pereira, C. Yarish, A. H. Buschmann, and I. Sousa-Pinto. 2011. study of the bioremediation potential of Codium fragile applied to seaweed IMTA with Gracilaria vermiculophylla: productivity and nutrient removal integrated multi-trophic aquaculture (IMTA) during the summer. Journal of performance of the seaweed in a land-based pilot scale system. Aquaculture Applied Phycology 20:183–190. 312:77–87. Kemp, W. M., W. R. Boynton, J. E. Adolf, D. F. Boesch, W. C. Boicourt, G. Boynton, W. R., J. H. Garber, R. Summers, and W. M. Kemp. 1995. Inputs, Brush, J. C. Cornwell, T. R. Fisher, P. M. Glibert, J. D. Hagy, L. W. Harding, transformations, and transport of nitrogen and phosphorus in Chesapeake E. D. Houde, D. G. Kimmel, W. D. Miller, R. I. E. Newell, M. R. Roman, Bay and selected tributaries. Estuaries 18:285–314. E. M. Smith, and J. C. Stevenson. 2005. Eutrophication of Chesapeake Bay: Buschmann, A. H., Md. C. Hernandez-Gonzalez, and D. Varela. 2008. Seaweed historical trends and ecological interactions. Marine Ecology-Progress Series future cultivation in Chile: perspectives and challenges. International Journal 303:1–29. of Environment and Pollution 33:432–456. Mao, Y., H. Yang, Y. Zhou, N. Ye, and J. Fang. 2009. Potential of the seaweed Chopin, T., A. H. Buschmann, C. Halling, M. Troell, N. Kautsky, A. Neori, G. P. Gracilaria lemaneiformis for integrated multi-trophic aquaculture with scal- Kraemer, J. A. Zertuche-Gonzalez, C. Yarish, and C. Neefus. 2001. Integrat- lop Chlamys farreri in North China. Journal of Applied Phycology 21:649– ing seaweeds into marine aquaculture systems: a key toward sustainability. 656. Journal of Phycology 37:975–986. Matos, J., S. Costa, A. Rodrigues, R. Pereira, and I. S. Pinto. 2006. Experi- Cognie, B., and L. Barille. 1999. Does bivalve mucus favour the growth of their mental integrated aquaculture of fish and red seaweeds in Northern Portugal. main food source, microalgae? Oceanologica Acta 22:441–450. Aquaculture 252:31–42. Cohen, R. A., and P. Fong. 2004. Physiological responses of a bloom-forming Middelboe, A. L., and K. Sand-Jensen. 2000. Long-term changes in macroalgal green macroalga to short-term change in salinity, nutrients, and light help communities in a Danish estuary. Phycologia 39:245–257. explain its ecological success. Estuaries 27:209–216. Msuya, F. E., and A. Neori. 2008. Effect of water aeration and nutrient load Diana, J. S., H. S. Egna, T. Chopin, M. S. Peterson, L. Cao, R. Pomeroy, level on biomass yield, N uptake and protein content of the seaweed Ulva M. Verdegem, W. T. Slack, M. G. Bondad-Reantaso, and F. Cabello. 2013. lactuca cultured in seawater tanks. Journal of Applied Phycology 20:1021– Responsible aquaculture in 2050: valuing local conditions and human inno- 1031. vations will be key to success. Bioscience 63:255–262. Mumford, T. F. 1990. Nori cultivation in North America: growth of the industry. FAO (Food and Agriculture Organiziation of the United Nations). 2009 The Hydrobiologia 204:89–98. state of the world fisheries and aquaculture 2008. FAO, Rome. Neori, A., T. Chopin, M. Troell, A. H. Buschmann, G. P. Kraemer, C. Halling, Glibert, P.M., and R. E. Magnien. 2004. Harmful algal blooms in the Chesapeake M. Shpigel, and C. Yarish. 2004. Integrated aquaculture: rationale, evolution Bay, U.S.A: common species, relationships to nutrient loading, management and state of the art emphasizing seaweed biofiltration in modem mariculture. approaches, successes, and challenges. Pages 48–55 in S. Hall, D. Anderson, Aquaculture 231:361–391. J. Kleindinst, M. Zhu, and Y. Zou, editors. Harmful algae management and Ouisse, V., P. Riera, A. Migne, C. Leroux, and D. Davoult. 2011. Freshwater mitigation. Asia-Pacific Economic Cooperation, Singapore. seepages and ephemeral macroalgae proliferation in an intertidal bay: I. Effect Glibert, P. M., R. Magnien, M. W. Lomas, J. Alexander, C. L. Fan, E. Haramoto, on benthic community structure and food web. Estuarine Coastal and Shelf M. Trice, and T. M. Kana. 2001. Harmful algal blooms in the Chesapeake Science 91:272–281. and coastal bays of Maryland, USA: comparison of 1997, 1998, and 1999 Taylor, R., R. L. Fletcher, and J. A. Raven. 2001. Preliminary studies on the events. Estuaries 24:875–883. growth of selected ‘green tide’ algae in laboratory culture: effects of irra- Halling, C., G. Aroca, M. Cifuentes, A. H. Buschmann, and M. Troell. 2005. diance, temperature, salinity and nutrients on growth rate. Botanica Marina Comparison of spore inoculated and vegetative propagated cultivation meth- 44:327–336. Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:16 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Effects of Dietary Protein and Fish Density on Performance and Production Economics of Golden Shiners in Pools R. T. Lochmanna, H. Phillipsa, D. Weldona, N. Stonea & C. Englea a Aquaculture/Fisheries Center of ExcellenceThe University of Arkansas at Pine Bluff, Pine Bluff, Arkansas 71601, USA Published online: 15 Apr 2014.

To cite this article: R. T. Lochmann, H. Phillips, D. Weldon, N. Stone & C. Engle (2014) Effects of Dietary Protein and Fish Density on Performance and Production Economics of Golden Shiners in Pools, North American Journal of Aquaculture, 76:2, 130-137, DOI: 10.1080/15222055.2013.864738 To link to this article: http://dx.doi.org/10.1080/15222055.2013.864738

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 North American Journal of Aquaculture 76:130–137, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2013.864738

ARTICLE

Effects of Dietary Protein and Fish Density on Performance and Production Economics of Golden Shiners in Pools

R. T. Lochmann,* H. Phillips, D. Weldon, N. Stone, and C. Engle Aquaculture/Fisheries Center of Excellence, The University of Arkansas at Pine Bluff, 1200 North University Drive, Mail Slot 4912, Pine Bluff, Arkansas 71601, USA

Abstract We conducted an outdoor feeding trial with Golden Shiners Notemigonus crysoleucas using diets with 22% or 28% protein from different sources. The 28% protein diet with soybean (SOY) meal was the control, and the three other diets contained 22% protein mainly from SOY, corn gluten feed (CGF), or corn distillers dried grains with solubles (DDGS). The 22% protein SOY diet was also offered as a meal or pellet in separate treatments. Each of the five diets was fed to fish at two densities: 250,000 fish/ha (low density) or 750,000 fish/ha (high density). Fish with an initial weight of 0.25 ± 0.003 g (mean ± SE) were stocked into four 4.1-m3 pools per treatment and fed daily to apparent satiation for 9 weeks. Diet effects were assessed by measuring growth, survival, feed conversion, condition index, and body composition. An enterprise budget analysis was conducted to determine the economic effects of different treatments. Fish weight gain was higher at the low density regardless of diet. Weight gain was also higher in fish fed the diet with 28% protein than in those fed other diets at either density. Relative weight (Wr) was higher at the high density and in fish fed the SOY diets compared with those fed the CGF or DDGS diets. Survival was higher at the high density, and no diet effects were apparent. Gross yield was higher at the high density and in the 28% protein SOY control. Feed conversion ratio was better at the low density. Whole-body lipid was higher in fish at the low density. The most economically beneficial diet was the 28% protein SOY diet, but producers must consider feed choice in the context of their overall business model. The 22% protein diets were inadequate for rapid growth, but did not compromise survival, indicating a possible use for maintenance feeding when fish reach market size.

The Golden Shiner Notemigonus crysoleucas is the main ﬁsh and Phillips 2012). However, even the commercial “economy” species cultured for bait in Arkansas, and in the United States diets with 28% or 32% protein, less soybean meal, and more $

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 generally. The total farm gate value of baitfish sold in the USA alternative ingredients still cost more than 400/ton. in 2007 was approximately US$ 21.5 million (NASS 2009). Natural foods can be a significant source of nutrition for Even though diet cost is a lower percentage of variable costs baitfish (Lochmann and Phillips 1996), and it is possible that in baitfish production than in food-fish production, recent high Golden Shiners could be reared profitably on supplemental di- feed costs have significantly reduced production profitability. ets that contain less protein than standard diets. Zooplankton is Feeding baitfish regularly with a lower-cost diet is preferable rich in protein and fat (Creswell 1993) and could supply the ex- to decreased feeding frequency, as the latter tends to degrade tra energy and amino acids needed to sustain fish performance fish condition (Stone et al. 1997). Diets with less expensive when using supplemental diets. For instance, Golden Shiners plant ingredients such as corn gluten feed and corn distillers fed diets containing 22% or 28% protein in ponds achieved dried grains with solubles have been used successfully in catfish similar final weights after 10 weeks (Lochmann and Phillips (Robinson and Li 2008; Li et al. 2010, 2011). Corn gluten feed 2009). Presumably, intake of natural foods minimized the dif- was also an adequate protein source for growth and survival of ferences in total nutrient intake of fish fed the different diets. Golden Shiners fed diets with 28% or 32% protein (Lochmann Both diets also contained high-quality protein sources including

*Corresponding author: [email protected] Received August 30, 2013; accepted November 5, 2013

130 LOW-PROTEIN DIETS FOR GOLDEN SHINERS 131

2% fish meal and at least 30% soybean meal. By comparison, then diets were prepared and pressure-pelleted as 3.2-mm pel- most alternative plant-protein sources contain lower quantities lets in the nutrition laboratory at UAPB as described previously of essential amino acids or reduced availability of these nutri- (Chen et al. 2003). The meal form of the 22% protein SOY diet ents (Francis et al. 2001). Diets with incomplete or imbalanced was produced by putting pellets in the blender and then siev- amino acid contents often reduce growth, increase feed conver- ing the powder to retain 2-mm particles. Diets were stored at sion, and can compromise fish health (Glencross et al. 2007). It −18◦C until just prior to feeding. Diets were analyzed for pro- is still possible to formulate all-plant diets with complementary tein (Kjeldahl procedure), dry matter, and ash content according alternative protein sources that will maintain good fish perfor- to standard methods (AOAC 1995). Total lipids were extracted mance (Wu et al. 1999; Li et al. 2003; Lochmann et al. 2004; and quantified gravimetrically (Folch et al. 1957). Crude fiber Sink et al. 2010). However, if total dietary protein is reduced was analyzed according to ANKOM filter bag technique (AOAC in conjunction with greater use of alternative protein sources, 2005). the likelihood of maintaining fish performance decreases unless Fish were fed twice on weekdays and once on weekends. there is a sufficient supply of high-quality natural food available Preweighed feed equivalent to 5–8% of body weight daily to prevent nutrient and energy deficiencies. The potential for was offered to the fish until they fed to apparent satiation, successful use of supplemental diets may also depend on fish which occurred in 30 min or less. The remaining feed from the stocking density, because more natural food is available per fish preweighed ration was weighed and subtracted from the total at a low density. This is especially pertinent to baitfish culture, to estimate feed consumption. Subsamples (25% of fish in each because fish density is commonly manipulated to control growth pool) were counted and weighed every 3 weeks to track growth. (Stone et al. 1997). The percentage of feed offered was reduced after each sam- To address these issues, we conducted a feeding trial with pling because the fish consumed a smaller percentage of feed Golden Shiners stocked at two densities to determine the effects as a percentage of body weight as they grew. Temperature and of supplemental diets with 22% protein from soybean meal dissolved oxygen (DO) were measured in the morning and after- (SOY) or alternative ingredients on fish performance and pro- noon daily in each pool using a dissolved oxygen meter (YSI 55 duction economics. A control diet with 28% protein and SOY DO, Yellow Springs Instruments, Yellow Springs, Ohio). Major as the main protein source was included for comparison. The zooplankton groups (rotifers, copepod nauplii, cyclopoid cope- alternative ingredients in the 22% protein diets were solvent- pods, and cladocerans) were also determined in each pool at extracted distillers dried grains with solubles (SE-DDGS) and weeks 4 and 8. Six 1-L water samples were obtained with a tube corn gluten feed (CGF). Diet and density effects were assessed sampler that encompassed the entire water column. The sam- by measuring growth, survival, feed conversion, condition in- ples were concentrated by straining them through a Wisconsin dex, and body composition. plankton net with 70-µm mesh and then preserving the resulting concentrate in 70% isopropyl alcohol. Samples were identified then quantified using a Sedgwick–Rafter cell and a microscope. METHODS The pH (pH meter, Denver Instruments, Colorado), total am- Feeding trial.—A 9-week feeding trial was conducted with monia nitrogen (TAN; salicylate–cyanurate method), and nitrite Golden Shiners in 2,946-L outdoor unfertilized pools at the were measured in each pool weekly with a Hach DR/890 col- University of Arkansas at Pine Bluff (UAPB). The aerated pools orimeter test laboratory (Hach Company, Loveland, Colorado). were filled with reservoir water approximately 1 week before Un-ionized ammonia was calculated from the results of the to- stocking and maintained as static systems during the trial to tal ammonia nitrogen and pH measurements. Hardness values

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 encourage plankton blooms. Groups of 82 fish corresponding were obtained using the U.S. Environmental Protection Agency to the low density of 250,000 fish/ha or 266 fish corresponding Manver 2 Buret Titration method (Hach method 8226) and alka- to the high density of 750,000 fish/ha were stocked into each of linity was measured with a Hach digital titrator (model 16900) four pools per treatment. The calculated individual initial weight using 1.6 N sulfuric acid and phenolphthalein as an indicator. of fish was 0.25 g. Water quality data are summarized in Table 2, and all parameters Four experimental diets were formulated (Table 1) to con- were within acceptable limits for Golden Shiners (Stone et al. tain either 28% protein with SOY as the main protein source, 1997). or 22% protein with SOY, CGF, SE-DDGS. The 28% protein At the end of the trial all fish were counted and weighed in SOY diet was the control. Because some baitfish producers also groups. One group of 25 fish per pool was retained for measure- use a meal form of diet rather than pellets, one diet (22% SOY) ments of lengths and weights of individual fish to determine was offered either as a pellet or meal in separate treatments. The relative weight (Liao et al. 1995). Each group was then ho- 22% SOY diet was chosen to test the meal form because there mogenized into a single pooled sample and used for proximate is very little information on the performance of Golden Shiner analysis as described for the diets. fed reduced-protein (<28%) meals. Dry diet ingredients were Economic analysis.—Enterprise budgets for Golden Shiner finely ground to a particle size of 1.0 mm in a Thomas–Wiley production were updated with 2012 costs and used to com- mill (Arthur H. Thomas Company, Philadelphia, Pennsylvania), pare the economic effects of the diets evaluated (Stone et al. 132 LOCHMANN ET AL.

TABLE 1. Ingredient (%, as fed, except where noted otherwise) and analyzed composition of diets with 22% or 28% protein fed to Golden Shiners at a low (250,000 ﬁsh/ha) or high (750,000 ﬁsh/ha) density in outdoor pools for 9 weeks.a

Ingredient 28-SOY (control) 22-SOYb 22-SE-DDGS 22-CGF Soybean meal (48%) 41.0 28.0 20.0 21.0 Cottonseed meal (41%) 10.0 8.0 8.0 8.0 DDGS 0.0 0.0 20.0 0.0 Corn gluten feed 0.0 0.0 0.0 23.0 Corn 25.0 43.0 32.0 27.0 Wheat bran 19.0 16.0 15.0 16.0 Poultry fat 2.0 2.0 2.0 2.0 Soybean oil 2.0 2.0 2.0 2.0 Vitamin premixc 0.5 0.5 0.5 0.5 Mineral premixd 0.5 0.5 0.5 0.5 Crude proteine 28.9 22.0 22.0 20.3 Lipide 5.5 6.9 6.2 6.9 Ashe 5.5 5.5 4.5 5.1 Moisture 12.5 14.7 14.0 16.0 Fibere 4.4 4.6 5.3 5.3 Nitrogen-free extractf 43.2 46.3 48.0 46.4 Energy : protein (kJ/g)g 48.8 63.7 63.8 67.7

aThe diet with 28% protein (28-SOY) was the control diet, which met the known nutrient requirements of Golden Shiners (Lochmann and Phillips 2009). The other diets contained 22% protein and were designated by their distinctive protein sources: SOY, soybean meal; SE-DDGS, solvent-extracted corn distillers dried grains with solubles; CGF, corn gluten feed. Diets with 22% protein were designed to be supplemental and contained suboptimal levels of protein for growth in the absence of natural foods (Lochmann and Phillips 1994). bThe 22% protein SOY diet was offered in the form of a pellet or meal (two different treatments) that had identical nutrient composition. cVitamin premix contains (as g/kg of premix): ascorbic acid, 50.0; d-calcium pantothenate, 5.0; choline chloride, 100.0; inositol, 5.0; menadione, 2.0; niacin, 5.0; pyridoxine HCl, 1.0; riboﬂavin, 3.0; thiamine HCl, 0.5; dl-α-tocopherol acetate (250 IU/g), 8.0; vitamin A acetate (20,000 IU/g), 5.0; vitamin micromix, 10.0; cellulose, 805.5. Vitamin micromix contains (as g/kg of micromix): biotin, 0.5; cholecalciferol (1 µg = 40 IU), 0.02; folic acid, 1.8; vitamin B12, 0.02; cellulose, 97.66. dMineral premix contains (as g/kg of premix): calcium phosphate (monobasic) monohydrate, 136.0; calcium lactate pentahydrate, 348.49; ferrous sulfate heptahydrate, 5.0; magnesium sulfate heptahydrate, 132.0; potassium phosphate (dibasic), 240; sodium phosphate (monobasic) monohydrate, 88.0; sodium chloride, 45.0; aluminum chloride hexahydrate, 0.15; potassium iodide, 0.15; cupric sulfate pentahydrate, 0.50; manganous sulfate monohydrate, 0.70; cobalt chloride hexahydrate, 1.0; zinc sulfate heptahydrate, 3.0; sodium selenite, 0.011. eAnalyzed composition (%, as fed). fNitrogen-free extract (NFE), calculated as NFE = 100 − (protein + lipid + moisture + ash + ﬁber), is an estimate of soluble carbohydrate. gEstimated energy content of the diets was based on values of 16.7, 16.7, and 37.7 kJ/g of carbohydrate, protein, and lipid, respectively (Serrano et al. 1992).

2008). The base budget was developed based on results of the fish size alone can affect body composition (Sutton et al. 2000). control diet (28-SOY). Separate budgets were developed for However, fish size did not have a significant effect on body each of the other diets for each of the stocking densities tested composition (P > 0.05), so size was excluded from the final by substituting values of production parameters that were signif- analysis. icantly different due to the diets evaluated in the relevant budget Water quality data, including rotifer abundance, were ana-

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 spreadsheet. In the base budget, the price used for floating pel- lyzed initially with repeated-measures ANOVA incorporating leted feed was $385/metric ton. Price differentials for the 22% diet and density as independent variables. Because there were protein diets, compared with that of the base control diet, were few diet or density effects, the effects over time are presented in estimated based on relative differences in individual ingredient the body of Table 2. In cases where there were diet or density ef- costs and were −$25/metric ton for the 22% protein SOY pel- fects (but not both), data were analyzed using one-way ANOVA let diet, −$56/metric ton for the 22% protein SOY meal diet, with either diet or density as the independent variable and the −$37/metric ton for the 22% CGF diet, and −$20/metric ton results are included in footnotes in Table 2. Treatment means for the 22% protein diet with SE-DDGS. were considered different at P < 0.05, and specific differences in Statistical analysis.—The experiment was a 2 × 5 factorial means were identified using Fisher’s least significant difference design. Percentage data were arcsine-square-root transformed test. Correlations between fish weight gain and each group of prior to statistical analysis. Performance data and body com- plankton were analyzed with a Z-test. Zooplankton abundance position were analyzed with a two-way ANOVA in Statview for groups other than rotifers was very low, so data were an- (SAS Institute, Cary, North Carolina) with diet and density as alyzed using descriptive statistics to obtain overall ranges for the independent variables. In addition, body composition data each group for the whole trial. These results are presented in the were analyzed using final fish weight as a covariate because text. LOW-PROTEIN DIETS FOR GOLDEN SHINERS 133

TABLE 2. Water quality conditions in a feeding trial with Golden Shiners fed RESULTS diets containing 22% or 28% protein with different protein sources in pools for 9 weeks.a Growth, Feed Utilization, and Influence of Natural Productivity Condition Mean ± SE P-value (time) Mean individual weight gain was higher in fish fed the 28% Temperature (◦C), a.m. 28.4 ± 0.03 <0.0001 (increased) protein SOY diet than in all other diets (Table 3). Weight gain Temperature (◦C), p.m. 32.0 ± 0.03 <0.0001 (fluctuated) was also higher at the low density regardless of diet. Feed intake Dissolved oxygen 7.0 ± 0.014 <0.0001 (fluctuated) was higher in fish fed the 28% protein SOY diet than in those (mg/L), a.m.b fed 22% pelleted diets with SE-DDGS, CGF, or SOY. Feed in- Dissolved oxygen 7.3 ± 0.02 <0.0001 (decreased) take was similar in fish fed the 28% protein SOY diet or 22% (mg/L), p.m. protein SOY meal diet, and was higher in fish at the low density Total ammonia nitrogen 0.03 ± 0.01 <0.0001 (decreased) than at the high density regardless of diet. Feed conversion ratio (mg/L) (FCR) was highest in fish fed the 22% protein SOY meal diet Nitrite (mg/L) 0.003 ± 0.000 <0.0001 (increased) and similar among fish fed the other diets (Table 3). The FCR pH 8.5 ± 0.02 <0.0001 (increased) was also lower in fish at the low density than at the high density. Hardness (mg/L)c 145.3 ± 1.9 <0.0001 (fluctuated) Survival was higher at the high density and there were no differ- Alkalinity (mg/L) 135.9 ± 2.4 <0.0001 (decreased) ences due to diet (Table 3). Gross yield was highest in fish fed Rotifers/L 72.6 ± 15.8 0.03 (decreased) the 28% protein SOY diet and higher in fish at the high density than at the low density (Table 3). Final fish weight gain was aData were analyzed using repeated measures ANOVA with diet, density, and time as the independent variables. Diet and density were not significant (P > 0.05) in most cases. not correlated with the abundance of rotifers, copepod nauplii, bMorning (a.m.) dissolved oxygen was higher at the low density (7.1 ± 0.02 mg/L) cyclopoid copepods, or cladocerans (P > 0.05). Rotifer abun- than the high density (6.9 ± 0.02 mg/L). dance (mean ± SE) was higher at the low density (162.5 ± cHardness differed by diet on the last two sampling dates, but the differences were not ± consistent and had no known biological significance (data not shown). 52/L) than at the high density (46.0 23.3/L) on the first

TABLE 3. Mean individual ﬁnal weight gain, relative weight, feed conversion, survival, and net yield of Golden Shiners fed diets with 22% or 28% proteinat two densities for 9 weeks. Abbreviations are: SOY, soybean as meal or pellet; SE-DDGS, solvent-extracted distillers grains with solubles; CGF, corn gluten feed.

Mean individual Mean individual Feed Survival Relative Gross yield a b c Diet and statistics weight gain (g) feed consumed (g, dry) conversion (%) weight (Wr) (kg/ha) Low density (250,000 ﬁsh/ha) 28-SOY (control) 1.5 2.2 1.5 91.2 102.5 380.1 22-SOY (pellet) 1.3 1.9 1.5 91.8 101.5 333.9 22-SOY (meal) 1.3 2.1 1.7 92.9 105.0 337.4 22-SE-DDGS 1.4 2.1 1.5 90.1 97.6 353.4 22-CGF 1.4 1.8 1.3 91.5 98.9 353.0 High density (750,000 ﬁsh/ha) 28-SOY (control) 0.8 1.5 1.8 99.5 105.5 694.8 22-SOY (pellet) 0.8 1.4 1.8 97.1 105.0 619.1

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 22-SOY (meal) 0.7 1.4 2.2 97.7 105.6 553.8 22-SE-DDGS 0.7 1.2 1.8 98.1 99.9 574.5 22-CGF 0.7 1.3 1.8 98.7 98.3 583.6 Pooled SE 0.34 0.06 0.05 0.80 3.74 18.12 Two-way ANOVA results: P (Diet) 0.01d 0.006e 0.0003f 0.73 <0.0001g 0.01h P (Density) <0.0001 <0.0001 <0.0001 <0.0001 0.01 <0.0001 (low > high) (low > high) (low < high) (low < high) (low < high) (low < high) P (Diet × Density) 0.43 0.52 0.36 0.43 0.26 0.24

aMean individual weight gain, feed conversion, and survival are means of four replicates per density consisting of 88–266 fish each. bDry feed consumed/fish weight gain. c Relative weight (Wr) = (weight/Ws)·100, where weight is in grams, logWs =−5.593 + 3.302 logTL, and TL is in millimeters. Relative weight is based on lengths and weights of 25 individual fish per replicate (100 per diet at each density). dWeight gain of fish fed the 28% protein SOY diet was higher than all others. eFeed intake was higher in fish fed the 28% protein SOY diet than those fed the 22% protein SOY pellet, 22% protein CGF, or 22% protein SE-DDGS diets. Feed intake of fish fed the 28% protein SOY diet or 22% protein SOY meal diet was similar. fFeed conversion was higher in fish fed the 22% protein SOY meal diet than all others. gRelative weight of fish fed any of the SOY diets was higher of that of fish fed the 22% protein CGF or 22% protein SE-DDGS diets. hGross yield of fish fed the 28% protein SOY diet was higher than all others. 134 LOCHMANN ET AL.

TABLE 4. Mean proximate composition (%, wet) of whole Golden Shiners sampling date, but over time rotifer abundance decreased across fed diets with 22% or 28% protein at two densities for 9 weeks. Each mean all diets at both densities (Table 2). There were no density effects represents four pooled samples consisting of 25 individual fish each. Abbrevia- tions are: SOY,soybean as meal or pellet; SE-DDGS, solvent-extracted distillers on other zooplankton groups. One-way ANOVA using diet as grains with solubles; CGF, corn gluten feed. the independent variable revealed some significant but transient effects on cyclopoids and cladocerans (data not shown). How- Diet and Dry ever, the abundance of zooplankton groups other than rotifers statistics Protein Lipid matter Ash was very low overall. Ranges for both sampling dates combined Low density (250,000 fish/ha) were: copepod nauplii, 0–10/L; cyclopoid copepods, 0–3/L; and cladocerans, 0–28/L. 28-SOY 12.9 8.7 25.4 2.6 Whole-body protein was similar in fish fed different diets at 22-SOY (pellet) 13.7 8.4 26.0 2.9 either density (Table 4). Whole-body lipid was higher in fish at 22-SOY (meal) 12.6 9.0 26.0 2.6 the low density, and there were no differences in lipid due to 22-SE-DDGS 12.3 9.0 27.1 3.1 diet (Table 4). There was a significant interaction between diet 22-CGF 12.7 8.5 25.6 2.9 and density on whole-body dry matter (Table 4). There were no High density (750,000 fish/ha) diet effects at the low density, while dry matter was higher in 28-SOY 12.9 7.9 26.5 2.8 fish fed any of the SOY diets compared with the CGF or SE- 22-SOY (pellet) 12.2 8.2 25.1 2.7 DDGS diets (Table 4). Whole-body ash was higher in fish fed 22-SOY (meal) 12.4 7.6 24.9 2.6 the SE-DDGS or CGF diets than in those fed the 22% protein 22-SE-DDGS 12.6 7.5 22.3 3.1 SOY meal or 28% protein SOY diets, and that of fish fed the 22-CGF 12.6 6.9 22.7 3.1 22% protein SOY pellet was intermediate (Table 4). There were Pooled SE 0.39 0.29 0.44 0.08 no differences in whole-body ash due to fish density. Two-way ANOVA results: Economic analysis.—The most economically beneficial diet P (Diet) 0.84 0.57 0.046a 0.0017b was the control diet (28% SOY protein) used as the basis of P (Density) 0.39 0.0003 0.0001a 0.41 comparison for both stocking densities evaluated (Table 5). In (low > high) the low-density treatment, the 22% protein diets showed reduced P (Diet × Density) 0.50 0.35 0.0007a 0.24 net returns that ranged from −$197/ha to −$363/ha, while the reduced net returns for the high-density treatment ranged from aBecause of the significant interaction between diet and density on dry matter, separate one-way ANOVA tests were used to analyze diet effects on dry matter at each density. −$575/ha to −$944/ha. There were no diet effects at the low density (P = 0.32). At the high density, dry matter was higher in fish fed any of the SOY diets compared with fish fed the CGF or SE-DDGS diets (P = 0.002). DISCUSSION bWhole-body ash was higher in fish fed the SE-DDGS or CGF diets than those fed the 22% protein SOY meal or 28% protein SOY diets, and that of fish fed the 22% protein SOY Growth of Golden Shiners fed diets with 22% protein and pellet diet was intermediate. different protein sources was lower than that of fish fed the 28%

TABLE 5. Results of enterprise budget analysis of economic effects of three 22% protein diets fed to Golden Shiners compared with a 28% protein SOY control diet. Abbreviations are: SOY, soybean as meal or pellet; SE-DDGS, solvent-extracted distillers grains with solubles; CGF, corn gluten feed.

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 Feed cost Total variable Total costs Net returns Difference from Treatment ($/metric ton) costs ($/ha) ($/ha)a ($/ha) control ($/ha) Low density (250,000 ﬁsh/ha) 28-SOY (control) 386 1,841 4,399 −1,006 0 22-SOY (pellet) 361 1,792 4,350 −1,369 −363 22-SOY (meal) 330 1,775 4,333 −1,321 −315 22-SE-DDGS 366 1,809 4,367 −1,212 −206 22-CGF 348 1,797 4,355 −1,203 −197 High density (750,000 ﬁsh/ha) 28-SOY (control) 386 2,320 4,878 1,325 0 22-SOY (pellet) 361 2,219 4,777 750 −575 22-SOY (meal) 330 1,909 4,467 477 −848 22-SE-DDGS 366 2,190 4,748 381 −944 22-CGF 348 2,174 4,732 477 −848

aTotal ﬁxed costs are a constant $2,558/ha because the only changes were in the variable inputs of feed quantity, prices, and stocking density. LOW-PROTEIN DIETS FOR GOLDEN SHINERS 135

protein SOY diet at two different stocking densities. Higher- dietary energy to protein (E:P ratio) can also affect growth. The protein diets usually promote better growth of small fish (Wil- optimal dietary E:P ratio for Golden Shiners is 40.6 kJ/g dietary son 2002), but there are few published reports of the effects of protein (Lochmann and Phillips 1994). The 28% protein SOY diets with less than 28% protein on Golden Shiners. In aquaria, control diet contained 48.8 kJ/g protein, while the 22% protein growth of Golden Shiners fed semipurified diets with 21% pro- diets contained much higher E:P ratios of 63.7–67.7 kJ/g protein. tein was lower than that of fish fed diets with 31% or 34.5% Excess dietary energy can reduce feed intake by producing satia- protein (Lochmann and Phillips 1994). However, there were tion quickly. Decreased consumption of a low-protein diet could no natural foods available to supplement the prepared diets. In limit intake of protein for growth even further (NRC 2011). Feed outdoor systems, this species acquires a significant amount of intake of fish fed the 28% protein SOY control diet was higher nutrition (40–83%) from natural foods (Lochmann and Phillips than that of fish fed any of the 22% protein pelleted diets, which 1996), which might justify the use of lower-protein supple- is consistent with this explanation. High-energy prepared diets mental feeds. Golden Shiners in ponds fed diets with 22% or in combination with protein-rich plankton could spare protein 28% protein had similar growth and yield after 10 weeks, even for growth, as observed in previous studies with Golden Shiners though growth of fish fed the 28% protein diet was higher at 4 and Goldfish Carassius auratus (Lochmann and Phillips 2009). weeks (Lochmann and Phillips 2009). Natural foods were not However, zooplankton abundance was low in the present study quantified in that study, but presumably they contributed to the and the low-protein diets did not support the same growth as the similar performance obtained in fish fed different diets. Fish in 28% protein SOY control diet with a more favorable E:P ratio. the pond study were also considerably larger at stocking (1.24 g) The only effects of feed form in this study were higher ap- than those in the current study (0.24 g), and higher protein levels parent feed intake of the meal compared with the pellet form are more critical for growth of smaller fish. of the 22% protein SOY diet, and higher FCR of fish fed the Diets in the current study were deliberately formulated with- 22% protein SOY meal diet. Because the composition and en- out animal protein sources to follow the current trend of using ergy content of the 22% protein SOY pellet and meal diets was mostly plant products in fish diets to reduce cost. A traditional identical, the feed form appeared to cause the effects. Channel high-quality protein source (soybean meal) was tested against al- Catfish and Common Carp Cyprinus carpio are known to use ternative plant proteins (CGF and DDGS), which have replaced pellets more efficiently than meals because they ingest fewer of some of the soybean meal in commercial diets for Channel Cat- the small meal particles (Lovell 1989). To the contrary, Golden fish Ictalurus punctatus and other aquatic species. In general, the Shiners fed the meal in this study appeared to consume more alternative ingredients have performed well in catfish (Robinson feed than those fed the pellet form of the same formula. It is et al. 2001; Robinson and Li 2008; Li et al. 2011; Lochmann possible that more of the meal particles sank before they could et al. 2012). Corn gluten feed at 20% of the diet also supported be consumed by the fish, which resulted in an overestimation good growth and survival of Golden Shiners fed diets with 28% of feed intake. This would also explain how fish fed the meal or 32% protein (Lochmann and Phillips 2012). However, weight appeared to consume as much feed as those fed the 28% protein gain was positively correlated with zooplankton abundance in SOY diet, but fish fed the meal had the poorest feed conver- the latter study, and in the current study zooplankton abundance sion. In addition, the high surface area of small feed particles was low and did not correlate with fish growth. The inability facilitates nutrient leaching, which can impair fish performance to predict or control the natural food supply makes it difficult (Lovell 1989). Overall, there did not appear to be any practical to formulate an appropriate supplemental feed for outdoor pro- advantage of using the meal compared with the pellet. How- duction systems (Hepher and Pruginin 1981). ever, commercial meals for fish are often made using different

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 The dietary protein requirement is influenced by the amino formulas and processing conditions than those used for pellets. acid composition of the protein sources, as well as the availabil- In that case, differences in diet ingredients and processing (e.g., ity of the amino acids (Wilson 2002). Corn gluten feed and dis- extrusion) will also influence fish performance. tiller’s grains contain substantially lower concentrations of the Feed consumption was higher in fish at the lower density essential amino acids lysine and methionine than soybean meal regardless of diet, even though all fish were fed to apparent (Robinson and Li 2013). The essential amino acid requirements satiation. It is not clear how density caused this effect, because of Golden Shiners are unknown, but quantitative requirements water quality was similar at both densities and all parameters are similar for most fish and deficiency rapidly reduces weight were acceptable for Golden Shiners (Stone et al. 1997). In a gain (NRC 2011). Because weight gain was lower in fish fed previous pond study with Golden Shiners, growth of fish at a all of the 22% protein diets (including the SOY diet) compared high density (750,000 fish/ha) was also limited by unknown with the 28% protein SOY control, it seems likely that growth density-dependent factors compared with that of the fish at a was reduced partly by an absolute amino acid deficiency. Zoo- lower density (375,000 fish/ha) (Lochmann et al. 2001). Fish plankton contain all essential amino acids (Aragao et al. 2004; were fed at a fixed percentage of body weight (4%) in the pond Mitra et al. 2007), but the limited supply of natural foods in study and feed intake was not measured. The current study this study did not compensate for nutrient deficiencies in the is the first to quantify higher feed intake in Golden Shiners supplemental diets with 22% protein. Differences in the ratio of at a low density compared with a higher one when feed was 136 LOCHMANN ET AL.

offered to apparent satiation and water quality conditions were 502 kg/ha in 2008 (Stone et al. 2008) at feed costs of $270/met- not limiting. The Golden Shiner is a shoaling species that tends ric ton. Since the low-density yields in this current study were to form groups of up to 250 individuals in nature (Krause et al. all less than 502 kg/ha, none of the low-density treatments 1996). The high-density treatment in this study had 266 fish were profitable. In the high-density treatments, all diets were per pool, and there were no indications that this density was profitable because all diets produced yields greater than the stressful except for their lower feed intake compared with that breakeven yield. However, the lower yields of the 22% protein in the low-density treatment (88 fish per pool). It is possible that diets resulted in lower net returns than the control 28% protein physiological indices of stress that we did not measure were SOY diet. The 22% protein diets were less expensive than the present in fish at the high density (as reviewed by Wedemeyer 28% protein SOY diet. However, the significantly lower yields 1997). However, condition factor and survival were higher at of the 22% protein diets resulted in reduced revenue that more the high density. The relationship between growth and survival than offset the reduced feed costs. in fish is complex, and can be altered by predation pressure, The economic analysis provides insight into the overall trends physiological stressors such as poor water quality, and increased identified in the tank study of yields obtained from the diets immune response in fish exposed to pathogens (Mangel and tested. However, other factors may affect diet performance and Stamps 2001). However, there was no evidence that any of yield in commercial ponds. The best choice of feed should in- these factors were present in the current study. Therefore, more clude consideration of feed cost and the yield of fish, but within research is needed to identify the specific density-dependent the context of the farm’s overall business model. effects on feed intake, growth, and survival in Golden Shiners. In summary, the highest gross yield was obtained in fish at Whole-body lipid was higher in fish at the low density re- the high density fed the 28% protein diet with soybean meal. gardless of diet. Golden Shiners consumed more feed at the low The most economically beneficial diet in this outdoor tank study density and gained more weight than fish at the high density. was also the 28% protein SOY diet. However, other factors may Although zooplankton abundance was low, it is also possible affect diet performance and yield in commercial ponds. There- that natural food intake per fish was higher at the low density. fore, feed choice must also be determined within the context Higher feed intake increases growth (Brett 1979) and lipid ac- of a farm’s overall business model. Reduced growth was the cumulates to a greater extent in rapidly growing fish (Jobling main effect of the 22% protein diets, so these diets would not 2001). Higher body fat might improve the postharvest resilience be adequate when fast growth is needed. However, survival was of Golden Shiners, as feeding is limited during transport and high on the lower-protein diets, indicating their potential use for retail display. Whole-body ash was higher in fish fed the SE- maintenance feeding when fish reach market size and must be DDGS diet than for those fed other diets. The reason for this is held in good condition until they are sold. not obvious, but the values obtained across diets (2.6–3.1%) are within the normal range for fish. Whole-body dry matter was higher in fish at the high density and fed any of the SOY diets ACKNOWLEDGMENTS compared with those fed the CGF or SE-DDGS diets. The soy- We thank Anderson’s Minnow Farm, Inc., for donating fish bean diets contained more digestible energy than the corn diets, for this study. Paxton Harper, Olivia Peek, and Steffani Hoffman which could increase dry matter retention in the carcass (NRC provided technical assistance. We thank Anita Kelly, Jeonghwan 2011). The effect was not observed at the low density, possi- Park, and Ramone Brown for reviewing this manuscript. bly because of the higher lipid accumulation in all fish at the lower density. Interestingly, relative weight (Wr; a measure of

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 fish condition) was higher in fish at the high density and higher REFERENCES in fish fed any of the SOY diets than it was in those fed the AOAC (Association of Official Analytical Chemists). 1995. Official methods of CGF or SE-DDGS diet. Again, the diet effects can be explained analysis, 16th edition. AOAC International, Arlington, Virginia. AOAC (Association of Official Analytical Chemists). 2005. Official methods of by higher levels of digestible energy in the soy diets, leading to analysis, 18th edition. AOAC International, Gaithersburg, Maryland. better fish condition. However, in previous studies with Golden Aragao, C., L. E. Conceicao, M. T. Dinis, and H. Fyhn. 2004. Amino acid pools Shiners and other species higher condition factor was associ- of rotifers and Artemia under different conditions: nutritional implications ated with higher whole-body lipid (Brown and Murphy 1991; for fish larvae. Aquaculture 234:429–445. Lochmann and Phillips 2009, 2012). Body composition fluctu- Brett, J. R. 1979. Environmental factors and growth. Pages 599–676 in W. S. Hoar, D. J. Randall, and J. R. Brett, editors. Fish physiology, volume 8. ates rapidly during ontogeny, making the relationship between Academic Press, New York. condition and body composition of small, fast-growing fish less Brown, M. L., and B. R. Murphy. 1991. Relationship of relative weight (Wr) predictable (Pangle and Sutton 2005). High survival (>97%) of to proximate composition of juvenile Striped Bass and hybrid striped bass. Golden Shiners at the high density indicated that the fish were Transactions of the American Fisheries Society 120:509–518. not compromised by having less body fat than those at the low Chen, R., R. Lochmann, A. Goodwin, K. Praveen, K. Dabrowski, and K.-J. Lee. 2003. Alternative complement activity and resistance to heat stress in density during the 9-week trial. Golden Shiners (Notemigonus crysoleucas) are increased by dietary vitamin The economics of baitfish production are sensitive to yields. C levels in excess of requirements for prevention of deficiency signs. Journal Breakeven yields of Golden Shiners were estimated to be of Nutrition 133:2281–2286. LOW-PROTEIN DIETS FOR GOLDEN SHINERS 137

Creswell, R. L. 1993. Aquaculture desk reference. Van Nostrand Reinhold, New Lochmann, R. T., and H. Phillips. 2012. Effects of diets with 28% or 32% protein York. and meat and bone meal or corn gluten feed on performance of Golden Shiner Folch, J., M. Lees, and G. H. Sloane-Stanley. 1957. A simple method for in pools. North American Journal of Aquaculture 74:457–462. the isolation and purification of total lipids from animal tissues. Journal of Lovell, T. 1989. Nutrition and feeding of fish. Van Nostrand Reinhold, New Biological Chemistry 226:497–509. York. Francis, G., H. P. S. Makkar, and K. Becker. 2001. Antinutritional factors in Mangel, M., and J. Stamps. 2001. Trade-offs between growth and mortality plant-derived alternate fish ingredients and their effects in fish. Aquaculture and the maintenance of individual variation in growth. Evolutionary Ecology 199:197–227. Research 3:583–593. Glencross, G. D., M. Booth, and G. L. Allan. 2007. A feed is only as good as Mitra, G., P. K. Mukhopadhyay, and S. Ayyappan, 2007. Biochemical com- its ingredients—a review of ingredient evaluation strategies for aquaculture position of zooplankton community grown in freshwater earthen ponds: nu- feeds. Aquaculture Nutrition 13:17–34. tritional implication in nursery rearing of fish larvae and early juveniles. Hepher, B., and Y. Pruginin 1981. Commercial fish farming. Wiley, New York. Aquaculture 272:346–360. Jobling, M. 2001. Nutrient partitioning and the influence of feed composition on NASS (National Agricultural Statistics Service). 2009. Census of agriculture body composition. Pages 354–375 in D. Houlihan, T. Boujard, and M. Jobling, (2007), United States, summary and state data, volume 1, geographic area editors. Food intake in fish. Blackwell Scientific Publications, Oxford, series part 51. U.S. Department of Agriculture, NASS AC-07-A-51, Wash- UK. ington, D.C. Krause, J., J.-G. Godin, and D. Brown. 1996. Size-assortativeness in multi- NRC (National Research Council). 2011. Nutrient requirements of fish and species fish shoals. Journal of Fish Biology 49:221–225. shrimp. National Academy Press, Washington, D.C. Li, M. H., E. H. Robinson, B. G. Bosworth, D. F. Oberle, and P. M. Lucas. Pangle, K. L., and T. M. Sutton. 2005. Temporal changes in the relationship 2011. Use of corn gluten feed and cottonseed meal to replace soybean meal between condition indices and proximate composition of juvenile Coregonus and corn in diets for pond-raised Channel Catfish. North American Journal artedi. Journal of Fish Biology 66:1060–1072. of Aquaculture 73:153–158. Robinson, E. H., and M. H. Li. 2008. Replacement of soybean meal in Channel Li, M. H., E. H. Robinson, D. F. Oberle, and P. M. Lucas. 2010. Effects of Catfish, Ictalurus punctatus, diets with cottonseed meal and distillers dried various corn distillers by-products on growth and feed efficiency of Channel grains with solubles. Journal of the World Aquaculture Society 39:521–527. Catfish, Ictalurus punctatus. Aquaculture Nutrition 16:188–193. Robinson, E. H., and M. H. Li. 2013. Comparative quality of foodstuffs for Li, M. H., D. J. Wise, B. B. Manning, and E. H. Robinson. 2003. Effect catfish feeds. Mississippi Agricultural and Forestry Experiment Station Tech- of dietary total protein and animal protein on growth and feed efficiency nical Bulletin 1205. of juvenile Channel Catfish Ictalurus punctatus and their response to Ed- Robinson, E. H., M. H. Li, and B. B. Manning. 2001. Evaluation of corn wardsiella ictaluri challenge. Journal of the World Aquaculture Society 34: gluten feed as a dietary ingredient for pond-raised Channel Catfish Ictulurus 223–228. punctatus. Journal of the World Aquaculture Society 32:68–71. Liao, H., C. L. Pierce, D. H. Wahl, J. B. Rasmussen, and W. C. Leggett. 1995. Serrano, J. A., G. R. Nematipour, and D. M. Gatlin III. 1992. Dietary protein Relative weight (Wr) as a field assessment tool: relationships with growth, requirement of the Red Drum (Sciaenops ocellatus) and relative use of dietary prey biomass, and environmental conditions. Transactions of the American carbohydrate and lipid. Aquaculture 101:283–291. Fisheries Society 124:387–400. Sink, T. D., R. Lochmann, and N. Kinsey. 2010. Growth and survival of Channel Lochmann, R., and H. Phillips. 1994. Dietary protein requirement of juvenile Catfish Ictalurus punctatus fry fed diets with 36 or 45% total protein and all- Golden Shiners (Notemigonus crysoleucas) and Goldfish (Carassius auratus) plant or animal-protein sources. Journal of the World Aquaculture Society in aquaria. Aquaculture 128:277–285. 40:124–129. Lochmann, R., and H. Phillips. 1996. Stable isotopic evaluation of the relative Stone, N., C. Engle, and E. Park. 2008. Production enterprise budget for Golden assimilation of natural and artificial foods by Golden Shiners Notemigonus Shiners. Southern Regional Aquaculture Center, Publication 122, Stoneville, crysoleucas in ponds. Journal of the World Aquaculture Society 27: Mississippi. 168–177. Stone, N., E. Park, L. Dorman, and H. Thomforde. 1997. Baitfish culture in Lochmann, R., and H. Phillips. 2009. Nutrition and feeding of baitfish. Univer- Arkansas: Golden Shiners, Goldfish, and Fathead Minnows. University of sity of Arkansas Cooperative Extension Program, University of Arkansas at Arkansas at Pine Bluff, Cooperative Extension Program, Publication MP Pine Bluff, ETB256-PD-2-09RV, Pine Bluff. 386, Pine Bluff. Lochmann,R.,C.Engle,G.Kumar,M.H.Li,J.L.Avery,B.G.Bosworth,and Sutton, S. G., T. P. Bult, and R. L. Haedrich. 2000. Relationships among fat C. S. Tucker. 2012. Multi-batch catfish production and economic analysis weight, body weight, water weight, and condition factors in wild Atlantic Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 using alternative (low-cost) diets with corn gluten feed and traditional diets Salmon parr. Transactions of the American Fisheries Society 129:527–538. with meat and bone meal. Aquaculture 366–367:34–39. Wedemeyer, G. A. 1997. Effects of rearing conditions on the health and physio- Lochmann, R., H. Phillips, S. Dasgupta, D. Gatlin, and S. Rawles. 2001. Stable logical quality of fish in intensive culture. Society for Experimental Biology carbon isotope ratios and standard production data as indices of Golden Seminar Series 62:35–71. Shiner, Notemigonus crysoleucas, performance in pond feeding trials. Journal Wilson, R. P. 2002. Amino acids and proteins. Pages 143–179 inJ. E. Halver of Applied Aquaculture 11:21–34. and R. W. Hardy, editors. Fish nutrition, 3rd edition. Academic Press, New Lochmann, R., N. Stone, and H. Phillips. 2004. Evaluation of 36%-protein di- York. ets with or without animal protein for rearing tank-hatched Golden Shiner Wu, Y. V., K. W. Tudor, P. B. Brown, and R. R. Rosati. 1999. Substitution of Notemigonus crysoleucas fry in ponds. North American Journal of Aquacul- plant proteins or meat and bone meal for fish meal in diets of Nile Tilapia. ture 66:271–277. North American Journal of Aquaculture 61:58–63. This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:16 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Size and ATP Content of Unfertilized Eggs from Farmed and Wild Atlantic Salmon in Newfoundland Lynn Lusha, Kimberley Burta, Dounia Hamoutenea, Nancy Camarillo-Sepulvedaa, Juan Carlos Perez-Casanovaa, Sharon Kennya, Pierre Gouleta, Ross Hinksb & Clyde Collierc a Fisheries and Oceans Canada, North Atlantic Fisheries Center, Post Office Box 5667, St. John's, Newfoundland and Labrador A1C 5X1, Canada b Natural Resources Miawpukek First Nations, Miawpukek Mi’kamawey Mawi’omi, Post Office Box 10, Conne River, Newfoundland and Labrador A0H 1J0, Canada c Gray Aqua Group Limited, Post Office Box 275, Conne River, Newfoundland and Labrador A0H 1J0, Canada Published online: 15 Apr 2014.

To cite this article: Lynn Lush, Kimberley Burt, Dounia Hamoutene, Nancy Camarillo-Sepulveda, Juan Carlos Perez- Casanova, Sharon Kenny, Pierre Goulet, Ross Hinks & Clyde Collier (2014) Size and ATP Content of Unfertilized Eggs from Farmed and Wild Atlantic Salmon in Newfoundland, North American Journal of Aquaculture, 76:2, 138-142, DOI: 10.1080/15222055.2014.886648 To link to this article: http://dx.doi.org/10.1080/15222055.2014.886648

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 North American Journal of Aquaculture 76:138–142, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886648

COMMUNICATION

Size and ATP Content of Unfertilized Eggs from Farmed and Wild Atlantic Salmon in Newfoundland

Lynn Lush,* Kimberley Burt, Dounia Hamoutene, Nancy Camarillo-Sepulveda, Juan Carlos Perez-Casanova, Sharon Kenny, and Pierre Goulet Fisheries and Oceans Canada, North Atlantic Fisheries Center, Post Office Box 5667, St. John’s, Newfoundland and Labrador A1C 5X1, Canada Ross Hinks Natural Resources Miawpukek First Nations, Miawpukek Mi’kamawey Mawi’omi, Post Office Box 10, Conne River, Newfoundland and Labrador A0H 1J0, Canada Clyde Collier Gray Aqua Group Limited, Post Office Box 275, Conne River, Newfoundland and Labrador A0H 1J0, Canada

the question of impacts of escapees on wild, genetically dis- Abstract tinct stocks. Throughout much of the literature, studies compar- In Newfoundland and Labrador (hereafter, Newfoundland), all ing wild and cultured salmonids have focused on populations farmed Atlantic Salmon Salmo salar originate from the Saint John generated from the same genetic source (e.g., McCormick and River, New Brunswick, strain, raising the question of potential impacts of escapees on wild, genetically distinct stocks. Unfertilized Bjornsson¨ 1994; Chittenden et al. 2010). The effect of rear- eggs of farmed and wild salmon were assessed for diameter, mass, ing environment on phenotypes and behavior outweigh genetic adenosine triphosphate (ATP) content, and fertilization success, differences between wild and hatchery reared Coho Salmon On- this is the first study to assess ATP content in Atlantic Salmon eggs. corhynchus kisutch (Chittenden et al. 2010); however, genetics Results demonstrated that farmed salmon egg mass (0.1046 g), di- still have influence on differences between wild and hatchery ameter (6.2 mm), and ATP content (0.0281 nmol/egg) were significantly lower than the same characteristics in wild salmon. Among populations (Riddell and Swain 1991). With rivers, such as Little and within females, variability in egg size was similar in both wild River on the south coast of Newfoundland, that experience low and farmed groups. This study lends some evidence to the fact that annual returns of spawners (DFO 2012), a better understanding in the Newfoundland context, eggs of escaped farmed salmon may of potential reproductive differences between Saint John River Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 be less likely to produce viable, large larvae compared with their farmed salmon and local stock is key. wild counterparts. Egg quality in salmonids has been assessed using characteristics such as mass, diameter, proximate analysis, and energy Escaped farmed salmon may share breeding grounds with content (e.g., Srivastava and Brown 1991). Svardson (1949) wild counterparts and potentially interbreed and produce hy- noted that the largest larvae hatch from the largest eggs and brids with reduced fitness (Fiske et al. 2006), which may have have higher survival and growth (Brown 1946). Research has negative impacts on wild populations (McGinnity et al. 2003; demonstrated that egg quality and embryo viability in Chinook Fraser et al. 2008). Wild stocks are believed to have devel- Salmon O. tshawytscha and steelhead O. mykiss (anadromous oped genotypic local adaptations and life history imperatives Rainbow Trout) may also depend on the adenosine triphos- (Saunders 1991). In Newfoundland and Labrador (hereafter, phate (ATP) content of the unfertilized egg (Wendling et al. Newfoundland), all farmed Atlantic Salmon Salmo salar origi- 2000, 2004); however, an assessment of ATP content of eggs nate from the Saint John River, New Brunswick, strain, raising in Atlantic Salmon has yet to be performed. This study is an

*Corresponding author: [email protected] Received September 17, 2013; accepted December 6, 2013

138 COMMUNICATION 139

assessment of differences between unfertilized eggs of farmed a farmed female. This procedure was repeated with different (Saint John River strain) and wild (Little River, Newfoundland) males resulting in 14 unique crosses for every female origin. female Atlantic Salmon. The documentation of size, energy After gentle mixing of oocytes with sperm, each cross was ex- content, and fertilization success of wild and farmed Atlantic posed to a 60-min freshwater flush to allow water hardening. Salmon eggs will add information to that already known about After this, each cross was placed in a monolayer on a screen the potential reproductive success of escapees in this region. in an incubation trough supplied with ambient river water from the same source as the wild broodstock collection. Fertiliza- tion success was measured 36 h later by collecting 50 eggs and METHODS clearing them in Stockard’s solution (5% formaldehyde, 4% Twenty-two 4- and 5-year-old wild Atlantic Salmon fe- glacial acetic acid, 6% glycerin, 85% distilled water). Fertiliza- males were caught in Little River, Newfoundland, and manually tion success was expressed as a percentage of fertilized eggs stripped of eggs by applying gentle pressure to the fish’s ventral of the total count: [(number of eggs showing cellular division / surface. Similarly, oocytes from twelve 5-year-old Saint John 50) × 100]. River strain farmed females held in sea cages were sampled. All statistical analyses were performed using SIGMA STAT Ten unfertilized eggs from each of the wild and farmed females software (Systat Software, Chicago, Illinois). One-way ANOVA were stored in 5% buffered formalin for size analysis (as per was employed to determine differences in unfertilized egg di- Fleming and Ng 1987). Egg diameters were measured (to the ameter, mass, and ATP content data with multiple comparisons nearest 0.1 mm) using a stereomicroscope fitted with a microm- tested using the Holm–Sidak method. Paired t-tests (same male) eter, and mass was determined (to the nearest 0.0001 g) using were used to compare fertilization success between farmed and an analytical balance. wild salmon (n = 4 for wild 5-year-old salmon; n = 10 for wild To determine ATP content, 4 mL of unfertilized eggs were 4-year-old salmon). Coefficients of variation were calculated as ◦ collected and flash frozen in dry ice and stored at −70 C until SD/mean × 100, and P-values were significant below α = 0.05 analysis. Extraction of ATP was performed as described by for all analyses. Boulekbache et al. (1989) by using five frozen eggs (∼0.5 g) homogenized (Polyton) in a 10-mL ice-cold extraction medium (0.5 M perchloric acid, 4 mM EDTA). A luciferine–luciferase RESULTS AND DISCUSSION assay of ATP was performed using a bioluminescence assay Results showed significant differences in unfertilized egg kit (Sigma-Aldrich, product number FL-AA). Bioluminescence mass and diameter between farmed and wild salmon (Table 1). was read with a microplate reader, and the ATP content for each Values for egg sizes were comparable to those documented by female was expressed as an average value determined from eggs other authors. Thorpe et al. (1984) described egg diameters for analyzed in triplicate. Concentrations of ATP were expressed as Atlantic Salmon ranging from 4.56 to 6.60 mm, and Srivastava pmoles/egg. In addition, the average egg mass for each female and Brown (1991) reported an egg mass range of 95–112 mg. was used to determine the ATP content per milligram of egg No differences in egg mass were detected among the two age- tissue and expressed as pmoles of ATP/mg egg. classes of wild females; however, 5-year-old females had eggs Fertilization success was assessed by fertilizing 750 eggs per with diameters greater than those for the 4-year-olds. Egg size female using sperm volumes equating to 107 sperm cells per varies with female age, although much of this appears to be egg (Scott and Baynes 1980; calculated following spermatocrit due to correlated effects of female size (Thorpe et al. 1984) determination as in Aas et al. 1991). Sperm from an individ- and juvenile growth rates in freshwater (Jonsson et al. 1996). In

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 ual male was used to fertilize eggs from a wild female and our study, farmed females were signiﬁcantly larger (P < 0.001)

TABLE 1. Mean ( ± SD) values for unfertilized egg mass, diameter, ATP content, and fertilization success of wild and farmed female Atlantic Salmon. Different lowercase letters accompanying values denote signiﬁcant differences (P < 0.05); NA = not available.

Oocyte ATP Age diameter ATP (pmol/mg Fertilization Origin (years) Length (cm) Oocyte mass (g) (mm) (pmol/egg) egg) success (%) 96.0 ± 3.9 (n = 4)a Farmed 575.8± 3.0 z 0.1046 ± 0.0088 z 6.2 ± 0.3 z 28.1 ± 2.4 z 0.386 ± 0.203 97.5 ± 2.6 z (n = 10) (n = 12) Wild 5 NA 0.1280 ± 0.0133 y 6.7 ± 0.3 y 49.6 ± 25.7 y 0.269 ± 0.0247 98.5 ± 1.0 (n = 4) (n = 5) Wild 454.1± 3.8 y 0.1170 ± 0.0097 y 6.5 ± 0.2 x 32.1 ± 9.2 zy 0.277 ± 0.08 99.2 ± 1.0 y (n = 10) (n = 17) aData separated for paired t-test (same males) to allow for proper comparisons of groups. 140 LUSH ET AL.

than female wild salmon, contradicting the common understand- using paired t-tests to compare wild and farmed crosses (n = ing that larger females produce larger eggs (e.g., Thorpe et al. 14) by grouping females of different ages, differences in fertil- 1984). However, hatchery rearing relaxes the natural selection ization success were significant (P = 0.020), and lower values that would favor large eggs and thus drives a rapid evolution of were found in farmed females (average fertilization success: small eggs (Heath et al. 2003). Therefore, despite their larger farmed = 96%; wild = 99%). While it is well established that size, farmed females may still produce smaller eggs. Our results variation in egg size within salmon can be primarily explained concur with Srivastava and Brown (1991), who also observed through offspring and maternal fitness optima (e.g., Fleming that wild Atlantic Salmon eggs have a greater mass and di- and Gross 1990; Einum et al. 2004), no studies have explored ameter than do their farmed counterparts. Differences in both variation in fish egg size with regard to fertilization ability (Mac- studies, while significant, are numerically small. For example, farlane et al. 2009). In a study on Sockeye Salmon O. nerka, Srivastava and Brown (1991) noted only a 0.17-mm difference Macfarlane et al. (2009) found that under conditions of sperm between wild and farmed Atlantic Salmon egg diameters, while limitation, fertilized eggs were significantly larger than eggs that our results showed a difference that ranged from 0.3 to 0.5 mm failed to be fertilized. Under conditions without sperm limita- between these groups. Also, Srivastava and Brown (1991) ob- tion (such as in our study), fertilization success was unaffected served a 0.017-g difference in egg mass between cultured and by egg size (Macfarlane et al. 2009), suggesting that size dif- wild salmon, while our study demonstrated a difference in egg ference may not always translate into different fertilization suc- mass that ranged from 0.012 to 0.023 mm. To determine the bi- cess in artificial conditions where sperm is not limiting. There ological significance of these differences in the Newfoundland is good evidence to show that sperm limitation occurs naturally context, more information on early life traits and survival is re- in externally fertilizing organisms (e.g., Levitan and Petersen quired to gain a better picture of the impacts of small differences 1995), including in fishes that release gametes in close prox- in egg size on offspring performance. imity (Petersen et al. 1992), such as salmonids. This overall Values for ATP measured in this study are lower than those trend in fertilization success and egg-size differences suggests measured in eggs from Chinook Salmon (2.61 nmol/egg) or that reproduction may not be as successful in farmed as in wild steelhead (1.92 nmol/egg) (Wendling et al. 2000, 2004). This salmon. may be because eggs were kept for up to 2 h at 1.4–2.3◦Cprior Coefficients of variation were calculated for egg size mea- to being frozen, and samples were analyzed after 9 months of surements within females (Table 2). Mean CVs were close to storage at −70◦C. This might have contributed to ATP degra- what is described in wild salmon by other authors, ∼4.0% (e.g., dation, affecting all samples in a similar fashion. Nonetheless, Einum and Fleming 2000). Interestingly, the size variability Aegerter and Jalabert (2004) found that ATP levels in Rain- of unfertilized eggs within a clutch is similar between a wild bow Trout remained constant 14 d after ovulation at 17◦C, and female salmon and a farmed female. This contradicts other lit- Wendling et al. (2004) found no changes in steelhead egg ATP erature showing that females in captivity produced more highly levels after 24 h of storage at 10◦C. To our knowledge, this is variable egg size per clutch (Einum and Fleming 2000; Leblanc the first investigation on ATP levels in Atlantic Salmon eggs; 2011). Einum and Fleming (2000) hypothesized that the lower therefore, more ATP measurements are necessary to make a variation in wild female egg size was related to evolutionary conclusion. Our results show that ATP levels per egg are lower adaptations resulting in the production of an optimum size to in farmed salmon than in wild salmon (Table 1), suggesting increase survival, size at hatch, and growth during early juvenile a potential reduction in viability and early development. The stages. However, there may be an advantage in maintaining egg process of egg activation (regardless of fertilization) is energeti- size variability within streams to face constraints such as low

Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 cally costly as it incorporates many intracellular and cytoskeletal oxygen (Rombough 2007) or the need for different emergence changes. Therefore, sufficient stores of ATP are required as ATP times to coincide with food availability (Einum and Fleming is depleted significantly within 2 h following activation and fer- 2002). Variability of egg mass among females was equivalent in tilization (Wendling et al. 2000). Nevertheless, when ATP was both farmed and wild groups (8.41, 10.39, and 8.29% for farmed, evaluated based on egg mass, there was no significant difference 5-year-old wild, and 4-year-old wild females, respectively). On between farmed and wild salmon, suggesting that the difference the contrary, variation in ATP content per egg among females in energy levels is likely related to the difference in mass. How- showed differences between wild and farmed salmon, with a CV ever, when we explored correlations between egg size (mass and diameter) and ATP content, only wild females showed a signif- TABLE 2. Percent CVs (mean ± SD) of unfertilized egg mass and diameter icant positive correlation between egg mass and ATP (data not within an egg clutch for wild and farmed Atlantic Salmon. presented). Despite differences in size and ATP content of eggs, paired Egg Wild females Farmed females t-tests showed that no significant difference in fertilization suc- characteristic (n = 22) (n = 12) cess was detected between farmed and wild Atlantic Salmon of Oocyte mass 6.41 ± 5.63 6.09 ± 1.89 the same age (5 years old); however, only four crosses for each Oocyte diameter 4.25 ± 2.02 3.75 ± 0.91 group were considered for this comparison (Table 1). When COMMUNICATION 141

of 8.54% for farmed fish and 51.81% and 28.67% for 5-year-old effects on hatchery-and wild-born Coho Salmon. PLoS ONE [online serial] and 4-year-old wild females, respectively. Wendling et al. (2000, 5(8):e12261. 2004) observed a ∼20% variability among females after mea- DFO (Department of Fisheries and Oceans). 2012. Stock assessment of New- foundland and Labrador Atlantic Salmon–2011. DFO, Canadian Science Ad- suring egg ATP content from hatchery-reared Chinook Salmon visory Secretariat Science Advisory Report 2011/077, Ottawa. and steelhead. Mean levels were higher in steelhead eggs ob- Einum, S., and I. A. Fleming. 2000. Highly fecund mothers sacrifice offspring tained late versus early in the collecting season (Wendling et al. survival to maximize fitness. Nature 405:565–567. 2004). The greater ATP variability among wild females could Einum, S., and I. A. Fleming. 2002. Does within-population variation in fish be related to a higher variability in spawning readiness and/or egg size reflect maternal influences on optimal values. American Naturalist 160:756–765. timing in wild populations. Culture conditions and breeding Einum, S., M. T. Kinnison, and A. P. Hendry. 2004. Evolution of egg size programs select for similar reproduction timing and length. On and number. Pages 126–153 in A. P. Hendry and S. C. Stearns, editors. the other hand, the onset of breeding in wild populations varies Evolution illuminated: salmon and their relatives. Oxford University Press, as an adaptation to favorable conditions for spawning, embryo New York. incubation, and juvenile feeding. (e.g., Fleming et al. 1996). Fiske, P., R. A. Lund, and L. P. Hansen. 2006. Relationships between the frequency of farmed Atlantic Salmon, Salmo salar L., in wild salmon population In conclusion, unfertilized eggs of farmed Atlantic Salmon and fish farming activity in Norway, 1989–2004. ICES Journal of Marine Sci- are smaller, weigh less, and contain less energy in the form of ence 63:1182–1189. ATP, and may have lower fertilization success than those of their Fleming, I. A., and M. R. Gross. 1990. Latitudinal clines–a trade-off between wild counterparts. However, while size values are indeed statis- egg number and size in Pacific salmon. Ecology 71:1–11. tically significantly different, numerical differences are not large Fleming, I. A., B. Jonsson, M. R. Gross, and A. Lamber. 1996. An experimental study of the reproductive behaviour and success of farmed and and therefore biological importance needs to be determined cau- wild Atlantic Salmon (Salmo salar). Journal of Applied Ecology 33:893– tiously, with the suggestion that further population studies be 905. undertaken. In addition to the findings of other authors who Fleming, I. A., and S. Ng. 1987. Evaluation of techniques for fixing, preserv- suggest that farmed salmon are less reproductively successful ing, and measuring salmon eggs. Canadian Journal of Fisheries and Aquatic due to inferior breeding behavior characteristics (Fleming et al. Sciences 44:1957–1962. Fraser, D. J., A. M. Cook, J. D. Eddington, P.Bentzen, and J. A. Hutchings. 2008. 1996; Jonsson 1997), our results lend some evidence to the fact Mixed evidence for reduced local adaptation in wild salmon resulting from that, in the Newfoundland context, eggs of escaped farmed At- interbreeding with escaped farmed salmon: complexities in hybrid fitness. lantic Salmon are less likely to produce large, viable larvae than Evolutionary Applications 1:501–512. are eggs of wild salmon. Heath, D. D., J. W. Heath, C. A. Bryden, R. M. Johnson, and C. W. Fox. 2003. Rapid evolution of egg size in captive salmon. Science 299:1738– 1740. Jonsson, B. 1997. A review of ecological and behavioural interactions between ACKNOWLEDGMENTS cultured and wild Atlantic Salmon. ICES Journal of Marine Science 54:1021– Funding for this project was provided by the Aquaculture 1039. Collaborative Research and Development Program of Fisheries Jonsson, N., B. Jonsson, and I. A. Fleming. 1996. Phenotypically plastic re- and Oceans Canada. Technical support was provided by John sponse of egg production to early growth in Atlantic Salmon. Functional Ecology 10:89–96. Murray of Fisheries and Oceans Canada, Jennifer Caines and Leblanc, C. A. 2011. The importance of egg size for the diversity of salmonids. Doug Caines of Northern Harvest Sea Farms, and Stephanie Doctoral dissertation. University of Iceland, Reykjavik. Gambin and Sherilynn Upshall of Memorial University’s Fish- Levitan, D. R., and C. Petersen. 1995. Sperm limitation in the sea. Trends in eries and Marine Institute. Ecology and Evolution 10:228–231. Macfarlane, C. P., D. J. Hoysak, N. R. Liley, and M. J. G. Gage. 2009. In vitro fertilization experiments using Sockeye Salmon reveal that bigger eggs are Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 more fertilizable under sperm limitation. Proceedings of the Royal Society B REFERENCES 276:2503–2507. Aas, G. H., T. Refstie, and B. Gjerde. 1991. Evaluation of milt quality of Atlantic McCormick,S.D.,andB.T.Bjornsson.¨ 1994. Physiological and hormonal Salmon. Aquaculture 95:125–132. differences among Atlantic Salmon parr and smolts reared in the wild, and Aegerter, S., and B. Jalabert. 2004. Effects of post-ovulatory oocyte ageing hatchery smolts. Aquaculture 121:235–244. and temperature on egg quality and on occurrence of triploid fry in Rainbow McGinnity, P., P. Prodohl, A. Ferguson, R. Hynes, N. O. Maoileidigh, N. Baker, Trout, Oncorhynchus mykiss. Aquaculture 231:59–71. and D. Cotter. 2003. Fitness reduction and potential extinction of wild pop- Boulekbache, H., J. Bastin, M. Andriamihaja, B. Lefebvre, and C. Joly. 1989. ulations of Atlantic Salmon, Salmo salar, as a result of interactions with Ageing of fish oocytes: effects on adenylic nucleotides content, energy charge escaped farm salmon. Proceedings of the Royal Society of London B Biolog- and viability of carp embryo. Comparative Biochemistry and Physiology B ical Sciences 270:2442–2450. 93:471–476. Petersen, C. W., R. R. Warner, S. Cohen, H. C. Hess, and A. T. Sewell. 1992. Brown, M. E. 1946. The growth of Brown Trout (Salmo trutta Linn.). I. Factors Variable pelagic fertilization success: implications for mate choice and spatial influencing the growth of trout fry. Journal of Experimental Biology 22:118– patterns of mating. Ecology 73:391–401. 129. Riddell, B. E., and D. P. Swain. 1991. Competition between hatchery and Chittenden, C. M., C. A. Biagi, J. Grimsrud Davidsen, A. Grimsrud Davidsen, wild Coho Salmon (Onchorhynchus kisutch): genetic variation for agonistic H. Kondo, A. McKnight, O. Pedersen, P. A. Raven, A. H. Rikardsen, J. behaviour in newly-emerged wild fry. Aquaculture 98:161–172. M. Shrimpton, B. Zuehlke, R. S. McKinley, and R. H. Devlin. 2010. Genetic Rombough, P. J. 2007. Oxygen as a constraining factor in egg size evolution in versus rearing-environment effects on phenotype: hatchery and natural rearing salmonids. Canadian Journal of Fisheries and Aquatic Sciences 64:692–699. 142 LUSH ET AL.

Saunders, R. L. 1991. Potential interaction between cultured and wild Atlantic Thorpe, J. E., M. S. Miles, and D. S. Keay. 1984. Developmental rate, fecundity, Salmon. Aquaculture 98:51–60. and egg size in Atlantic Salmon, Salmo salar L. Aquaculture 43:289–305. Scott, A. P., and S. M. Baynes. 1980. A review of the biology, handling Wendling, N. C., D. C. Bencic, J. J. Nagler, J. G. Cloud, and R. L. Inger- and storage of salmonids spermatozoa. Journal of Fish Biology 17:707– man. 2000. Adenosine triphosphate levels in Chinook Salmon (Oncorhynchus 739. tshawytscha) eggs following in vitro maintenance and activation/fertilization. Srivastava, R. K., and J. A. Brown. 1991. The biochemical characteristics and Fish Physiology and Biochemistry 22:217–223. hatching performance of cultured and wild Atlantic Salmon (Salmo salar) Wendling, N. C., D. C. Bencic, J. J. Nagler, J. G. Cloud, and R. L. Ingermann. eggs. Canadian Journal of Zoology 69:2436–2441. 2004. Adeosine triphosphate levels in steelhead (Onchorhynchus mykiss) Svardson, G. 1949. Natural selection and egg number in ﬁsh. Reproductive eggs: an examination of turnover, localization and role. Comparative Bio- Institute Freshwater Research Drottningholm 29:115–122. chemistry and Physiology Part A 137:739–748. Downloaded by [Department Of Fisheries] at 20:16 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:17 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Plasma Components and Hepatic Insulin-like Growth Factor Expression Indicate Nutritional Status in Yellowtail Seriola quinqueradiata M. Kawanagoa, S. Takemurab, R. Ishizukaa & I. Shioyaa a Aquaculture Fundamental Research GroupCentral Research Laboratory, Nippon Suisan Kaisha, Ltd. 1-32-3, Nanakuni, Hachioji, Tokyo 192-0991, Japan b Research Support and Administration SectionCentral Research Laboratory, Nippon Suisan Kaisha, Ltd. 1-32-3, Nanakuni, Hachioji, Tokyo 192-0991, Japan Published online: 15 Apr 2014.

To cite this article: M. Kawanago, S. Takemura, R. Ishizuka & I. Shioya (2014) Plasma Components and Hepatic Insulin- like Growth Factor Expression Indicate Nutritional Status in Yellowtail Seriola quinqueradiata, North American Journal of Aquaculture, 76:2, 143-152, DOI: 10.1080/15222055.2014.886649 To link to this article: http://dx.doi.org/10.1080/15222055.2014.886649

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 North American Journal of Aquaculture 76:143–152, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886649

ARTICLE

Plasma Components and Hepatic Insulin-like Growth Factor Expression Indicate Nutritional Status in Yellowtail Seriola quinqueradiata

M. Kawanago* Aquaculture Fundamental Research Group, Central Research Laboratory, Nippon Suisan Kaisha, Ltd. 1-32-3, Nanakuni, Hachioji, Tokyo 192-0991, Japan S. Takemura Research Support and Administration Section, Central Research Laboratory, Nippon Suisan Kaisha, Ltd. 1-32-3, Nanakuni, Hachioji, Tokyo 192-0991, Japan R. Ishizuka and I. Shioya Aquaculture Fundamental Research Group, Central Research Laboratory, Nippon Suisan Kaisha, Ltd. 1-32-3, Nanakuni, Hachioji, Tokyo 192-0991, Japan

Abstract We examined the effect of feeding ration and refeeding time on plasma biochemical components and hepatic insulin-like growth factor gene (IGF) mRNA expression to evaluate nutritional indicators in juvenile Yellowtail Seriola quinqueradiata (also known as Buri). Fish were fed a diet for 10 d with the following feeding rations: 80, 40, 20, and 10% satiation. Satiated feeding was 4.65% of body weight. At day 10, the plasma concentrations of albumin, calcium, inorganic phosphate, and total protein, activities of alkaline phosphatase and amylase, and hepatic IGF1 mRNA and IGF2 mRNA expression were significantly correlated with feeding ration. In the analysis of refeeding time, fish were refed after a 24-h fasting period and plasma biochemical components and hepatic IGFs mRNA were analyzed at 0, 3, 6, and 9 h after refeeding. Plasma concentrations of total bilirubin, inorganic phosphate, creatinine, glucose, and potassium, and activities of amylase were significantly correlated with refeeding time. The correlation between refeeding time and gene expression was significant for IGF2 mRNA (r = 0.70, P < 0.05) but not for IGF1 mRNA (r = 0.67, P = 0.07). Comparing the results of the two experiments, the plasma concentrations of total protein, albumin, and calcium reflected only the feeding ration. The plasma concentration of inorganic phosphate, activities of alkaline phosphatase and amylase, and hepatic IGF1 mRNA and IGF2 mRNA reflected both the feeding ration Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 and refeeding time. These observations suggest that these components reflected the nutritional status of Yellowtail, but sampling time should be carefully examined because of the fluctuating feature of these components.

Plasma biochemical components are applicable as a health condition and some metabolites or biochemical components status index. Because feeding methods and the nutritional status has been reported in some fish species. Fasting significantly affect these values, measuring plasma biochemical components decreases plasma concentrations of biochemical components may be useful for assessing the nutritional condition of fish. The such as total protein and glucose in Common Carp Cyprinus nutritional condition of an organism can be measured directly or carpio (Shimeno et al. 1997). indirectly by examining energy reserves such as metabolites and The plasma concentrations of some metabolites and activi- metabolizing enzymes. A relationship between the nutritional ties of some enzymes are also affected by the feeding rhythm,

*Corresponding author: [email protected] Received September 29, 2013; accepted December 14, 2013 143 144 KAWANAGO ET AL.

and this has been closely examined in mammals (Escobar et al. In this study, we evaluated (1) nutritional indicators by an- 1998). In fish, daily rhythms of biochemical components, such alyzing plasma biochemical components that reflect nutritional as glucose, amino acids, protein, and aspartate–alanine transam- or metabolic status and (2) whether hepatic IGF1 mRNA and inase have been examined in Rainbow Trout Oncorhynchus IGF2 mRNA expression reflect the nutritional status in Yellow- mykiss (Polakof et al. 2007a, 2007b). A few reports have tail. Experimental nutritional conditions were designed using been published on fasting or restricted feeding and postpran- the feeding ration and time course model studied in Gilthead dial changes in plasma biochemical components; e.g., orally Seabream (Meton´ et al. 2000). Plasma biochemical components administered glucose affects the plasma glucose concentration and hepatic IGF mRNA expression were measured and com- in Yellowtail Seriola quinqueradiata (also known as Buri [Fu- pared with feeding ration and the refeeding time course. ruichi and Yone 1981]). However, plasma biochemical components should be investigated further to assess the relationship between nutritional status and blood biochemical components in Yellowtail. Because aquaculture production of this species METHODS is higher than that of other cultured finfish species in Japan, Experiment 1: effect of feeding ration on plasma biochemical evaluation of the nutritional or health status of Yellowtail is components and hepatic IGF1 mRNA and IGF2 mRNA expres- needed. sion.—Juvenile Yellowtail (average body weight, 45 g) were Insulin-like growth factors (IGFs), which are mitogenic hor- distributed evenly among eight, circular, 100-L polycarbonate mones expressed in multiple organs, regulate somatic growth tanks (10 fish/tank) and acclimated to the experimental condi- in vertebrates (Jones and Clemmons 1995). The fish IGF sys- tions in a flow-through seawater system for a week prior to the tem is also regulated by nutritional conditions. The feeding experiments. The water temperature was maintained at 24.5◦C ration affects hepatic IGF1 mRNA and blood IGF1 levels in throughout the acclimation and experimental periods. The ex- Coho Salmon O. kisutch (Beckman et al. 2004) and Gilthead perimental diet, which consisted of 52.2% crude protein, 14.9% Seabream Sparus aurata (Meton´ et al. 2000). The relationship crude fat, and 2.4% moisture, was purchased from Nippon Su- between fasting and the hepatic IGF1 mRNA level has been isan Kaisha (Tokyo, Japan). Before starting the experiments, demonstrated in many fish species, such as Gilthead Seabream all fish were fed the experimental diet until satiated as deter- (Meton´ et al. 2000), Grouper Epinephelus coioides (Pedroso mined visually by the personnel distributing the feed for four et al. 2006), and Mozambique Tilapia Oreochromis mossambi- consecutive days, and the amount of feed consumed per tank cus (Uchida et al. 2003). In Yellowtail, hepatic IGF1 mRNA was recorded daily. These data were used to calculate the feed- correlates with specific growth ratios using a feed restriction ing ration at satiation, which was determined to be an average model (Fukada et al. 2012). However, a fasting treatment either of 4.65 ± 0.31% (mean ± SD) of body weight. Before and increases (Pedroso et al. 2009) or does not affect (Fukada et al. after satiated-feeding periods, all fish were anesthetized with 2012) hepatic IGF1 mRNA in Yellowtail and is interpreted as a 0.015% 2-phenoxyethanol in seawater and their body weight potential species-specific effect. Therefore, IGF mRNA expres- was determined. A total of four experimental groups were then sion should be investigated further to elucidate the nutritional randomly assigned to eight tanks (two replicate 100-L circular regulation not only of IGF1 mRNA but also of IGF2 mRNA in tanks used per experimental group) with the following feeding order to interpret the role of these two anabolic molecules. rations: 80% of the satiation feeding ration (80% sat), 40% of the Diurnal or feeding rhythm regulation of hepatic IGF1 mRNA satiation feeding ration (40% sat), 20% of the satiation feeding and IGF2 mRNA expression has been examined in Rabbitfish ration (20% sat), and 0% of the satiation feeding ration (fasting,

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 Siganus guttatus (Ayson et al. 2007) and in Gilthead Seabream 0% sat). (Meton´ et al. 2000). These reports suggest that feed intake regu- Twenty-four hours after the last feeding on day 10, all fish lates short-term hepatic IGF1 mRNA expression and then subse- were anesthetized and their body weights determined. A total of quently play an important role in anabolic function. No data have four fish per experimental group were randomly selected from been published on postprandial hepatic IGF1 mRNA or IGF2 each group for further analysis. The fish were killed, and the mRNA in Yellowtail. Investigation of the changes in these hep- blood was collected with a heparinized syringe. Plasma was atic postprandial IGF mRNAs is likely necessary to understand separated from whole blood by centrifugation, and then stored anabolism after feeding in Yellowtail. at −80◦C in deep freezer for later analysis of the biochemical The carangids, Yellowtail and Amberjack Seriola dumerili, components. The liver was also dissected, weighed, and im- are the major commercially cultured fishes in Japan. Annual mediately frozen and stored at −80◦C in deep freezer for later production of these Seriola species reached approximately 159, gene expression analysis. Growth performance was evaluated 300 metric tons in 2011 in Japan (MAFF 2011). Artificial feeds by calculating (1) the total feed intake (grams per fish) for feed have been developed to improve feeding and economic effi- acceptance, (2) the feed conversion efficiency: weight gain : ciency. Evaluation of nutritional components is an important feed ratio for feed utilization, (3) hepatosomatic index (HSI) tool to monitor nutritional condition of fish provided with dif- calculated as 100 × liver (g)/body weight (g) (expressed as %) ferent test diets or fed with artificial diet. and (4) growth ratio calculated as {[mean final body weight (g) NUTRITIONAL STATUS AND PLASMA BIOCHEMISTRY IN YELLOWTAIL 145

− mean initial body weight (g)] / mean initial body weight (g) ward: GGAGAGAGAGGCTTTTATTTCAGTAAAC, reverse: / experimental days (d)} × 100 (expressed as %). CGTGACCGCCGTGCAT. The GenBank accession numbers Experiment 2: time course change in plasma biochemical were AB179839.1 for beta-actin, AB439208 for IGF1, and components and hepatic IGF1 mRNA and IGF2 mRNA levels AB179839 for IGF2. All PCRs were run in duplicate with a after refeeding.—Because feeding status affects biochemical nontemplate control in order to comfirm that only specific prod- components and hepatic IGF mRNAs as described in the result uct was amplified with specific primer pair. All the PCR data of experiment 1, experimental fish of the same group that was were normalized to beta-actin using the 2(− CT) method used in experiment 1 were reared with continuous feeding in (Livak and Schmittgen 2001). In our previous reports, we eval- order to control nutritional status. Juvenile Yellowtail (average uated the nutritional effect on hepatic IGF1 mRNA and IGF2 body weight, 204 g) were distributed between two, circular, mRNA using the primer pairs with beta-actin as a normal- 1,000-L polycarbonate tanks (20 fish/tank) and acclimated to ization gene (Kawanago et al., in press). Therefore, we con- the experimental conditions in a flow-through seawater system sider that our originally designed primer pairs are recognized for 1 month prior to the experiments. The body weights (mean generally. ± SD) in the two replicated tanks were 203 ± 38 g and 207 ± Statistical analysis.—Data were analyzed by means of one- 43 g, respectively. Twenty-four hours after the last feeding, all way ANOVA and Tukey’s multiple variance test using SPSS fish were fed at 80% of the satiation feeding ration at 0700 hours 15.0J for Windows software. Data are presented as the mean ± and sampled 0, 3, 6, and 9 h after refeeding. At each time, six SD. Linear regression analysis for the feeding-ration experiment fish were anesthetized, and plasma and livers were collected and and multiple regression analysis for the time-course experiment analyzed as described in experiment 1. We confirmed that all were performed to analyze the correlation between the feeding sampled fish had taken feed by visually checking their stomach regimen and various components. Data were considered signif- contents. Unfed fish were not sampled in this experiment. icant at P < 0.05 and P < 0.01. Analysis of plasma biochemical components.—The concentration of plasma components were determined using the au- tomated metabolites analyzer Vetscan (Abaxis) equipped with RESULTS a multirotor. The 14 components analyzed were the concentrations (g/dL) of serum albumin, total bilirubin, blood urea nitro- Growth Performance in Experiment 1 gen, calcium, inorganic phosphate, creatinine, glucose, sodium, Body weights significantly increased in the 40% and 80% sat potassium, total protein, and globulin, and the activities (IU/L) groups compared with the 0% and 20% sat groups on day 10 of alkaline phosphatase, alanine aminotransferase, and amylase. (Table 1). The body weight of Yellowtail on day 10 and feeding Quantitative real-time PCR for IGF1 mRNA and IGF2 ration were positively correlated (r = 0.62, P < 0.01). The mRNA.—Total RNA was isolated from the liver using an RNeasy growth ratios were −0.8, −0.2, 0.9, and 3.4 for the 0, 20, 40, mini kit (Qiagen). The RNA concentration was measured with and 80% sat groups, respectively, and were positively correlated a spectrophotometer (Hitachi-Hightechnologies). Synthesis of with the feeding ration (r = 0.98, P < 0.01). The increase in complementary DNA (cDNA) was carried out using a Prime- liver weight (as indicated by the HSI) was also correlated with script first-strand synthesis kit (Takarabio). Quantitative real- the feeding ration (r = 0.67, P < 0.01). time PCR (qPCR) was carried out using the ABI7500 Real-Time PCR system (Applied Biosystems) with SYBR premix Extaq Plasma Biochemical Components in Experiment 1 (Takarabio) detection reagents to validate the effect of feeding Of the analyzed biochemical components, plasma albumin,

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 ration and refeeding time on gene expression. Gene-speciﬁc total protein, calcium, and inorganic phosphate concentration primer sequences are as follows: beta-actin, forward: CCT- and activity of alkaline phosphatase and amylase were signif- GCGTCTGGACTTGGCCG, reverse: GGAGGAGGCAGCG- icantly affected by the feeding ration according to one-way GTACCCA; IGF1, forward: CGACACGCTGCAGTTTGTGT, ANOVA and post hoc tests. All six of these components were reverse: CCATAGCCTGTTGGTTTACTGAAA; IGF2,for- also correlated with the feeding ration according to regression

TABLE 1. Body weight (BW) and liver weight (as HSI) changes (mean ± SD) in Yellowtail fed to varying percentages of satiation (sat). Values accompanied by different lowercase letters are signiﬁcantly different among the experimental groups (P < 0.05); NS = not signiﬁcant.

Experimental group Body measurement 0% sat 20% sat 40% sat 80% sat Signiﬁcance by ANOVA BW on day 0 (g) 46.5 ± 12.7 46.1 ± 9.4 50.0 ± 8.0 47.5 ± 6.0 NS BW on day 10 (g) 42.7 ± 11.0 x 45.2 ± 9.3 x 54.6 ± 9.3 y 66.6 ± 8.3 z P < 0.01 HSI (%) 0.69 ± 0.1 y 0.73 ± 0.1 y 0.82 ± 0.1 yz 1.1 ± 0.1 z P < 0.01 146 KAWANAGO ET AL.

TABLE 2. Plasma biochemical components (mean ± SD, n = 4) in Yellowtail fed to varying percentages of satiation (sat) and linear regression analysis with ration size and plasma components. Values accompanied by different lowercase letters are signiﬁcantly different among the experimental groups (P < 0.05); ND = not detected; NS = not signiﬁcant. Concentrations of albumin, globulin, total protein, glucose, calcium, sodium, potassium, total bilirubin, blood urea nitrogen, and creatinine are expressed as mg/dL. Plasma activities of alkaline phosphatase, amylase, and alanine amino transferase are expressed as IU/L.

Correlation with Experimental group ration size Signiﬁcance Plasma component 0% sat 20% sat 40% sat 80% sat by ANOVA r-value P-value Albumin 0.7 ± 0.3 y 1.6 ± 0.5 z 1.9 ± 0.1 z 2.0 ± 0.2 z P < 0.001 0.74 P < 0.01 Globulin ND 1.5 ± 0.1 1.5 ± 0.1 1.4 ± 0.5 NS −0.20 NS Total protein 2.5 ± 0.2 y 2.9 ± 0.6 yz 3.3 ± 0.1 z 3.4 ± 0.4 z P < 0.05 0.67 P < 0.01 Glucose 137.3 ± 20.1 114.2 ± 13.3 135.0 ± 8.8 140.1 ± 20.7 NS 0.21 NS Calcium 12.8 ± 0.7 y 13.9 ± 1.7 z 14.5 ± 0.1 z 14.6 ± 1.1 z P < 0.05 0.53 P < 0.05 Inorganic phosphate 10.0 ± 0.7 10.5 ± 0.9 10.3 ± 0.6 12.0 ± 1.6 P < 0.05 0.65 P < 0.01 Sodium 172.0 ± 2.4 173.6 ± 4.2 175.0 ± 5.8 177.5 ± 5.0 NS 0.44 NS Potassium 3.5 ± 0.8 5.6 ± 1.1 4.8 ± 0.8 4.7 ± 0.9 NS 0.18 NS Total bilirubin 0.4 ± 0.0 0.4 ± 0.1 0.4 ± 0.1 0.4 ± 0.1 NS −0.39 NS Blood urea nitrogen 23.0 ± 2.2 22.4 ± 4.3 22.3 ± 9.1 29.3 ± 1.3 NS 0.45 NS Creatinine 0.6 ± 0.2 0.7 ± 0.4 0.8 ± 0.4 1.0 ± 0.8 NS 0.30 NS Alkaline phosphatase 21.8 ± 3.1 y 27.6 ± 6.0 y 39.3 ± 4.6 y 39.3 ± 4.8 z P < 0.001 0.76 P < 0.01 Amylase 53.3 ± 20.9 x 101.0 ± 31.0 yz 109.7 ± 16.0 y 151.0 ± 23.7 z P < 0.001 0.74 P < 0.01 Alanine amino 15.8 ± 6.3 19.6 ± 3.0 25.0 ± 12.7 19.5 ± 3.4 NS 0.18 NS transferase

analysis (Table 2). The plasma globulin concentration was lower was 19.8 ± 0.4 cycles (range, 19.1–20.6) and did not differ than the detection limit in the 0% sat fish group. statistically among the four feeding ration groups assessed by one-way ANOVA with a post hoc test. There were no interwell Hepatic IGF1 mRNA and IGF2 mRNA Expression in differences in mean Ct values for PCR reaction plates; Ct values Experiment 1 (mean ± SD) were 19.9 ± 0.5 and 19.8 ± 0.5. Therefore, we Hepatic IGF1 mRNA expression was significantly increased conclude that hepatic beta-actin was sufficient to be stable as a in the 20, 40, and 80% sat group compared with the 0% sat normalizing gene in experiment 1. group (P < 0.05; Table 3). Hepatic IGF2 mRNA expression was significantly increased in the 80% sat group compared with the other three groups (P < 0.05; Table 3). The correlation Plasma Biochemical Components in Experiment 2 between the feeding ration and gene expression was significant Of the analyzed biochemical components, plasma concen- for IGF1 mRNA (r = 0.77, P < 0.01) and IGF2 mRNA (r = trations of glucose, inorganic phosphate, potassium, total biliru- 0.62, P < 0.05). Hepatic gene expression and body weight at bin, blood urea nitrogen, and creatinine and activity of alkaline Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 day 10 were significantly correlated: r = 0.52 for IGF1 mRNA phosphatase, alanine aminotransferase, and amylase were sig- (P < 0.05) and r = 0.79 for IGF2 mRNA (P < 0.01). The nificantly affected by the refeeding time according to one-way mean threshold cycle value of all the samples analyzed by the ANOVA. Among these nine components, plasma concentrations autothreshold algorithm in the software for beta-actin mRNA of glucose, inorganic phosphate, potassium, total bilirubin, and

TABLE 3. Hepatic IGF1 mRNA and IGF2 mRNA expression (mean ± SD, n = 4) in Yellowtail fed to varying percentages of satiation (sat) with respect to ration size and linear regression analysis of ration size and gene expression on day 10. Values accompanied by different lowercase letters are signiﬁcantly different among the experimental groups (P < 0.05). Hepatic gene expression level was calculated as relative abundance compared with the 0% sat group (expressed as 1.0).

Correlation with Experimental group ration size Gene 0% sat 20% sat 40% sat 80% sat r-value P-value IGF1 mRNA 1.0 ± 0.2 x 1.6 ± 0.3 z 1.8 ± 0.5 z 2.5 ± 0.8 z 0.77 P < 0.01 IGF2 mRNA 1.0 ± 0.5 y 1.4 ± 0.6 y 2.1 ± 0.9 y 2.8 ± 1.5 z 0.62 P < 0.05 NUTRITIONAL STATUS AND PLASMA BIOCHEMISTRY IN YELLOWTAIL 147

TABLE 4. Plasma biochemical components (mean ± SD, n = 5) in Yellowtail at 0, 3, 6, and9hafterrefeedingandmultiple regression analysis with refeeding time. Values accompanied by different lowercase letters are signiﬁcantly different among the experimental groups (P < 0.05); NS = not signiﬁcant.

Correlation Time after refeeding (h) with time Signiﬁcance Plasma component 0 3 6 9 by ANOVA r-value P-value Albumin 1.9 ± 0.2 1.9 ± 0.4 2.1 ± 0.1 1.9 ± 0.2 NS 0.17 NS Globulin 1.3 ± 0.1 1.8 ± 0.7 1.4 ± 0.2 1.4 ± 0.1 NS 0.38 NS Total protein 3.2 ± 0.1 3.6 ± 0.5 3.6 ± 0.3 3.3 ± 0.2 NS 0.53 NS Glucose 123.7 ± 4.6 y 188.8 ± 23.5 z 164.2 ± 27.2 z 166.0 ± 7.2 z P < 0.001 0.68 P < 0.01 Calcium 13.5 ± 0.4 14.0 ± 0.9 14.2 ± 0.5 14.1 ± 0.4 NS 0.46 NS Inorganic phosphate 8.2 ± 0.5 x 13.0 ± 1.8 yz 14.5 ± 2.5 z 10.6 ± 2.5 xy P < 0.001 0.79 P < 0.01 Sodium 173.3 ± 5.2 172.7 ± 5.9 171.3 ± 4.3 175.0 ± 5.5 NS 0.23 NS Potassium 5.8 ± 0.8 x 3.5 ± 1.0 y 1.9 ± 0.3 z 3.5 ± 0.8 y P < 0.001 0.87 P < 0.01 Total bilirubin 0.3 ± 0.0 y 0.4 ± 0.1 yz 0.5 ± 0.1 z 0.4 ± 0.1 yz P < 0.05 0.48 P < 0.05 Blood urea nitrogen 14.8 ± 1.5 y 17.7 ± 2.6 z 16.0 ± 1.1 yz 17.5 ± 2.0 z P < 0.05 0.38 NS Creatinine 0.5 ± 0.1 x 2.1 ± 0.3 y 3.2 ± 0.3 z 3.6 ± 0.5 z P < 0.001 0.97 P < 0.01 Alkaline phosphatase 28.2 ± 4.3 y 41.2 ± 3.7 z 41.6 ± 6.3 z 37 ± 5.1 z P < 0.001 0.77 NS Amylase 94.8 ± 25.0 y 135.3 ± 35.0 yz 151.2 ± 22.6 z 57.0 ± 54.1 y P < 0.01 0.82 P < 0.01 Alanine amino 6.7 ± 1.5 y 8.2 ± 1.5 yz 8.8 ± 1.2 yz 9.0 ± 2.8 z P < 0.05 0.48 NS transferase

creatinine and activities of amylase were also signiﬁcantly cor- that hepatic beta-actin was suitable for a normalizing gene in related with refeeding time according to regression analysis experiment 2. (Table 4).

DISCUSSION Hepatic IGF1 mRNA and IGF2 mRNA Expression in Experiment 2 Feeding Ration and Growth Hepatic IGF1 mRNA and IGF2 mRNA expression was sig- We showed a significant correlation between the feeding ra- nificantly affected by refeeding time (one-way ANOVA: P < tion and body weight, the growth ratio, and the liver weight (as 0.05). Hepatic IGF1 mRNA expression was significantly in- HSI) in Yellowtail after 10 d of the feeding regimen using fish creased 3 h after refeeding compared with 9 h after refeeding (P with initial body weight of 45 g in 100-L tanks. By comparison, < 0.05: Table 5). A similar increase was seen for hepatic IGF2 fasting decreased the body weight after 2 weeks and the liver mRNA expression (P < 0.05; Table 5). Refeeding time and gene weight after 1 week in Yellowtail with an initial body weight expression were significantly correlated for IGF2 mRNA (r = of 96.5 g using 1,000-L tanks (Fukada et al. 2012). Restricted 0.70, P < 0.05) but not for IGF1 mRNA (r = 0.67 [NS], P = feeding of 1.0, 2.0, and 3.0% of body weight was also signifi- Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 0.07). The Ct of all the samples analyzed for beta-actin gene was cantly correlated with body weight after 2 weeks using 111.4-g 19.9 ± 0.5 (mean ± SD; range, 18.9–20.9) and did not differ fish in 500-L rearing tanks (Fukada et al. 2012). Other studies statistically among the four sampling time groups assessed by on feeding ration and growth have been reported, for exam- one-way ANOVA with post hoc test. Therefore, we conclude ple, in Gilthead Seabream (Perez-S´ anchez´ et al. 1995; Meton´

TABLE 5. Hepatic IGF1 mRNA and IGF2 mRNA expression (mean ± SD, n = 6) in Yellowtail at 0, 3, 6, and 9 h after refeeding and multiple regression analysis of refeeding time and gene expression. Values accompanied by different lowercase letters are signiﬁcantly different among the experimental groups (P < 0.05). Hepatic gene expression level was calculated as relative abundance compared with refeeding in the 0-h group (as expressed 1.0).

Correlation Time after refeeding (h) with time Gene 0 3 6 9 Signiﬁcance by ANOVA r-value P-value IGF1 mRNA 1.0 ± 0.1 yz 1.6 ± 0.5 z 0.9 ± 0.2 yz 0.7 ± 0.3 y P < 0.05 0.67 P = 0.07 IGF2 mRNA 1.0 ± 0.3 yz 2.1 ± 0.6 z 1.1 ± 0.4 yz 0.7 ± 0.1 y P < 0.05 0.70 P < 0.05 148 KAWANAGO ET AL.

et al. 1999, 2000), Arctic Char Salvelinus alpinus (Cameron ration in Yellowtail, suggesting that plasma total protein could et al. 2007), and Common Carp (Shimeno et al. 1997). Our be used as a nutritional indicator in Yellowtail. experimental periods were shorter than those in other reports. Globulin is important for immunological responses. The However, we confirmed a clear relationship between feeding plasma total globulin concentration was higher in Nile Tilapia ration and body weight even in a 10-d experiment. Oreochromis niloticus that received certain herbal ingredients and showed increased growth performance (Goda 2008). An Plasma Biochemical Components increase in globulin and total protein indicates that fish are im- We first reported that the plasma concentrations of albumin, munologically strong (Nayak et al. 2004). In the 0% sat group total protein, calcium, and inorganic phosphate and activities in our study, the plasma globulin concentration was below the of alkaline phosphatase and amylase were significantly affected level of detection. This may mean that fasting has a negative by the feeding ration in Yellowtail. These six components were effect on the immunological status in Yellowtail. also significantly correlated with the feeding ration according to Alkaline phosphatase is an intracellular membrane- linear regression analysis in Yellowtail. In particular, the plasma associated protein that is found in all tissues and is believed to albumin concentration and alkaline phosphatase and amylase function in the transport of ions and absorption of water across activities were strongly correlated, suggesting that changes in cell membranes. As liver-derived isozymes were believed to these components may reflect the feeding ration. By assessing be the major contributor to plasma alkaline phosphatase ac- biochemical components in the refeeding time model, we first tivity in Atlantic Salmon Salmo salar (Johnston et al. 1994), reported that the following nine components were significantly increased plasma alkaline phosphatase activity may be related different among the groups or were correlated with refeeding to increased processing of energy substrates by the liver, lead- time in Yellowtail: glucose, inorganic phosphate, potassium, ing to increased leakage into the blood (Congleton and Wagner total bilirubin, blood urea nitrogen, and creatinine and activities 2006). In our study, plasma alkaline phosphatase activity was of alkaline phosphatase, amylase, and alanine aminotransferase. significantly affected by both the feeding ration and refeeding Comparison of the results from our two experiments showed time. Therefore, using plasma alkaline phosphatase as an nutri- that plasma albumin, total protein, and calcium concentrations tional indicator should be approached with caution because its were affected only by the feeding ration, suggesting that these activity is quite variable in response to the feeding condition. three components are especially useful as nutritional indicators The serum amylase concentration reflects the balance be- that reflect growth in Yellowtail. Plasma alkaline phosphatase tween the rates of amylase entry into and removal from the and amylase activities and the concentration of inorganic phos- blood. The pancreas and salivary glands, and also liver, probably phate were affected by both feeding ration and refeeding time. account for almost all of the serum amylase activity in humans Because these components vary with time, the sampling time (Bhutta and Rahman 1971; Pieper-Bigelow et al. 1990). How- should be carefully considered when using them as nutritional ever, the clinical relevance of low serum amylase is still poorly indicators. understood in humans (Muneyuki et al. 2012). The serum albumin concentration is nutritionally regulated The effects of disease and nutrition on blood amylase have (Don and George 2004). Chronic protein restriction significantly not been closely examined in fish. In Atlantic Salmon, the cDNA decreased the mRNA levels of serum albumin in rat liver (Straus sequence of α-amylase has been cloned, and the mRNA is ex- and Takemoto 1990; Straus et al. 1994). Dietary protein reple- pressed in pancreatic tissue, liver, and heart (Frøystad et al. tion significantly increased hepatic albumin and plasma albumin 2006). A high starch-containing diet increased intestinal amy- concentrations in protein-restricted rats (Qu et al. 1996). Fasting lase activity in Cobia Rachycentron canadum (Ren et al. 2011).

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 also significantly decreased hepatic albumin mRNA in rabbits Whether plasma amylase originates in these tissues in fish is (Rothschild et al. 1968). Compared with studies in mammals, unclear. No reports have been published on the effect of feeding nutritional regulation of albumin is not well understood in fish ration or the time course on blood amylase activity in fish. but is important because plasma albumin may be a useful indi- Plasma amylase activity was affected by both the feeding cator of the nutritional status in fish. ration and refeeding time in Yellowtail. Plasma amylase and Plasma albumin and the total protein concentration were pos- liver weight tended to correlate but not significantly (r = 0.49 itively correlated with the feeding ration in Yellowtail. The [NS], P = 0.06). Plasma amylase activity was also tended to plasma total protein concentration decreased in fasted Rain- correlate with refeeding time. The highest value was observed bow Trout, Chinook Salmon Oncorhynchus tshawytscha (Con- 6 h after refeeding, suggesting an effect of digestive activity. gleton and Wagner 2006), Brown Trout Salmo trutta (Navarro As discussed, plasma amylase activity was affected by both et al. 1992), Common Carp (Shimeno et al. 1997), and Gilthead the feeding ration and refeeding time. Experimental conditions Seabream (Sala-Rabanal et al. 2003). The plasma total protein should thus be carefully determined because of the fluctuating also increased with increasing feeding ration in Common Carp nature of plasma amylase activity. (Shimeno et al. 1997) and Rainbow Trout (Storebakken et al. An increase in the plasma concentration of glucose after 1991). Comparing the results of experiments 1 and 2 in our feeding has been reported in many fish species such as Rain- study, the plasma total protein also mainly reflected the feeding bow Trout (Polakof et al. 2007b), Common Carp (Hoseini and NUTRITIONAL STATUS AND PLASMA BIOCHEMISTRY IN YELLOWTAIL 149

Ghelichpour 2013), Tench Tinca tinca (De Pedro et al. 2005), Hepatic IGF1 mRNA expression was positively correlated and Common Dentex Dentex dentex (Pavlidis et al. 1999), for with the growth ratio in Nile Tilapia with restricted feeding reg- example. Yellowtail also exhibited a postprandial increase in imens, and quantification of hepatic IGF1 levels has been pro- plasma glucose 3–9 h after refeeding, with the highest value posed to be an indicator that can assess growth in this species observed 3 h after refeeding. Although the postprandial insulin (Vera Cruz et al. 2006). The hepatic IGF1 mRNA level was also concentration has not been examined in Yellowtail, changes in correlated with the specific growth ratio using a feed restriction the postprandial plasma glucose concentration may be partly af- model in Yellowtail, but the data were limited to only IGF1 fected by insulin, because insulin administration decreases the mRNA (Fukada et al. 2012). In this study, we are the first to de- blood glucose concentration in Yellowtail (Furuichi and Yone scribe that hepatic IGF2 mRNA was also significantly correlated 1982), and glucose administration increases the blood insulin with body weight in the feeding ration model in Yellowtail. concentration (Furuichi and Yone 1981). In contrast to our study, hepatic IGF1 mRNA was signifi- Half of the calcium in plasma is ionized and half is bound cantly higher in livers of 3–9-d fasted Yellowtail (Pedroso et al. to plasma proteins (Andreasen 1985). A decline in plasma pro- 2009) or was not affected by 3 weeks of fasting (Fukada et al. teins in fasting fish was also seen along with a lower plasma 2012). Reasons for these paradoxical results are not known but calcium concentration (Congleton and Wagner 2006). We did may represent species-specific responses to adapt to some nutri- not observe a correlation between the plasma concentration of tional conditions (Pedroso et al. 2009). Those two reports used calcium and total protein in Yellowtail (data not shown). a 18S rRNA gene as a normalization gene. The mean Ct value The plasma inorganic phosphorus concentration was affected of 18S rRNA is stable under continuous feeding, fasting, and by both the feeding ration and refeeding time in Yellowtail. refeeding conditions in Yellowtail liver (Pedroso et al. 2009). Phosphorus intake differed among the groups because of the Fukada et al. (2012) discussed that appropriate normalization different feeding rations. However, the plasma inorganic phos- is needed to use hepatic IGF1 mRNA as an indicator of nutri- phate concentration was significantly increased only in the 80% tional status in Yellowtail. The beta-actin gene is ubiquitously sat group. In Rainbow Trout, a feeding regimen that contained expressed in most animal cell types and has long been used as various phosphorus levels did not affect the plasma phospho- an internal standard for RNA quantification (Small et al. 2008). rus concentration (Skonberg et al. 1997). Our data are partially In Channel Catfish Ictalurus punctatus,thebeta-actin gene is consistent with that earlier study in Rainbow Trout. not a useful normalization gene under fasting condition because The plasma potassium concentration was significantly de- fasting treatment significantly decreases the mean Ct value of creased 3 and 6 h after refeeding. Insulin treatment decreased beta-actin (Small et al. 2008). Small et al. (2008) suggested that the blood potassium concentration in humans, domestic dogs rRNA (i.e., 18S) is a relatively stable normalization gene for Canis familiaris (DeFronzo et al. 1975), and Norway rats Rat- qPCR data, with some exceptions. In contrast with the previous tus norvegicus (Clausen and Kohn 1977). As discussed above, studies using Yellowtail and Channel Catfish, the mean Ct value the plasma insulin concentration may be increased after refeed- for beta-actin gene was not significantly affected either feeding ing, and the decreased plasma potassium concentration may be ration or feeding time in our study with Yellowtail. Therefore, related to the actions of insulin in Yellowtail. Whether insulin we suggest that the paradoxical changes of hepatic IGF1 mRNA affects the plasma potassium concentration in fish has not, how- affected by the fasting treatment was not due to differences of ever, been examined. the normalization gene because the mean Ct values of both beta- actin and 18S rRNA were stable during the nutritional treatment. This also indicates that not only 18S rRNA but beta-actin mRNA

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 Hepatic IGF1 and IGF2 Expression is a useful normalization gene for qPCR assay for nutritional Both hepatic IGF1 mRNA and IGF2 mRNA expression sig- studies using Yellowtail liver. However, the nutritional effect nificantly correlated with feeding ration in Yellowtail in experi- on normalization stability is not fully understood in Yellowtail. ment 1. Hepatic IGF2 mRNA was significantly affected by and Therefore, selection of a stable normalization gene is suggested correlated with the refeeding time in experiment 2; hepatic IGF1 to be optimized for every examination in Yellowtail. mRNA showed a trend to correlate with refeeding time, but this Examination of the hepatic IGF1 mRNA and IGF2 mRNA was not significant (P = 0.07). These data may indicate that responses to refeeding has been reported for longer refeeding both hepatic IGF1 mRNA and IGF2 mRNA expression reflect periods than those used in this study, such as for days or weeks the feeding ration and feeding time. in Rainbow Trout (Gabillard et al. 2006), Grouper (Pedroso Considering the fluctuation in the gene expression level, in- et al. 2006), and Yellowtail (Fukada et al. 2012). Few data ex- vestigating the sampling time will be necessary because gene ist regarding hepatic IGF1 mRNA and IGF2 mRNA expression expression was relatively dynamic after 3 h and was more stable within 24 h of refeeding. After a postprandial period of 9 h, hep- at 0 h (24 h after the last feeding the day before). These will be atic IGF1 mRNA levels increased to a peak that was equivalent important considerations when assessing the nutritional status to 1.9-fold the value at 0 h in Gilthead Seabream (Meton´ et al. of Yellowtail, especially the growth status, using the expression 2000). Hepatic IGF1 mRNA also reached a peak level 5–6 h of these two genes. after refeeding in Rabbitfish (Ayson et al. 2007). Hepatic IGF1 150 KAWANAGO ET AL.

mRNA increased 2 h after feeding and gradually decreased with ganic phosphate, because these components vary with time, time in Channel Catfish (Pohlenz et al. 2013). Refeeding after sampling time should be carefully considered when using them fasting rapidly increased hepatic IGF1 mRNA and IGF2 mRNA as nutritional indicators. along with plasma IGF1 concentration in hybrid striped bass We also demonstrated that hepatic IGF2 mRNA correlated (White Bass Morone chrysops × Striped Bass M. saxalis), and with both feeding ration and refeeding time in Yellowtail. Hep- these responses are suggested to be nutritional characteristics in atic IGF1 mRNA significantly correlated with feeding ration this hybrid species (Picha et al. 2008). The increase in the an- and, although not significant, had a trend to correlate with abolic endocrine molecule transcripts, IGF1 mRNA and IGF2 refeeding time in Yellowtail. Considering the fluctuation in the mRNA, may be a contributing factor to the accelerated growth gene expression level, it will be important to consider the sam- characteristic of a compensatory growth response (Picha et al. pling time when using the expression of these two genes. Fur- 2008). The plasma IGF1 concentration significantly increased ther approaches, such as measuring the plasma IGF1 and IGF2 2 and 10 h after feeding in Coho Salmon (Shimizu et al. 2009). concentration, are needed to develop nonlethal methods for de- Those investigators discussed that the plasma IGF1 concentra- termining nutritional status. tion response was related to longer-term integrated trends in nutritional status and efficacy of plasma IGF1 as a growth indicator (Shimizu et al. 2009). Similar trends have been reported in ACKNOWLEDGMENTS domestic chickens Gallus gallus in which refeeding increased We thank Tsuyoshi Goto, Naoko Umeda, Erika Abe, and hepatic IGF1 mRNA expression and the plasma IGF1 concen- Noritaka Hirazawa, the fish disease control team, for advising tration 2 h after refeeding (Kita 1998). Our data showed upreg- us about fish rearing methods, and Takashi Hara of the Yellowtail ulated expression of IGF1 mRNA (1.6-fold) and IGF2 mRNA seed production team for giving us the experimental fish. (2.1-fold) 3 h after refeeding in Yellowtail, and these responses may contribute to a postprandial anabolic response. REFERENCES Treatment with IGF1 also significantly increased protein syn- Andreasen, P. 1985. Free and total calcium concentrations in the blood of thesis and decreased protein degradation in Rainbow Trout pri- Rainbow Trout, Salmo gairdneri, during ‘stress’ conditions. Journal of Ex- mary myocytes (Cleveland and Weber 2010), suggesting that perimental Biology 118:111–120. IGF1 promotes protein accretion. Meton´ et al. (2000) and Ayson Ayson, F. G., E. G. deJesus-Ayson, and A. Takemura. 2007. mRNA expression et al. (2007) hypothesized that a postprandial increase in hepatic patterns for GH, PRL, SL, IGF-I and IGF-II during altered feeding status in Rabbitfish, Siganus guttatus. General and Comparative Endocrinology IGF1 mRNA indicates the importance of IGF1 in anabolic func- 150:196–204. tion. This hypothesis is also supported by the results of our study Beckman, B. R., M. Shimizu, B. A. Gadberry, and K. A. Cooper. 2004. Response showing a postprandial trend in increased hepatic IGF1 mRNA of the somatotropic axis of juvenile Coho Salmon to alterations in plane of in Yellowtail. Hepatic IGF2 mRNA also showed postprandial nutrition with an analysis of the relationships among growth rate and circu- changes in our study, which may indicate interspecific differ- lating IGF-I and 41 kDa IGFBP. General and Comparative Endocrinology 135:334–344. ences compared with Rabbitfish, for example. Further studies Bhutta, I. H., and M. A. Rahman. 1971. Serum amylase activity in liver disease. should be performed to reveal the function of hepatic IGF2 Clinical Chemistry 17:1147–1149. mRNA in response to nutritional status in fish. In addition, the Cameron, C., R. Moccia, P. A. Azevedo, and J. F. Leatherland. 2007. Effect method of hepatic IGF1 mRNA and IGF2 mRNA determination of diet and ration on the relationship between plasma GH and IGF-1 con- Salvelinus alpinus requires lethal sampling. Therefore, another approach, such as centrations in Arctic Charr, (L.) Aquaculture Research 38:877–886. measuring the plasma IGF1 and IGF2 concentrations, is needed Clausen, T., and P. G. Kohn. 1977. The effect of insulin on the transport of

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 to assess the nutritional status with live fish as suggested by sodium and potassium in rat soleus muscle. Journal of Physiology 265: Pedroso et al. (2009). Presently, methods for the measurement 19–42. of IGF1 and IGF2 in Yellowtail have not been reported. Thus, Cleveland, B. M., and G. M. Weber. 2010. Effects of insulin-like growth factor- further study to establish these methods should be investigated. I, insulin, and leucine on protein turnover and ubiquitin ligase expression in Rainbow Trout primary myocytes. American Journal of Physiology Regula- tory Integrative and Comparative Physiology 298:R341–350. Congleton, J. L., and T. Wagner. 2006. Blood-chemistry indicators of nutritional CONCLUSION status in juvenile salmonids. Journal of Fish Biology 69:473–490. We evaluated nutritional indicators in Yellowtail by analyz- De Pedro, N., A. I. Guijarro, M. A. Lopez-Pati´ no,˜ R. Mart´ınez-Alvarez,´ and M. J. Delgado. 2005. Daily and seasonal variations in haematological and ing plasma biochemical components that reflect nutritional or blood biochemical parameters in the Tench, Tinca tinca Linnaeus, 1758. metabolic status and whether hepatic IGF1 mRNA and IGF2 Aquaculture Research 36:1185–1196. mRNA expression reflect the nutritional status. We showed that DeFronzo, R. A., R. S. Sherwin, M. Dillingham, R. Hendler, W. W. Tamborlane, plasma albumin, total protein, and calcium concentrations were and P. Felig. 1978. Influence of basal insulin and glucagon secretion on affected only by the feeding ration, suggesting that these three potassium and sodium metabolism. Journal of Clinical Investigation 61:472– 479. components are especially useful as nutritional indicators that Don, B. R., and K. George. 2004. Poor nutritional status and inflammation: reflect Yellowtail growth. In the case of plasma alkaline phos- serum albumin: relationship to inflammation and nutrition. Seminars in Dial- phatase and amylase activities and the concentration of inor- ysis 17:432–437. NUTRITIONAL STATUS AND PLASMA BIOCHEMISTRY IN YELLOWTAIL 151

Escobar,C.,M.D´ıaz-Munoz,˜ E. Fabiola, and R. Aguilar-Roblero. 1998. Persis- Nayak, A. K., B. K. Das, M. P. S. Kohli, and S. C. Mukherjee. 2004. The im- tence of metabolic rhythmicity during fasting and its entrainment by restricted munosuppressive effect of α-permethrin on Indian Major Carp, Rohu (Labeo feeding schedules in rats. American Journal of Physiology Regulatory Inte- rohita Ham.). Fish and Shellfish Immunology 16:41–50. grative and Comparative Physiology 274:R1309–R1316. Pavlidis, M., L. Greenwood, M. Paalavuo, H. Mols¨ a,¨ and J. T. Laitinen. 1999. Frøystad, M. K., E. Lilleeng, A. Sundby, and Å. Krogdahl. 2006. Cloning The effect of photoperiod on diel rhythms in serum melatonin, cortisol, glu- and characterization of α-amylase from Atlantic Salmon (Salmo salar L.). cose, and electrolytes in the Common Dentex, Dentex dentex. General and Comparative Biochemistry and Physiology Part A Molecular and Integrative Comparative Endocrinology 113:240–250. Physiology 145:479–492. Pedroso, F. L., E. G. deJesus-Ayson, H. H. Cortado, S. Hyodo, and F. G. Ayson. Fukada, H., K. Murashita, T. Furutani, and T. Masumoto. 2012. Yellowtail 2006. Changes in mRNA expression of grouper (Epinephelus coioides) insulin-like growth factor 1: molecular cloning and response to various nu- growth hormone and insulin-like growth factor I in response to nutritional tritional conditions. Domestic Animal Endocrinology 42:220–229. status. General and Comparative Endocrinology 145:237–246. Furuichi, M., and Y.Yone.1981. Change of blood sugar and plasma insulin levels Pedroso, F. L., H. Fukada, and T. Masumoto. 2009. Molecular characteriza- of fishes in the glucose tolerance test. Nippon Suisan Gakkaishi 47:761–764. tion, tissue distribution patterns and nutritional regulation of IGFBP-1, -2, (In Japanese.) -3 and -5 in Yellowtail, Seriola quinqueradiata. General and Comparative Furuichi, M., and Y. Yone. 1982. Effect of insulin on blood sugar levels of fishes Endocrinology 161:344–353. [carp, red seabream and Yellowtail]. Nippon Suisan Gakkaishi 48:1289– Perez-S´ anchez,´ J., H. Mart´ı-Palanca, and S. J. Kaushik. 1995. Ration size 1291. and protein intake affect circulating growth hormone concentration, hepatic Gabillard, J. C., B. B. Kamangar, and N. Montserrat. 2006. Coordinated reg- growth hormone binding and plasma insulin-like growth factor-1 immunore- ulation of the GH/IGF system genes during refeeding in Rainbow Trout activity in marine teleost, the Gilthead Seabream (Sparus aurata). Journal of (Oncorhynchus mykiss). Journal of Endocrinology 191:15–24. Nutrition 125:546–552. Goda, A. M. A.-S. 2008. Effect of dietary ginseng herb (ginsana G115) sup- Picha, M. E., M. J. Turano, C. K. Tipsmark, and R. J. Borski. 2008. Regulation of plementation on growth, feed utilization, and hematological indices of Nile endocrine and paracrine sources of Igfs and Gh receptor during compensatory Tilapia, Oreochromis niloticus (L.), fingerlings. Journal of the World Aqua- growth in hybrid striped bass (Morone chrysops × Morone saxatilis). Journal culture Society 39:205–214. of Endocrinology 199:81–94. Hoseini, S. M., and M. Ghelichpour. 2013. Effects of pre-sampling fasting on Pieper-Bigelow, C., A. Strocchi, and M. D. Levitt. 1990. Where does serum serum characteristics of Common Carp (Cyprinus carpio L.). International amylase come from and where does it go? Gastroenterology Clinics of North Journal of Aquatic Biology 1:6–13. America 19:793–810. Johnston, C. E., B. S. Horney, S. Deluca, A. MacKenzie, J. G. Eales, and R. Pohlenz, C., B. Alejandro, M. Thomas, B. C. Small, D. S. MacKenzie, and Angus. 1994. Changes in alkaline phosphatase isoenzyme activity in tissues D. M. Gatlin III. 2013. Effects of dietary arginine on endocrine growth and plasma of Atlantic Salmon (Salmo salar) before and during smoltification factors of Channel Catfish, Ictalurus punctatus. Comparative Biochemistry and gonadal maturation. Fish Physiology and Biochemistry 12:485–497. and Physiology Part A Molecular and Integrative Physiology 166:215– Jones, J. I., and D. R. Clemmons. 1995. Insulin-like growth factors and their 221. binding proteins: biological actions. Endocrine Reviews 16:3–34. Polakof, S., R. M. Ceino, B. Fernandez-Dur´ an,J.M.M´ ´ıguez, and J. L. Soengas. Kawanago, M., S. Takemura, R. Ishizuka, and I. Shioya. In press. Dietary 2007a. Daily changes in parameters of energy metabolism in brain of Rainbow branched-chain amino acid supplementation affects growth and hepatic Trout: dependence on feeding. Comparative Biochemistry and Physiology insulin-like growth factor gene expression in Yellowtail, Seriola quinquera- Part A Molecular and Integrative Physiology 146:265–273. diata. Aquaculture Nutrition. DOI: 10.1111/anu.12141. Polakof, S., J. M. M´ıguez, and J. L. Soengas. 2007b. Daily changes in pa- Kita, K. 1998. Refeeding increases hepatic insulin-like growth factor-I (IGFI) rameters of energy metabolism in liver, white muscle, and gills of Rainbow gene expression and plasma IGF-I concentration in fasted chicks. British Trout: dependence on feed. Comparative Biochemistry and Physiology Part Poultry Science 39:679–682. A Molecular and Integrative Physiology 147:363–374. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression Qu,Z.,P.R.Ling,J.C.Chow,P.A.Burke,R.J.Smith,andB.R.Bistrian. data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. 1996. Determinants of plasma concentrations of insulin-like growth factor-I Methods 25:402–408. and albumin and their hepatic mRNAs: the role of dietary protein content and MAFF (Ministry of Agriculture, Forestry and Fisheries, Japan). 2011. Fiscal tumor necrosis factor in malnourished rats. Metabolism 45:1273–1278. year 2011. Annual report on forest and forestry in Japan. MAFF, Tokyo. Ren, M., Q. Ai, M. Kangsen, H. Ma, and X. Wang. 2011. Effect of dietary carbo- Available: http://www.maff.go.jp/e/index.html. (February 2014). hydrate level on growth performance, body composition, apparent digestibil- Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 Meton,´ I., A. Caseras, E. Canto,´ F. Fernandez,´ and I. V. Baanante. 2000. Liver ity coefficient and digestive enzyme activities of juvenile Cobia, Rachycentron insulin-like growth factor-I mRNA is not affected by diet composition or canadum L. Aquaculture Research 42:1467–1475. ration size but shows diurnal variations in regularly-fed Gilthead Seabream Rothschild, M. A., M. Oratz, J. Mongelli, and S. S. Schreiber. 1968. Effects of (Sparus aurata). Journal of Nutrition 130:757–760. a short-tern fast on albumin synthesis studied in vivo, in the perfused liver, Meton,´ I., D. Mediavilla, A. Caseras, E. Canto,F.Fernandez,´ and I. V.Baanante. and on amino acid incorporation by hepatic microsomes. Journal of Clinical 1999. Effect of diet composition and ration size on key enzyme activities of Investigation 47:2591–2599. glycolysis–gluconeogenesis, the pentose phosphate pathway and amino acid Sala-Rabanal, M., J. Sanchez,´ A. Ibarz, J. Fernandez-Borr´ as,` J. Blasco, and M. metabolism in liver of Gilthead Seabream (Sparus aurata). British Journal of A. Gallardo. 2003. Effects of low temperatures and fasting on hematology and Nutrition 82:223–232. plasma composition of Gilthead Seabream (Sparus aurata). Fish Physiology Muneyuki, T., K. Nakajima, A. Aoki, M. Yoshida, H. Fuchigami, H. Munakata, and Biochemistry 29:105–115. S. E. Ishikawa, H. Sugawara, M. Kawakami, S. Momomura, and M. Kakei. Shimeno, S., T. Shikata, H. Hosokawa, T. Masumoto, and D. Kheyyali. 1997. 2012. Latent associations of low serum amylase with decreased plasma insulin Metabolic response to feeding rates in Common Carp, Cyprinus carpio. Aqua- levels and insulin resistance in asymptomatic middle-aged adults. Cardiovas- culture 151:371–377. cular Diabetology [online serial] 11:80. Shimizu, M., K. A. Cooper, W. W. Dickhoff, and B. R. Beckman. 2009. Post- Navarro, I., J. Gutierrez,´ and J. Planas. 1992. Changes in plasma glucagon, in- prandial changes in plasma growth hormone, insulin, insulin-like growth sulin and tissue metabolites associated with prolonged fasting in Brown Trout factor (IGF)-I, and IGF-binding proteins in coho salmon fasted for varying (Salmo trutta fario) during two different seasons of the year. Comparative periods. American Journal of Physiology Regulatory Integrative and Com- Biochemistry and Physiology Part A Physiology 102:401–407. parative Physiology 297:R352–R361. 152 KAWANAGO ET AL.

Skonberg, D. I., Y. Leon, W. H. Ronald, and F. M. Dong. 1997. Metabolic transthyretin nuclear transcripts in rat liver. Journal of Nutrition 124:1041– response to dietary phosphorus intake in Rainbow Trout (Oncorhynchus 1051. mykiss). Aquaculture 157:11–24. Straus, D. S., and C. D. Takemoto. 1990. Effect of dietary protein deprivation on Small, B. C., C. A. Murdock, A. L. Bilodeau-Bourgeois, B. C. Peterson, and G. insulin-like growth factor (IGF)-I and -II, IGF binding protein-2, and serum C. Waldbieser. 2008. Stability of reference genes for real-time PCR analyses albumin gene expression in rat. Endocrinology 127:1849–1860. in Channel Catﬁsh (Ictalurus punctatus) tissues under varying physiological Uchida, K., S. Kajimura, L. G. Riley, T. Hirano, K. Aida, and E. G. Grau. conditions. Comparative Biochemistry and Physiology Part B Biochemistry 2003. Effects of fasting on growth hormone/insulin-like growth factor I axis and Molecular Biology 151:296–304. in the tilapia, Oreochromis mossambicus. Comparative Biochemistry and Storebakken, T., S. S. O. Hung, C. C. Calvert, and E. M. Plisetskaya. 1991. Physiology Part A Molecular and Integrative Physiology 134:429–439. Nutrient partitioning in Rainbow Trout at different feeding rates. Aquaculture Vera Cruz, E. M., C. L. Brown, J. A. Luckenbach, M. E. Picha, R. B. Bolivar, and 96:191–203. R. J. Borski. 2006. Insulin-like growth factor-I cDNA cloning, gene expres- Straus, D. S., N. W. Marten, J. M. Hayden, and E. J. Burke. 1994. Pro- sion and potential use as a growth rate indicator in Nile Tilapia, Oreochromis tein restriction speciﬁcally decreases the abundance of serum albumin and niloticus. Aquaculture 251:585–595. Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:17 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Methods and Accuracy of Sexing Sockeye Salmon Using Ultrasound for Captive Broodstock Management Deborah A. Frosta, W. Carlin McAuleya, Bryon Kluvera, Mike Wastela, Desmond Maynarda & Thomas A. Flagga a National Marine Fisheries Service, Northwest Fisheries Science Center, Environmental and Fisheries Science Division, Manchester Research Station, Post Office Box 130, Manchester, Washington 98353, USA Published online: 15 Apr 2014.

To cite this article: Deborah A. Frost, W. Carlin McAuley, Bryon Kluver, Mike Wastel, Desmond Maynard & Thomas A. Flagg (2014) Methods and Accuracy of Sexing Sockeye Salmon Using Ultrasound for Captive Broodstock Management, North American Journal of Aquaculture, 76:2, 153-158, DOI: 10.1080/15222055.2014.886647 To link to this article: http://dx.doi.org/10.1080/15222055.2014.886647

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 North American Journal of Aquaculture 76:153–158, 2014 American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886647

TECHNICAL NOTE

Methods and Accuracy of Sexing Sockeye Salmon Using Ultrasound for Captive Broodstock Management

Deborah A. Frost,* W. Carlin McAuley, Bryon Kluver, Mike Wastel, Desmond Maynard, and Thomas A. Flagg National Marine Fisheries Service, Northwest Fisheries Science Center, Environmental and Fisheries Science Division, Manchester Research Station, Post Ofﬁce Box 130, Manchester, Washington 98353, USA

spleen, stomach, and gonads. With fish, ultrasound imaging is Abstract often used for identifying sex and reproductive readiness, in- Ultrasound has been widely used to noninvasively examine the cluding gonad volume, fecundity, egg condition, and spawn con- internal anatomy and reproductive status of many fish species. dition (see Novelo and Tiersch 2012 for review). Martin et al. Since 2003, ultrasonography has been an integral part of broodstock management for Endangered Species Act–listed endangered (1983) were among the first to observe maturing salmonids Redfish Lake Sockeye Salmon Oncorhynchus nerka. Initial ultra- (Coho Salmon Oncorhynchus kisutch) with ultrasound. Other sound screenings for each year’s broodstock were conducted 3 to researchers have reported use of ultrasonic examinations to ob- 4 months prior to anticipated maturation, before external signs tain gonad sizes (Blythe et al. 1994), prespawn versus postspawn of maturation developed. In keeping with the natural life history condition of Steelhead O. mykiss (anadromous Rainbow Trout) of these fish, this permitted the separation of nonmaturing fish from maturing fish so the immature fish could maintain feeding kelts (Evans et al. 2004), and egg size and fecundity (Bryan and growth. It also permitted maturing seawater-reared fish to be et al. 2005). Some authors have reported using ultrasonogra- transferred to freshwater for final maturation. Designing a spawn- phy for serial monitoring of gonad development (Newman et al. ing matrix to maintain genetic diversity requires knowledge of the 2008). Since 2003, we have used ultrasound technology to aid sex of each fish before spawning. Approximately 1 month prior to broodstock husbandry management and development of breed- spawning, a second ultrasound was performed to verify sex identification and thus ensure the accuracy of the spawning matrix. Over ing scenarios for a captive broodstock gene rescue program for a 9-year period, 2,662 fish were examined. The accuracy of sexing the most endangered stock of salmon in the Pacific Northwest, age-3 adult Sockeye Salmon ranged from 94.0% to 100%, with Endangered Species Act (ESA)-listed endangered Redfish Lake an average of 97.7%, while the accuracy of sexing age-4 adult fish Sockeye Salmon O. nerka (Waples 1991; Flagg et al. 2004). ranged from 92.0% to 100%, with an average of 94.4%. The av- A key component of the Redfish Lake Sockeye Salmon gene erage accuracy was similar for fish examined in seawater or fresh- Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 water (98.1% versus 97.3%, respectively). Ultrasound was shown rescue program is production of gametes for use in restoration to be an accurate and effective tool for managing the husbandry of activities at Redfish Lake (Idaho). Annually, several hundred our captive broodstock life history. adults are reared from each broodstock lineage (Maynard et al. 2012). These fish may mature at ages 2 to 5 years, typically at age 3 or age 4 in captivity. Their management and husbandry Ultrasound technology has been widely used to noninvasively practices are conducted in a manner similar to their natural life examine internal anatomy and reproductive status for many fish history. For instance, maturing fish need to be transitioned off species. Ultrasonography uses acoustic energy to create cross- their feed in late spring (up to 4 to 5 months before maturity sectional moving images of internal anatomies. Different inter- in October), while nonmaturing fish in the same brood group nal tissues have different structural densities and provide dif- need to be maintained on their feeding and growth schedule. ferential ultrasound echo reception. This allows for ultrasound Fish maturing in seawater environments at our laboratory need imaging to provide a visual assessment of size, shape, and con- to be transitioned to freshwater in late spring/early summer. Ad- dition of internal organs, including such things as heart, liver, ditionally, annual breeding matrix designs to ensure maximum

*Corresponding author: [email protected] Received September 13, 2013; accepted January 2, 2014 153 154 FROST ET AL.

avoidance of inbreeding need to be developed in advance of minations of maturing fish were made at the initial ultrasound spawning (Kozfkay et al. 2008). in June. Only a few more maturing fish were identified when the This complicated husbandry scenario necessitated an ability assessments repeated in July. All nonmaturing fish remained in to determine whether fish are initiating sexual maturity prior their rearing locations until they matured in subsequent years. to the development of visual cues (Swanson et al. 2008). We A final ultrasound examination was conducted in late August initially tried to make all these fish culture–management deci- or early September on all presumed maturing fish to verify the sions based solely on judging external secondary sexual char- initial maturation and sex assessments for calculating the rate of acteristics and color changes. However, this required multiple accuracy of those assessments. The accuracy of these data was examinations of the fish, usually later in summer, which de- vital to the development of the spawning matrix, and ultimately, layed both freshwater transition for maturing seawater-reared to maintaining genetic diversity in this highly managed stock. fish and growth trajectories for the nonmaturing fish. The use Sex was ultimately confirmed at or near spawning by obvious of ultrasound technology has greatly simplified the process of secondary sexual characteristics (i.e., body shape) or spawning maturation sorting for our broodstock program. (expression of gametes). In this communication, we relate our methods and accuracy The ultrasound platform we used consisted of a plastic bin using ultrasound as a long-term management tool that serves two (51 × 38 × 19 cm) filled with anesthetic-treated water to purposes: (1) husbandry and life history management protocols about 2.5 cm below the top rim. A V-shaped “cradle” (67- of separating maturing fish from nonmaturing fish and moving cm-long slats 12 cm wide at the top slanted to a 3.8-cm-wide maturing seawater-reared fish to freshwater when needed, and opening at the bottom) was positioned diagonally in the bin (2) accuracy of the spawning matrix for breeding fish in cap- lengthwise (Figure 1). Small batches of fish (two to six) were tivity. We hope this summary of our success with ultrasound dipnetted to a 1.0-m-diameter tank holding about 200 L of water imaging will serve as a useful case history and blueprint for mixed with about 180–200 mL of a stock solution (50 g/L) of managing an intensive broodstock or gene-rescue program. tricaine methanesulfonate (MS-222). After about 2–3 min, the fish lost equilibrium and did not react to grasping of the caudal peduncle. Anesthetizing small batches ensured that fish were METHODS not under anesthesia for too long. Fish were observed closely These examinations were conducted at the National Marine for gill movement. Fisheries Service (NMFS) Northwest Fisheries Science Cen- Each fish was hand-carried and laid in the cradle with the head ter’s Manchester Research Station, near Seattle, Washington. slanting down in the water, ventral side up, and their dorsal fin This report covers our ultrasound work between 2003 and 2011. aligning through the 3.8 cm opening in the bottom of the cradle Juvenile fish were hatched and initially reared in freshwater. (Figure 2). Larger fish (typically >500 g) were placed on the As parr, the fish were marked with passive integrated transpon- cradle with their head submerged at least to the gills. Smaller fish der (PIT) tags for individual identification and fin-clipped for (<500 g) were held submerged in the palm of the hand during DNA analysis for use in constructing future breeding matrices scanning, as they tended to sink through the opening between the (Kozfkay et al. 2008). As smolts, a portion were transferred slats in the cradle. No fish restraints were used in this procedure, to tanks supplied with filtered seawater for rearing to matu- other than holding the anesthetized fish by the caudal peduncle. rity, while the remainder continued to be reared to maturity in Most ultrasound examinations were conducted inside a building, freshwater (see Maynard et al. 2012 for details). In the wild, ma- but some of the final maturation checks were done in an enclosed turing Sockeye Salmon cease feeding before freshwater reentry greenhouse type structure that covered our raceways. The bright

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 in June to July and spawn in their natal lakes in September and sunlight there washed out the contrast, making it sometimes October. To approximate this sequence, feed rations for the 3- difficult to distinguish the gonads in the images. We found it and 4-year-old fish (their typical ages of maturation in captivity) essential to shade the ultrasound machine to increase visibility were reduced through May and then eliminated by mid-June. of the screen. We constructed a shaded “booth” out of polyvinyl In late June, an initial ultrasound assessment was conducted to chloride pipe and black plastic sheeting, large enough for a determine whether fish were maturing, as evidenced by enlarg- person to stand inside with the machine and see the images in a ing gonads in the ultrasound images, and if so, their sex was darkened environment. recorded. Fish determined to be maturing were separated from An Aloka model SSD-500V ultrasound system (www. the nonmaturing fish so the immature fish could resume feeding. alokavet.com) was the unit primarily used to examine each Maturing seawater-reared fish were transported to freshwater for fish individually. It was used in “B” mode, with a focus depth final maturation. Maturing freshwater-reared fish were placed between 11 and 22 cm. In a few instances, a Sonosite Vet in separate tanks from the immature fish. 180 Plus, with an L38/10-5 MHz transducer (probe), was also In some years, a second ultrasound was performed on the used; however, the operation was similar for either system. remaining immature fish in July to reassess for maturing fish The Aloka model SSD-500V ultrasound machine included a that may have been missed at the initial examination. However, variable gain, between 30 and 90 db, but was primarily used the inconsistency of these assessments led us to group these data in midrange (i.e., 66 db). The linear probe (UST-5561-7.5; with the initial ultrasound assessments. Generally, most deter- 7.5 MHz) had a 38-mm wide field of view and was used TECHNICAL NOTE 155

FIGURE 1. The ultrasound equipment includes the console, probe, holding pan, and cradle. Elevating and shielding the console increases visibility and reduces splash interference.

uncovered and without transmission gel, because the fish’s wet an average accuracy rate of 97.7% for age-3 (N = 2,572) and skin provided good transmission. The probe could be used 94.4% for age-4 (N = 90) fish (Table 1). Overall, 64 of the fish submerged or in air as long as the fish’s skin was wet. The probe examined (2.4%) were incorrectly identified as to sex or ma- was placed perpendicular to the body axis on the fish’s ventral turity status (Table 2). The majority (62.5%) of the erroneous surface between the pectoral fins, corresponding interiorly calls were males that were mistakenly called females. The sec- to the pericardial wall. While holding the fish by the caudal ond largest error (21.9%) was immature fish that were mistaken

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 peduncle with one hand, the observer moved the probe with the for maturing males. other hand posteriorly along the ventral surface of the abdomen During the Sockeye Salmon maturation windows in 2010 toward the pelvic fins and viewed the image on the screen. For and 2011, we made direct comparisons of sex error rates for each fish, date of examination, PIT-tag number, status (either freshwater-reared versus seawater-reared fish. The error rates maturing or immature), and, if maturing, the sex were recorded. between freshwater-reared adults (2.7%) versus seawater-reared Research and handling of these ESA-listed fish is limited by adults (1.9%) were consistent with the overall error rate for all federal permits. We did not subject them to prolonged handling fish examined between 2003 and 2011 (2.4%). by attempting to identify the stage of maturation or gametoge- Ultrasound images appeared on the screen in real time. Each nesis. The Aloka SSD-500V has an optional peripheral 3.5-in scan took between 5 to 10 s to locate and identify the internal floppy drive for recording images; typically, we did not record structures and gonads. Typically, we located the kidney first and images but occasionally saved an image with good visual quality proceeded to the gonads from there. Maturing female ovaries for future reference. appeared as paired, bright, granular masses of smaller circular units (the eggs), while the male testes appeared as generally grey or darker amorphous lobes (Figure 3). Detailed descriptions by RESULTS AND DISCUSSION Martin et al. (1983) of the distinct echoic nature of the go- We examined 2,662 fish over 9 years and were able to iden- nads in maturing salmon were helpful in establishing references tify maturing Sockeye Salmon in our captive broodstocks at markers for our assessments. Evans et al. (2004) also provided 156 FROST ET AL.

FIGURE 2. Fish being ultrasonically examined. Note ventral presentation with head in water, and probe perpendicular to the body axis. Downloaded by [Department Of Fisheries] at 20:17 15 June 2014

FIGURE 3. Ultrasound images of Sockeye Salmon, with large arrows showing (A) maturing ovaries in a female and (B) maturing testes in a male. The small arrows point to the kidney, which is used as a reference point for orientation. TECHNICAL NOTE 157

TABLE 1. Numbers (N) of Sockeye Salmon spawned at age 3 and age 4 from ovaries due to the hypoechoic (nonreflective) nature of and accuracy of ultrasound in determining sex approximately 4 months before both tissues. In our observations, the testes appeared dark and spawning. The total numbers and average accuracy for each age-class are in bold italics. uniform in texture, as sometimes immature ovaries did also. As the eggs began to mature and enlarge with fluid, their mem- Brood NNNAccuracy branes became hyperechoic, reflecting a brighter white on the year (males) (females) (errors) (%) ultrasound output. The eggs also became individually distin- guishable as small circular organelles on the ultrasound image. Age 3 As the testes matured, they either remained dark or sometimes 2000 110 204 0 100 lightened to a shaded grey, leading to misinterpretation. How- 2001 115 152 16 94.0 ever, testes also increased in diameter and length as the fish 2002 197 135 7 97.9 matured: it was usually this relative size difference that was 2003 170 204 5 98.7 ultimately used to distinguish between maturing testes and im- 2004 174 129 11 96.4 mature ovaries. 2005 181 125 5 98.4 We found the late June time frame to be the earliest we could 2006 90 92 6 96.7 achieve reliable sex and maturation determination through the 2007 104 121 7 96.9 use of ultrasonography. This is at least 2 months before visual 2008 130 139 2 99.3 external signs of maturation reliably distinguish males from fe- 1,271 1,301 59 97 7 . males, based upon our prior experience utilizing only visual Age 4 cues with this captive broodstock. Dual ultrasound assessments 2001 6 0 0 100 (initial and final) allowed us to recognize the changes in the 2003 0 2 0 100 gonads as they matured. Increases in gonad size and shading 2006 39 18 3 94.7 changes ranging from subtle to distinct helped us to recognize 2007 22 3 2 92.0 variations in the ultrasound images of the fish and to combine 67 23 5 94.4 both of these factors to improve our image reading. We reduced the time the fish were handled by quickly obtaining clear ultrasound images and by knowledgeable interpretation of those useful descriptions of the ultrasonic properties of fish tissue. images. Images were read by one of our two or three trained, In our observations, maturing gonads were clearly visible. Be- experienced people. We did not track results between differ- cause maturing gonads developed anteriorly first, they started ent readers. Sex determinations on questionable images were out in the ultrasound image as relatively large in diameter and generally made after the readers conferred with each other. diminished in size as the probe was moved toward the ventral Dual assessments also enabled verification of our original fins. In immature fish, undeveloped gonad tissue either was not maturity assessments. We found this ability to be an important observed, or if observed, lacked definition and faded away much management tool for developing an accurate breeding profile by sooner when the probe was moved. These fish were generally allowing us to make corrections before creating the spawning returned to the immature population with the anticipation that matrix. These breeding designs have been quite successful in further development, if they were indeed maturing, would be retaining a large portion (>95%) of the original founding genetic clearer on the next maturity check on immature fish in July or variability of the population (Kalinowski et al. 2012). August. Other studies that reported determinations of sex and repro-

Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 For the most part in our examinations, the gonads were easy ductive status had accuracies greater than 90%, similar to our to identify. However, in a few instances where errors occurred, results. Blythe et al. (1994) were able to achieve a 95% mean testes in earlier stages were sometimes difficult to distinguish accuracy in female sex determination in Striped Bass Morone saxatilis and were able to track maturation status through the TABLE 2. Number and types of ultrasound errors (missed calls) in Redfish annual reproductive cycle. Masoudifard et al. (2011) were able Lake Sockeye Salmon captive broodstock maturity checks, verified at spawning. to achieve 97.5% accuracy for sex determination in Beluga Stur- Total geon Huso huso. Newman et al. (2008) observed an overall 96% Error type Age 3 Age 4 errors % Error accuracy across several age-classes of Murray Cod Maccul- lochella peelii, with a few smaller and younger fish accounting Female, called male 8 0 8 12.5 for most of their misinterpretations, as gonads were smaller and Male, called female 37 3 40 62.5 had less echogenic contrast. Newman et al., however, interpreted Immature, called male 12 2 14 21.9 most of their males by the lack of ovarian detection, whereas Immature, called female 2 0 2 3.1 in our studies, we generally used size and shade of the gonads and presence or absence of egg-like organelles to characterize Total 59 5 64 sex of maturing fish. Our ability to detect gonad maturation 4 158 FROST ET AL.

months before spawning is consistent with the ultrasound de- Evans, A. F., M. S. Fitzpatrick, and L. K. Siddens. 2004. Use of ultra- terminations of Reimers et al. (1987), who were able to detect sound imaging and steroid concentrations to identify maturational status in impending maturation 5 months before spawning. This is an adult steelhead. North American Journal of Fisheries Management 24:967– 978. important aspect of our successful management of this captive Flagg, T. A., C. W. McAuley, P. K. Kline, M. S. Powell, D. Taki, and J. C. Gisla- broodstock. son. 2004. Application of captive broodstocks to preservation of ESA-listed In conclusion, the use of ultrasound examination has been stocks of Pacific salmon: Redfish Lake Sockeye Salmon case example. Pages critical to the successful broodstock management for the Red- 387–400 in M. J. Nickum, P. M. Mazik, J. G. Nickum, and D. D. MacKin- fish Lake Sockeye Salmon captive broodstock efforts. With an lay, editors. Propagated fish in resource management. American Fisheries Society, Symposium 44, Bethesda, Maryland. average accuracy rate of 97.5%, our experience using ultra- Kalinowski, S. T., D. M. Van Doornik, C. C. Kozfkay, and R. S. Waples. sonography for sorting our captive broodstock for maturity has 2012. Genetic diversity in the Snake River Sockeye Salmon captive brood- helped us maintain important life-cycle elements similar to those stock program as estimated from broodstock records. Conservation Genetics in the natural environment. This versatile, noninvasive tool also 13:1183–1193. permitted development of accurate annual breeding profiles for Kozfkay, C., M. Campbell, J. Heindel, D. Baker, P. Kline, M. Powell, and T. Flagg. 2008. A genetic evaluation of relatedness for broodstock management maintaining the genetic diversity of our ESA-listed fish. This of captive, endangered Snake River Sockeye Salmon, Oncorhynchus nerka. communication provides useful techniques and information for Conservation Genetics 9:1421–1430. using ultrasound to determine sex and maturity. It may also pro- Martin, R. W., J. Myers, S. A. Sower, D. J. Phillips, and C. McAuley. 1983. vide guidance to others about managing some aspects of the Ultrasonic imaging, a potential tool for sex determination of live fish. North husbandry and breeding of an at-risk population. American Journal of Fisheries Management 3:258–264. Masoudifard, M., A. R. Vajhi, M. Moghim, R. M. Nazari, A. R. Naghavi, and M. Sohrabnejad. 2011. High validity sex determination of three year ACKNOWLEDGMENTS old cultured Beluga Sturgeon Huso huso using ultrasonography. Journal of Applied Ichthyology 27:643–647. This Redfish Lake Sockeye Salmon gene rescue program is Maynard, D. J., T. A. Flagg, W. C. McAuley, D. A. Frost, B. Kluver, M. R. a cooperative between NMFS, the Idaho Department of Fish Wastel, J. E. Colt, and W. W. Dickhoff. 2012. Fish culture technology and and Game, the Bonneville Power Administration (BPA), the practices for captive broodstock rearing of ESA-listed salmon stocks. NOAA Shoshone Bannock Tribes, and others involved with salmon con- Technical Memorandum NMFS-NWFSC-117. servation in the Columbia River Basin. The program is funded Newman, D. M., P. L. Jones, and B. A. Ingram. 2008. Sexing accuracy and indicators of maturation status in captive Murray Cod Maccullochella peelii by the BPA and coordinated through the Northwest Power and peelii using non-invasive ultrasonic imagery. Aquaculture 279:113–119. Conservation Council’s Fish and Wildlife Program. Mention of Novelo, N. D., and T. R. Tiersch. 2012. A review of the use of ultrasonography specific products does not constitute endorsement by NOAA. in fish reproduction. North American Journal of Aquaculture 74:169–181. Reimers, E., P. Landmark, T. Sorsdal, E. Bohmer, and T. Solum. 1987. Deter- mination of salmonids’ sex, maturation and size: an ultrasound and photocell REFERENCES approach. Aquaculture Magazine 13(November/December):41–44. Blythe, B., L. A. Helfrich, W. E. Beal, B. Bosworth, and G. S. Libey. 1994. Swanson, P., B. Campbell, K. Shearer, J. Dickey, B. Beckman, D. Larsen, L. Determination of sex and maturational status of Striped Bass Morone saxatilis Park, and B. Berejikian. 2008. Application of reproductive technologies to using ultrasonic imaging. Aquaculture 125:175–184. captive breeding programs for conservation of imperiled stocks of Pacific Bryan, J. L., M. L. Wildhaber, and D. B. Noltie. 2005. Examining Neosho salmon. Cybium 32(Supplement 2):279–282. Madtom reproductive biology using ultrasound and artificial photothermal Waples, R. S., O. W. Johnson, and R. P. Jones Jr. 1991. Status review for Snake cycles. North American Journal of Aquaculture 67:221–230. River Sockeye Salmon. NOAA Technical Memorandum NMFS-F/NWC-195. Downloaded by [Department Of Fisheries] at 20:17 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:17 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Decline in Feeding Activity of Female Cultured Delta Smelt Prior to Spawning Tien-Chieh Hunga, Kai J. Ederb, Alireza Javidmehrc & Frank J. Loged a Fish Conservation and Culture Laboratory, Biological and Agricultural Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA b Civil and Environmental Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA c Aquatic Health Program, Veterinary Medicine: Anatomy, Physiology and Cell Biology, University of California, Davis, One Shields Avenue, Davis, California 95616, USA d Civil and Environmental Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA Published online: 15 Apr 2014.

To cite this article: Tien-Chieh Hung, Kai J. Eder, Alireza Javidmehr & Frank J. Loge (2014) Decline in Feeding Activity of Female Cultured Delta Smelt Prior to Spawning, North American Journal of Aquaculture, 76:2, 159-163, DOI: 10.1080/15222055.2014.886650 To link to this article: http://dx.doi.org/10.1080/15222055.2014.886650

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 North American Journal of Aquaculture 76:159–163, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886650

TECHNICAL NOTE

Decline in Feeding Activity of Female Cultured Delta Smelt Prior to Spawning

Tien-Chieh Hung* Fish Conservation and Culture Laboratory, Biological and Agricultural Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA Kai J. Eder Civil and Environmental Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA Alireza Javidmehr Aquatic Health Program, Veterinary Medicine: Anatomy, Physiology and Cell Biology, University of California, Davis, One Shields Avenue, Davis, California 95616, USA Frank J. Loge Civil and Environmental Engineering Department, University of California, Davis, One Shields Avenue, Davis, California 95616, USA

sitive fish; however, their limited number in the field hinders Abstract research purposes. The Fish Conservation and Culture Labora- An observed decline in feeding behavior of Delta Smelt tory at the University of California, Davis (UC Davis), plays an Hypomesus transpacificus due to spawning is presented here. At important role in culturing Delta Smelt as a reliable supply for three different temperatures to which examined fish acclimated (10.2, 14.1, and 18.1◦C), female Delta Smelt consistently showed research programs and developing a genetically managed refu- significantly higher stomach content than male Delta Smelt when gial population (Lindberg et al. 2013). The cultured smelt have carrying early-stage eggs. However, a significant decline of feed allowed for the development of a bioenergetic model of Delta intake was observed in female Delta Smelt when eggs ripened and Smelt (Eder et al. 2012), which has also enabled observations were ready for spawning. These results indicate that female Delta on feeding throughout the spawning season. Smelt tend to have high food consumption at the egg development stages but change behavior as eggs ripen. Possible causes for the The feeding activity of Delta Smelt close to spawning is

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 change in feeding behavior are discussed. of interest because most fish change their appetite and feeding behavior during sexual maturation, spawning migration, and spawning (Hoskins and Volkoff 2012). A decline in feeding is Delta Smelt Hypomesus transpacificus are small, primarily often seen during the reproductive stages of fish (Krumsick and annual fish, once abundant in the upper San Francisco Bay–Delta Rose 2012; Ozyurt et al. 2012), and sometimes can be observed Estuary in Northern California. However, the number of Delta in both sexes (Fordham and Trippel 1999; Skjæraasen et al. Smelt has declined dramatically since the 1980s (Bennett and 2004). Good nutrition and ample food for the mother fish is es- Moyle 1996); they were listed federally as threatened in 1993 sential to developing greater energy reserves during early stages (U.S. Office of the Federal Register 1993) and further listed as of egg formation; therefore, females tend to continue feeding endangered in 2009 (CFGC 2009) by the state of California. as they approach spawning (Skjæraasen et al. 2004) or start The Delta Smelt are considered an indicator of the ecosystem feeding sooner after spawning than do male fish and consume condition of the Delta since they are an environmentally sen- more food than males (Fordham and Trippel 1999). Those two

*Corresponding author: [email protected] Received October 9, 2013; accepted January 11, 2014

159 160 HUNG ET AL.

feeding–reproductive patterns are believed to be related to the as they approach spawning has not been described previously; seasons in which the fish spawn (Link and Burnett 2001). By therefore, we present observations in this note. studying the change of feeding activities, important information could be provided to the bioenergetic model, and maturity of eggs could be predicted by observing the change of food METHODS residual/feces ratio in each culturing tank. Generally speaking, feeding activity and reproductive behav- Fish collection and care.—Approximately 1,560 cultured adult ior are controlled by a complex system of hormones, including Delta Smelt were obtained from the UC Davis Fish Conservation appetite stimulators and inhibitors (Volkoff et al. 2010). The and Culture Laboratory at Byron, California, and transported to brain and gastrointestinal tract are the two major sources of the Putah Creek Facility of the Center for Aquatic Biology and appetite-regulating hormones (Hoskins and Volkoff 2012). The Aquaculture on the UC Davis campus. The fish were distributed change of feeding behavior could be caused by altered hormone equally among three parallel recirculating aquaculture systems. level inside the fish, and the change is different among species. Each system has an independent water circuit and three circu- For example, Atlantic Salmon Salmo salar stop eating long be- lar holding tanks. The fish were acclimated to the systems for fore spawning, but Haddock Melanogrammus aeglefinus cease 9 days at 12.5◦C, and the system temperatures were then in- eating just before spawning (Luquet and Watanabe 1986). The creased or decreased to target temperatures of 18.1◦C (System difference in timing of cessation of feeding may also be caused 1), 14.1◦C (System 2), and 10.2◦C (System 3). The temperature by other matters such as anadromous migration nature (Bardon- was changed at a rate of 1◦C per 2 days. The fish were held for net and Bagliniere` 2000). In addition, the behavior of searching a total of 37 days before the experiment began. The light/dark for microhabitat may cause fish to go off feed (Gladstone 2007a, regime during the experiment was 14:10 (with lights on from 2007b). The change of feeding activity of cultured Delta Smelt 0600 to 2000 hours). Downloaded by [Department Of Fisheries] at 20:18 15 June 2014

FIGURE 1. Organs/tissues of a female Delta Smelt. (a) Early-stage eggs, (b) ripe-stage eggs, (c) stomach, and (d) intestine. This is one sample of some cultured Delta Smelt that do not release eggs naturally and cause them to become overripe. Only 2 females of the 1,080 examined Delta Smelt were observed to carry both early-stage and ripe-stage eggs, and those two ﬁsh were not included in any category of the examined ﬁsh. TECHNICAL NOTE 161

Stomach content and egg maturation.—We present here a rela- diately. The obtained stomach content samples were dried for tionship between stomach content and the maturation of eggs 24hat58◦C and cooled in a desiccator before weighing. The carried by the female Delta Smelt. The fish were provided with stomach content is presented as the percentage of the fish wet excess food (approximately 2% of the fish wet weight per day) weight. continuously using automatic feeders. The food offered was a Because this study was carried out during a food consumption mixture of formulated 4/6 NRD diet (INVE Aquaculture, Salt study, a full set of samples was gathered on day 5 only and partial Lake City, Utah) and 370 Hikari plankton food (By-Rite Pet sets were gathered on the rest of the days. A total of 491 adult Supply, Hayward, California) at a 2:1 volume ratio. Twelve fish Delta Smelt (240 male, 245 female, and 6 unknown) were in- were sampled from each system every 3 h between 0600 and volved in this study. The females were separated into two groups 2100 hours (72 samples per system per day) for five successive based on maturation stages of their eggs: early stage (Stage II days. Sampled fish were weighed and killed by severing the to III) and ripe stage (Stage IV to V) as shown in Figure 1. spinal cord, and their stomach contents were extracted imme- These stages are based on the macrocharacteristics described in

(A) (B) All examined Delta Smelt Male Delta Smelt 1.0 1.0 18.1° C (n=12,23,24,36,36,36) 18.1° C (n= 6, 8,11,19,16,11) 0.9 14.1° C (n=12,24,12,36,36,36) 0.9 14.1° C (n= 8,13, 6,16,23,18) 10.2° C (n=12,24,24,36,36,36) 10.2° C (n= 6,12, 9,19,18,21) 0.8 0.8

0.7 0.7

0.6 0.6

0.5 0.5

0.4 0.4

0.3 0.3

0.2 0.2

0.1

Stomach content (% of fish wet weight) of fish wet content (% Stomach 0.1 Stomach contentStomach of fish wet (% weight)

0.0 0.0 0600 0900 1200 1500 1800 2100 0600 0900 1200 1500 1800 2100 Sampling time Sampling time (C) (D) Female Delta Smelt with early-stage eggs Female Delta Smelt with ripe-stage eggs 1.0 1.0 18.1° C (n= 3,12,12,12,19,22) 18.1° C (n= 2, 3, 1, 5, 1, 3) 0.9 14.1° C (n= 4, 9, 6,16,12,15) 0.9 14.1° C (n= 0, 2, 0, 4, 1, 3) 10.2° C (n= 5, 8,12,12,13,13) 10.2° C (n= 0, 2, 1, 5, 5, 2) 0.8 0.8

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 0.7 0.7

0.6 0.6

0.5 0.5

0.4 0.4

0.3 0.3

0.2 0.2

0.1

Stomach content (% of fish wet weight) of fish wet content (% Stomach 0.1 Stomach content (% of fish wet weight) of fish wet content (% Stomach

0.0 0.0 0600 0900 1200 1500 1800 2100 0600 0900 1200 1500 1800 2100 Sampling time Sampling time

FIGURE 2. Stomach contents of examined Delta Smelt collected at various sampling times and temperatures. (A) All examined fish, (B) male fish, (C) female fish with early-stage eggs, and (D) female fish with ripe-stage eggs. The numbers shown in the legends represent the number of samples taken at each sequential sampling time, and the error bars represent standard deviations. 162 HUNG ET AL.

the California Department of Fish and Wildlife’s classification tion, one-way analysis of variance (ANOVA) method followed system (CDFW 2013; Mager et al. 2004). At the early stage, by an all pairs Tukey’s test was used to compare the stomach ovaries are visible to the naked eyes, and oocytes are slightly contents of fish on day 5 of all the three categories (male, female orange and smaller than 0.5 mm in diameter. The fish with en- with early-stage eggs, and female with ripe-stage eggs). All re- larged abdomen and bright orange oocytes (about or larger than ported P-values are from the mentioned statistical approach. 1 mm in diameter) are considered to carry ripe-stage eggs. RESULTS Statistical analysis.—The subsample of fish mentioned above Among the fish examined, the percentages of male and fe- has been used to study the association between egg matura- male were 48.9% and 49.9%, respectively. Based on the egg tion and gut content. Gender of fish was considered as another maturation identification method used in this study, 83.7% of ex- factor to discover any significant difference between male and amined female Delta Smelt carried early-stage eggs and 16.3% female samples. For this purpose, 187 fish were subsampled carried ripe-stage eggs. Stomach contents of all the examined from day 5 for statistical analysis. Demographic characteristics fish are shown in Figure 2. The six smelt with unknown sex of subsampled fish were similar to the rest of samples, and the were included only in the “all examined Delta Smelt” category result could be generalized to the experimental study. Samples (Figure 2A). The results show that the Delta Smelt in all cate- from six different sampling events were combined since sam- gories tended to feed less at the first sampling time of the day pling time was not an interesting factor in this study. Two-tailed (0600 hours). Although Delta Smelt are mostly sensitive to light paired Student’s t-tests were used to compare the stomach con- at their early stages (Lindberg et al. 2013), the light regime could tent between all the examined male Delta Smelt and female still affect the feeding of adult Delta Smelt. The stomach content Delta Smelt with early-stage eggs at each temperature. In addi- in the female Delta Smelt with early-stage eggs appears to be Downloaded by [Department Of Fisheries] at 20:18 15 June 2014

FIGURE 3. One-way ANOVA of fish samples taken on day 5. Fish are grouped by the acclimated temperature at (A) 18.1◦C, (B) 14.1◦C, and (C) 10.2◦C. F-early indicates the female fish carries early-stage eggs while F-ripe indicates the female fish with ripe-stage eggs. Diamonds show mean, one SD, and 2SD for each group. Width and height of diamonds indicate sample size and stomach contents, respectively. Each dot is one data point. Circles for Tukey’s tests show upper and lower 95% confidence levels. TECHNICAL NOTE 163

slightly but not significantly more than that of the male Delta J. T. Hollibaugh, editor. San Francisco Bay: the ecosystem. American Asso- Smelt (P > 0.05) at all three system temperatures (Figure 2B ciation for the Advancement of Science, San Francisco. versus Figure 2C). CDFW (California Department of Fish and Wildlife). 2013. Spring Kodiak survey: egg stages. CDFW, Sacramento. Available: http://www.dfg.ca.gov/ The ANOVAresults (Figure 3) show that although there is no delta/data/skt/eggstages.asp. (February 2014). significant difference between the stomach content in male fish CFGC (California Fish and Game Commission). 2009. Uplisting the Delta Smelt ◦ ◦ and that in female fish with ripe-stage eggs at 14.1 C and 10.2 C, to endangered species status: amending California Endangered Species Act, a significant difference was observed between males and females Title 14, CCR, Section 670.5. with early-stage eggs (P < 0.0001) at all three temperatures Eder, K. J., J. C. Lindberg, N. A. Fangue, and F. J. Loge. 2012. Longfin smelt ◦ bioenergetics. University of California–Davis, Report to the U.S. Bureau of tested: 18.1, 14.1, and 10.2 C. A significant difference was also Reclamation, R10AC20107, Davis. found between the stomach contents of female fish with early- Fordham, S. E., and E. A. Trippel. 1999. Feeding behaviour of cod (Gadus stage and those with ripe-stage eggs at all three temperatures morhua) in relation to spawning. Journal of Applied Ichthyology 15:1–9. (P < 0.0001, P = 0.0011, and P = 0.0047, respectively). Gladstone, W. 2007a. Temporal patterns of spawning and hatching in a spawning P-values were calculated by using the Tukey–Kramer honestly aggregation of the temperate reef fish Chromis hypsilepis (Pomacentridae). Marine Biology 151:1143–1152. significant difference method. Gladstone, W. 2007b. Selection of a spawning aggregation site by Chromis hypsilepis (Pisces: Pomacentridae): habitat structure, transport poten- DISCUSSION tial, and food availability. Marine Ecology Progress Series 351:235– When carrying early-stage eggs, female Delta Smelt showed 247. Hoskins, L. J., and H. Volkoff. 2012. The comparative endocrinology of feeding higher stomach content than males. However, the growth in in fish: insights and challenges. General and Comparative Endocrinology length of female Delta Smelt slows during spawning season 176:327–335. (Moyle 2002). This suggests that the female fish is using gained Kirkwood, T. B., and R. Holliday. 1979. The evolution of ageing and longevity. energy for gonadal development. The trade-off between using Proceedings of the Royal Society of London Series B Biological Sciences energy for reproduction or somatic cell growth is commonly 205:531–546. Krumsick, K. J., and G. A. Rose. 2012. Atlantic Cod (Gadus morhua) feed seen in fish (Kirkwood and Holliday 1979; Valdesalici and during spawning off Newfoundland and Labrador. ICES Journal of Marine Cellerino 2003). Additionally, female Delta Smelt apparently Science 69:1701–1709. stopped feeding when their eggs became ripe and were ready Lindberg, J. C., G. Tigan, L. Ellison, T. Rettinghouse, M. M. Nagel, and K. M. to be delivered. The decline of fish feeding activity at spawn- Fisch. 2013. Aquaculture methods for a genetically managed population of ing has been observed in many species (Luquet and Watanabe endangered Delta Smelt. North American Journal of Aquaculture 75:186– 196. 1986; Skjæraasen et al. 2004). At this stage, the eggs expand Link, J., and J. Burnett. 2001. The relationship between stomach contents and and occupy most of the available space in the fish. The reduced maturity state for major northwest Atlantic fishes: new paradigms? Journal space in the body cavity for food could explain the change of of Fish Biology 59:783–794. feeding activity (Fordham and Trippel 1999). Reduced feed- Luquet, P., and T. Watanabe. 1986. Interaction “nutrition-reproduction” in fish. ing could also be caused by hormone levels triggering anorexia Fish Physiology and Biochemistry 2(1–4):121–129. Mager, R. C., S. I. Doroshov, J. P. Eenennaam, and R. L. Brown. 2004. Early (Fordham and Trippel 1999; Volkoff et al. 2010; Hoskins and life stages of Delta Smelt. Pages 169–180 in F.Feyrer,L.R.Brown,R.L. Volkoff 2012) or by behavioral changes that supersede feeding Brown, and J. J. Orsi, editors. Early life history of fishes in the San Fran- activity, such as searching for spawning microhabitat. Further cisco Estuary and watershed. American Fisheries Society, Symposium 39, study is needed to reveal the major cause of feeding suppression Bethesda, Maryland. in these fish. Moyle, P. B. 2002. Inland fishes of California. University of California Press, Berkeley. Ozyurt, C. E., S. Mavruk, and V. B. Kiyaga. 2012. Effects of predator size and Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 ACKNOWLEDGMENTS gonad maturation on food preference and feeding intensity of Sander luciop- This study was funded by the U.S. Bureau of Reclamation un- erca (Linnaeus, 1758). Turkish Journal of Fisheries and Aquatic Sciences der contract R10AC20107. Any opinions, findings, and conclu- 12:315–322. Skjæraasen, J. E., A. G. V. Salvanes, Ø. Karlsen, R. Dahle, T. Nilsen, and sions or recommendations expressed in this material are those B. Norberg. 2004. The effect of photoperiod on sexual maturation, appetite of the authors and do not necessarily reflect the views of the and growth in wild Atlantic Cod (Gadus morhua L.). Fish Physiology and supporting agency. Assistance in daily system maintenance and Biochemistry 30:163–174. sampling provided by Rose U. Hoang is gratefully acknowl- U.S. Office of the Federal Register. 1993. Endangered and threatened wildlife edged. and plants: determination of threatened status for the Delta Smelt. Federal Register 58:42(5 March 1993):12854–12864. Valdesalici, S., and A. Cellerino. 2003. Extremely short lifespan in the annual REFERENCES fish Nothobranchius furzeri. Proceedings of the Royal Society of London Bardonnet, A., and J. L. Bagliniere.` 2000. Freshwater habitat of Atlantic Salmon Series B Biological Sciences 270(Supplement 2):S189–S191. (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences 57:497– Volkoff, H., L. J. Hoskins, and S. M. Tuziak. 2010. Influence of intrinsic signals 506. and environmental cues on the endocrine control of feeding in fish: po- Bennett, W. A., and P. B. Moyle. 1996. Where have all the fishes gone? Interac- tential application in aquaculture. General and Comparative Endocrinology tive factors producing fish declines in the Sacramento–San Joaquin estuary. 167:352–359. This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:18 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Effect of Stocking Density on Growth and Survival of the Prawn Macrobrachium tenellum Cultured in a Cage- Pond System Fermín López-Uriosteguia, Jesús T. Ponce-Palafoxa, José L. Arredondo-Figueroab, Mario A. Benítez-Mandujanoc, Manuel García-Ulloa Gómezd, Sergio Castillo Vargasmachucae & Héctor M. Esparza-Lealf a Estudiante de Posgrado de Ciencias Biológicas y Agropecuarias, Universidad Autónoma de Nayarit, Escuela Nacional de Ingeniería Pesquera, Laboratory Bioingeniería Costera, Tepic, Nayarit 63155, Mexico b Universidad Autónoma de Aguascalientes, Centro de Ciencias Agropecuarias, km 3 de la carretera Jesús María-La Posta, Jesús María, Aguascalientes 20131, Mexico c Universidad Juárez Autónoma de Tabasco, División Académica Multidisciplinaria De Los Ríos, Av. Universidad s/n, Zona de la Cultura, Col. Magisterial, Centro, Tabasco 86040, Mexico d Universidad de Guadalajara, San Patricio, Melaque, Jalisco 48900, Mexico e Universidad Autónoma de Nayarit, Escuela Nacional de Ingeniería Pesquera, Bahía de Matanchen, San Blas, Nayarit 63000, Mexico f Instituto Politecnico Nacional Centro Interdisciplinario de Investigación para el Desarrollo Integral Unidad Regional Sinaloa, Bulevar Juan de Dios Bátiz Paredes 250, Colonia San Joaquin, Guasave, Sinaloa 81101, Mexico Published online: 15 Apr 2014.

To cite this article: Fermín López-Uriostegui, Jesús T. Ponce-Palafox, José L. Arredondo-Figueroa, Mario A. Benítez- Mandujano, Manuel García-Ulloa Gómez, Sergio Castillo Vargasmachuca & Héctor M. Esparza-Leal (2014) Effect of Stocking Density on Growth and Survival of the Prawn Macrobrachium tenellum Cultured in a Cage-Pond System, North American Journal of Aquaculture, 76:2, 164-169, DOI: 10.1080/15222055.2014.886646

To link to this article: http://dx.doi.org/10.1080/15222055.2014.886646

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 Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 North American Journal of Aquaculture 76:164–169, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886646

ARTICLE

Effect of Stocking Density on Growth and Survival of the Prawn Macrobrachium tenellum Cultured in a Cage-Pond System

Ferm´ın Lopez-Uriostegui´ and Jesus´ T. Ponce-Palafox* Estudiante de Posgrado de Ciencias Biologicas´ y Agropecuarias, Universidad Autonoma´ de Nayarit, Escuela Nacional de Ingenier´ıa Pesquera, Laboratory Bioingenier´ıa Costera, Tepic, Nayarit 63155, Mexico Jose´ L. Arredondo-Figueroa Universidad Autonoma´ de Aguascalientes, Centro de Ciencias Agropecuarias, km 3 de la carretera Jesus´ Mar´ıa-La Posta, Jesus´ Mar´ıa, Aguascalientes 20131, Mexico Mario A. Ben´ıtez-Mandujano Universidad Juarez´ Autonoma´ de Tabasco, Division´ Academica´ Multidisciplinaria De Los R´ıos, Av. Universidad s/n, Zona de la Cultura, Col. Magisterial, Centro, Tabasco 86040, Mexico Manuel Garc´ıa-Ulloa Gomez´ Universidad de Guadalajara, San Patricio, Melaque, Jalisco 48900, Mexico Sergio Castillo Vargasmachuca Universidad Autonoma´ de Nayarit, Escuela Nacional de Ingenier´ıa Pesquera, Bah´ıa de Matanchen, San Blas, Nayarit 63000, Mexico Hector´ M. Esparza-Leal Instituto Politecnico Nacional Centro Interdisciplinario de Investigacion´ para el Desarrollo Integral Unidad Regional Sinaloa, Bulevar Juan de Dios Batiz´ Paredes 250, Colonia San Joaquin, Guasave, Sinaloa 81101, Mexico Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 Abstract Freshwater prawns, Macrobrachium tenellum, were reared at four stocking densities (6, 12, 18, and 24 prawns/m3), with three replicates each, in 12 bottom cages of 3 m3 capacity that were placed in a 1,422-m2 earthen pond. The growth, weight gain, production, speciﬁc growth rate (SGR), feed conversion ratio (FCR), and survival of the prawns were determined. We stocked 540 juveniles that had an average weight of 1.57 ± 0.09 g (mean ± SE) for 180 d and fed them twice a day with commercial shrimp pellets containing 35% crude protein. Water quality variables were measured during the study. All of the growth and production parameters were affected by stocking density (P < 0.05). The mean weight and SGR increased at low densities. The lowest mean weight (17.2 ± 2.0 g) was observed at the higher density (24 prawns/m3), although production increased at high densities and varied from 1,307.2 kg/ha at a density of 6 prawns/m3 to 2,013.3 kg/ha at a density of 24 prawns/m3. Survival varied from 79 ± 1% at a density of 6 prawns/m3 to 47.5 ± 0.6% at a density of 24 prawns/m3. The overall results suggested that stocking density affected the growth and survival of M. tenellum cultured in the cage-pond system. The initial stocking density represents a very important

*Corresponding author: [email protected] Received July 11, 2013; accepted December 27, 2013

164 EFFECT OF STOCKING DENSITY ON PRAWN SURVIVAL 165

culture variable in terms of marketing for this freshwater prawn because at all tested densities the prawns reached their individual market size.

Freshwater prawn culture on the Mexican Pacific coast is a polyethylene screen to exclude predators and prevent prawns restricted to the exotic species, the giant river prawn Macro- from escaping. The earthen pond was rectangular, and the cages brachium rosenbergii. However, certain native species of the were placed in two parallel lines, each consisting of six cages same genus, such as the longarm river prawn M. tenellum and occupying one-half of the pond. Water exchange (5% of the cauque river prawn M. americanum, offer a high potential for pond water) was performed nightly from 1900 to 2100 hours to use in aquaculture (Ponce-Palafox et al. 2002). The distribu- assure adequate water circulation and prevent stratification of tion of M. tenellum in North and South America ranges from the water column. Baja California, Mexico (27◦N) to the Chira River, Peru (5◦S) Water variables.—Temperature and dissolved oxygen were (Holthuis 1980). Since the late 1970s, M. tenellum has been measured daily with a YSI-85 handheld meter (Yellow Springs considered a good candidate for cultivation. It is found at high Instruments) and pH was measured daily with a digital pH me- densities under natural conditions, is not aggressive, and can ter (Bernauer F-1002) in the cage area at a depth of 30 cm at tolerate a wide range of fluctuating temperatures, salinity, and 1000 and 1700 hours. Water transparency was measured with a oxygen concentrations (Ponce-Palafox et al. 2013). In addition, Secchi disk. Samples of water were collected monthly from the it has weak pincers, is unable to harm organisms that it manip- subsurface water near the cages to monitor the following vari- + ulates, and is evidently unable to leave the water because its ables: alkalinity, ammonia nitrogen (NH4 -N), nitrite-nitrogen − − pereiopods are not strong enough to lift the weight of its body. (NO2 -N), nitrate-nitrogen (NO3 -N), and total phosphorus (to- Many studies on the technological aspects of M. tenellum tal P). All of these analyses were performed with a YSI 9000 culture have recently been conducted and have included re- photometer (Yellow Springs Instruments). search on diseases in the commercial production of prawn in Experimental procedures.—Macrobrachium tenellum juve- rural farms (Ponce-Palafox et al. 2005), nutrition (Garc´ıa-Ulloa niles (n = 540; weight, 1.57 ± 0.09 g [mean ± SD]) were et al. 2008; Espinosa-Chaurand et al. 2012), and grow-out (Vega- stocked in 12 cages at densities of 6, 12, 18, and 24 prawns/m3. Villasante et al. 2011; Ponce-Palafox et al. 2013). Nevertheless, Three replicates, randomly distributed in the central area of the several external culture-related factors, such as density, remain pond, were used per treatment. The prawns were fed twice daily to be studied to facilitate optimal freshwater prawn production. with a commercial shrimp diet (35% crude protein, 8% lipid, The culture of freshwater prawns has traditionally been per- and 12% moisture; size 2–2.5 × 6–7 mm; Nutripec Camaronina formed in ponds (New and Singholka 1985; Correia et al. 2002; XT, Agribrands, Purina, Mexico). The daily feed ratio was de- Weimin and Xianping 2002; Lam et al. 2006; Murthy et al. 2012) termined based on wet weight and gradually adjusted from 10% and cage culture (Stanley and Moore 1983; Sagi et al. 1986; at the beginning of the experiment to 5% after 90 d of growth. Marques et al. 2000; Cuvin-Aralar et al. 2007), primarily with The 5% ratio continued to be fed until the end of the experi- M. rosenbergii. The aim of this study was to evaluate the influ- ment (180 d). The daily feed ratios were based on feed demand ence of various stocking densities on the growth performance, and determined after periodic monitoring of the feed trays. The production, and survival of M. tenellum reared in a pond in a feeding transitions occurred at the same time for all treatments.

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 system of bottom cages. All of the prawns were collected and weighed monthly to evaluate their growth and adjust the amount of feed supplied. After 180 d of culture, the prawns were individually counted and METHODS weighed to determine mean weight (g), weight gain (g/week), Locale and experimental structure.—The experiment was production (kg/ha), specific growth rate (SGR; %/d), feed con- performed in a 1,422-m2 earthen pond located at the San version efficiency (FCE; %) and survival (%). Both SGR and Cayetano Aquaculture Center, the property of the Fish- FCE were determined as follows: eries Department of the Nayarit State Government, Mex- ◦ ◦ / = × − / − ico (21 27 24.23 N, 104 49 29.67 W). Twelve bottom cages SGR (% d) 100 [(ln W2 ln W1) t2 t1)] and 3 (3 × 1 × 1 m), each with a capacity of 3 m , were placed in FCE (%) = 100 × (W2 − W1)/C, the pond. The cages were constructed from polyethylene mesh (0.7 mm diameter) and iron frames. Circular feeding trays 50 cm where W2 and W1 are mean body weights (g) at times t2 and t1 in diameter made of polyethylene screening were placed in the (d), respectively, and C is the wet weight of food consumed (g). cages, and synthetic mesh bags (raffia bags) were provided as Statistical analysis.—The prawn performance data were ana- refuges for the prawns. The cages were tied to stakes, placed lyzed for normality and homoscedasticity with a Shapiro–Wilk on the pond bottom at a water depth of 1 m, and covered with test and a Bartlett test, respectively. As the requirements of 166 LOPEZ-URIOSTEGUI´ ET AL.

± TABLE 1. Pond water quality variables (mean SD) in cage-pond system The maximum density (dmax) was obtained as follows: of M. tenellum during grow-out (180 d). Within a row, means accompanied by different lowercase letters are significantly different (P < 0.05). β d = 1 , max β (2) Water quality variables Morning Afternoon 2 2 ◦ ± ± Temperature ( C) 27.0 1.2 x 28.8 1.2 x where dmax = maximum density. The maximum biomass was Transparency (Secchi 82.5 ± 0.5 given by disk depth, cm) ± ± β2 β Dissolved oxygen 8.2 1.6 x 8.6 0.6 x 1 1 Biomassmax = β0 + − . (3) (mg/L) 2β2 2 pH 8.5 ± 0.2 x 8.4 ± 0.1 x Alkalinity 110.0 ± 0.90 x 135 ± 1.3 y (mg CaCO /L) 3 RESULTS NH –N (mg/L) 0.19 ± 0.05 x 0.15 ± 0.08 x 3 In general, the temperature, dissolved oxygen, pH, ammo- NO −–N (mg/L) 0.55 ± 0.07 x 0.42 ± 0.09 x 2 nia, nitrites, and nitrates did not differ significantly (P > 0.05) NO −–N (mg/L) 1.58 ± 0.15 x 1.32 ± 0.35 x 3 between the morning and afternoon in all cages and thus among Total P (mg/L) 0.56 ± 0.09 x 0.38 ± 0.10 y treatments (Table 1). Significant differences (P < 0.05) were found only in alkalinity and total P concentration. The minimum temperature and oxygen concentration were 25.0◦C and 7.0 mg/L, respectively. During the summer, the temperature these tests were satisfied by both the prawn performance data remained at approximately 27.9◦C (1–149 d in culture). From and the data on the water quality variables, a paired t-test was November to December (150–180 d in culture, fall–winter under used to compare the morning and afternoon means of these vari- the temperate climate of the study locale) the mean temperature ables. All data from all treatments were subjected to an ANOVA was 26.4◦C (the minimum was 25◦C). Dissolved oxygen showed (Montgomery 1997). Differences among treatments were eval- higher values during the grow-out phase (8.2–8.6 mg/L). The uated with a Tukey multiple comparison test of the means. The pH values varied slightly and remained close to 8.5. The other results were evaluated at the 5% significance level. Values ex- water quality variables are documented in Table 1. pressed as percentages were square-root arcsine-transformed The growth, production, and survival data showed signifi- prior to analysis, but the nontransformed values are presented cant differences (P < 0.05) among densities (Table 2). The fi- for ease of interpretation. The analyses were conducted using nal weight decreased as prawn density increased, ranging from Statistica package version10 (StatSoft, Tulsa, Oklahoma). The 27.0 ± 1.9 g in the 6-prawns/m3 group to 17.2 ± 2.0 g in the 24- density–biomass relationship produced a quadratic equation of prawns/m3 group (Figure 1). Production varied from 1,307.0 ± the form 151.3 kg/ha in the 6-prawns/m3 group to 2,013.0 ± 220.1 kg/ha in the 24-prawns/m3 group. Survival decreased with density, = β + β + β , Biomass 2d2 (1) varying from 79 ± 1% in the 6-prawns/m3 group to 47.5 ± 0.6% in the 24-prawns/m3 group. The SGR was significantly 3 where β2 is a quadratic coefficient (other than 0), β1 is the linear higher (P < 0.05) in the 6-prawns/m group (1.5 ± 0.1). The coefficient, β0 is the intercept, and d is the density. FCR differed significantly (P < 0.05) among treatments. The Downloaded by [Department Of Fisheries] at 20:18 15 June 2014

TABLE 2. Data (mean ± SD) for cage-pond system on growth performance and production of M. tenellum cultured at different stocking densities. Within a row, means accompanied by different lowercase letters are signiﬁcantly different (P < 0.05).

Density (prawns/m3) Parameters 6 12 18 24 Initial weight (g) 1.6 ± 0.3 z 1.5 ± 0.5 z 1.6 ± 0.4 z 1.6 ± 0.2 z Final weight (g) 27.0 ± 1.9 z 21.6 ± 1.2 y 19.0 ± 2.3 yx 17.2 ± 2.0 x Biomass (g/m3) 130.7 ± 05.1 z 173.5 ± 10.1 y 192.0 ± 12.7 yx 201.3 ± 14.5 x Weight gain (kg/week) 1.05 ± 0.20 x 0.84 ± 0.09 y 0.72 ± 0.05 z 0.65 ± 0.02 z Production (kg/ha) 1,307.2 ± 151.3 1,734.6 ± 163.4 1,920.2 ± 189.0 2,013.3 ± 220.1 z SGR (%/d) 1.5 ± 0.06 z 1.4 ± 0.05 y 1.3 ± 0.06 x 1.3 ± 0.08 x FCR 1.2 ± 0.1 w 1.8 ± 0.1 x 2.1 ± 0.1 y 2.4 ± 0.1 z Survival (%) 79.0 ± 1.0 z 65.5 ± 0.9 y 55.5 ± 0.3 x 47.5 ± 0.6 w EFFECT OF STOCKING DENSITY ON PRAWN SURVIVAL 167

at the 30-prawns/m3 density, where growth showed density dependence. Substitution of the values obtained in the ﬁeld (Fig- ure 2) in the equation for the parabolic curve yielded the following relationship:

Biomass = 74.98 + 10.806 d − 0.2322 d2. (4)

Equations (2) and (3) were then applied, and a density of 24 prawns/m3 was found to produce the maximum biomass.

DISCUSSION The values of the water quality variables were within the range recommended for the cultivation of freshwater prawns (New 2002). Because M. tenellum tolerates a wide range of varying environmental conditions (Ponce-Palafox et al. 2013), FIGURE 1. Growth of M. tenellum at different stocking densities in a cage- this result suggests that the temperature was the only sampled pond system. Stocking densities followed by different lowercase letters in the environmental variable that could directly influence prawn per- legend are significantly different at P < 0.05. formance in this experiment. The stocking densities used in this study ranged from semi- FCR was low at densities of 6 and 12 prawns/m2, with values of intensive (6 prawns/m3) to intensive (24 prawns/m3) culture sys- 1.2 ± 0.1 and 1.8 ± 0.1, respectively. Prawn survival differed tems. These values are higher than those traditionally used for significantly among treatments (P < 0.05). Survival was greater prawn pond culture (Cuvin-Aralar et al. 2007). The mean indi- than 50% in the 6- and 18-prawns/m3 treatments (55.5–79.0%). vidual weight of the cultured prawns in this study did not attain However, higher levels of production were recorded in the 18- the desired market size established for M. rosenbergii (35 g) and 24-prawns/m3 treatments (1,920.0–2,013.0 kg/ha). for the principal shellfish markets in Mexico (Ponce-Palafox Biomass increased up to 201.3 kg/m3 at the 24-prawns/m3 1997). Nevertheless, the prawns attained the commercial size density (Figure 2) and then decreased to lower than 190.2 kg/m3 (25 g) recognized for this species in local markets in Mexico Downloaded by [Department Of Fisheries] at 20:18 15 June 2014

FIGURE 2. Biomass of M. tenellum at different stocking densities in a cage-pond system. 168 LOPEZ-URIOSTEGUI´ ET AL.

(Ponce-Palafox et al. 2002). In fact, the commercial size ob- gest that density-dependent processes regulate the final weight, tained in this study for M. tenellum is comparable with the weight gain, FCR, and survival at densities of 30 prawns/m3 or commercial shrimp size. Compared with the results of a previ- more. This pattern has also been found in wild shrimp popula- ous experiment (Ponce-Palafox et al. 2013), the greater size of tions at high densities (Perez-Casta´ neda˜ and Defeo 2005) the cage-reared prawn (27.0 g) at harvest in the treatment with Farming of M. tenellum in a cage-pond system is feasible, 6prawns/m3 was presumably due to both the improved water especially in terms of survival rates, which were found to be quality and the feeding strategies applied in this study. The mean similar to or better than those obtained in conventional earthen daily weight gain in this experiment was similar to that achieved ponds (Ponce-Palafox et al. 2013). These data suggest that stock- in the previous experiment (Ponce-Palafox et al. 2013) at den- ing juveniles in cages at high densities of 30 prawns/m3 or more sities of 12–14 prawns/m3. The most rapid growth and highest for the secondary nursery phase is be a good strategy for reduc- survival occurred at 6 prawns/m3. The decreases in growth found ing the costs of the nursery phase because cages can be placed at high density (24 prawns/m3) appeared to be due to the higher either within grow-out ponds or in other appropriate water bod- biomass of caged prawns in the ponds and density-dependent ies available on the farm. Further research should evaluate the effects (Nagarathinam et al. 2000; Cuvin-Aralar et al. 2007). effect of introducing additional surfaces and shelters inside the Macrobrachium tenellum juveniles cultured at an initial cages as well as the use of cages for grow-out culture. Cage stocking size of 0.18–0.35 g in ponds furnished with plastic sizes and other appropriate materials for nets and frames also canvas grew to 9.3 g in 66 d after stocking at 8 prawns/m3 (Vega- need to be considered to reduce the costs and allow applications Villasante et al. 2011). Juveniles of the same species reared at at a commercial scale. an initial stocking size of 1.9–2.5 g in earthen ponds grew to Lastly, we found that density affects the growth and sur- 19.3 g in 144 d after stocking at 14 prawns/m3 (Ponce-Palafox vival of M. tenellum. The effect is greater after 150 d of culture et al. 2013), results that are similar to those of this study. for densities greater than 12 prawns/m3. The choice of a suit- Menasveta and Piyatiratitvokul (1982) found that the growth able density represents a very important culture strategy for this of M. rosenbergii in cages was lower than that observed in freshwater prawn in terms of marketing because the animals at ponds and ditches. However, survival was significantly better all tested densities reached the individual market size. in cages. Marques et al. (2000) and Ang et al. (1992) obtained results similar to those of this study in nursery systems and REFERENCES in the grow-out phase of M. rosenbergii in cages, respectively. Ang, K. J., C. S. Komilus, and S. H. Cheah. 1992. Culture of Macrobrachium 3 They reported that a stocking density of 10 postlarvae/m pro- rosenbergii in cages. Page 127 in L. M. Chou, editor. The third Asian fish- vided better growth than 20 or 50 postlarvae/m3 although the eries forum: proceedings of the third Asian fisheries forum. Asian Fisheries final biomass was highest at the greatest stocking density. Al- Society, Selangor, Malaysia. though the mean final weight obtained in this experiment was Correia, E. S., J. A. Pereira, M. O. Apolinario, and A. Horowitz. 2002. Effect of comparable with that obtained by Ponce-Palafox et al. (2013) pond aging on natural food availability and growth of the freshwater prawn Macrobrachium rosenbergii. Aquaculture Engineering 26:61–69. 3 at 14 prawns/m in a semi-intensive system, the yields in the Cuvin-Aralar, M. L. A., E. V.Aralar, M. Laron, and R. Westley. 2007. Culture of current study were superior for all treatments. A comparison of Macrobrachium rosenbergii (De Man 1879) in experimental cages in a fresh- both studies suggests that this difference is due to differences in water eutrophic lake at different stocking densities. Aquaculture Research water quality and feeding. 38:288–294. An inverse relationship between stocking density and growth Espinosa-Chaurand, L., C. Flores-Zepeda, H. Nolasco-Soria, O. Carrillo-Farnes, and F. Vega-Villasante. 2012. Efecto del nivel proteico de la dieta sobre el has been reported for various species of Macrobrachium in desarrollo de juveniles de Macrobrachium tenellum (Smith, 1871). [Effect of

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 different culture systems (Marques et al. 2000, 2010, 2012). dietary protein level on the development of juvenile Macrobrachium tenellum Freshwater prawns of the genus Macrobrachium show territo- (Smith, 1871).] Revista MVZ Cordoba´ 17:3140–3146. rial behavior (New 2002). At high densities, even if food is Garc´ıa-Ulloa, M. G., L. A. Lopez-Aceves,´ J. T. Ponce-Palafox, H. Rodr´ıguez- ´ available, access to the food may be negatively affected due to Gonzalez, and J. L. Arredondo-Figueroa. 2008. Growth of fresh-water prawn Macrobrachium tenellum (Smith, 1871) juveniles fed isoproteic diets sub- stress caused by aggressive behavior (Wickins and Lee 2002). stituting fish meal by soya bean meal. Brazilian Archives of Biology and Comparisons among Macrobrachium species show that male Technology 51:57–65. M. rosenbergii cultured in cages reached 35.6 g in 150 d at Holthuis, L. B. 1980. Shrimps and prawns of the world. FAO (Food and Agri- densities of 15 prawns/m3 (Cuvin-Aralar et al. 2007), whereas culture Organization of the United Nations) Fisheries Synopsis 125(1). males of M. tenellum reached 27.8 g at 150 d at 12 prawns/m3. Lam, M. L., D. N. Long, and J. C. Micha. 2006. The effects of densities and feed types on the production of Macrobrachium rosenbergii (de Man) in the Cuvin-Aralar et al. (2007) reported an FCR of 2.1 for M. rosen- rotational rice–prawn system. Aquaculture Research 37:1297–1304. bergii reared at densities similar to those used in this study, in Marques, H. L. A., H. P. Barros, M. Mallasen, M. V. Boock, P. M. C. M. Valenti. which the mean final FCR was 1.9. These differences can be 2012. Influence of stocking densities in the nursery phase on the growth of partially explained because M. tenellum was fed twice daily and Macrobrachium amazonicum reared in net pens. Aquaculture 358/359:240– M. rosenbergii once. Stocking densities of M. tenellum up to 245. Marques, H. L., J. V. Lombardi, and M. V. Boock. 2000. Stocking densities for 3 6prawns/m in the experimental cages did not affect survival or nursery phase culture of the freshwater prawn Macrobrachium rosenbergii in performance during the growth phase. However, the results sug- cages. Aquaculture 187:127–132. EFFECT OF STOCKING DENSITY ON PRAWN SURVIVAL 169

Marques, H. L. A., J. V. Lombardi, M. Mallasen, H. P. Barros, and M. V. niques for the cultivation of native freshwater shrimp to the American Pa- Boock. 2010. Stocking densities in cage rearing of Amazon River prawn cific Macrobrachium tenellum (Smith, 1871) and M. americanum (Bate, (Macrobrachium amazonicum) during nursery phases. Aquaculture 307:201– 1968).] Pages 534–546 in I. de Blas, editor. Comunicacion´ cient´ıfica. Vir- 205. tual Iberoamerican Congress on Aquaculture, Zaragoza, Spain. Available: Menasveta, P., and S. Piyatiratitvokul. 1982. Effects of different culture systems www.revistaaquatic.com/civa2002/coms/pdf/67.pdf. (June 2013). on growth survival and production of the giant freshwater prawn Macro- Ponce-Palafox, J. T., R. Gonzalez-Salas,´ O. Romero-Cruz, I. Febrero-Toussaint, brachium rosenbergii. Pages 175–189 in M. B. New, editor. Giant prawn J. L. Arredondo-Figueroa, and H. Esparza-Leal. 2005. Enfermedades del farming. Elsevier, Amsterdam. camaron´ de agua dulce Macrobrachium tenellum y M. rosenbergii du- Montgomery, D. C. 1997. Design and analysis of experiments. Wiley, New rante el cultivo comercial en estanques rusticos,´ en empresas rurales [The York. freshwater shrimp Macrobrachium tenellum and M. rosenbergii diseases Murthy, H. S., R. Kumarswamy, K. J. Palaksha, H. R. Sujatha, and R. Shankar. in commercial culture ponds in rural communities.] Revista Electronica´ 2012. Effect of different types of shelters on survival and growth of giant de Veterinaria REDVET [online serial] 6(12): Available: www.veterinaria. freshwater prawn, Macrobrachium rosenbergii. Journal of Marine Science org/revistas/redvet/n121205/120507.pdf (July 2012). and Technology 20:153–157. Ponce-Palafox, J. T., F. L. Uriostegui, M. A. Ben´ıtez-Mandujano, S. Castillo- Nagarathinam, N., J. S. Sampath-Kumar, and V. Sundararaj. 2000. Influence Vargasmachuca, A. Ben´ıtez-Valles, J. Gomez-Gurrola,´ and J. L. Arredondo- of stocking density on growth, production and survival of Macrobrachium Figueroa. 2013. Comparative growth performance of male and female fresh- rosenbergii (de Man) in a monoculture grow-out pond. Indian Journal of water prawn Macrobrachium tenellum (Decapoda: Palaemonidae) cultured Fisheries 47:103–108. in tropical earthen ponds. International Journal of Fisheries and Aquaculture New, M. B. 2002. Farming freshwater prawns- a manual for the culture 5:26–28. of the giant river prawn (Macrobrachium rosenbergii). FAO (Food and Sagi, A., Z. Raanan,´ D. Cohen, and Y. Wax. 1986. Production of Mac- Agriculture Organization of the United Nations) Fisheries Technical Paper robrachium rosenbergii in monosex population: yield characteristics un- 428. der intensive monoculture conditions in cages. Aquaculture 51:265– New, M. B., and S. Singholka. 1985. Freshwater prawn farming. A manual 275. for the culture of Macrobrachium rosenbergii. FAO (Food and Agriculture Stanley, R. W., and L. B. Moore. 1983. The growth of Macrobrachium rosen- Organization of the United Nations) Fisheries Technical Paper 225. bergii fed commercial feeds in pond cages. Journal of the World Mariculture Perez-Casta´ neda,˜ R., and O. Defeo. 2005. Growth and mortality of transient Society 14(1–4):174–184. shrimp populations (Farfantepenaeus spp.) in a coastal lagoon of Mexico: Vega-Villasante, F., E. A. Mart´ınez-Lopez,´ L. D. Espinosa-Chaurand, M. C. role of the environment and density-dependence. ICES Journal of Marine Cortes-Lara,´ and H. Nolasco-Soria. 2011. Crecimiento y supervivencia del Science 62:14–24. langostino (Macrobrachium tenellum) en cultivos experimentales de verano Ponce-Palafox, J. T. 1997. Current and future trend in freshwater prawn culture yotono˜ en la costa tropical del Pac´ıfico Mexicano. [Growth and survival of in Mexico. Pages 252–254 in L. Creswell and Y. Harache, editors. Martinique prawn (Macrobrachium tenellum) in experimental cultures during summer ’97: island aquaculture and tropical aquaculture. European Aquaculture So- and autumn in the tropical Mexican Pacific coast.] Tropical and Subtropical ciety, Oostende, Belguim. Agroecosystems 14:581–588. Ponce-Palafox, J. T., F. Arana-Magallon,´ H. Cabanillas-Beltran,´ and H. Esparza- Weimin, M., and G. Xianping. 2002. Freshwater prawn culture in China: an Leal. 2002. Bases biologicas´ y tecnicas´ para el cultivo de los camarones overview. Aquaculture Asia 7:9–12. de agua dulce nativos del Pac´ıfico Americano Macrobrachium tenellum Wickins, J. F., and D. O. Lee. 2002. Crustacean farming: ranching and culture. (Smith, 1871) y M. americanum (Bate, 1968). [Biological aspects and tech- Blackwell Scientific Publications, Oxford, UK. Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 This article was downloaded by: [Department Of Fisheries] On: 15 June 2014, At: 20:18 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

North American Journal of Aquaculture Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/unaj20 Stress Responses in Pallid Sturgeon Following Three Simulated Hatchery Stressors Lucas R. Nelsona & Brian C. Smalla a Center for Fisheries, Aquaculture, and Aquatic Sciences, Southern Illinois University, Department of Animal Science, Food, and Nutrition, 1125 Lincoln Drive, Mail Code 6511, Carbondale, Illinois 62901, USA Published online: 15 Apr 2014.

To cite this article: Lucas R. Nelson & Brian C. Small (2014) Stress Responses in Pallid Sturgeon Following Three Simulated Hatchery Stressors, North American Journal of Aquaculture, 76:2, 170-177, DOI: 10.1080/15222055.2014.886651 To link to this article: http://dx.doi.org/10.1080/15222055.2014.886651

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 North American Journal of Aquaculture 76:170–177, 2014 C American Fisheries Society 2014 ISSN: 1522-2055 print / 1548-8454 online DOI: 10.1080/15222055.2014.886651

ARTICLE

Stress Responses in Pallid Sturgeon Following Three Simulated Hatchery Stressors

Lucas R. Nelson and Brian C. Small* Center for Fisheries, Aquaculture, and Aquatic Sciences, Southern Illinois University, Department of Animal Science, Food, and Nutrition, 1125 Lincoln Drive, Mail Code 6511, Carbondale, Illinois 62901, USA

Abstract Because sturgeon populations are in critical decline, many species are artificially propagated in hatcheries with the goal of increasing wild stocks. Capture, transport, handling, confinement, and environmental fluctuations are just a few of the possible stressors Pallid Sturgeon Scaphirhynchus albus can be exposed to in conservation hatcheries. The objectives of the present study were to elucidate the effects of acute exposure to high ammonia concentrations, low dissolved oxygen (DO) concentrations, and crowding on stress responses in Pallid Sturgeon, quantified by changes in plasma cortisol and glucose. Exposure to un-ionized ammonia-nitrogen concentrations of 0.6 mg/L for 24 h resulted in no significant (P > 0.05) change in plasma cortisol concentrations, suggesting these fish do not perceive the buildup of toxins. Exposure to 2 mg/L DO over a 0.5-h period resulted in a significant (P < 0.05) increase in plasma cortisol concentrations during the low DO challenge, with levels decreasing (P > 0.05) within 30 min of recovery and returning to prestress levels within 2.5 h of recovery. Plasma cortisol concentrations increased (P < 0.05) within 0.25 h of crowding and remained elevated after 0.5 h of crowding. Removal of the crowding stressor did not result (P > 0.05) in decreased circulating cortisol within 30 min. Regardless of the stressor, Pallid Sturgeon plasma glucose levels were unaffected (P > 0.05). As such, glucose does not appear to be a reliable indicator of stress in Pallid Sturgeon. Defining the effects of stress on Pallid Sturgeon physiology is a critical first step toward understanding how stress affects fish health and well being and optimization of propagation practices.

The Pallid Sturgeon Scaphirhynchus albus is a federally en- factors, in conjunction with relatively slow growth and long dangered North American species native to the lower Missis- time to sexual maturity, have severely impacted native popula-

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 sippi and entire Missouri river drainages (Bailey and Cross 1954; tions of this species (Keenlyne and Jenkins 1993; Snyder 2002; Kallemeyn 1983). Extensive dam construction has restricted Hrabik et al. 2007; USFWS 2007). In response, the U.S. Fish Pallid Sturgeon movement (Kynard et al. 2007; Braaten et al. and Wildlife Service placed the Pallid Sturgeon on the endan- 2008), and main channel river modiﬁcation for barge passage gered species list, and it is now protected throughout its native (Funk and Robinson 1974), along with increased organic loads, range (USFWS 1990). has led to the destruction of native habitat and spawning areas. Stocking of Pallid Sturgeon by state and federal agencies has There has been a substantial negative impact on water tem- increased drastically in the last 20 years (Secor et al. 2000, perature regimes, food availability, turbidity, and hydrography, 2002; Irelands et al. 2002; Smith et al. 2002), and studies which has led to poor recruitment (Keenlyne and Evenson 1989; have suggested that maintenance of Pallid Sturgeon popula- USFWS 1993; Pegg et al. 2003). Hybridization with the much tions is likely to be completely reliant on these efforts (Snyder more common Shovelnose Sturgeon S. platorynchus has also 2002; Hrabik et al. 2007; USFWS 2007). Relatively low calcu- negatively affected wild stocks (Keenlyne et al. 1994; Campton lated survival rates of released hatchery-reared Pallid Sturgeon et al. 2000; Tranah et al. 2004; Boley and Heist 2011). These (Steffensen et al. 2010) have led to the need to optimize hatchery

*Corresponding author: [email protected] Received December 12, 2013; accepted January 19, 2014

170 STRESS RESPONSES IN PALLID STURGEON 171

culture activities while reducing fish stress and maintaining fish ment, including Scaphirhynchus sturgeons (Barton et al. 2000; health. Webb et al. 2007), Beluga Sturgeon Huso Huso (Falahatkar et al. Stress factors like high ammonia concentrations, low dis- 2009), Green Sturgeon A. medirostris (Lankford et al. 2005), and solved oxygen (DO) concentrations, and crowding are common Paddlefish Polyodon spathula (Barton et al. 1998). in intensive culture settings, and the effects of these stressors With increasing interest in the propagation of Pallid Sturgeon have been extensively studied with cultured food fishes. How- for population restoration and reestablishment, it is important ever, the effects of these factors on hatchery-reared Pallid Stur- to understand negative physiological responses to environmen- geon have not been well studied. Ammonia occurs naturally in tal stressors associated with hatchery practices. Stress associated culture settings as a product of the breakdown of dietary proteins with high ammonia concentrations, low DO concentrations, and and amino acids (Sharma and Ahlert 1977). As the primary ni- crowding stress may impact the health and welfare of cultured trogenous waste product in a large majority of fish (Russo 1985; Pallid Sturgeon and negatively affect stocking efforts. As such, Jobling 1994), ammonia concentrations can rapidly increase in the objectives of the present study were to elucidate the ef- culture situations. High production densities along with the use fects of acute exposure to high ammonia concentrations, low of high protein diets to speed up growth can expedite system- DO concentrations, and crowding on stress responses in Pallid wide ammonia accumulation (Kaushik 1980; Tomasso 1994; Sturgeon. Thomas and Piedrahita 1998). Several studies have demonstrated that ammonia can reach lethal levels under intensive culture conditions for multiple species (Wise et al. 1989; Mazik et al. 1991; Ashe et al. 1996), and exposure of Channel Catfish METHODS Ictalurus punctatus to elevated levels of ammonia has been ob- Research fish.—All fish were handled in accordance with ap- served to elicit a cortisol stress response (Tomasso et al. 1981b; proval from the Institutional Animal Care and Use Committee at Small 2004). Southern Illinois University Carbondale under Protocol 10–034. Analyses of plasma cortisol and plasma glucose concentra- Age-0 Pallid Sturgeon were obtained from Gavin’s Point Na- tions are common methods of assessing physiological stress tional Fish Hatchery, South Dakota, and reared on the Southern responses in fish (Tomasso 1981a, 1981b; Barton 2002; Small Illinois University–Carbondale campus for 10 months prior to 2004; Weber et al. 2008). Increased plasma cortisol, an im- the study. Fish were stocked, according to the experimental de- portant stress hormone in most vertebrates, indicates a primary signs outlined below, into a recirculating culture system that con- response to stress exposure (Donaldson 1981; Wendelaar Bonga sisted of 76-L glass aquaria with central biological and mechan- 1997; Mommsen et al. 1999; Barton 2002). An increase in ical filtration. At the time of stocking, the average individual fish plasma glucose concentration signifies a secondary response weight was 84.6 ± 9 g (mean ± SE). The fish were allowed to to stress exposure as glycogen is mobilized and metabolized to acclimate to the system for 3 weeks. All fish were fed a commer- provide energy for body processes necessary to evade and adapt cially available fish diet (AquaMax Fry Starter 200, Purina Mills, to stress responses (Pickering 1981; Iwama et al. 1997). St. Louis, Missouri) to satiation once daily. All fish were fasted Dissolved oxygen levels are affected by numerous factors the day prior to sampling. A 12-h photoperiod was maintained like water temperature, dissolved solid concentrations, water throughout the study using overhead fluorescent lighting. Water turbulence, and stocking density (Boyd 1982; Fries et al. 1993; temperature was maintained at 21◦C and tanks were individually Abdalla and Romaire 1996), and exposure to low DO levels aerated throughout the study. Temperature and DO were mea- can elicit a physiological stress response in Channel Catfish sured using a model 550A digital meter (YSI, Yellow Springs,

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 (Small 2004). Acute exposure of juvenile Shortnose Sturgeon Ohio). Total ammonia nitrogen (TAN) and nitrite concentra- Acipenser brevirostrum to DO levels of 2.5 mg/L resulted in an tions were maintained at 0 mg/L by biofiltration, and water ex- 86% mortality rate (Jenkins et al. 1995), and chronic exposure changes were performed daily to maintain nitrate concentrations to DO levels below 5.3 mg/L can cause mortality and decreased below 10 mg/L. Total ammonia nitrogen was measured by direct overall growth in White Sturgeon A. transmontanus (Cech et al. Nesslerization (APHA 1989). Total ammonia nitrogen, pH, and 1984). temperature data were used to calculate un-ionized ammonia ni- Exposure to crowded conditions is another common stressor trogen (Emerson et al. 1975). Nitrite and nitrate were measured found in hatcheries. Tanks are often stocked in high densities using a Smart 3 Colorimeter (LaMotte Company, Chestertown, to optimize space (Boyd 1982), and haulers are overstocked Maryland). during transport for stocking (Fries et al. 1993). Short-term ef- Plasma collection and cortisol and glucose analyses.—The fects like physical damage, psychological stress, and reduced respective stress challenges (high ammonia, acute oxygen deple- feed intake can lead to long-term effects including reduced tion, and crowding) were conducted separately over three con- growth and increased disease susceptibility in multiple teleost secutive days. For all blood samples, fish were sedated to a “han- fish species (Iwama et al. 1992; Fevolden et al. 1993; Davis dleable” state (Summerfelt and Smith 1990) in a 70-mg/L aer- et al. 2002; Small and Bilodeau 2005). Stress responses have ated bath of tricaine methanesulfonate (MS-222; TRICAINE-S, also been observed in chondrosteans exposed to severe confine- Western Chemical, Ferndale, Washington) buffered with sodium 172 NELSON AND SMALL

bicarbonate. All blood samples were collected within 5 min low DO stressor (2.5 h following the return to pretreatment of sedation ventrally from the caudal vasculature using 1-mL concentrations). heparinized syringes, dispensed into 1.5-mL centrifuge tubes, Crowding stress.—Six tanks were each stocked with eight and stored on ice (<1 h) until centrifuged (3,000 × g,10min, fish. Water volume and total fish mass were determined for 4◦C) to separate the plasma. Plasma was extracted and stored at each tank. At t = 0 h (before application of the stressor), blood −20◦C until analyses. Plasma glucose concentrations were an- samples were collected from one fish from all six tanks. Fish that alyzed using the glucose hexokinase method (Pointe Scientific, were bled were not returned to experimental tanks. Individuals Canton, Michigan), and plasma cortisol concentrations were removed from each tank were weighed and subtracted from the evaluated with a time-resolved fluoroimmunoassay according total fish weight per tank, and all tanks were then crowded to a to Small and Davis (2002) and validated for Pallid Sturgeon density of 36 g fish/L of water using a plastic grating (crowder) (Fenn et al. 2013). similar to that used in Small (2004). After 0.25 h of crowding, Ammonia stress.—Nine tanks were stocked with four fish blood samples were collected from two fish per tank from three each and randomly assigned to one of three sample times: 0, 24, tanks. These tanks were not sampled again because of the change and 26 h after addition of ammonia, using three replicate tanks in density. At 0.5 h, blood samples were collected from two fish per time point. Before application of the stressor (time, 0 h), per tank from three different tanks, and the crowders were then blood was collected from two fish per tank from three tanks. removed from all tanks. At 0.75 and 1 h (15 and 30 min after the Individual tank water volumes of the remaining tanks were cal- removal of the crowding stressor), blood samples were again culated, water flow was stopped while maintaining aeration, and collected from two fish per tank from three tanks. ammonium chloride was added to obtain a final un-ionized am- Statistical analysis.—Significant differences in plasma cor- monia nitrogen concentration of 0.6 mg/L, based on the LC50 tisol and glucose concentrations were determined by ANOVA (concentration lethal to 50% of test organisms in a specified with treatment as the fixed effect and tank within treat- time) for Shortnose Sturgeon (Isely and Tomasso 1998). Fish ment as the random effect using SAS 9.2 statistical software were exposed to 0.6 mg/L ammonia nitrogen for 24 h. At 24 h, (SAS Institute, 2002–2008, Software Release 9.2, Cary, North blood was collected from two fish per tank from the three corre- Carolina). When significant differences were found using sponding tanks. Also at this time, water flow was restored to the ANOVA, pairwise contrasts were made using a Fisher’s least remaining three tanks and dry ammonia remover (PROLINE, squares difference (LSD) test to identify significant differences Pentair Aquatic Eco-Systems, Apopka, Florida) was introduced at the 5% level. Assumptions for homogeneity of variance and to rapidly remove and detoxify the ammonia. At 26 h (2 h after normality of the data were tested by examination of correla- the ammonia was removed), blood was again collected from tion between absolute residuals and predicted values and the two fish per tank from the three corresponding tanks. Ammonia Shapiro–Wilk test for normality. nitrogen concentration was verified at 24 h prior to the addition of PROLINE and did not differ from 0.6 mg/L. The ammonia nitrogen concentration prior to the start and at the conclusion of RESULTS the experiment was 0 mg/L. Ammonia Stress Dissolved oxygen stress.—Twelve tanks were stocked with Exposure to un-ionized ammonia concentrations of 0.6 mg/L four fish per tank and randomly assigned to one of four sample for 24 h resulted in no differences (F2, 6 = 0.67, P = 0.55) in times, 0, 30, 60, and 180 min, using three replicate tanks per time plasma glucose concentrations relative to prestress levels in Pal- = point. At t 0 h (before application of the stressor), blood was lid Sturgeon (Figure 1a). Furthermore, there were no significant

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 collected from two fish per tank in three tanks. Water flow and (F2, 6 = 0.15, P = 0.86) changes in plasma cortisol concen- aeration was stopped to the remaining nine tanks, and purified trations after the ammonia challenge relative to prestress levels nitrogen gas was bubbled through a manifold into these tanks to (Figure 1b). decrease DO levels to 2 mg/L, based on LC50 values estimated for Shortnose Sturgeon (Campbell and Goodman 2004). This Dissolved Oxygen Stress DO concentration was maintained for 30 min, and each tank Exposure to a DO concentration of 2 mg/L over a 0.5-h pe- was checked every 2 min and adjusted with air and nitrogen riod resulted in no differences (F3, 8 = 0.03, P = 0.99) in plasma flow into the tanks accordingly. At 30 min after the reduction glucose concentrations relative to prestress levels in Pallid Stur- of DO, blood was collected from two fish per tank from the geon (Figure 2a); however, plasma cortisol concentrations in- three corresponding tanks. At the same time, nitrogen flow to creased significantly (F3, 8 = 29.00, P ≤ 0.0001) during the low the tanks was stopped, and water and airflow were restored to DO challenge (Figure 2b). Circulating cortisol levels averaged return DO to the system-wide pretreatment concentration. At 20.3 ± 5.2 ng/mL after 0.5 h compared with 5.1 ± 0.3 ng/mL 60 min after the reduction of DO (30 min following the return before the stress. Plasma cortisol levels decreased significantly to pretreatment concentrations), blood was again collected from (t8 =−4.41, P = 0.0023) within 30 min of recovery (between two fish per tank from the three corresponding tanks. A final 0.5 and 1 h) and returned to prestress levels (t8 =−1.23, P = blood sample was collected 180 min after the initiation of the 0.26) within 2.5 h of recovery (3 h). STRESS RESPONSES IN PALLID STURGEON 173

a) 100 90 80 70 60 50 40 30 20 10

Plasma Glucose (mg/dL) 0 02426

b) 10 9 8 7 6 5 4 3 2

Plasma Cortisol (ng/mL) 1 0 02426 Time (h)

FIGURE 1. Mean ( ± SE) plasma concentrations of (a) glucose and (b) cortisol in Pallid Sturgeon in response to a 24-h exposure to 0.6 mg/L un-ionized ammonia nitrogen. The dashed line indicates the time the stressor was removed. No signiﬁcant differences (P > 0.05) were observed for either plasma component.

Crowding Stress FIGURE 2. Mean ( ± SE) plasma concentrations of (a) glucose and (b) cortisol Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 in Pallid Sturgeon in response to a 0.5-h exposure to 2 mg/L dissolved oxygen. No significant differences (F4, 13 = 1.02, P = 0.43) were The dashed line indicates the time the stressor was removed. No significant observed in circulating glucose concentrations as a result of differences (P > 0.05) were observed for plasma glucose. Different letters above confinement (Figure 3a). Plasma cortisol concentrations were bars indicate significant differences (P < 0.05) in plasma cortisol between time significantly (F4, 13 = 13.60, P ≤ 0.0001) affected by crowd- points. ing (Figure 3b). Cortisol increased (t13 =−5.97, P < 0.0001) within 0.25 h of crowding and remained elevated (t13 =−4.90, P < 0.0003) after 0.5 h of crowding. After 0.25 h of confine- role in fish population restoration, juvenile Pallid Sturgeon may ment, Pallid Sturgeon experienced an average 3.4-fold increase also be exposed to stressors under hatchery conditions. The data in plasma cortisol levels. Removal of the crowder did not result presented here are the first to demonstrate the acute effects of t =− P = ( 13 0.07, 0.95) in a decrease in circulating cortisol low DO on stress responses in Pallid Sturgeon and a lack of from 0.5 to 1 h (30 min after the crowder was removed). stress response in this species to an acute elevation in water ammonia concentrations. These data also corroborate previous DISCUSSION results on crowding-induced stress in Pallid Sturgeon (Barton For fish, abiotic and biotic stressors occur daily in the natural et al. 2000; Webb et al. 2007), but demonstrate a rapid cortisol environment. Because hatchery propagation plays an important increase within the first 15 min. 174 NELSON AND SMALL

Pallid Sturgeon exhibited an approximately fourfold increase in plasma cortisol concentrations following exposure to low DO for 30 min, although circulating glucose concentrations did not change. While this study is the first to demonstrate that low DO elicits an acute stress response in Pallid Sturgeon, hypoxia has been demonstrated to induce a robust cortisol stress response in cultured food fish, such as Channel Catfish (Tomasso et al. 1981a). Catfish also had a rapid cortisol response similar to that in Shovelnose Sturgeon; however, the catfish recovered within 30 min of return to normal DO concentrations (Tomasso et al. 1981a; Small 2004). Sturgeon appear to take somewhat longer torecoverascortisollevelsreturnedtobaselineby1hafter the return to normoxia. This may indicate a prolonged stress response in Pallid Sturgeon or a relatively longer half-life of cortisol in circulation. Ammonia toxicity has been demonstrated in Shortnose Stur- geon, but not in Scaphirhynchus species. The 96-h LC50 for un-ionized ammonia nitrogen is reported to be 0.58 ± 0.213 mg/mL (Fontenot et al. 1998) and was the basis for the concentration of ammonia exposure used in the present study. Ammonia is the primary metabolic waste product from nitrogen metabolism and in hatchery situations, where high densities of fish are often held, ammonia concentrations can become lethal (Wise et al. 1989; Tomasso 1994; Ashe et al. 1996). While sublethal levels of ammonia can reduce growth in Channel Catfish (Robinette 1976), no information exists for sturgeon beyond the LC50 results of Fontenot et al. (1998). Understanding how ammonia affects sturgeon physiology is critical to good hatchery management practices. A stress response was expected in Pallid sturgeon exposed to 0.6 mg/mL un-ionized ammonia nitrogen for 24 h; however, no response was observed. This is very different from what has been reported for Channel Catfish (Tomasso et al. 1981b; FIGURE 3. Mean ( ± SE) plasma concentrations of (a) glucose and (b) cortisol Small 2004). Small (2004) reported about a fivefold increase in Pallid Sturgeon in response to a 0.5-h crowding stressor. The dashed line in cortisol after 24 h. Even so, the cortisol response seen in indicates the time the stressor was removed. No significant differences (P > catfish exposed to physical stressors is typically much higher, 0.05) were observed for plasma glucose. Different letters above bars indicate increasing 10- to 20-fold over baseline concentrations (Small < significant differences (P 0.05) in plasma cortisol between time points. et al. 2008), indicating a less robust response to elevated ammo-

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 nia concentrations in Pallid Sturgeon. Tomasso et al. (1981b) Hypoxia and the sensitivity of sturgeons to low DO have stated their results suggest cortisol is released into circulation been studied in several of the Acipenser species. In Shortnose in response to a physiological dysfunction brought about by the Sturgeon, Jenkins et al. (1995) observed age-related differences toxins and not simply their presence in the environment. If so, in susceptibility, with juveniles being more susceptible to low the level and time of toxin built up in the present study may not DO. There was roughly a 60% increase in mortality among 64- have reached a critical point necessary to elicit physiological d-old fish exposed to 2.5 mg/L DO at 22.5◦C and 5% salinity dysfunction and a cortisol stress response. Although the con- compared with 100-d-old fish. More recently, Campbell et al. centration used was based on the previously published LC50, (2004) estimated a 24-h LC50 for young of the year Shortnose that LC50 was determined over 96 h and for a different species, Sturgeon to be between 2.2 and 3.1 mg/L across the range of Shortnose Sturgeon (Fontenot et al. 1998). temperature and salinity expected in their natural habitat during Physical stressors are among the most commonly studied spring and summer. These results were the basis for selecting for assessing physiological responsiveness in fish (Barton and 2.5 mg/L DO in the present study. Iwama 1991), and this holds true for the small amount of re- The present research focused on the stress responses of Pallid search conducted on sturgeon stress (Webb et al. 2007). As a Sturgeon associated with an acute exposure to low DO, simu- result of these studies, a typical stress response has been demon- lating what many would consider a common hatchery stressor. strated across many species of teleost fish, where cortisol serves STRESS RESPONSES IN PALLID STURGEON 175

as the primary stress hormone and is often the index used to (2000) reported that Pallid Sturgeon were unresponsive to being determine the magnitude of response to a perceived stressor held in the air in a net and as such, had little muscular activity (Billard et al. 1981; Barton and Iwama 1991). The typical cor- compared with their observations of salmonid and percid fishes. tisol stress response observed in teleost fishes appears to also Although not quantified in either study, fish in the present study occur in chondrosteans following confinement stress, albeit at a also exhibited little reaction to being crowded. Once the crowder much lower magnitude that in most teleosts (Barton et al. 2000; was in its final position, the fish remained still until the crow- Webb et al. 2007). As a result, cortisol has been identified as the der was removed. This apparent lethargy may also account for primary glucocorticoid synthesized and secreted in response to the relatively lower cortisol response in Pallid Sturgeon when a stressor in Pallid Sturgeon (Webb et al. 2007). A similar obser- compared with most teleost species, which generally exhibit a vation was made for Paddlefish (Barton et al. 1998), suggesting maximum cortisol response of 100–200 ng/mL (Barton 2002). cortisol is likely the major glucocorticoid common across all In brief, glucose does not appear to be a reliable indicator of chondrosteans. pallid Sturgeon stress. In the present study, Pallid Sturgeon were exposed to an acute crowding stressor for 30 min and the rate of increase and de- Conclusions crease in plasma cortisol monitored. A rapid increase in plasma Using plasma cortisol as a stress indicator, Pallid Sturgeon cortisol was observed within 15 min; however, plasma levels re- seem to be tolerant of elevated un-ionized ammonia nitrogen mained elevated for at least 1.5 h after the crowder was removed. levels of 0.6 mg/L for 24 h; however, further research should in- Barton et al. (2000) observed a similar increase in cortisol re- vestigate the effects of ammonia exposure on fish health and sur- sulting from a 6-h crowding stressor and observed a return to vival since these fish may not perceive the buildup of toxins until baseline cortisol levels by 6 h after removal of the stressor; cor- serious physiological damage has occurred. Acute exposure to tisol was not measured in the interim. Webb et al. (2007) did low DO and high fish density both elicited significant cortisol not measure cortisol beyond the stress period. Relative to sev- stress responses with relatively prolonged recovery times. More eral well-studied teleost species, such as sunshine bass (Striped research is needed to explore Pallid Sturgeon stress responses to Bass Morone saxatilis × White Bass M. chrysops) and Channel a wide range of biotic and abiotic stressors in order to optimize Catfish that recover from an acute physical stressor in less than culture efforts. Defining the effects of both acute and chronic 1 h (Davis and Small 2006), the return of circulating cortisol stressors on Pallid Sturgeon physiology is a critical first step concentrations is slower in Pallid Sturgeon. Again, this appears toward understanding how stress affects fish health and well to indicate a prolonged stress response in Pallid Sturgeon or being. potentially a longer half-life of cortisol in circulation. For sturgeon propagation, it is common to have higher stock- Regardless of the stressor the Pallid Sturgeon were exposed to ing densities in the hatchery than would be experienced in the in the present study, there was no change in plasma glucose con- wild and to handle or otherwise disturb fish during routine hus- centration. The lack of a glucose response to any of the stressors bandry procedures (Bayunova et al. 2002). Capture, transport, applied was not entirely unexpected. Although in teleost fishes handling, confinement, and environmental fluctuations are just elevated plasma cortisol levels following a stressful event lead a few of the possible stressors these fish can be exposed to to secondary and tertiary stress responses, including gluconeo- in conservation hatcheries. Propagation-related stressors in the genesis, this is not consistently observed in sturgeon. Bayunova hatchery affect fish health and have even been linked to repro- et al. (2002) observed a nearly 300% increase in plasma glucose ductive dysfunctions and poor gamete quality in wild-caught levels in Russian Sturgeon A. gueldenstaedtii 20 h following a sturgeon broodfish (Bayunova et al. 2002). Because sturgeon

Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 30-min air exposure, and Lankford et al. (2005) observed a simi- populations are in critical decline, many species are artificially lar response in Green Sturgeon 14 d into a chronic stress regime. propagated in hatcheries with the goal of increasing wild stocks. However, both Barton et al. (2000) and Webb et al. (2007) re- For this to be successful before it becomes too late, best man- ported no change in plasma glucose levels in Pallid Sturgeon agement practices must be developed that reduce the effects of following a confinement stressor lasting up to 12 and 24 h, re- common propagation stressors to ensure the release of healthy spectively. Fish density in the study by Webb et al. (2007) was fish. It is also important to assess the long-term effects of expo- approximately twice that used by Barton et al. (2000) and in the sure to these stressors on the health and fitness of stocked Pallid present study. Several factors may be affecting the glucose re- Sturgeon. sults reported for the various sturgeon species, including species or even genus differences and differences in experimental design, such as temperature, water quality, handling, fish size, and ACKNOWLEDGMENTS maturity. The authors sincerely thank Carlin Fenn and Elliott Kittel Another possible explanation why plasma glucose concen- for their assistance with data collection and fish husbandry and trations in Pallid Sturgeon were not significantly affected by any Julie Schroeter and Jacob Bledsoe for editing the manuscript. of the stressors administered is that relative energy needs during Gratitude is also extended to the Center for Fisheries, Aqua- or following an acute stress exposure are lower. Barton et al. culture, and Aquatic Sciences of Southern Illinois University 176 NELSON AND SMALL

Carbondale for supplying the facilities necessary for this re- Donaldson, E. M. 1981. The pituitary-interrenal axis as an indicator of stress in search. Funds for this work were provided, in part, through the fish. Pages 11–47 in A. D. Pickering, editor. Stress and fish. Academic Press, undergraduate assistantship (UGA) program of Southern Illinois New York. Emerson, K., R. C. Russo, R. E. Lund, and R. V. Thurston. 1975. Aqueous University. ammonia equilibrium calculations: effects of pH and temperature. Journal of the Fisheries Research Board of Canada 32:2379–2383. REFERENCES Falahatkar, B., S. Poursaeid, M. Shakoorian, and B. Barton. 2009. Responses Abdalla, A. A. F., and L. P. Romaire. 1996. Effects of timing and duration to handling and confinement stressors in juvenile Great Sturgeon Huso huso. of aeration on water quality and production of Channel Catfish. Journal of Journal of Fish Biology 75:784–796. Applied Aquaculture 6:1–10. Fenn, C. M., D. C. Glover, and B. C. Small. 2013. Efficacy of AQUI-S 20E APHA (American Public Health Association), American Water Works Associ- as a sedative for handling and cortisol suppression in Pallid Sturgeon. North ation, Water Pollution Control Federation. 1989. Standard methods for the American Journal of Fisheries Management 33:1172–1178 examination of water and wastewater, 17th edition. APHA, Washington, DC. Fevolden, S.-E., T. Refstie, and B. Gjerde. 1993. Genetic and phenotypic param- Ashe, D., J. R. Tomasso, A. G. Eversole, and T. I. J. Smith. 1996. Toxicity of eters for cortisol and glucose stress response in Atlantic Salmon and Rainbow ammonia to fingerling White Bass: effect of selected environments on uptake Trout. Aquaculture 118:205–216. dynamics. Progressive Fish-Culturist 58:277–280. Fontenot, Q. C., J. J. Isely, and J. R. Tomasso. 1998. Acute toxicity of ammo- Bailey, R. M., and F. B. Cross. 1954. River sturgeons of the American genus nia and nitrite to Shortnose Sturgeon fingerlings. Progressive Fish-Culturist Scaphirhynchus: characters, distribution, and synonymy. Papers of the Michi- 60:315–318. gan Academy of Science Arts and Letters 39:169–208. Fries, J. N., C. S. Berkhouse, J. C. Morrow, and G. J. Carmichael. 1993. Eval- Barton, B. A. 2002. Stress in fishes: a diversity of responses with particular ref- uation of an aeration system in a loaded fish-hauling tank. Progressive Fish- erence to changes in circulating corticosteroids. Integrative and Comparative Culturist 55:187–190. Biology 42:517–525. Funk, J. L., and J. W. Robinson. 1974. Changes in the channel of the lower Barton, B. A., H. Bollig, B. L. Hauskins, and C. R. Jansen. 2000. Juvenile Missouri River and effects on fish and wildlife. Missouri Department of pallid (Scaphirhynchus albus) and hybrid pallid × shovelnose (S. albus × Conservation, Aquatic Series 11, Jefferson City. platorynchus) sturgeons exhibit low physiological responses to acute handling Hrabik, R. A., D. P. Herzog, D. E. Ostendorf, and M. D. Petersen. 2007. Lar- and severe confinement. Comparative Biochemistry and Physiology Part A vae provide first evidence of successful reproduction by Pallid Sturgeon, Molecular and Integrative Physiology 126:125–134. Scaphirhynchus albus, in the Mississippi River. Journal of Applied Ichthyol- Barton, B. A., and G. K. Iwama. 1991. Physiological changes in fish from stress ogy 23:436–443. in aquaculture with emphasis on the response and effects of corticosteroids. Irelands, S. C., R. C. P. Beamesderfer, V. L. Paragamian, V. D. Wakkinen, Annual Review of Fish Diseases 1:3–26. and J. T. Siple. 2002. Success of hatchery-reared juvenile White Sturgeon Barton, B. A., A. B. Rahn, G. Feist, H. Bollig, and C. B. Schreck. 1998. Phys- (Acipenser transmontanus) following release in the Kootenai River, Idaho, iological stress responses of the freshwater chondrostean Paddlefish (Polyo- USA. Journal of Applied Ichthyology 18:642–650. don spathula) to acute physical disturbances. Comparative Biochemistry and Isely, J. J., and J. R. Tomasso. 1998. Acute toxicity of ammonia and nitrite to Physiology Part A 120:355–363. Shortnose Sturgeon fingerlings. Progressive Fish-Culturist 60:315–318. Bayunova, L., I. Barannikova, and T. Semenkova. 2002. Sturgeon stress reac- Iwama, G. K., J. C. McGeer, and N. J. Bernier. 1992. The effects of stock tions in aquaculture. Journal of Applied Ichthyology 18:397–404. and rearing density on the stress response in juvenile Coho Salmon (On- Billard, R., C. Bry, and C. Gillet. 1981. Stress, environment and reproduction corhynchus kisutch). ICES Marine Science Symposia 194:67–83. in teleost fish. Pages 185–208 in A. D. Pickering, editor. Stress and fish. Iwama, G. K., A. D. Pickering, J. P. Sumpter, and C. B. Schreck, editors. Academic Press, New York. 1997. Fish stress and health in aquaculture. Society for Experimental Biology Boley, R. M., and E. J. Heist. 2011. Larval surveys indicate low levels of Seminar Series 62. endangered Pallid Sturgeon reproduction in the middle Mississippi River. Jenkins, W. E., T. Smith, L. Heyward, and D. M. Knott. 1995. Tolerance of Transactions of the American Fisheries Society 140:604–1612. Shortnose Sturgeon, Acipenser brevirostrum, juveniles to different salinity Boyd, C. 1982. Managing water quality in Channel Catfish. Journal of Soil and and dissolved oxygen concentrations. Proceedings of the Annual Confer- Water Conservation 37:207–209. ence Southeastern Association of Fish and Wildlife Agencies 47(1993):476– Braaten, P. J., D. B. Fuller, L. D. Holte, R. D. Lott, W. Viste, T. F. Brandt, and 484. R. G. Legare. 2008. Drift dynamics of larval Pallid Sturgeon and Shovelnose Jobling, M. J. 1994. Fish bioenergetics. Chapman and Hall, London. Downloaded by [Department Of Fisheries] at 20:18 15 June 2014 Sturgeon in a natural side channel of the upper Missouri River, Montana. Kallemeyn, L. 1983. Status of the Pallid Sturgeon Scaphirhynchus albus. North American Journal of Fisheries Management 28:808–826. Fisheries 8(1):3–9. Campbell, J. G., and L. R. Goodman. 2004. Acute sensitivity of juvenile Short- Kaushik, S. J. 1980. Influence of nutritional status on the daily patterns of nose Sturgeon to low dissolved oxygen concentrations. Transactions of the nitrogen excretion in the carp (Cyprinus carpio L.) and the Rainbow Trout American Fisheries Society 133:772–776. (Salmo gairdneri). Reproduction Nutrition Development 20:1751–1765. Campton, D. E., A. L. Bass, F. A. Chapman, and B. W. Bowman. 2000. Genetic Keenlyne, K. D., and P. D. Evenson. 1989. Standard and relative weight for the distinction of Pallid, Shovelnose and Alabama sturgeon: emerging species Pallid Sturgeon (Scaphirhynchus albus). Proceedings of the South Dakota and the U.S. Endangered Species Act. Conservation Genetics 1:17–32. Academy of Sciences 2:41–49. Cech, J. J., S. J. Mitchell, and T. E. Wragg. 1984. Comparative growth of Keenlyne, K. D., and L. G. Jenkins. 1993. Age at sexual maturity of the Pallid juvenile White Sturgeon and Striped Bass: effects of temperature and hypoxia. Sturgeon. Transactions of the American Fisheries Society 122:393–396. Estuaries 7:12–18. Keenlyne, K. D., C. J. Henry, A. Tews, and P. Clancey. 1994. Morphometric Davis, K. B., B. R. Griffin, and W. L. Gray. 2002. Effect of handling stress comparisons of upper Missouri River sturgeons. Transactions of the American on susceptibility of Channel Catfish Ictalurus punctatus to Ichthyophthirius Fisheries Society 123:779–785. multifiliis and Channel Catfish virus infection. Aquaculture 214:55–66. Kynard, B., E. Parker, D. Pugh, and T. Parker. 2007. Use of laboratory studies Davis, K. B., and B. C. Small. 2006. Rates of cortisol increase and decrease in to develop a dispersal model for Missouri River Pallid Sturgeon early life Channel Catfish and sunshine bass exposed to an acute confinement stressor. intervals. Journal of Applied Ichthyology 23:365–374. Comparative Biochemistry and Physiology Part C Toxicology and Pharma- Lankford, S. E., T. E. Adams, R. A. Miller, and J. J. Cech Jr. 2005. The cost cology 143:134–139. of chronic stress: impacts of a nonhabituating stress response on metabolic STRESS RESPONSES IN PALLID STURGEON 177

variables and swimming performance in sturgeon. Physiological and Bio- P. J., Anders, D. H. Secor, and D. A. Dixon, editors. Biology, management, chemical Zoology 78:599–609. and protection of North American sturgeon. American Fisheries Society, Mazik, P. M., M. L. Hinman, D. A. Winkelmann, S. J. Klaine, B. A. Simco, and Symposium 28, Bethesda, Maryland. N. C. Parker. 1991. Influence of nitrite and chloride concentrations on survival Snyder, D. E. 2002. Pallid and Shovelnose sturgeon larvae – morphological and hematological profiles of Striped Bass. Transactions of the American description and identification. Journal of Applied Ichthyology 18:240–265. Fisheries Society 120:247–254. Steffensen, K. D., L. A. Powell, and J. D. Koch. 2010. Assessment of hatchery- Mommsen, T. P., M. M. Vijayan, and T. W. Moon. 1999. Cortisol in teleosts: reared Pallid Sturgeon survival in the lower Missouri River. North American dynamics, mechanisms of action, and metabolic regulation. Reviews in Fish Journal of Fisheries Management 30:671–678. Biology and Fisheries 9:211–268. Summerfelt, R. C., and L. S. Smith. 1990. Anesthesia surgery and related Pegg, M. A., C. L. Pierce, and A. Roy. 2003. Hydrological alteration along techniques. Pages 213–272 in C. B. Schreck and P.B. Moyle, editors. Methods the Missouri River basin: a time series approach. Journal of Aquatic Science for fish biology. American Fisheries Society, Bethesda, Maryland. 65:63–72. Thomas, S. L., and R. H. Piedrahita. 1998. Apparent ammonia-nitrogen pro- Pickering, A. D. editor. 1981. Stress and fish. Academic Press, New York. duction rates of White Sturgeon (Acipenser transmontanus) in commercial Robinette, H. R. 1976. Effect of selected sublethal levels of ammonia on the aquaculture systems. Aquaculture Engineering 17:45–55. growth of Channel Catfish (Ictalurus punctatus). Progressive Fish-Culturist Tomasso, J. R. 1994. Toxicity of nitrogenous wastes to aquaculture animals. 38:26–29. Reviews in Fisheries Science 2:291–314. Russo, R. C. 1985. Ammonia, nitrite, and nitrate. Pages 455–471 in G. M. Rand Tomasso, J. R., K. B. Davis, and N. C. Parker. 1981a. Plasma corticosteroid and S. R. Petrocelli, editors. Fundamentals of aquatic toxicology. Hemisphere, dynamics in Channel Catfish (Ictalurus punctatus) during and after oxygen New York. depletion. Journal of Fish Biology 18:519–526. Secor, D. H., P. J. Anders, W. Van Winkle, and D. A. Dixon. 2002. Can we study Tomasso, J. R., K. B. Davis, and B. A. Simco. 1981b. Plasma corticosteroid sturgeons to extinction? What we do and don’t know about the conservation dynamics in Channel Catfish (Ictalurus punctatus) exposed to ammonia and of North American sturgeons. Pages 3–10 in W. Van Winkle, P. J. Anders, nitrite. Canadian Journal of Fisheries and Aquatic Sciences 38:1106–1112. D. H. Secor, and D. A. Dixon, editors. Biology, management, and protection Tranah, G., D. E. Campton, and B. May. 2004 genetic evidence for hybridization of North American sturgeon. American Fisheries Society, Symposium 28, of Pallid and Shovelnose sturgeon. Journal of Heredity 95:474–480. Bethesda, Maryland. USFWS (U.S. Fish and Wildlife Service). 1990. Endangered and threatened Secor, D. H., V. Arevyev, A. Nikolaev, and A. Sharov. 2000. Restoration of wildlife and plants: determinations of endangered status for the Pallid Stur- sturgeons: lessons from the Caspian Sea sturgeon ranching program. Fish geon. Federal Register 55:173(6 September 1990):36641–36647. and Fisheries 1:215–230. USFWS (U.S. Fish and Wildlife Service). 1993. Recovery plan for the Pallid Sharma, B., and R. C. Ahlert. 1977. Nitrification and nitrogen removal. Water Sturgeon (Scaphirhynchus albus). USFWS, Bismarck, North Dakota. Resources 11:897–925. USFWS (U.S. Fish and Wildlife Service). 2007. Pallid Sturgeon Small, B. C. 2004. Effect of isoeugenol sedation on plasma cortisol, glucose, (Scaphirhynchus albus) 5-year review: summary and evaluation. USFWS, and lactate dynamics in Channel Catfish Ictalurus punctatus exposed to three Billings, Montana. stressors. Aquaculture 238:469–481. Webb, M. A., J. A. Allert, K. M. Kappenman, J. Marcos, G. W. Feist, C. B. Small, B. C., and A. L. Bilodeau. 2005. Effects of cortisol and stress on Channel Schreck, and C. H. Shackleton. 2007. Identification of plasma glucocorticoids Catfish (Ictalurus punctatus) pathogen susceptibility and lysozyme activity in Pallid Sturgeon in response to stress. General and Comparative Endocrinol- following exposure to Edwardsiella ictaluri. General and Comparative En- ogy 154:98–104. docrinology 142:256–262. Weber G. M., R. L. Vallejo, S. E. Lankford, J. T. Silverstein, and T. J. Welch. Small, B. C., and K. B. Davis. 2002. Validation of a time-resolved flouroim- 2008. Cortisol response to a crowding stress: heritability and association with munoassay for measuring plasma cortisol in Channel Catfish (Ictalurus punc- disease resistance to Yersinia ruckeri in Rainbow Trout. North American tatus). Journal of the World Aquaculture Society 33:184–187. Journal of Aquaculture 70:425–433. Small, B. C., K. B. Davis, and B. C. Peterson. 2008. Elucidating the effects of Wendelaar Bonga, S. E. 1997. The stress response in fish. Physiological Reviews cortisol and stress on economically important traits in Channel Catfish. North 77:591–625. American Journal of Aquaculture 70:223–235. Wise, D. J., C. R. Weireich, and J. R. Tomasso. 1989. Toxicity of ammonia to Smith, T. I. J., M. C. Collins, W. C. Post, and J. W. McCord. 2002. Stock en- Red Drum Sciaenops ocellatus fingerlings with information on uptake and hancement of Shortnose Sturgeon: a case study. Pages 31–44 in Van Winkle, depuration. Journal of the Wildlife Society 20:188–192. Downloaded by [Department Of Fisheries] at 20:18 15 June 2014