ARTICLE IN PRESS

Water Research 38 (2004) 2847–2850

Lethal and sublethal toxicity of ammonia to native, invasive, and parasitised freshwater amphipods John Prentera,*, Calum MacNeila, Jaimie T.A. Dicka, Gillian E. Riddella, Alison M. Dunnb a

School of Biology and Biochemistry, Queen’s University Belfast, Medical Biology Centre, 97 Lisburn Road, Belfast BT9 7BL, N. Ireland, UK b Centre for Biodiversity and Conservation, School of Biology, University of Leeds, Leeds LS2 9JT, UK Received 26 June 2003; received in revised form 15 December 2003; accepted 26 March 2004

Abstract In lethal and sublethal ammonia toxicity tests, we examined differences in tolerance of three species of freshwater amphipods, one native and two invasive in Ireland. The native Gammarus duebeni celticus was slightly less tolerant to ammonia than the invasive G. pulex (96 h LC50=1.155 and 1.544 mg l!1, respectively), while another invader, Crangonyx pseudogracilis, had the lowest tolerance (LC50=0.36 mg l!1). Parasitism of G. pulex by the acanthocephalan Echinorhynchus truttae greatly reduced the tolerance of the invader to ammonia (LC50=0.381 mg l!1). Further, precopula pair disruption tests indicated that G. d. celticus was more sensitive to ammonia than G. pulex at sublethal levels. We discuss these results in the context of the ecological replacements of native by invader amphipods. r 2004 Elsevier Ltd. All rights reserved. Keywords: Acanthocephalan; Ammonia; Amphipod; Invasion; Species replacement

1. Introduction In the last century, freshwaters in Ireland, such as the River Lagan, have been invaded by several amphipod species [1], with the native Gammarus duebeni celticus replaced in many areas by G. pulex transplanted from England [2]. A second invader, the North American Crangonyx pseudogracilis now occupies the most polluted areas of the same rivers, where neither Gammarus spp. are found [3]. Amphipod invasions and replacements are frequently accompanied by environmental degradation [4], with distribution patterns indicating that G. d. celticus is replaced in poorer water quality areas by invaders such as G. pulex and C. pseudogracilis, implying a greater physicochemical tolerance of the *Corresponding author. Tel.: +44-28-9097-2065; fax: +4428-9097-5877. E-mail address: [email protected] (J. Prenter).

invaders [3,5]. Dick [2] emphasized the role of intraguild predation (IGP) of G. d. celticus and C. pseudogracilis by G. pulex in determining distribution, whereas MacNeil et al. [3] emphasized the greater tolerance to organic pollution of C. pseudogracilis. Despite the wealth of such field data, however, no study has tested directly if these invaders are more tolerant of common pollutants than the native. Further, acanthocephalan parasites can cause elevated ventilation rates in amphipods, along with greater activity levels [6] and potentially faster metabolism [7]. Consequently, apart from the debilitating effects on host physiology, parasitised amphipods may have increased exposure to pollutants. The interaction of parasite and toxicant effects could, therefore, mediate the invasion process if the native G. d. celticus is less frequently parasitised than G. pulex. This is the case in the River Lagan, with fewer populations of G. d. celticus harbouring Echinorhynchus truttae than populations of

0043-1354/$ - see front matter r 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2004.03.042

ARTICLE IN PRESS 2848

J. Prenter et al. / Water Research 38 (2004) 2847–2850

G. pulex, and with lower prevalence in the native than the invader [C. MacNeil, pers. obs.]. Ammonia is an important inorganic contaminant of freshwater [8] with the ionised form of ammonia (NH+ 4 ) largely inert in aquatic environments [9], but the unionised form (NH3) highly toxic [10]. We adopted acute (lethal) and chronic (sublethal) toxicity tests to assess differences in tolerance of the Irish native G. d. celticus and invasive G. pulex and C. pseudogracilis (acute test only) to ammonia. We also examine the effect of E. truttae parasitism on the acute toxicity of ammonia to G. pulex.

2. Materials and methods 2.1. Collection, holding and acclimation of test organisms G. d. celticus, G. pulex, G. pulex parasitised with E. truttae, and C. pseudogracilis were collected in March 2003 by kick sampling the River Lagan (UK grid refs. J306645, J325686, and J338698, respectively). Water temperatures were 12–13" C, pH 7.8–8.1 (Gallen Kampf meter), conductivities 305–330 ms cm!1 (Dist WP; Hanna Instruments), dissolved oxygen 8.9–9.5 mg l!1 (Jenway DO meter 9071) and total ammonia levels o0.25 mg l!1 (o0.012 mg l!1 NH3) (TetraWerke test kit). Animals of each species were acclimated separately for 7 days in tanks (65 # 40 # 20 cm3, l # w # d) of River Lagan water at 12" C supplied with stream conditioned leaf oak, Quercus robur, and alder, Alnus glutinosa. 2.2. Toxicity tests Adult G. d. celticus (14–16 mm), G. pulex with and without E. truttae (13–15 mm), and C. pseudogracilis (4– 6 mm) were subjected to nominal ammonia concentrations (see below) in continuous exposure, static conditions. We followed standard methodology [11–13], but with filtered River Lagan water at 12" C, 12 h D:12 h L and pH 7.5. Following range finding tests, three replicates (each with five individuals) were performed: G. d. celticus, parasitised G. pulex and C. pseudogracilis at 1, 5, 10 and 20 mM NH4Cl (0.12, 0.62, 1.23 and 2.462 mg l!1 NH3, see [14]); unparasitised G. pulex at 1, 5, 10, 15, 20 and 25 mM NH4Cl (0.12, 0.62, 1.23, 1.85, 2.462 and 3.08 mg l!1 NH3). 96 h LC50 values were determined with ToxCalct (Tidepool Scientific Software). Further, we adapted the sublethal G. Pulex Precopula Separation (GaPPS) test [15] to examine behavioural indications of stress to ammonia. Precopula pairs (G. d. celticus, n ¼ 10 and G. pulex, n ¼ 20; all unparasitised) were exposed to 100 ml of test solution in 180 ml plastic containers (0, 1 and 10 mM NH4Cl—0, 0.12, 1.23 mg l!1 NH3) for 9 h. We used Fisher’s exact probability tests to

compare frequencies of precopula separation between the two species at the three concentrations.

3. Results Table 1 shows ammonia LC50 values. G. d. celticus was slightly less tolerant than unparasitised G. pulex, with C. pseudogracilis the least tolerant. However, parasitism greatly reduced the tolerance of G. pulex to ammonia, making it substantially less tolerant than G. d. celticus. Precopula pair separation was clearly more frequent in response to ammonia in G. d. celticus than G. pulex (Table 2), with pairs of the former splitting early in the experiment, particularly at 1.23 mg l!1 (10 mM) NH3 (Fig. 1).

4. Discussion Assumed differences in tolerance to pollutants in the native and invasive amphipod species examined here are principally derived from species distributions and correlations with habitat conditions [3,5]. Under controlled laboratory conditions, we have shown variation in tolerance of these species to a common pollutant and indicated that parasitism could decrease this tolerance. Our ammonia LC50 estimate for G. pulex is similar to that of Williams et al. [16] of 2.05 mg l!1 NH3. Our results indicate that the native G. d. celticus is only Table 1 Ammonia 96 h LC50 values (mg l!1) for Gammarus spp. and C. pseudogracilis Species

G. G. G. C.

LC50

1.155 1.544 0.381 0.360

d. celticus pulex pulex (parasitised) pseudogracilis

95% Conf. limits Upper

Lower

1.625 1.817 0.568 0.587

0.821 1.098 0.221 0.169

Table 2 Separation frequencies of precopula pairs of Gammarus spp [NH4Cl]

Paired Separated

0 mg l!1

0.12 mg l!1

1.23 mg l!1 Gdc

Gdc

Gp

Gdc

Gp

10 0 ns

20 0

6 20 4 0 p ¼ 0:008

Gp

3 17 7 3 p ¼ 0:005

Gdc=G. duebeni celticus, Gp=G. pulex, ns denotes nonsignificant Fisher’s exact probability test.

ARTICLE IN PRESS J. Prenter et al. / Water Research 38 (2004) 2847–2850

2849

Cumulative percentage separation

80 G. d. celticus (1.23 mg/l) 60

G. d. celticus (0.12 mg/l)

40

20

0

G. pulex (1.23 mg/l)

G. pulex (0.12 mg/l) 0

60

120

180

240

300

360

420

480

540

Time (min)

Fig. 1. Cumulative percentage precopula pair separation in Gammarus spp.

slightly less tolerant to ammonia than the invader G. pulex. E. truttae parasitism, however, greatly reduces the tolerance of the invader to ammonia. Therefore, although G. pulex is more successful in mutual predatory interactions with G. d. celticus and thus replaces the native in rivers and lakes [2,5], high levels of parasitism in G. pulex may prevent its incursions where water quality is low, or short-term ‘pulses’ of pollution may remove parasitised invaders. Furthermore, parasite-induced behavioural changes in G. pulex make this invader more vulnerable than the native to shared predators, such as salmonids [6]. Previous field studies of amphipod invasions in The Netherlands, Germany and N. Ireland have all indicated that C. pseudogracilis is more tolerant of organic pollution than Gammarus spp. [3,17]. Our finding that C. pseudogracilis is less tolerant of ammonia pollution than either Gammarus spp. is, therefore, unexpected. Indeed, our estimate for C. pseudogracilis adults is an order of magnitude lower that that of Arthur et al. [18] for adults in spring (3.29–3.56 mg l!1 NH3 at pH=8– 8.2,13–13.3" C). C. pseudogracilis used in our experiments were very small adults (4–6 mm body length) and it is possible that the greater surface to volume ratio may have increased their sensitivity to ammonia pollution, as larger animals are generally more tolerant to toxicants [19]. In addition, interspecies differences in the thickness, toughness and permeability of the cuticular and gill membranes to salts [20,21] may contribute to differences in tolerance to ammonia. Our study, like that of McCahon et al. [22], found that precopula pair separation was a sensitive and rapid indicator of stress to raised ammonia levels, as higher concentrations of pollutant caused more pairs to separate. The more rapid pair separation in G. d. celticus than G. pulex (Table 2, Fig. 1) supports the finding that G. pulex is more tolerant to ammonia pollution at sublethal levels. Disruption of reproduction

at sublethal pollutant levels may contribute to the loss of species and a decrease in invertebrate community diversity and density [22]. The disruption of precopula in Gammarus spp. is particularly likely to have population implications as females that go without copulations at moult are not available for reproduction until another full period of the moult cycle has elapsed (around 1 month). G. pulex has a higher reproductive output than G. d. celticus [2], and the greater susceptibility of G. d. celticus to precopula disruption may mean G. pulex populations have a further competitive advantage in polluted water. We have shown parasitism of G. pulex by E. truttae reduced host tolerance to ammonia. Similarly, McCahon et al. [22] and McCahon and Poulton [23] found parasitism by the fish acanthocephalan parasite Pomphorynchus laevis exacerbated the toxic effects of ammonia, acid and aluminium pollution on G. pulex, with parasitised animals showing reduced feeding rates and earlier mortality [22]. Bauer et al. [24] noted that P. laevis, as with E. truttae, makes G. pulex photophilic. In European freshwaters where G. pulex is the native and G. roeseli is the invader and not affected by the parasite, the invasion process is being facilitated by increased exposure of the parasitised native to fish predation [24]. Our results also have important implications for the invasion process when parasite prevalence or effect differs significantly between native and invader. For instance, in areas with high ammonia levels, G. d. celticus may be able to withstand predatory attacks from G. pulex weakened by parasitism. Also, although this ‘intraguild predation’ or ‘IGP’ tends to be in favour of G. pulex, it is mutual and G. pulex with pollution stress exacerbated by parasitism may be vulnerable to predation by the native. Furthermore, reduced feeding levels in parasitised animals [22] and perhaps lower growth rates could also alter the balance in IGP.

ARTICLE IN PRESS 2850

J. Prenter et al. / Water Research 38 (2004) 2847–2850

5. Conclusions The native G. d. celticus and G. pulex had similar tolerances to ammonia, with the native being slightly less tolerant than the invader. However, G. pulex parasitised by E. truttae was substantially less tolerant to ammonia than unparasitised G. pulex or G. d. celticus. Greater precopula disruption in response to sublethal ammonia levels was also observed in G. d. celticus compared to G. pulex. Our results indicate that assumptions derived from species survey data about the relative tolerances of native and invasive amphipods to many common pollutants must be questioned and investigated further by empirical study. High levels of parasitism in G. pulex may prevent its incursions in areas of low water quality and, therefore, mediate the replacement of the native G. d. celticus.

[11]

[12]

[13]

[14]

[15]

Acknowledgements Thanks to Mark Briffa for comments and suggestions. This study was supported by NERC grant GR3/12871.

[16]

References [1] Strange CD, Glass GB. The distribution of freshwater gammarids in Northern Ireland. Proc Roy Irish Acad 1979;79(B):145–53. [2] Dick JTA, Montgomery WI, Elwood RW. Replacement of the indigenous amphipod Gammarus duebeni celticus by the introduced G. pulex: differential cannibalism and mutual predation. J Anim Ecol 1993;62:79–88. [3] MacNeil C, Dick JTA, Elwood RW. Differential physicochemical tolerances of amphipod species revealed by field transplantations. Oecologia 2000;124:1–7. [4] Conlan KE. Amphipod crustaceans and environmental disturbance: a review. J Nat Hist 1994;28:519–54. [5] MacNeil C, Montgomery WI, Dick JTA, Elwood RW. Factors influencing the distribution of native and introduced Gammarus spp. in Irish river systems. Arch Hydrobiol 2001;151:353–68. [6] MacNeil C, Dick JTA, Elwood RW, Hatcher MJ, Dunn AM. Differential drift and parasitism in invading and native Gammarus spp (Crustacea: Amphipoda). Ecography 2003;6:467–73. [7] Poulin R. ‘‘Adaptive’’ changes in the behaviour of parasitized animals: a critical review. Int J Parasitol 1995 ;25:1371–83. [8] Mason CF. Biology of freshwater pollution. London: Longman; 1981. p. 356. [9] Tabata K. Toxicity of ammonia to aquatic animals with reference to the effect of pH and carbon dioxide. Bull Tokai Regional Fish Res Lab 1962;34:67–74. [10] Russo RC. Ammonia, nitrite and nitrate. In: Rand GM, Petrocelli SR, editors. Fundamentals of aquatic toxicol-

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

ogy: methods and applications. London: Hemisphere Publishing Corp; 1985. p. 455–71. ASTM Standards. Standard guide for conducting acute toxicity tests on aqueous ambient samples and effluents with fishes, macroinvertebrates, and amphibians. Designation E1192-97. In: Annual book of ASTM standards, 1997, 11.05: 440–452. ASTM Standards. Standard guide for behavioural testing in aquatic toxicology. Designation E1604–94. In: Annual book of ASTM standards, 1994, 11.04: 983–990. Girling AE, Pascoe D, Janssen CR, Peither A, Wenzel A, Scha¨fer H, Neumeier B, Mitchell GC, Taylor EJ, Maund SJ, Lay JP, Ju¨ttner I, Crossland NO, Stephensen RR, Persoone G. Development of methods for evaluating toxicity to freshwater ecosystems. Ecotoxicol Environ Safety 2000;45:148–76. Emerson K, Russo RC, Lund RE, Thurston RV. Aqueous ammonia equilibrium calculations: effect of pH and temperature. J Fish Res Board Canada 1995;32:2379–83. Poulton M, Pascoe D. Disruption of precopula in Gammarus pulex (L.)—development of a behavioural bioassay for evaluating pollutant and parasite induced stress. Chemosphere 1990;20:403–15. Williams K, Green D, Pascoe D. Toxicity testing with freshwater macroinvertebrates: methods and application in environmental management. In: Pascoe D, Edwards RW, editors. Freshwater biological monitoring. Oxford: Pergamon; 1984. p. 81–91. Pinkster S, Smit H, Brandse-De Jong N. The introduction of the alien amphipod Gammarus tigrinus SEXTON, 1939, in the Netherlands and its competition with indigenous species. Crustaceanna Supplement 1977;4:91–105. Arthur JW, West CW, Allen KN, Hedtke SF. Seasonal toxicity of ammonia to five fish and nine invertebrate species. Bull Environ Contam Toxicol 1987;38:324–31. Naylor C, Pindar L, Calow P. Inter- and intraspecific variation in sensitivity to toxins: the effects of acidity and zinc on the freshwater crustaceans Asellus aquaticus (L.) and Gammarus pulex (L.). Water Res 1990;24:757–62. Sutcliffe DW. Sodium regulation and adaptation to freshwater in Gammarid crustaceans. J Exp Biol 1968;48: 359–80. Shires R, Lane NJ, Inman CBE, Lockwood APM. Microtubule systems associated with the separate junctions of the gill cells in four gammarid amphipods. Tissue Cell 1994;27:3–12. McCahon CP, Poulton MJ, Thomas PC, Xu Q, Pascoe D, Turner C. Lethal and sublethal toxicity of field simulated farm waste episodes to several freshwater invertebrate species. Water Res 1991;25:661–71. McCahon CP, Poulton MJ. Lethal and sublethal effects of acid, aluminium and lime on Gammarus pulex during repeated simulated episodes in a Welsh stream. Freshwater Biol 1991;25:169–78. Bauer A, Trouve S, Gregoire A, Bollache L, Cezilly F. Differential influence of Pomphorhynchus laevis (Acanthocephala) on the behaviour of the native and invader gammarid species. Int J Parasitology 2000;30:1453–7.