Gammarus: Important Taxon in Freshwater and Marine Changing EnvironmentsView this Special Issue
Dietary Preference of Gammarus pulex and Asellus aquaticus during a Laboratory Breeding Programme for Ecotoxicological Studies
An investigation was undertaken to establish if Gammarus pulex and Asellus aquaticus preferred a diet of unconditioned, artificially or naturally conditioned alder leaves (Alnus glutinosa). Standardised, 24 hour ex situ feeding assays were undertaken with both species to determine their food preference. The results showed that A. aquaticus ate more leaf material compared to G. pulex (Z 23.909, P 0.001) when exposed to all three test variables. Also, both G. pulex and A. aquaticus demonstrated a preference for naturally conditioned leaves compared to the other two variables, with unconditioned leaves proving the least popular food option for both macroinvertebrates (Z 18.803, ). However, both species ate varying amounts of all the leaf treatments (Z 136.399, ). Subsequently, the author outlined a feeding methodology for natural alder leaf conditioning that could be used during a laboratory breeding programme.
What is the best diet for Gammarus pulex and Asellus aquaticus during a laboratory breeding programme and/or ecotoxicological study? Over the years, scientists have used a variety of nutritional supplements to feed macroinvertebrates during breeding programmes and experiments, including dog food , baby, and fish food . If the macroinvertebrates were being bred for ecotoxicological studies (or as test subjects within bioassays) they need to be representative of wild specimens, and it is well documented that a test, animals response could be affected by their past history, diet, life stage, disease and so forth [3, 4]. Therefore, by feeding the animals with an unnatural diet, which may not contain the appropriate nutritional requirements, they could display a false negative/positive response during a test. Most workers, however, have gone down the more traditional route of using detritus to feed detritivores .
The role of allochthonous organic matter (e.g., leaves, wood) in streams and rivers has been extensively documented . Freshly fallen leaves and other plant detritus that enter the water are rapidly colonized by microorganisms, a process referred to as conditioning . There is considerable experimental evidence that shredders fed on detritus show preferences for and survive better on substrata that has been previously colonized by fungi, for example, Bueler . It has been assumed that microbial colonization improves the nutritional quality of detritus through fungi having a differential ability to eliminate plant allelochemicals , fungal synthesis of micronutrients, production of mycotoxins , and/or the ability of detritivores to utilize acquired fungal enzymes . Graca et al.  also demonstrated that G. pulex and A. aquaticus both discriminated between fungal mycelia and either fungally colonized or uncolonized leaf material. However, whereas A. aquaticus fed by scraping the leaf surface, thereby, selectively ingesting fungal mycelia, G. pulex nibbled the leaf, consuming both fungal and leaf matrix.
The food quality of detritus has been defined in terms of chemical (e.g., nitrogen and lignin), physical (e.g., resistance), and biological (e.g., microbial biomass) parameters. High-quality food has a low C : N ratio, low lignin content, low resistance, and high microbial biomass ; therefore, alder would be described as a high-quality food. When G. pulex have been offered the choice between alder (Alnus glutinosa), beech (Fagus sylvatica), oak (Quercus robur), elm (Ulmus glabra), ash (Fraxinus excelsior), and willow (Salix caprea), the alder leaves were ingested at a much faster rate .
In contrast, Willoughby and Sutcliffe  found that the best diet for G. pulex was a mixture of conditioned elm and oak leaves. On this diet, the animals had a growth rate of approximately 150 μg day−1 at 10°C in specimens of less than 16 mg body weight. In larger specimens, the rate apparently increased to about 350 μg day−1. However, workers such as Nilsson  found that, at 15°C, an average of 1928.7 calories were produced from alder leaves g−1 day−1, which is considerably greater than other leaves, for example, beech (197.6 calories were produced from beech leaves g−1 day−1). The growth rate for Nilsson’s smaller G. pulex specimens, which were fed on alder leaves was similar to the rate of 130.8 μg day−1 at 15°C obtained by Willoughby and Sutcliffe  with a diet of oak and elm.
Researchers have previously used artificial  and natural [4, 13] methods to condition leaf material. The aim of this paper is to establish if the macroinvertebrates G. pulex and A. aquaticus prefer a diet of artificially or naturally conditioned alder leaves by undertaking ex situ feeding assays. In addition, the animals preference for conditioned and unconditioned leaf material will be assessed.
2. Materials and Methods
The G. pulex and A. aquaticus used in this study were obtained from a standardised laboratory breeding programme. The breeding programme’s founder population originated from an unpolluted river source. Animals were captured, transported to the laboratory, and maintained under standardised conditions. The specimens were allowed to randomly copulate and the subsequent F1, F2, F3 generations, and so forth were used for experimental purposes .
G. pulex (12–15 mg dry mass) and A. aquaticus (7–10 mg dry mass) males were used in the experiments. 24 hours prior to the test, 300 A. aquaticus were removed from the culturing tank and divided equally between 30, 500 mL sterile plastic pots (with screw lids), which contained 500 mL of deionised water. The animals were maintained under oxygen-depleting conditions without nutritional supplements at 15°C. For 16 hours per day, the animals were illuminated with a fluorescent light (with a specification for freshwater invertebrates), to simulate on a small scale the macroinvertebrates natural climatic conditions. The glow mimicked the thermal warmth and daytime illumination obtained from the sun radiation. The same procedure was also undertaken with 300 G. pulex. Bloor et al.  previously showed that in a deionised water test media (without aeration) both G. pulex and A. aquaticus could survive for several weeks without mortalities.
Alder leaves (Alnus glutinosa) were collected during the autumn fall (from Hillier’s Arboretum, Romsey, UK), air dried, and stored in refuge bags (in a dry location) until required. The leaf material was cut into 1800 squares (length 2.0 cm and width 2.0 cm). 600 squares (1.16 g) were placed in 500 mL of enriched water (Table 1), inoculated with a standard amount of Cladosporium fungus (fungi : leaves, 1 : 20) and incubated for 10 days . 600 squares (1.16 g) were soaked in 500 mL of river water containing 0.50 g of decaying detritus for 10 days (river water and detritus were collected from the River Itchen, Southampton, UK). The remaining squares were saturated in 500 mL of deionised water for 10 days.
The 1800 squares were then air dried for 24 hours and weighed. 120 squares from each treatment were put into the separate 500 mL test pots containing A. aquaticus and fed to the animals (the deionised water was not changed, and aeration was not applied); therefore, each test was replicated 10 times. After 24 hours, the squares were removed, air dried (for 24 hours), and reweighed. The amount of consumed detritus was then calculated by subtracting the final leaf weight from the conditioned weight. The same investigation was then repeated with G. pulex.
The data was analysed using PASW 18 statistical software. Initially, the Kolmogorov-Smirnov test was used to determine normality (). As the data was normally distributed, a parametric paired t-test was applied to establish if there was a significant difference between the initial and final weights of the leaves (). Finally, a general linear model was undertaken to investigate which leaf type was preferred by G. pulex and A. aquaticus.
Application of Kolmogorov-Smirnov test indicated that there was no departure from normal distribution for the G. pulex unconditioned leaves (Z 0.160, P 0.757), natural conditioned leaves (Z 0.211, P 0.385), or artificially conditioned leaves (Z 0.151, P 0.411). The A. aquaticus data was also normally distributed (unconditioned leaves (Z 0.195, P 0.574), natural conditioned leaves (Z 0.163, P 0.621), or artificially conditioned leaves (Z 0.184, P 0.199)).
This enabled application of the parametric paired t-test (), which showed that there was a significant difference between the initial and final weight of unconditioned leaves (Z 8.157, ), natural conditioned leaves (Z 34.259, ), and artificially conditioned leaves (Z 9.918, ) for G. pulex and also A. aquaticus (unconditioned leaves (Z 11.420, ), natural conditioned leaves (Z 66.002, P 0.001), and artificially conditioned leaves (Z 35.146, )).
Finally, a general linear model demonstrated that there was a significant difference between the amount of leaf material consumed by G. pulex and A. aquaticus (Z 23.909, P 0.001), the type of leaf treatment consumed (Z 18.803, ), and the amount each species consumed of each leaf type (Z 136.399, ). Figures 1 and 2 illustrate that G. pulex and A. aquaticus consumed the leaf material in the order of naturally conditioned > artificially conditioned > unconditioned.
Bacteria and fungi are important components of the detritivore diet , G. pulex and A. aquaticus both discriminated between fungal mycelia and either fungally colonized or uncolonized leaf material , which was illustrated by this study. The results clearly demonstrate that both species of macroinvertebrates preferred a diet of conditioned leaf material over unconditioned leaves, with natural conditioning being the favoured conditioning option. On comparing the initial and final weights of the natural and artificially conditioned leaf material, it can be concluded that natural conditioning produced heavier and noticeably softer leaves, which could be attributed to the colonization of micro-organisms. Research has demonstrated that A. aquaticus feed, by scraping the leaf surface, thereby, selectively ingesting fungal mycelia, which would explain why these animals preferred the naturally conditioned leaves . In contrast, G. pulex nibbles the leaf, consuming both fungal and leaf matrix . As such, it might have been expected that the G. pulex would not discriminate between the natural and artificial leaves, but the results of this study showed that natural conditioning was the diet choice for both species.
Studies have shown that G. pulex and A. aquaticus would grow to sexual maturity and reproduce on a diet of decaying leaves . Few studies have measured or compared the rates of growth on different diets, but some authors have claimed that decaying leaves with rich flora of bacteria and fungi are more palatable and support faster growth of G. pulex than leaves without microorganisms .
However, workers, such as Graca et al.,  demonstrated that although the growth of A. aquaticus was reduced when unconditioned leaves were provided, leaf conditioning does not influence G. pulex growth. This is because G. pulex has the ability to compensate for a low-energy uptake by reducing its energy expenditure. The mechanism behind this principle remains unclear but is probably linked to a decline in activity . Whatever the mechanism, the outcome of this difference in response is that reduction in food quality has a greater impact on the energy balance of A. aquaticus than that of G. pulex, resulting in less energy being available. G. pulex may also resort to cannibalism in experimental situations when insufficient/inappropriate nutritional supplements are available , which could hinder a laboratory breeding programme.
When establishing a laboratory breeding programme for ecotoxicological studies, it is important that the animals are maintained in standardised and repeatable conditions. The animals need to remain stress-free or their toxicological response could be manipulated . The animals diet is an important factor in maintaining a healthy and stress-free population, and consequently, it is important to keep the animals in the most natural environment as possible. By providing a diet that mimics their natural food source and contains the appropriate nutritional requirements for growth and reproduction, the animals would be representative of wild stocks during ecotoxicological studies. Therefore, the author would suggest that naturally conditioned alder leaves are an excellent diet choice for G. pulex and A. aquaticus populations within a laboratory breeding programme. The presented research supports the use of the feeding methodology outlined in Bloor , in order to breed and maintain healthy populations of both macroinvertebrates during a breeding programme.
Bloor  discussed that abscised alder leaves (Alnus glutinosa) should be collected during the autumn fall (from one tree), air dried, and stored. As such, the food source would be standardised as all the leaves were collected from the same tree on the same day. 10 L of river water and a handful of organic detritus should be collected from an unpolluted source and transferred to the laboratory in a lidded plastic container. On return to the laboratory, the water and detritus should be poured into a 15 L plastic box (the box should not be sealed with a lid). Handfuls of the precollected alder leaves should be submerged in the water and mixed with the precollected organic detritus (no precise measurements), which would inoculate the alder leaves with bacteria and fungus. The leaves should be conditioned for at least 10 days. After that time and when required, leaves should be extracted from the box and placed in the aquariums (excess liquid should be squeezed from the leaves to reduce the level of organic enrichment applied to the water). Additional air-dried leaves should then be immersed in the conditioning box to replace the utilised ones.
The leaves should be liberally scattered in the culture and rearing aquariums, to fulfil the animals nutritional requirements and replaced at regular intervals (enough leaves to cover the aquarium floor to a depth of approximately 50 mm). The juveniles should, however, be supplied with conditioned alder leaves for shelter and grazing but also fed upon adult faeces that should be syringed from the culture aquariums (when required), until the animals can feed entirely upon conditioned leaves (after about 25 days).
In summary, when undertaking a laboratory breeding programme with G. pulex and A. aquaticus, naturally conditioned alder leaves would be the recommended food source. As such, a feeding methodology was outlined that could be utilised during a breeding programme. The author would recommend that a priority for future research would be to investigate if the diet/health of laboratory populations of G. pulex and A. aquaticus could be improved by feeding a mixed diet.
L. G. Willoughby and D. W. Sutcliffe, “Experiments on feeding and growth of the amphipod Gammarus pulex (L.) related to its distribution in the River Duddon,” Freshwater Biology, vol. 6, no. 6, pp. 577–586, 1976.View at: Google Scholar
M. C. Bloor, “Animal standardisation for mixed species ecotoxicological studies: establishing a laboratory breeding programme of Gammarus pulex and Asellus aquaticus,” Zoologica Baetica, vol. 21, pp. 179–190, 2010.View at: Google Scholar
N. H. Anderson and J. R. Sedell, “Detritus processing by macroinvertebrates in stream ecosystems,” Annual Review of Entomology, vol. 24, pp. 351–377, 1979.View at: Google Scholar
S. W. Gollady, J. R. Webster, and E. F. Benfield, “Factors affecting food utilization by a leaf shredding aquatic insect: leaf species and conditioning time,” Ecography, vol. 6, no. 2, pp. 157–162, 1983.View at: Google Scholar
C. M. Bueler, “Feeding preference of Pteronarcys pictetii (Plecoptera: Insecta) from a small, acidic, woodland stream,” The Florida Entomologist, vol. 67, no. 3, pp. 393–401, 1984.View at: Google Scholar
G. A. Rosenthal and D. H. Janzen, Herbivores: Their Interaction with Secondary Plant Metabolites, Academic Press, New York, NY, USA, 1979.
M. A. S. Graca, L. Maltby, and P. Calow, “Comparative ecology of Gammarus pulex (L.) and Asellus aquaticus (L.) population dynamics and microdistribution,” Hydrobiologia, vol. 281, no. 3, pp. 155–162, 1994.View at: Google Scholar
T. M. Iversen, “Ingestion and growth in Sericostoma personatum (Trichoptera) in relation to the nitrogen content of ingested leaves,” Oikos, vol. 25, no. 3, pp. 278–282, 1974.View at: Google Scholar
L. M. Nilsson, “Energy budget of a laboratory population of Gammarus pulex (Amphipoda),” Oikos, vol. 25, no. 1, pp. 35–42, 1974.View at: Google Scholar
C. P. McCahon and D. Pascoe, “Culture techniques for three freshwater macroinvertebrate species and their use in toxicity tests,” Chemosphere, vol. 17, no. 12, pp. 2471–2480, 1988.View at: Google Scholar
F. Barlocher and B. Kendrick, “Dynamics of the fungal population on leaves in a stream,” Journal of Ecology, vol. 62, pp. 761–791, 1974.View at: Google Scholar
M. A. S. Graca, Observations on the feeding biology of two stream-dwelling detritivores: Gammarus pulex (L.) and Asellus aquaticus (L.), Ph.D. thesis, University of Sheffield, South Yorkshire, UK, 1990.