Trash fish is common feed for caged fish in marine aquaculture. Most feed is not eaten and enters the water surrounding fish farms. The organic matter in trash fish is a nutrient source contributing to aquatic eutrophication impacts such as algal blooms and low oxygen. The objective of this study was to examine whether phytoplankton utilized organic matter of trash fish directly. Fifteen microalgae species were cultured in the medium made of open oceanic water with f/2 enrichment (Treatment A, f/2 medium), with only silicate added (Treatment B, for diatom species) and with fish tissue+silicate (Treatment C). Four species grew significantly faster on fish tissue than in f/2 medium, another 4 species had similar growth rates between the two treatments (A and C). Growth rates of Pyramimonas sp. on different amount of fish tissue appeared to increase initially with increasing added fish tissue weight, reached a maximum at an intermediate amount of fish tissue additions. Time course of batch culture of Chaetoceros curvisetus showed that the species was capable of utilizing organic nitrogen released from fish tissue and grew well. Dissolved organic nitrogen (DON) was lower in the batch culture with algae and fish tissue than that with fish tissue but without algae in the medium, which indicated a direct utilization of DON during algal growth. These results indicated that fish tissue could be a direct nutrient source to support phytoplankton growth and coastal management needs to pay attention to organic contamination from trash fish.
Fish farms in coastal waters contribute to a large proportion of seafood production. Fish farms especially in east-south Asian countries use trash fish to feed caged fish and hence, discharge a large amount of nutrients to costal waters. When Epinephelus areolatus was cultured in an open-sea-cage farm, only 8.6% of trash fish nitrogen was harvested in caged fish, indicating that 91.4% of trash fish nitrogen was lost into the surrounding water (Leung et al., 1999). In western Sweden, only 21 to 22% of the total carbon input of feed to the farm was recovered in harvest, which indicated that 78 to 79% of total trash fish carbon (or 878 to 952 kg C per tonne of fish produced) was lost to the aquatic environment (Hall et al., 1990). It was reported that 75–80% nitrogen in fish diet was lost to environment and 90% of that loss entered the water column (Kaspar et al., 1988). It was estimated that 1,325 tons of nitrogen and 250 tons of phosphorus entered sea water from every 10,000 tons of fish production yearly (Islam, 2005). Obviously the trash fish in aquaculture was an important contamination source to coastal waters in terms of eutrophication.
The water near fish farms is exposed to high organic contamination before the organic matter is decomposed and ultimately re-mineralized into inorganic nutrients. The main organic contamination includes organic protein in trash fish body and farmed fish feces. The un-eaten trash fish is broken down to small pieces, detritus, dissolved organic matter (including dissolved organic nitrogen [DON]; dissolved organic phosphorus [DOP]; dissolved organic carbon [DOC] and is eventually decomposed by bacteria to dissolved inorganic nutrients. It has been reported that some algal species could use dissolved organic matter (DOM) during its decomposition (Berg et al., 1997; Mulholland et al., 2002; Lomas et al., 2004; Heil, 2005; Sun et al., 2006). Mixotrophic nutrition helps to supplement nitrogen, carbon and phosphorus requirements for phytoplankton species when inorganic nutrients or light are not sufficient (Bockstahler and Coats, 1993; Jacobson and Anderson, 1996). However, it was not clear whether DOM during the course of trash fish decomposition was utilized by algae directly. Particularly when different phytoplankton species have different capacity of utilizing organic matter, trash fish waste may result in a change in phytoplankton species composition in the surrounding water. The objective of our study was to examine direct utilization of dissolved organic nitrogen by growing phytoplankton species in the medium with trash fish tissue.
Materials and Methods
Culture medium and culture conditions
Oceanic water (salinity = 35.5) was collected at the surface in the open water of the South China Sea (117°E, 18°N), which is very oligotrophic. The collected ocean water was filtered by polycarbonate membrane (47 mm, 0.2 μm) and autoclaved for 20 min under 120°C and 0.1 Mpa before being used as the culture medium. The algal species as the stock were cultured in f/2 medium (Guillard, 1975) before the experiments started. The culture temperature, light intensity and light period were 22.5 ± 0.5°C, 40 μmol photons m−2s−1 and 14L:10D, respectively.
Golden Threadfin Bream (Nemipterus virgatus), a type of trash fish used widely as feed for fish aquaculture in China, was purchased from the marketplace. The flesh under skin was cut into pieces, and a piece was wrapped in a 0.2 μm polycarbonate membrane filter and was placed in the culture tube bottom. The polycarbonate membrane was used to filter out bacteria from the wrapped fish tissue pieces so that dissolved organic matter decomposed from the fish tissue pieces could leak into the culture medium without bacteria.
Three experiments were conducted to achieve the objective.
Growth rates vs. different fish tissue weight
Preliminary experiments showed that Pyraminonas sp. grew well on fish tissue, and therefore it was used to examine the relationship between growth rate and fish tissue weight. The relationship would also allow us to select a suitable weight of trash fish for cultures of other algal species in the other two experiments as discussed in Species difference in growth rates and Direct utilization of dissolved organic nitrogen. Batch cultures were set up in 50 ml columnar glass test tubes containing the oceanic water with different amount of fish tissue wrapped with 0.2 μm polycarbonate membrane. Seven treatments (Table 1) were designed with each treatment in three replicates. In vivo fluorescence was used to measure changes in algal biomass and to estimate algal growth rate. In general, for the same species, the in vivo fluorescence yield was linearly related to cell number (Wood et al., 2005). In vivo fluorescence has been used to estimate growth of many species of dinoflagellates (Uchida et al., 1999; Doblin et al., 2000), diatom and picoplankton (Maldonado and Price, 2001; Wood et al., 2005). In this study, in vivo fluorescence was measured twice daily (14:00 and 20:00 h) on a fluorometer, TD-700 (Tuner Designs).
Growth rate in the exponential growth period was calculated based on the equation μ = (ln(Nt)-ln(N0))/t, where Nt and N0 were in vivo fluorescence at time t and the initial time, respectively (Wood et al., 2005). The same calculation method for μ was used throughout this study. One-way ANOVA was used to test the effect of fish tissue weight on algal growth rate. Duncan's multiple comparison test was used to examine the significant difference (P < 0.05) in μ among the treatments when the result of one-way ANOVA was significant. The equation:
Species difference in growth rates
Batch cultures of 15 microalgal species, including 4 diatom species, were grown in 50 ml columnar glass tubes for examining species differences in growth rates. For the diatom species, three treatments were used. Treatment A contained f/2 medium. Treatment B contained only Si (Na2SiO3). Treatment C (Fish group) contained fish tissue+Si. The final concentration of Si was 42.4, 26.5 and 53 μmol l−1 in Treatments A, B and C, respectively. For the non-diatom species, only Treatments A and C (Fish group) were used. The Si concentrations, 26.5 and 53 μmol l−1, did not significantly affect the 15 species’ growth in our preliminary study. We did not test the optimum amount of fish tissue for each species’ maximum growth rate. We used the optimum (210 mg l−1, see Results and Discussion) fish tissue weight obtained from the growth of Pyraminonas sp. for the batch cultures of the 15 species. The statistical data analysis (One-way ANOVA and Duncan's multiple comparison test) for the algal growth rates was the same as described above.
Direct utilization of dissolved organic nitrogen
A diatom species, Chaetoceros curvisetus, was used for the experiment on direct utilization of DON as the species had relatively high growth rate as described above (see Results and Discussion). Batch cultures of the treatment with both algal cells and fish tissue and control with fish tissue only in triplicate of 3 L glass flasks were set up. Both the experimental treatment and control had 2 L seawater with 106 μmol l−1 Si being added in each flask. A piece of fish tissue (0.5 ± 0.02 g) was wrapped by 0.2 μm polycarbonate membrane filter and was put on the bottom of the flasks. Water samples (50 ml) were collected daily for measurements of in vivo fluorescence and DON, with no new medium being added. The DON samples were filtered by GF/F membrane. Total dissolved nitrogen (TN) was determined by TOC-VCPH (SHIMADZU) and the dissolved inorganic nitrogen (DIN: NO3, NO2, NH4) by SAN System (Skalar, 1995, The Netherlands). The DON was the difference between TN and DIN. When a comparison is made between the treatment with algae+fish tissue, and the treatment with fish tissue only, the DON concentration in each treatment was normalized to the initial weight of fish tissue in the treatment (the unit is μmol DON l−1 (g fish tissue)−1).
A model was used to describe the release of DON vs time: DON = DONmax T/(Kt + T), where DONmax was the maximum of DON concentration and T was time, and Kt was time when DON reached half the maximum, respectively. The utilized DON by the algal cells was the differences in DON concentrations between the control (fish tissue only) and the culture of C. curvisetus cells with fish tissue, and the accumulative differences during a number of days would be the total DON utilized by the algal species during the incubation days. Linear regression was used to simulate the relationship between in vivo fluorescence and uptake of DON. Student's t-test was used to test the difference of DON concentration between the treatment and control.
Results and Discussion
Effect of different amount of fish tissue on algal growth
Growth rates of Pyramimonas sp. with different fish tissue weight were shown in Table 1. Growth rates were significantly different among the seven fish weights (treatments) (P < 0.001). The lowest growth rate was 0.27 ± 0.12 d−1 without fish tissue (Control). This result was similar to another study which reported that low nutrient concentration caused low growth rate of Prorocentrum micans (Weng et al., 2007). Growth rates increased initially with tissue weight and reached the highest growth rate (Treatment E) of 0.94 ± 0.06 d−1, and remained in the similar level (no significant differences from the highest growth rate) when tissue weight further increased.
The relationship between growth rate and fish tissue weight follows the hyperbolic function as in Equation (1), which is similar to the Michael-Menton's equation at a steady state:
Different growth rates among the 15 algal species
Growth rates of the 4 algal species (Pavlova viridis, Dunaliella tertioleta, Prymnesium patelliferium, Nitzschia closterium) grown in the medium with fish tissue were significantly higher than that of Treatment A (f/2) without fish tissue (Table 2). Another 4 species (Phaeodactylum tricornutum, Chaetoceros curvisetus, Platymonas helgolandica, and Heterosigma akashiwo) had similar growth rates between treatment with fish tissue and treatment A (f/2) without fish tissue. Growth rates of the 3 diatom species (Phaeodactylum tricornutum, Skeletonema costatum, and Nitzschia closterium) in the fish tissue treatment were significantly higher than Treatment B. In the treatment with fish tissue, Chattonella marina did not grow and Chattonella ovate had the second lowest growth rate. The highest growth rate was displayed by Phaeodactylum tricornutum. The averaged growth rate in the treatment with fish tissue was 0.56 ± 0.19(n = 4), 0.40 ± 0.23(n = 3), 0.50 ± 14(n = 3) and 0.25 ± 0.22(n = 4) d−1 for Bacillariophyta, Haptophyta, Chlorophyta and Chrysophyta species, respectively.
Higher growth rates on fish tissue than on f/2 medium suggested that dissolved organic matter might have stimulated more growth of these species. Similar growth rate between the treatment with fish tissue and Treatment A without fish tissue indicated that these species were capable of utilizing nutrients (organic+inorganic) released by fish tissue. Differences in growth rates among the species reflected species’ different ability in using nutrients released from fish tissue. The result indicated fish could be a direct nutrient source for phytoplankton growth in fish farm areas, as there is always some wasted trash fish in surrounding waters of fish farms. Species differences in utilizing trash fish could contribute to algal species composition change and blooms of a particular species.
Direct utilization of dissolved organic nitrogen
The growth trend of C. curvisetus is shown in Figure 2. The species grew slowly (lag phase) during days 0–8 (in vivo fluorescence 4.5–5.8), grew exponentially during days 8–24 (in vivo fluorescence 5.8–237) and kept relatively constant growth rate (stationary phase) during days 24–32 (in vivo fluorescence 237–295.8). Growth rate was 0.23 ± 0.01 d−1 in the exponential phase.
The highest ratio of DON to total dissolved nitrogen was 93.0% in the Control. The trend of change in DON concentrations is shown in Figure 3. DON concentration in the Control (with fish tissue but without algae) increased gradually and reached 1,909.2 ± 221.3 on day 24.8 and 1788.3 ± 949.1 μmol l−1 g−1 on day 31.8, respectively. However, when C. curvisetus was grown in the medium, DON concentration decreased. The decrease in DON concentration between the two treatments was obvious after day 12, with DON decreasing to 1167.4 ± 63.6 on day 24.8 and 1,214.7 ± 3.7 μmol l−1 g−1 on day 31.8, respectively. The difference on day 24.8 between the treatment and Control was significant (P < 0.05), which indicated the utilization of DON 741.8 μmol l−1 g−1 accounting for 38.85% of the Control.
The results from the DON release equation between the control and the culture were shown in Table 3. Both DONmax and Kt in the culture with C. currrisetus were significantly lower (P < 0.05) than the Control. Those differences between Control and C. currrisetus suggested DON was utilized directly by this species. The relationship between in vivo fluorescence and used DON was linear (Figure 4), indicating that large phytoplankton biomass could be supported by fish released DON. It was speculated that heterotrophic nutritional mode could be started or mixtrophic nutritional mode could be strengthened when phytoplankton were in high DON concentration environment. It was observed that amino acids supported growth of P. tricornutum (Saks et al., 1976) and Pfiesteria sp. (Glibert et al., 2006). In fact, it is known that some alga can use DON in various pathways. A complex in Photosystem II had the L-amino acid oxidase activity, which enabled the algae Synechococcus sp. to utilize DON (Meyer and Pistorius, 1987). Another enzyme on cell surface of some marine phytoplankton, particularly the genus Pleurochrysis, oxidized many L-amino acids to produce H2O2, NH3, and an α-keto acid extracellularly. The NH4+ was subsequently taken up and used for growth (Palenik and Morel, 1990). It is likely that an increase in trash fish DOM can increase cell density of phytoplankton in the eutrophication water directly.
Mao et al. (2007) reported that the species (C. curvisetus) had a growth rate of 0.146 d−1 under optimal factors (temperature 20°C, light intensity 78.12 μmol photon m−2 s−1 and salinity 25 g l−1). The growth rate in our study was higher than this reported value, suggesting stimulation of growth of this species by fish organic matter.
Organic nutrients were released from fish tissue gradually, which was similar to nutrients release from some sediment or dissolved organic matters discharging from land and rivers (Hu et al., 2001; Moller and Riisgard, 2007). Algal blooms could be caused by high concentrations of DOM in aquaculture area, river mouth and the other euthrophic zones. Relationship between DOM and algal growth should be further studied.
There appeared to be an optimal fish amount for algae to reach a maximum growth rate; however, there were differences in growth rates among the algal species in utilizing fish tissue. Some algae could utilize fish tissue but others could not; while dissolved organic nitrogen could be utilized directly by algae.
Fish feed will increase dissolved organic matter in fish farm areas. Fish-DOM could be a direct contamination source for phytoplankton growth in fish farm area. This has great implications for environmental assessment and management.
The study was supported by SCSIO Youth Research Grant (SQ200815), and NSFC (41176129, U1033002 and ARC DP110103155).