- Email: [email protected]
Effects of (Margaritifera margaritifera) glochidial infection on performance of tank-reared Atlantic salmon (Salmo salar)
Aquaculture 256 (2006) 74 – 79 www.elsevier.com/locate/aqua-online
Effects of (Margaritifera margaritifera) glochidial infection on performance of tank-reared Atlantic salmon (Salmo salar) James W. Treasurer a,⁎, Lee C. Hastie b , Dougie Hunter c , Fiona Duncan c , Claire M. Treasurer d a
b
Viking Ardtoe Marine Laboratory, Acharacle, Argyll PH36 4LD, UK University of Aberdeen, Department of Zoology, Tillydrone Avenue, Aberdeen AB24 2TZ, UK c Marine Harvest, Blar Mhor Industrial Estate, Fort William PH33 7PT, UK d 9 Zetland Avenue, Fort William PH33 6LL, UK Received 6 January 2006; accepted 7 February 2006
Abstract Experimental infection of pearl mussel glochidia on the performance of Atlantic salmon hosts was examined. Infection intensity was 1392 ± 641 SD glochidia per fish six weeks after challenge but declined significantly 15 weeks after infection to means of 50 and 112 fish− 1 in two trial tanks and, by 6 months numbers were b 1 fish− 1. This loss was attributed to fungal treatments with a combination of malachite green which was still the authorised treatment during the study in 2001 and formalin. The weight of infected fish was significantly lower than controls at 15 weeks but this was not significant after 1 year. The condition factor of infected and naïve fish was not significantly different. Lactate was measured as a possible indicator of stress in infected fish but there was no significant difference with controls. Fish that had been previously infected with glochidia were re-infected in their second year and comparison made with infection of naïve fish to determine whether glochidial infection elicits an immune response. Although there was no significant difference in glochidial numbers in both groups at around 9000 glochidia host− 1 3 weeks after infection, numbers of glochidia in naïve fish did not change to 15 weeks after challenge, whereas there was a significant reduction in previously infected fish to 116 fish− 1. Various treatments were used to provoke closure of the glochidial valves. It is concluded that infection of salmon with glochidia levels in the current study had no significant effect on salmon performance, condition and stress as measured by assay of lactate. © 2006 Elsevier B.V. All rights reserved. Keywords: Pearl mussel; Margaritifera margaritifera; Glochidia; Salmonid; Salmo salar
1. Introduction The endangered freshwater pearl mussel, Margaritifera margaritifera, has a dispersive parasitic larval (glochidial) stage and large numbers are shed by adult ⁎ Corresponding author. Tel.: +44 1397 875000; fax: +44 1397 875001. E-mail address: [email protected] (J.W. Treasurer). 0044-8486/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2006.02.031
mussels in mid-summer. These attach to the gills of salmonids where they encyst and remain for c. 12 months before excysting and dropping to the river bed where they settle (Hastie and Young, 2001). Compared to other parasitic infections of salmon and trout, glochidiosis is of little economic importance. However, in certain situations, pearl mussels can cause problems for fish farmers (Hastie and Young, 2003). Most studies of the mussel– host relationship have focussed on distribution and
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
ecology (Young and Williams, 1984a,b; Hastie and Young, 2001, 2003), effects of previous challenges on glochidial survival (Young et al., 1987), and comparative susceptibility of different fish hosts to infection (Meyers and Millemann, 1977; Bauer, 1987). Relatively little attention has focussed on the impacts of glochidia on the performance and survival of the host (Treasurer and Turnbull, 2000). While some authorities have suggested that glochidiosis has little effect on host growth and survival, others have indicated that high levels of infection in farmed situations may induce mortalities and anglers and fishery managers have implied that pearl mussels may be detrimental to endangered salmonid stocks. The impact of pearl mussel glochidia on their host is difficult to measure and monitor in the field. In the present study the potential impact of infection was assessed by the artificial infection of cultured Atlantic salmon and by measurement of juvenile salmon in replicated tanks with comparison with naive fish. Although immune responses of fish to bacterial and viral infection have been frequently documented, measurement of host immune response to parasitic infection has been limited (Meyers et al., 1980; Bauer and Vogel, 1987). It has been suggested that older fish previously challenged with glochidia have lower infection intensity and, although it has been inferred that fish mount an immune response, this has not been verified (Bauer, 1987; Hastie and Young, 2001). Therefore previously challenged fish were re-infected experimentally, and compared with infection of naïve control fish, to assess if an immune response could be elicited. 2. Methods 2.1. Infection Uninfected juvenile salmon of 5.6 ± 1.5 SD g mean weight were transferred from a salmon hatchery in Lochaber in June 2001 to an outdoor experimental hatchery. 300 fish were infected by cohabiting with five spawning pearl mussels checked for spawning condition by carefully opening their shell valves with special opening tongs, and checking for the presence of glochidia in the modified gill structures of the female mussels (Young and Williams, 1984a). Salmon were cohabited in a tank overnight with pearl mussels to permit natural infection. The fish were evenly divided between two 1 m2 × 40 cm glassfibre tanks. Flow rate was 10l min− 1, and water temperatures were recorded daily and were in the range 3 to 17 °C during the first year of the infection study. A total of 300 non-infected control fish were maintained in two other adjacent tanks. Fish were fed twice daily to satiation with 2 mm Trouw diet.
75
Mortalities were removed daily and recorded. Fifty fish from each tank were anaesthetised in 20 mg l− 1 benzocaine and measured two weeks after infection and then at 6, 10, 15, 25, 30 and 36 weeks to fork length mm and weight to 0.01 g. Ten of these fish were killed, five from each tank, by overdose in benzocaine and gills from the left side of the fish excised and examined at ×10 magnification and glochidia counted on the first gill arch. As previous studies have found no significant difference in glochidia numbers between gill arches (Treasurer and Turnbull, 2000) the total number of glochidia per fish was calculated by a multiplier of 8. Glochidia were measured at ×40 magnification laterally across the carapace using an ocular micrometer. Blood samples were collected from the caudal fin from infected and control fish 15 weeks after commencement of the experiment and samples tested for lactate as a measurement of stress. Plasma lactate was determined enzymatically using Sigma Diagnostic kits (Sigma; Proc. no. 735). The condition of salmon was measured as (body mass g 100/length3). A sample of fish used in the first infection experiment was also examined at the donor hatchery from the main stock 5 months after natural infection in the hatchery. Twenty fish were measured and gills examined as above. 2.2. Re-infection experiment One year post-infection, the remaining fish in both trial and control groups were infected with glochidia on 15 July 2002 by experimental challenge. Fish were sampled 6 and 12 weeks after infection and numbers of glochidia counted on all gills in 5 fish from each treatment. 2.3. Methods for promoting closure of glochidia An attempt was made to prevent infection of hatchery reared salmon by glochidia by causing closure of the glochidia before they passed through the hatchery. Collected spat of 0.5 ml volume were introduced to 5 ml capacity six well plates containing test solutions. Spat were assessed for closure of valves, immediately and after exposure to test solutions for 5 min. 10 ml of unchlorinated water were added to the mussel suspension at this point and mussel viability was assessed after 24 h. The treatments were: NaCl at 200, 2000, 5000, 10,000 and 50,000 mg l− 1, hydrogen peroxide at 17,500, 35,000, 175,000 and 350,000 mg l− 1, formaldehyde at 50, 100, 200, 400, 1000, 2000, and 20,000 mg l− 1, and copper sulphate at 250, 500, 5000, 50,000 mg l− 1.
76
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79 180
Data on infection intensity, growth and lactate values were tested for homogeneity of variances by Bartlett's test. As variances were mainly heterogeneous statistical comparison of treatments was carried out with the nonparametric Mann Whitney and Kruskal–Wallis tests. Differences were taken as significant when P b 0.05. The variances of condition factor values were highly variable and data were transformed to common logarithm values prior to ANOVA. 3. Results
Fork length mm
2.4. Statistical analyses
160 140 120 100
Infected T1 Infected T2 Control T3 Control T4
80 60 0
1
2
3
4
5
6
7
8
9
10
Months post infection
Fig. 2. A comparison of growth in length of infected (tanks T1 and T2) and control (T3 and T4) juvenile salmon. N = 50 fish per tank at each sampling point.
3.1. Course of infection and mussel dimensions Two weeks after infection mean glochidia numbers were 818 ± 377 SD per fish (Fig. 1) with a wide range in abundance, from 352 to 1184 glochidia fish − 1 . Glochidia numbers were initially stable with 1392 ± 641 SD glochidia fish− 1 6 weeks post-infection and 1385 ± 553 glochidia fish− 1 10 weeks after infection. However, glochidia numbers dropped significantly in October, at 15 weeks post-infection, to 50 ± 45 glochidia fish− 1 in one tank and 112 ± 197 glochidia fish− 1 in the other tank (Z scores of 4.39 and 4.35, P b 0.001, respectively, Mann Whitney) (Fig. 1). By 14 January, 25 weeks after infection, glochidia numbers were negligible, zero in 10 fish examined from tank 1 and 0.64 ± 2.32 glochidia fish− 1 in the second tank. No glochidia were seen in fish examined in February and April. Mean glochidium diameter was 296 ± 13 μm on infection, 348 ± 13 μm by 10 weeks post-infection and 365 ± 28 μm at 15 weeks post-infection. Only limited data were available from naturally infected fish from the salmon hatchery in 2000.
N glochidia fish-1
2500
Tank 1 Tank 2
2000 1500 1000 500 0 0
1
2
3
4
5
6
7
8
9
10
Months post infection
Fig. 1. Infection intensity of glochidia in two tanks of artificially challenged salmon juveniles. Fish were experimentally infected in July 2001. N = 10 fish at each sampling point for each treatment. Bars, indicate SD, are shown in only one tank from each treatment to enhance clarity.
Prevalence was 40% and intensity of infection was only 1.7 glochidia five months after natural infection. 3.2. Host impacts of infection 3.2.1. Length Mean length of newly infected salmon on 27 July 2001 was 76.8 ± 6.4 mm fork length (FL) and not significantly different from uninfected fish, 76.2 ± 5.0 mm (Mann Whitney test Z = 0.66, P = 0.51). By 15 weeks post-infection the infected fish were smaller than the controls (Fig. 2), namely 113.4 ± 14.7 and 114.0 ± 14.6 mm compared with 115.6 ± 13.8 and 121.4 ± 14.4 mm in uninfected fish, and this was significant (Kruskal–Wallis, H = 17.49, P = 0.00056). Following 6 months post-infection the salmon in one tank of each treatment were slightly longer than the other (H = 18.3, P = 0.000434), but by 9 months post-infection there was no significant difference (Z = 1.57, P = 0.1172) between infected and control fish, 157 ± 9 and 154 ± 9 mm, respectively. 3.2.2. Weight The pattern in weights mirrored that of length with starting weights of 5.6 ± 1.5 g in infected fish and 5.5 ± 1.3 g in uninfected controls (Mann Whitney, Z = 0.46, P = 0.64) (Fig. 3). There was no significant difference in weight after 6 and 10 weeks post-infection and by 15 weeks the weight of infected fish was slightly lower at 18.1 ± 5.8 and 18.4 ± 6 g compared with 19.1 ± 5.9 and 22.8 ± 6.7 g in controls and this was significant (Kruskal–Wallis, H = 23.78, P = 0.000028). By six and seven months weights were not significantly different between tanks and at nine months post-infection infected fish were 36.9 ± 6.4 g compared with 35.4 ± 6.7 g in controls with no significant difference in weight (Z = 1.51, P = 0.13).
3.2.3. Condition factor There was no significant difference (P N 0.05) in condition factor (K) between infected and trial fish on any of the sampling dates. However, on 22 February control fish in one tank and infected fish in another had higher condition factors. However, when data from the tanks in each treatment were combined, these were found to be highly variable, and data were log transformed prior to statistical analysis. Although K in the control group was higher, this was not significant. As the fish had been subject to Saprolegnia infection, and required treatment, it was concluded that differences between tanks were due to fungal infection rather than glochidia, particularly as the majority of glochidia had been shed by this time.
N glochidia fish-1
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
77
15000 10000 5000 0 0
5
10
15
20
Weeks post infection Previously infected
Naïve fish
Fig. 4. The pattern of glochidia infection in naïve and previously infected salmon in their second year. Fish infected in July 2002. Bars indicate SD.
There was no significant difference in lactate levels between fish infected with glochidia (6.7 ± 3.4 SD) and uninfected (7.3 ± 2.9 SD) fish (n = 5 each sample, P = 0.15).
glochidia fish− 1 in naïve fish, and this was not significant (Mann Whitney Z = 0.2268, P = 0.82) (Fig. 4). Seven weeks post-infection glochidia numbers remained high in the naïve fish, 7407 ± 3286 glochidia fish− 1, but glochidial intensity had significantly declined to 5260 ± 1793 glochidia fish− 1 in the previously infected fish (Z = 2.65, P = 0.008), although the glochidia numbers between treatment groups were not significantly different (Z = 0.36, P = 0.17). By 16 November 2002, when the remaining fish were sacrificed, 15 weeks after challenge, glochidia numbers in the previously infected group had declined markedly to 116 ± 141 glochidia fish− 1 while infection intensity was unchanged in the naïve fish at 8120 ± 5355 glochidia fish− 1. Neither mean length, 228 ± 17 v. 226 ± 12 mm (Z = 0.23, P = 0.82), nor weight, 126 ± 33 v. 121 ± 20 g (Z = 0.27, P = 0.79), of re-infected and naïve fish were significantly different.
3.5. Re-infection experiment
3.6. Preventing infection
The variation in infection intensity was high between fish in each treatment group. Mean intensity 3 weeks after infection was 9096 ± 3046 SD glochidia fish− 1 in previously infected fish compared with 8932 ± 3719
Various salt and other solutions were used to try to elicit closing of shell valves, thus preventing attachment to fish in hatcheries. All solutions induced closure of valves within 5 to 6 min, with the exception of formaldehyde where they remained open. Glochidia exposed to hydrogen peroxide, copper sulphate and formaldehyde did not appear viable after 24 h but most glochidia exposed to NaCl solutions from 200 to 2000 mg l− 1 were still viable and showed movement in the shell valves.
3.3. Mortalities Mortalities over the one year study period were compared between two tanks of infected fish and controls. There were no significant differences between the numbers of mortalities in any of the tanks to February 2002. Thereafter there was a large mortality affecting the tanks, due to pump failure, and further comparison of survival was not possible. 3.4. Lactate measures
Mean weight g
50 40 30 20
Infected T1 Infected T2 Control T3 Control T4
10 0
0
2
4
6
8
10
Months post infection
Fig. 3. Comparison of growth in weight of infected (T1 and T2) and non-infected (T3 and T4) juvenile salmon.
4. Discussion 4.1. Pattern of infection The overall prevalence and mean glochidial loads dropped dramatically by 15 weeks post-infection, from 1393 ± 641 glochidia fish− 1 to 0.64 ± 2.32. This
78
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
decline following infection has been reported in other studies (Hastie and Young, 2001) and during the present study a loss of N99% to January was observed. This feature is typical, and has been attributed to incomplete attachment (Bruno et al., 1988) or a specific host response (Hastie and Young, 2001). The extent of the host response varies between salmonid species (Meyers et al., 1980). The decline observed during the present study was high, and a complete loss of glochidia had occurred 6 months after infection, so other factors may have been responsible. While it has been reported that treatments with therapeutants such as copper sulphate, salt and Roccal are ineffective in controlling glochidial infection in farmed salmonids (Bruno et al., 1988), the decline in the present study suggests that therapeutic treatments may have affected glochidia numbers. Fish were treated monthly for Ichthyobodo infection with 200 mg l − 1 formalin and concurrently with malachite green. The latter is now discontinued from use with food chain species but during the first year of the study in 2001 it was the recognised treatment for Saprolegnia. In the last year of the study when malachite green was withdrawn from use fish were therefore treated for fungus with Pyceze (50% w/v bronopol, Novartis Animal Vaccines) and this appeared to have no effect on glochidial numbers, and this may therefore be used as an alternative treatment for fish used for collection of glochidia for restocking purposes. 4.2. Impact on hosts This study shows that a relatively high level of infection of 1393 glochidia per 5 g salmon juvenile had no significant effect on salmon survival and only a small transient effect on growth in length and weight. While the mean weight of uninfected fish initially showed a trend to a higher weight than infected fish the fish grew at the same rate following loss of glochidia to three months after infection. In a previous study of infected farmed salmon with loads of up to 480 glochidia fish− 1 , no significant differences in survival, growth and condition factor of infected and non-infected salmon smolts stocked in seawater cages were observed (Treasurer and Turnbull, 2000), indicating no measurable effects of glochidia on the host at this level of infection. Rather than a threat to endangered salmonid stocks it is likely that pearl mussels improve water quality in rivers and clean gravel beds and this may benefit salmon spawning and nursery habitat (Ziuganov et al., 1994; Hastie and Cosgrove, 2001).
4.3. Immune response This study demonstrates that there is an immune response as, in previously infected salmon, glochidia numbers declined to low numbers 15 weeks after infection, although numbers in naïve fish did not decline significantly. It is therefore possible that a vaccine could be developed to minimise the effects of infection with glochidia in the farm setting. It has been shown that transformation success in a variety of unionid mussels was significantly lower on previously primed fish (Dodd et al., 2005). Further, primed fish showed binding of serum antibodies with glochidia proteins. 4.4. Implications for fish farm and fishery management This study indicates that, at the levels of infection present, glochidia settlement had no significant effect on the growth in length and weight, and condition factor of juvenile salmon, nor of stress measured as lactate, and did not cause further mortality compared with controls. Effects of glochidial infection on salmonid hosts are therefore limited. If these findings are replicated in wild juvenile salmon production, then healthy populations of pearl mussels are unlikely to have a significant effect on survival and performance of salmon fry. Glochidia numbers declined rapidly from 1384 per fish 10 weeks after infection in the present study and this was attributed to the combined treatments for Saprolegnia infection with malachite green (now withdrawn from use) and formalin. Pyceze (Bronopol, Novartis Animal Vaccines Ltd.) is now the standard treatment for Saprolegnia in salmonid culture and there was no reduction in glochidia numbers associated with these treatments in the present study. If farmers are concerned about heavy glochidial infestation the alternative controls remain stocking hatcheries with salmon fry after the period of glochidial release and filtration of incoming water. An alternative, as shown to be effective here, would be to encourage valve closure in glochidia as they pass through a hatchery. Acknowledgments We thank Marine Harvest and Andrew MacLean of Sunbeam Aquaculture for the fish and facilities used in this study and for staff support. We are grateful for the interest and support given in the laboratory methods by Tony Laidler, Helen Morrison, and Carol Cox of Marine Harvest, and to Simon Wild, Charles MacLean, and John MacPhee for assistance with the tank based trials.
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
References Bauer, G., 1987. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.). III. Host relationships. Arch. Hydrobiol. 76, 413–423. Bauer, G., Vogel, C., 1987. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.). I. Host response to glochidiosis. Arch. Hydrobiol. 76, 393–402. Bruno, D.W., McVicar, A.H., Waddell, I.F., 1988. Natural infection of farmed Atlantic salmon, Salmo salar L., parr by glochidia of the freshwater pearl mussel. Bull. Eur. Assoc. Fish Pathol. 8, 23–26. Dodd, B.J., Bernhart, M.C., Rogers-Lowey, C.L., Fobian, T.B., Dimock, R.V., 2005. Cross-resistance of largemouth bass to glochidia of unionid mussels. J. Parasitol. 91, 1064–1072. Hastie, L.C., Cosgrove, P.J., 2001. The decline of migratory salmonid stocks: a new threat to pearl mussels in Scotland. Freshw. Forum 15, 85–96. Hastie, L.C., Young, M.R., 2001. Freshwater pearl mussel (Margaritifera margaritifera) glochidiosis in wild and farmed salmonid stocks in Scotland. Hydrobiologia 445, 109–119. Hastie, L.C., Young, M.R., 2003. Conservation of the freshwater pearl mussel 2: relationship with salmonids. Conserving Natura 2000 Rivers Conservation Techniques, vol. 2. English Nature, Peterborough. 44 pp. Meyers, T.R., Millemann, R.E., 1977. Glochidiosis of salmonid fishes. I. Comparative susceptibility to experimental infection with
79
Margaritifera margaritifera (L.) (Pelecypoda: Margaritiferidae). J. Parasitol. 63, 728–733. Meyers, T.R., Millemann, R.E., Fustish, C.A., 1980. Glochidiosis of salmonid fishes IV: Humoral and tissue responses of Coho and Chinook salmon to experimental infection with Margaritifera margaritifera (L.) (Pelecypoda: Margaritiferidae). J. Parasitol. 66, 274–281. Treasurer, J.W., Turnbull, T., 2000. The pathology and seawater performance of farmed Atlantic salmon infected with glochidia of Margaritifera margaritifera. J. Fish Biol. 57, 858–866. Young, M.R., Williams, J.C., 1984a. The reproductive biology of the freshwater pearl mussel. I. Field studies. Arch. Hydrobiol. 99, 405–422. Young, M.R., Williams, J.C., 1984b. The reproductive biology of the freshwater pearl mussel. II. Laboratory studies. Arch. Hydrobiol. 100, 29–43. Young, M.R., Purser, J., Al-Mousawi, B., 1987. Infection and successful reinfection of Brown trout (Salmo trutta (L.)) with glochidia of Margaritifera margaritifera (L.). Am. Malacol. Bull. 5, 125–128. Ziuganov, V., Zotin, A., Nezlin, L., Tretiakov, V., 1994. The Freshwater Pearl Mussels and their Relationships with Salmonid Fish. VNIRO. Russian Federal Research Institute of Fisheries and Oceanography, Moscow. 104 pp.
Effects of (Margaritifera margaritifera) glochidial infection on performance of tank-reared Atlantic salmon (Salmo salar) James W. Treasurer a,⁎, Lee C. Hastie b , Dougie Hunter c , Fiona Duncan c , Claire M. Treasurer d a
b
Viking Ardtoe Marine Laboratory, Acharacle, Argyll PH36 4LD, UK University of Aberdeen, Department of Zoology, Tillydrone Avenue, Aberdeen AB24 2TZ, UK c Marine Harvest, Blar Mhor Industrial Estate, Fort William PH33 7PT, UK d 9 Zetland Avenue, Fort William PH33 6LL, UK Received 6 January 2006; accepted 7 February 2006
Abstract Experimental infection of pearl mussel glochidia on the performance of Atlantic salmon hosts was examined. Infection intensity was 1392 ± 641 SD glochidia per fish six weeks after challenge but declined significantly 15 weeks after infection to means of 50 and 112 fish− 1 in two trial tanks and, by 6 months numbers were b 1 fish− 1. This loss was attributed to fungal treatments with a combination of malachite green which was still the authorised treatment during the study in 2001 and formalin. The weight of infected fish was significantly lower than controls at 15 weeks but this was not significant after 1 year. The condition factor of infected and naïve fish was not significantly different. Lactate was measured as a possible indicator of stress in infected fish but there was no significant difference with controls. Fish that had been previously infected with glochidia were re-infected in their second year and comparison made with infection of naïve fish to determine whether glochidial infection elicits an immune response. Although there was no significant difference in glochidial numbers in both groups at around 9000 glochidia host− 1 3 weeks after infection, numbers of glochidia in naïve fish did not change to 15 weeks after challenge, whereas there was a significant reduction in previously infected fish to 116 fish− 1. Various treatments were used to provoke closure of the glochidial valves. It is concluded that infection of salmon with glochidia levels in the current study had no significant effect on salmon performance, condition and stress as measured by assay of lactate. © 2006 Elsevier B.V. All rights reserved. Keywords: Pearl mussel; Margaritifera margaritifera; Glochidia; Salmonid; Salmo salar
1. Introduction The endangered freshwater pearl mussel, Margaritifera margaritifera, has a dispersive parasitic larval (glochidial) stage and large numbers are shed by adult ⁎ Corresponding author. Tel.: +44 1397 875000; fax: +44 1397 875001. E-mail address: [email protected] (J.W. Treasurer). 0044-8486/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.aquaculture.2006.02.031
mussels in mid-summer. These attach to the gills of salmonids where they encyst and remain for c. 12 months before excysting and dropping to the river bed where they settle (Hastie and Young, 2001). Compared to other parasitic infections of salmon and trout, glochidiosis is of little economic importance. However, in certain situations, pearl mussels can cause problems for fish farmers (Hastie and Young, 2003). Most studies of the mussel– host relationship have focussed on distribution and
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
ecology (Young and Williams, 1984a,b; Hastie and Young, 2001, 2003), effects of previous challenges on glochidial survival (Young et al., 1987), and comparative susceptibility of different fish hosts to infection (Meyers and Millemann, 1977; Bauer, 1987). Relatively little attention has focussed on the impacts of glochidia on the performance and survival of the host (Treasurer and Turnbull, 2000). While some authorities have suggested that glochidiosis has little effect on host growth and survival, others have indicated that high levels of infection in farmed situations may induce mortalities and anglers and fishery managers have implied that pearl mussels may be detrimental to endangered salmonid stocks. The impact of pearl mussel glochidia on their host is difficult to measure and monitor in the field. In the present study the potential impact of infection was assessed by the artificial infection of cultured Atlantic salmon and by measurement of juvenile salmon in replicated tanks with comparison with naive fish. Although immune responses of fish to bacterial and viral infection have been frequently documented, measurement of host immune response to parasitic infection has been limited (Meyers et al., 1980; Bauer and Vogel, 1987). It has been suggested that older fish previously challenged with glochidia have lower infection intensity and, although it has been inferred that fish mount an immune response, this has not been verified (Bauer, 1987; Hastie and Young, 2001). Therefore previously challenged fish were re-infected experimentally, and compared with infection of naïve control fish, to assess if an immune response could be elicited. 2. Methods 2.1. Infection Uninfected juvenile salmon of 5.6 ± 1.5 SD g mean weight were transferred from a salmon hatchery in Lochaber in June 2001 to an outdoor experimental hatchery. 300 fish were infected by cohabiting with five spawning pearl mussels checked for spawning condition by carefully opening their shell valves with special opening tongs, and checking for the presence of glochidia in the modified gill structures of the female mussels (Young and Williams, 1984a). Salmon were cohabited in a tank overnight with pearl mussels to permit natural infection. The fish were evenly divided between two 1 m2 × 40 cm glassfibre tanks. Flow rate was 10l min− 1, and water temperatures were recorded daily and were in the range 3 to 17 °C during the first year of the infection study. A total of 300 non-infected control fish were maintained in two other adjacent tanks. Fish were fed twice daily to satiation with 2 mm Trouw diet.
75
Mortalities were removed daily and recorded. Fifty fish from each tank were anaesthetised in 20 mg l− 1 benzocaine and measured two weeks after infection and then at 6, 10, 15, 25, 30 and 36 weeks to fork length mm and weight to 0.01 g. Ten of these fish were killed, five from each tank, by overdose in benzocaine and gills from the left side of the fish excised and examined at ×10 magnification and glochidia counted on the first gill arch. As previous studies have found no significant difference in glochidia numbers between gill arches (Treasurer and Turnbull, 2000) the total number of glochidia per fish was calculated by a multiplier of 8. Glochidia were measured at ×40 magnification laterally across the carapace using an ocular micrometer. Blood samples were collected from the caudal fin from infected and control fish 15 weeks after commencement of the experiment and samples tested for lactate as a measurement of stress. Plasma lactate was determined enzymatically using Sigma Diagnostic kits (Sigma; Proc. no. 735). The condition of salmon was measured as (body mass g 100/length3). A sample of fish used in the first infection experiment was also examined at the donor hatchery from the main stock 5 months after natural infection in the hatchery. Twenty fish were measured and gills examined as above. 2.2. Re-infection experiment One year post-infection, the remaining fish in both trial and control groups were infected with glochidia on 15 July 2002 by experimental challenge. Fish were sampled 6 and 12 weeks after infection and numbers of glochidia counted on all gills in 5 fish from each treatment. 2.3. Methods for promoting closure of glochidia An attempt was made to prevent infection of hatchery reared salmon by glochidia by causing closure of the glochidia before they passed through the hatchery. Collected spat of 0.5 ml volume were introduced to 5 ml capacity six well plates containing test solutions. Spat were assessed for closure of valves, immediately and after exposure to test solutions for 5 min. 10 ml of unchlorinated water were added to the mussel suspension at this point and mussel viability was assessed after 24 h. The treatments were: NaCl at 200, 2000, 5000, 10,000 and 50,000 mg l− 1, hydrogen peroxide at 17,500, 35,000, 175,000 and 350,000 mg l− 1, formaldehyde at 50, 100, 200, 400, 1000, 2000, and 20,000 mg l− 1, and copper sulphate at 250, 500, 5000, 50,000 mg l− 1.
76
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79 180
Data on infection intensity, growth and lactate values were tested for homogeneity of variances by Bartlett's test. As variances were mainly heterogeneous statistical comparison of treatments was carried out with the nonparametric Mann Whitney and Kruskal–Wallis tests. Differences were taken as significant when P b 0.05. The variances of condition factor values were highly variable and data were transformed to common logarithm values prior to ANOVA. 3. Results
Fork length mm
2.4. Statistical analyses
160 140 120 100
Infected T1 Infected T2 Control T3 Control T4
80 60 0
1
2
3
4
5
6
7
8
9
10
Months post infection
Fig. 2. A comparison of growth in length of infected (tanks T1 and T2) and control (T3 and T4) juvenile salmon. N = 50 fish per tank at each sampling point.
3.1. Course of infection and mussel dimensions Two weeks after infection mean glochidia numbers were 818 ± 377 SD per fish (Fig. 1) with a wide range in abundance, from 352 to 1184 glochidia fish − 1 . Glochidia numbers were initially stable with 1392 ± 641 SD glochidia fish− 1 6 weeks post-infection and 1385 ± 553 glochidia fish− 1 10 weeks after infection. However, glochidia numbers dropped significantly in October, at 15 weeks post-infection, to 50 ± 45 glochidia fish− 1 in one tank and 112 ± 197 glochidia fish− 1 in the other tank (Z scores of 4.39 and 4.35, P b 0.001, respectively, Mann Whitney) (Fig. 1). By 14 January, 25 weeks after infection, glochidia numbers were negligible, zero in 10 fish examined from tank 1 and 0.64 ± 2.32 glochidia fish− 1 in the second tank. No glochidia were seen in fish examined in February and April. Mean glochidium diameter was 296 ± 13 μm on infection, 348 ± 13 μm by 10 weeks post-infection and 365 ± 28 μm at 15 weeks post-infection. Only limited data were available from naturally infected fish from the salmon hatchery in 2000.
N glochidia fish-1
2500
Tank 1 Tank 2
2000 1500 1000 500 0 0
1
2
3
4
5
6
7
8
9
10
Months post infection
Fig. 1. Infection intensity of glochidia in two tanks of artificially challenged salmon juveniles. Fish were experimentally infected in July 2001. N = 10 fish at each sampling point for each treatment. Bars, indicate SD, are shown in only one tank from each treatment to enhance clarity.
Prevalence was 40% and intensity of infection was only 1.7 glochidia five months after natural infection. 3.2. Host impacts of infection 3.2.1. Length Mean length of newly infected salmon on 27 July 2001 was 76.8 ± 6.4 mm fork length (FL) and not significantly different from uninfected fish, 76.2 ± 5.0 mm (Mann Whitney test Z = 0.66, P = 0.51). By 15 weeks post-infection the infected fish were smaller than the controls (Fig. 2), namely 113.4 ± 14.7 and 114.0 ± 14.6 mm compared with 115.6 ± 13.8 and 121.4 ± 14.4 mm in uninfected fish, and this was significant (Kruskal–Wallis, H = 17.49, P = 0.00056). Following 6 months post-infection the salmon in one tank of each treatment were slightly longer than the other (H = 18.3, P = 0.000434), but by 9 months post-infection there was no significant difference (Z = 1.57, P = 0.1172) between infected and control fish, 157 ± 9 and 154 ± 9 mm, respectively. 3.2.2. Weight The pattern in weights mirrored that of length with starting weights of 5.6 ± 1.5 g in infected fish and 5.5 ± 1.3 g in uninfected controls (Mann Whitney, Z = 0.46, P = 0.64) (Fig. 3). There was no significant difference in weight after 6 and 10 weeks post-infection and by 15 weeks the weight of infected fish was slightly lower at 18.1 ± 5.8 and 18.4 ± 6 g compared with 19.1 ± 5.9 and 22.8 ± 6.7 g in controls and this was significant (Kruskal–Wallis, H = 23.78, P = 0.000028). By six and seven months weights were not significantly different between tanks and at nine months post-infection infected fish were 36.9 ± 6.4 g compared with 35.4 ± 6.7 g in controls with no significant difference in weight (Z = 1.51, P = 0.13).
3.2.3. Condition factor There was no significant difference (P N 0.05) in condition factor (K) between infected and trial fish on any of the sampling dates. However, on 22 February control fish in one tank and infected fish in another had higher condition factors. However, when data from the tanks in each treatment were combined, these were found to be highly variable, and data were log transformed prior to statistical analysis. Although K in the control group was higher, this was not significant. As the fish had been subject to Saprolegnia infection, and required treatment, it was concluded that differences between tanks were due to fungal infection rather than glochidia, particularly as the majority of glochidia had been shed by this time.
N glochidia fish-1
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
77
15000 10000 5000 0 0
5
10
15
20
Weeks post infection Previously infected
Naïve fish
Fig. 4. The pattern of glochidia infection in naïve and previously infected salmon in their second year. Fish infected in July 2002. Bars indicate SD.
There was no significant difference in lactate levels between fish infected with glochidia (6.7 ± 3.4 SD) and uninfected (7.3 ± 2.9 SD) fish (n = 5 each sample, P = 0.15).
glochidia fish− 1 in naïve fish, and this was not significant (Mann Whitney Z = 0.2268, P = 0.82) (Fig. 4). Seven weeks post-infection glochidia numbers remained high in the naïve fish, 7407 ± 3286 glochidia fish− 1, but glochidial intensity had significantly declined to 5260 ± 1793 glochidia fish− 1 in the previously infected fish (Z = 2.65, P = 0.008), although the glochidia numbers between treatment groups were not significantly different (Z = 0.36, P = 0.17). By 16 November 2002, when the remaining fish were sacrificed, 15 weeks after challenge, glochidia numbers in the previously infected group had declined markedly to 116 ± 141 glochidia fish− 1 while infection intensity was unchanged in the naïve fish at 8120 ± 5355 glochidia fish− 1. Neither mean length, 228 ± 17 v. 226 ± 12 mm (Z = 0.23, P = 0.82), nor weight, 126 ± 33 v. 121 ± 20 g (Z = 0.27, P = 0.79), of re-infected and naïve fish were significantly different.
3.5. Re-infection experiment
3.6. Preventing infection
The variation in infection intensity was high between fish in each treatment group. Mean intensity 3 weeks after infection was 9096 ± 3046 SD glochidia fish− 1 in previously infected fish compared with 8932 ± 3719
Various salt and other solutions were used to try to elicit closing of shell valves, thus preventing attachment to fish in hatcheries. All solutions induced closure of valves within 5 to 6 min, with the exception of formaldehyde where they remained open. Glochidia exposed to hydrogen peroxide, copper sulphate and formaldehyde did not appear viable after 24 h but most glochidia exposed to NaCl solutions from 200 to 2000 mg l− 1 were still viable and showed movement in the shell valves.
3.3. Mortalities Mortalities over the one year study period were compared between two tanks of infected fish and controls. There were no significant differences between the numbers of mortalities in any of the tanks to February 2002. Thereafter there was a large mortality affecting the tanks, due to pump failure, and further comparison of survival was not possible. 3.4. Lactate measures
Mean weight g
50 40 30 20
Infected T1 Infected T2 Control T3 Control T4
10 0
0
2
4
6
8
10
Months post infection
Fig. 3. Comparison of growth in weight of infected (T1 and T2) and non-infected (T3 and T4) juvenile salmon.
4. Discussion 4.1. Pattern of infection The overall prevalence and mean glochidial loads dropped dramatically by 15 weeks post-infection, from 1393 ± 641 glochidia fish− 1 to 0.64 ± 2.32. This
78
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
decline following infection has been reported in other studies (Hastie and Young, 2001) and during the present study a loss of N99% to January was observed. This feature is typical, and has been attributed to incomplete attachment (Bruno et al., 1988) or a specific host response (Hastie and Young, 2001). The extent of the host response varies between salmonid species (Meyers et al., 1980). The decline observed during the present study was high, and a complete loss of glochidia had occurred 6 months after infection, so other factors may have been responsible. While it has been reported that treatments with therapeutants such as copper sulphate, salt and Roccal are ineffective in controlling glochidial infection in farmed salmonids (Bruno et al., 1988), the decline in the present study suggests that therapeutic treatments may have affected glochidia numbers. Fish were treated monthly for Ichthyobodo infection with 200 mg l − 1 formalin and concurrently with malachite green. The latter is now discontinued from use with food chain species but during the first year of the study in 2001 it was the recognised treatment for Saprolegnia. In the last year of the study when malachite green was withdrawn from use fish were therefore treated for fungus with Pyceze (50% w/v bronopol, Novartis Animal Vaccines) and this appeared to have no effect on glochidial numbers, and this may therefore be used as an alternative treatment for fish used for collection of glochidia for restocking purposes. 4.2. Impact on hosts This study shows that a relatively high level of infection of 1393 glochidia per 5 g salmon juvenile had no significant effect on salmon survival and only a small transient effect on growth in length and weight. While the mean weight of uninfected fish initially showed a trend to a higher weight than infected fish the fish grew at the same rate following loss of glochidia to three months after infection. In a previous study of infected farmed salmon with loads of up to 480 glochidia fish− 1 , no significant differences in survival, growth and condition factor of infected and non-infected salmon smolts stocked in seawater cages were observed (Treasurer and Turnbull, 2000), indicating no measurable effects of glochidia on the host at this level of infection. Rather than a threat to endangered salmonid stocks it is likely that pearl mussels improve water quality in rivers and clean gravel beds and this may benefit salmon spawning and nursery habitat (Ziuganov et al., 1994; Hastie and Cosgrove, 2001).
4.3. Immune response This study demonstrates that there is an immune response as, in previously infected salmon, glochidia numbers declined to low numbers 15 weeks after infection, although numbers in naïve fish did not decline significantly. It is therefore possible that a vaccine could be developed to minimise the effects of infection with glochidia in the farm setting. It has been shown that transformation success in a variety of unionid mussels was significantly lower on previously primed fish (Dodd et al., 2005). Further, primed fish showed binding of serum antibodies with glochidia proteins. 4.4. Implications for fish farm and fishery management This study indicates that, at the levels of infection present, glochidia settlement had no significant effect on the growth in length and weight, and condition factor of juvenile salmon, nor of stress measured as lactate, and did not cause further mortality compared with controls. Effects of glochidial infection on salmonid hosts are therefore limited. If these findings are replicated in wild juvenile salmon production, then healthy populations of pearl mussels are unlikely to have a significant effect on survival and performance of salmon fry. Glochidia numbers declined rapidly from 1384 per fish 10 weeks after infection in the present study and this was attributed to the combined treatments for Saprolegnia infection with malachite green (now withdrawn from use) and formalin. Pyceze (Bronopol, Novartis Animal Vaccines Ltd.) is now the standard treatment for Saprolegnia in salmonid culture and there was no reduction in glochidia numbers associated with these treatments in the present study. If farmers are concerned about heavy glochidial infestation the alternative controls remain stocking hatcheries with salmon fry after the period of glochidial release and filtration of incoming water. An alternative, as shown to be effective here, would be to encourage valve closure in glochidia as they pass through a hatchery. Acknowledgments We thank Marine Harvest and Andrew MacLean of Sunbeam Aquaculture for the fish and facilities used in this study and for staff support. We are grateful for the interest and support given in the laboratory methods by Tony Laidler, Helen Morrison, and Carol Cox of Marine Harvest, and to Simon Wild, Charles MacLean, and John MacPhee for assistance with the tank based trials.
J.W. Treasurer et al. / Aquaculture 256 (2006) 74–79
References Bauer, G., 1987. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.). III. Host relationships. Arch. Hydrobiol. 76, 413–423. Bauer, G., Vogel, C., 1987. The parasitic stage of the freshwater pearl mussel (Margaritifera margaritifera L.). I. Host response to glochidiosis. Arch. Hydrobiol. 76, 393–402. Bruno, D.W., McVicar, A.H., Waddell, I.F., 1988. Natural infection of farmed Atlantic salmon, Salmo salar L., parr by glochidia of the freshwater pearl mussel. Bull. Eur. Assoc. Fish Pathol. 8, 23–26. Dodd, B.J., Bernhart, M.C., Rogers-Lowey, C.L., Fobian, T.B., Dimock, R.V., 2005. Cross-resistance of largemouth bass to glochidia of unionid mussels. J. Parasitol. 91, 1064–1072. Hastie, L.C., Cosgrove, P.J., 2001. The decline of migratory salmonid stocks: a new threat to pearl mussels in Scotland. Freshw. Forum 15, 85–96. Hastie, L.C., Young, M.R., 2001. Freshwater pearl mussel (Margaritifera margaritifera) glochidiosis in wild and farmed salmonid stocks in Scotland. Hydrobiologia 445, 109–119. Hastie, L.C., Young, M.R., 2003. Conservation of the freshwater pearl mussel 2: relationship with salmonids. Conserving Natura 2000 Rivers Conservation Techniques, vol. 2. English Nature, Peterborough. 44 pp. Meyers, T.R., Millemann, R.E., 1977. Glochidiosis of salmonid fishes. I. Comparative susceptibility to experimental infection with
79
Margaritifera margaritifera (L.) (Pelecypoda: Margaritiferidae). J. Parasitol. 63, 728–733. Meyers, T.R., Millemann, R.E., Fustish, C.A., 1980. Glochidiosis of salmonid fishes IV: Humoral and tissue responses of Coho and Chinook salmon to experimental infection with Margaritifera margaritifera (L.) (Pelecypoda: Margaritiferidae). J. Parasitol. 66, 274–281. Treasurer, J.W., Turnbull, T., 2000. The pathology and seawater performance of farmed Atlantic salmon infected with glochidia of Margaritifera margaritifera. J. Fish Biol. 57, 858–866. Young, M.R., Williams, J.C., 1984a. The reproductive biology of the freshwater pearl mussel. I. Field studies. Arch. Hydrobiol. 99, 405–422. Young, M.R., Williams, J.C., 1984b. The reproductive biology of the freshwater pearl mussel. II. Laboratory studies. Arch. Hydrobiol. 100, 29–43. Young, M.R., Purser, J., Al-Mousawi, B., 1987. Infection and successful reinfection of Brown trout (Salmo trutta (L.)) with glochidia of Margaritifera margaritifera (L.). Am. Malacol. Bull. 5, 125–128. Ziuganov, V., Zotin, A., Nezlin, L., Tretiakov, V., 1994. The Freshwater Pearl Mussels and their Relationships with Salmonid Fish. VNIRO. Russian Federal Research Institute of Fisheries and Oceanography, Moscow. 104 pp.