Lipid biosynthesis from [14C]-glucose in the coral Montipora digitata

ABSTRACT:     Glucose has been implicated in functioning as a form of carbon translocated from symbiont zooxanthellae to the host coral cell. The present paper describes the lipid biosynthesis from [¹⁴C]-glucose in the coral tissue. To study the incorporation of [¹⁴C]-glucose into lipids, the branch tips of the coral Montipora digitata were incubated with [¹⁴C]-glucose or another radiolabeled substrates. The lipid biosynthesis from [¹⁴C]-glucose was dependent on light, and was decreased by dark conditions or by photosystem II inhibitor, 3-(3,4-dichlorophenyl)-1,1-dimethylurea. Of the lipid classes, the light dependency was more pronounced with the biosynthesis of triacylglycerols (TG) and wax compared with phospholipids. Examination of [¹⁴C]-label distribution in the glycerolipids suggested that [¹⁴C]-glucose supplied mainly the fatty acid moiety of newly synthesized TG, while [¹⁴C]-glucose provided evenly the fatty acid moiety and the glycerol skeleton of phospholipids. The comparison of [¹⁴C]-labeling of lipid from host coral tissue and symbiont zooxanthellae suggested that [¹⁴C]-glucose entered the coral cell and was processed in parallel in the zooxanthellae and host cells. Furthermore, the coral cells used various [¹⁴C]-labeled sugars for lipid synthesis with similar lipid labeling profile as was the case for glucose. The current study thus supports the view that the low-molecular-weight compound, sugars and amino acids, once translocated from zooxanthellae to host cell were metabolized toward lipogenesis as well as glycerol.     

Use of stable isotopes to distinguish farmed from wild Atlantic salmon, Salmo salar

Abstract –  Stable isotopes of carbon and nitrogen ( δ ¹³C and δ ¹⁵N) were examined in wild and aquaculture origin Atlantic salmon, Salmo salar , to evaluate their utility to identify escaped farmed fish. Samples of muscle tissue obtained from wild Conne River, Newfoundland, salmon were significantly more enriched in nitrogen ( δ ¹⁵N: mean = 12.75; SD ± 0.38‰) but depleted in lipid corrected carbon ( δ ¹³C': mean = −20.51; SD ± 0.23‰) by comparison with aquaculture specimens obtained from Bay d'Espoir, Newfoundland ( δ ¹⁵N = 10.96 ± 0.19‰;  δ ¹³C' = −19.25 ± 0.17‰) resulting in a complete separation of the two groups. Aquaculture specimens differed in δ ¹³C' from analyses of commercial salmon diet by 0.24‰, within the enrichment range associated with trophic transfers, while the δ ¹⁵N values in salmon muscle were enriched by 5.01‰. Although differences occurred in direct comparisons of white muscle and adipose tissue ( N  = 49), the average δ ¹³C' and δ ¹⁵N signatures varied in absolute amounts by only 0.5‰, supporting the use of adipose tissue as a nonlethal means to determine isotopic signatures of Atlantic salmon.     

Effects of lipid extraction on stable carbon and nitrogen isotope analyses of fish tissues: potential consequences for food web studies

Abstract –  We examined whether solvent-based lipid extractions, commonly used for stable isotope analysis (SIA) of biota, alters δ ¹⁵N or δ ¹³C values of fish muscle tissue or whole juvenile fish. Lipid extraction from muscle tissue led to only small (<1‰) isotope shifts in δ ¹³C and δ ¹⁵N values. By contrast, ecologically significant shifts (+3.4‰ for δ ¹³C and +2.8‰ for δ ¹⁵N) were observed for whole juvenile fish. Sample variance was not affected by lipid extraction. For tissue-specific SIA, two sample aliquots may be required: a lipid-extracted aliquot for stable carbon isotope analysis when differing lipid content among tissues is a concern, and a nonextracted aliquot for δ ¹⁵N determination. Whole organism SIA is not recommended because of the mix of tissues having different turnover times; for very small fish, we recommend that fish be eviscerated, decapitated, and skinned to minimise differences with samples of muscle tissue.     

Natural abundance of 15N and 13C in fish tissues and the use of stable isotopes as dietary protein tracers in rainbow trout and gilthead sea bream

For developing efficient diets, two sets of experiments examined whether the use and allocation of dietary protein can be traced by labelling with stable isotopes (¹⁵N and ¹³C) in two culture fish ( Oncorhynchus mykiss and Sparus aurata) . In the first experiment, natural abundance and tissue distribution of these isotopes were determined, by measuring the δ¹³C and δ¹⁵N values by isotopic ratio mass spectrometry, in fingerlings (14–17 g) adapted to diets differing in the percentage of fish meal replacement by plant protein sources. For both species, δ¹⁵N and δ¹³C were greater in tissues with higher protein and lower lipid content. Delta ¹⁵N of diets and tissues decreased as replacement increased, suggesting δ¹⁵N can be used as a marker for dietary protein origin. The ¹⁵N fractionation (δ¹⁵N fish − δ¹⁵N diet) differed between groups, and could thus be used to indicate protein catabolism. In the second experiment, fish (75–90 g) of each species ingested a diet enriched with ¹⁵N-protein (10 g kg⁻¹ diet) and ¹³C-protein (30 g kg⁻¹ diet). These proportions were suitable for determining that the delta values of tissue components were high enough above natural levels to allow protein allocation to be traced at 11 and 24 h after feeding, and revealed clear metabolic differences between species.     

14C-glucose injection in Atlantic cod, Gadus morhua, metabolic responses and excretion via the gill membrane

In the present study cod were separately placed in fish metabolic chambers. Ten individuals were injected with a dose of 1 μCi ¹⁴C-glucose, in addition to 1 g pure glucose per kg live weight, and 10 individuals were kept under the same conditions and sham injected with 9 g L⁻¹ (0.9%) sterile saline solution.
Of the injected ¹⁴C-glucose, 3% was recovered in plasma, 2.9% in liver, 1.7% in red muscle, 18.2% in white muscle, 0.2% in heart, 1.9% in gills and 12.1% in gill-surrounding water. Of the 12.1% in gill-surrounding water, 9.1% was found to be in the form of CO₂, leaving 3% to pure glucose and lactate. These results indicate a relatively high production of CO₂ from glucose showing a utilization for energy, and a possible route of excretion of excess glucose via the gill membrane.
Activity of ¹⁴C-glucose on a concentration basis showed in descending order the following organs to be metabolically active in glucose utilization: gills > plasma (as a transporter) > heart > red muscle > liver > white muscle.
Of the total amount of ¹⁴C-glucose injected, only 0.3% was recovered in the liver lipid fraction, and mainly as triacylglyceroles; minor ¹⁴C-activity was found in mono- and diacylglyceroles and free fatty acids. These results show very low activity of de novo syntheses of lipid from injected glucose in the cod.     

In vitro hydroxylation of vitamin D3 and 25-hydroxy vitamin D3 in tissues of Atlantic salmon Salmo salar, Atlantic mackerel Scomber scombrus, Atlantic halibut Hippoglossus hippoglossus and Atlantic cod Gadus morhua

The in vitro metabolism of ¹⁴CD₃ and ³H25OHD₃ was investigated in different tissues from Atlantic salmon Salmo salar , Atlantic mackerel Scomber scombrus , Atlantic halibut Hippoglossus hippoglossus and Atlantic cod Gadus morhua . The tissues analysed were liver, kidney, head kidney, gills, spleen and intestine. The metabolites were extracted in methanol–chloroform and separated by normal-phase high-pressure liquid chromatography (HPLC) followed by scintillation counting. Identification of the metabolites was by comigration with standards on normal and reversed-phase HPLC systems and by protein-binding assays. All tissues from all species analysed produced hydroxylated derivatives identified as 25OHD₃, 24,25(OH)₂D₃ and 1,25(OH)₂D₃. In addition, some unidentified derivatives were recorded, one probably being 25,26(OH)₂D₃. Organs producing great amounts of one metabolite also produced considerable amounts of the other possible derivatives, suggesting a lower degree of specificity in fish organs than in human organs. The predominating metabolite was 24,25(OH)₂D₃ in all organs from salmon and mackerel during incubation with ¹⁴CD₃ and within most organs from all species during ³H25OHD₃ incubation. The latter observation probably results from the need for decreasing rather than increasing the calcium absorption in these species, which live at least some periods of life in a marine environment.     

Pharmacokinetics of Dietary 13C-Labeled Docosahexaenoic Acid and Docosapentaenoic Acid in Red Sea Bream Chrysophrys major

The objectives of this study were to investigate: 1) the pharmacokinetics of dietary docosahexaenoic acid (DHA) and docosapentaenoic acid (DPA) using ¹³C-labeled fatty acids; 2) the interorgan transport of DHA in the red sea bream by monitoring the DHA level of several organs; and 3) the relationship between the plasma DHA level and optimum dietary DHA level in the plasma of the red sea bream Chrysophrys major . For this purpose, a mixture of 38.5% of [¹³C]DHA, 8.5% of [¹³C]DPA, and 4.2% of [¹³C]palmitic acid were given to the red sea bream at dose level of 8.0, 16.0, and 47.9 mg/kg by a single oral administration. For [¹³C]DHA, the maximum plasma concentration (t_max) occurred at 2.00–3.00 h after the oral administration. The peak plasma concentration (C_max) and the area under the plasma concentration-time curve to 24 h (AUC_0-24 for [¹³C]DHA level linearly increased with respect to dosage. [¹³C]DHA appeared in each organ (plasma, erythrocyte and the fat body of the orbit, liver, intestine, skin, brain, heart and muscle) at 0.5 h and was observed until 24 h. From the values determined for the pharmacokinetic parameters, the range of the effective plasma DHA level for normal growth of the red sea bream was suggested to be between 21.0 and 40.3 μg/mL. For [¹³C]DPA, the AUC_0-24 and C_max values also linearly increased with the dosage, but t_max did not depend on it.     

Carbon source and trophic position of pelagic fish in coastal waters of south-eastern Izu Peninsula,Japan, identified by stable isotope analysis

ABSTRACT:   It is important to clarify trophic dynamics in marine ecosystems for management of the fishing ground. Organic carbon sources and trophic position of pelagic fishes in the coastal waters of the south-eastern Izu Peninsula, Japan, were examined on the basis of carbon and nitrogen stable isotope distributions. The δ¹³C of the fishes was mostly distributed from −19 to −16‰ for nektonic fishes (13 species of adults and immatures) and planktonic fishes (10 species of larvae and juveniles), close to the δ¹³C values of particulate organic matter and planktonic decapods. These δ¹³C signatures for the inhabitants of the water column were in contrast with the high δ¹³C values (mainly −16 to −13‰) for demersal fishes of Scorpaeniformes and benthic polychaetes collected in the surf zone. These results indicate that nektonic and planktonic fishes depend on phytoplankton for carbon supply. The δ¹⁵N signatures suggest that the trophic position ranged 3.1–4.5 for the nektonic fishes and 2.9–3.7 for the planktonic fishes, premised on trophic level 3 for larval Japanese anchovy Engraulis japonicus . Thus, planktivorous fishes should be mainly assigned to trophic levels 3 and 4 in this area.     

Evaluation of shrimp polyculture system in Thailand based on stable carbon and nitrogen isotope ratios

ABSTRACT: To quantify the contribution by cocultured animals to waste assimilation in an intensive shrimp farm in Thailand, the food web structures of the macrobenthos in a reservoir pond, a shrimp culture pond and water treatment ponds were examined using the stable C and N isotope ratio technique. Seawater for aquaculture was drawn from a creek, and stored in a reservoir pond, used for farming the banana prawn Fenneropenaeus merguiensis in culture ponds, and then recycled through treatment ponds where the green mussel Perna viridis was cultured to remove organic wastes discharged from the farming. The clam worm Nereididae sp. and the mud creeper Cerithideopsilla cingulata in the culture pond had δ ¹³C values of −21.0‰ and −18.4‰, respectively, suggesting that shrimp feed (mean δ ¹³C = −20.7‰) was the main food source for these species. The δ ¹³C analysis also suggested that sediments (−23.7‰) in the reservoir pond and particulate organic matter (POM) (−24.0‰) and/or sediments (−25.0‰) in the treatment pond supplied carbon for most macrobenthic animals. However, green mussels in the treatment pond had a mean δ ¹³C value of −20.5‰, suggesting that shrimp feed was the main food source for this species.     

Inhibition of post-mortem muscle softening following in situ perfusion of protease inhibitors in tilapia

ABSTRACT:     A method of introducing protease inhibitors into fish muscle through the bulbus arteriosus was developed using an in situ perfusion technique. Perfusion efficiency was initially tested using eosin and [³⁵S]-methionine. Visible fluorescence was observed in the gill, liver, intestine and dorsal muscle of the eosin-treated tilapia, and the occurrence of eosin in the blood vessels of the dorsal muscle was confirmed under a fluorescence stereoscopic microscope with ultraviolet light. The radioactivity of [³⁵S]-methionine was taken into the dorsal muscle and liver at a concentration of 7.8 Bq/g and 70.2 Bq/g, respectively, after perfusion with 1000 Bq/mL solution. Using the perfusion technique with four kinds of protease inhibitors dissolved in physiological saline, the type of proteases implicated in the post-mortem muscle softening in tilapia (867 ± 195 g, n  = 10/protease inhibitor) was investigated. After the perfusion of leupeptin (serine and cysteine protease inhibitor), benzyloxycarbonyl-Val-Ala-Asp-fluoromethylketone (Z-VAD-fmk; caspase inhibitor), chymostatin (serine protease inhibitor) and ο -phenanthroline (metalloprotease inhibitor), the breaking strength of the perfused muscle was measured as a parameter of the meat toughness and compared with that of the control fish, which were perfused with physiological saline only. The reduction of breaking strength during storage was inhibited by the perfusion of leupeptin and Z-VAD-fmk.     

Mitochondrial and peroxisomal β-oxidation capacities in various tissues from Atlantic salmon Salmo salar

In order to investigate the capacities of different tissues to oxidize fatty acids, total β-oxidation (mitochondrial and peroxisomal) of [1–¹⁴C]palmitoyl-CoA was determined in liver and red- and white muscle from adult and juvenile Atlantic salmon Salmo salar. By including potassium cyanide (KCN) in the assay medium, it was possible to differentiate between mitochondrial and peroxisomal β-oxidation capacities. Mitochondrial β-oxidation dominated in all tissues except in livers from juvenile fish where the peroxisomal β-oxidation dominated. In general, the red muscle possesses the highest fatty acid oxidation capacity, however, by taking into consideration the fact that white muscle occupies approximately 60% of the total body weight, this study demonstrates that the white muscle is an important tissue in the overall fatty acid catabolism.     

Physiological responses of juvenile wedge sole Dicologoglossa cuneata (Moreau) to high stocking density

Physiological responses to a high stocking density were tested in juvenile wedge sole Dicologoglossa cuneata (Moreau). Fish were kept at low (1 kg m⁻²), medium (3 kg m⁻²) and high (9 kg m⁻²) stocking densities for 22 days. No differences in the weight, length, survival and hepatosomatic index were observed among treatments. Basal plasma cortisol and osmolality were found to be directly and positively related to stocking density. A mild increase in plasma glucose was seen at medium density, and plasma protein was elevated at medium and high densities. The liver glucose and glycogen content was inversely related to stocking density. The liver triglyceride level was significantly elevated at the highest density, and the α-amino acid content decreased at the highest density. In muscle, glucose levels were significantly higher in fish kept at the lowest density; the α-amino acid content was elevated in fish kept at high density. In conclusion, plasma cortisol levels indicated an increasing stress level depending on the culture density, but significant changes in energy reserves did not occur in tissue (mainly liver and muscle glycogen and glucose reserves were significantly affected).     

Ecological and genetic differentiation among the Arctic charr of Lake Aigneau, Northern Québec

Abstract –  Two modal size groups of sexually mature Arctic charr ( Salvelinus alpinus ) differing in shape and found at different depths in Lake Aigneau in the Canadian sub-Arctic are described and tested for genetic and ecological differentiation. Forms consisted of a small littoral resident, mean size 21.7 cm, and a large profundal resident, mean size 53.9 cm. Mitochondrial DNA analysis indicated that seven of eight haplotypes were diagnostic for either the littoral or profundal fish, with 66.6% of the variation being found within form groupings. Pairwise tests of microsatellite data indicated significant differences in nine of 12 loci and a significant difference between the forms across all tested loci. Molecular variation was partitioned to 84.1% within and 15.9% between forms and suggestive of either restricted interbreeding over time or different allopatric origins. Stable isotope signatures were also significantly different, with the profundal fish having higher δ¹³C and δ¹⁵N values than the littoral fish. Overlap and separation, respectively, in the range of form δ¹³C and δ¹⁵N signatures indicated that carbon was obtained from similar sources, but that forms fed at different trophic levels. Littoral fish relied on aquatic insects, predominantly chironomids. Profundal fish were largely piscivorous, including cannibalism. Predominantly empty stomachs and low per cent nitrogen muscle-tissue composition among profundal fish further indicated that the feeding activity was limited to the winter when ice-cover increases the density of available prey at depth. Results provide evidence of significant differences between the modal groups, with origins in both genetics and ecology.     

Spring viraemia of carp virus (SVCV): comparison of immunoperoxidase, fluorescent antibody and cell culture isolation techniques for detection of antigen

Abstract. The sensitivity of the immunoperoxidase (IP) and fluorescent antibody (FA) techniques applied to frozen sections of organs of carp infected with spring viraemia virus (SVCV) was similar, both in respect of the intensity of the reaction and in the detection rate of the antigen. Seven days after a waterborne infection both methods detected the antigen in the kidneys and to a lesser extent in the liver and spleen. Quantification of virus gave a titre of 10^2.8 to 10^5.5 TCID₅₀/g of kidney, whereas the agent could not be isolated from liver and spleen tissue. In contrast, at 24¹/₂ days post–infection the viral antigen could be readily detected in liver and spleen but only occasionally in the kidneys using the IP and FA techniques. At this time, liver and spleen showed infectivity titres of 10^3.8 to 10^7.2 TCID₅₀/g tissue, whereas the kidneys were found to be free of infective virus. It is concluded that using the IP and FA techniques SVCV can be detected most frequently in the kidneys from 7 to 14 days post–infection and in the liver and spleen from 14 to 24 days post-infection.     

Stable carbon and nitrogen isotopic evidence for ecotonal coupling between boreal forests and fishes

Abstract— As a result of water turbulence effects on boundary layer diffusion resistance and carbon isotopic discrimination, the δ¹³C values (ratios of ¹³C: ¹²C) of attached algae may often overlap those of terrestrial plants, thereby making it impossible to distinguish between the relative importance of these two potential food sources for aquatic animals. The present study used a dual isotope approach (δ¹³C and δ¹⁵N) to refine measurements of the incorporation of allochthonous organic matter into freshwater fishes. The dependence of five species of littoral fishes on terrestrial detritus for part of their energy sustenance was demonstrated. The littoral zones of boreal Canadian Shield lakes are, therefore, not isolated from their surrounding riparian forests in terms of carbon flow as present day timber management guidelines erroneously assume, but instead exhibit a measurable degree of ecotonal land-water coupling. As a result, clearcut logging of riparian forests to the lakeshore edge, permissible by law in most Canadian provinces containing boreal forests, may have to be reassessed as a forest harvesting strategy.     

Carbon sources of Amazonian fisheries

Variation in the seasonal and spatial isotopic composition of plant C₄ (aquatic macrophytes) and C₃ (forest, C₃ aquatic macrophytes and algae), and that of fish [ Prochilodus nigricans Agassiz, Mylossoma duriventre (Cuvier), Colossoma macropomum (Cuvier), Semaprochilodus insignis (Schomburgk) and S. taeniurus Steindachner in the Amazon floodplain were analysed to test whether the fisheries deliver plant carbon to the population of Manaus in the same proportion as it is available in the floodplain. The contribution of C₄ plants was significantly lower in ¹³C during the season of high water levels and increased toward the west of the basin. Mylossoma duriventre and C. macropomum changed δ¹³C levels, while the δ¹³C of P. nigricans and C. macropomum shifted spatially. The contribution of C₄ to the fisheries yield was small. C³ plants (excluding phytoplankton) also contributed less than expected. This was explained by the importance of detritivores to the yield of the fisheries and the dependence of these species on algal carbon.     

The effect of dietary phosphatidylcholine and its constituent fatty acids on microdiet ingestion and fatty acid absorption rate in gilthead sea bream, Sparus auratus, larvae

This study tested the effect of the level of dietary phosphatidylcholine (PC) and its constituent medium-chain fatty acids on microdiet ingestion (μg diet larva⁻¹ h⁻¹) and the absorption rate of the free fatty acid [¹⁴C]16:0 (pmole larva⁻¹ h⁻¹) in 15, 20, 21, 25, 26, 30 and 31-day-old gilthead sea bream, Sparus auratus L., larvae. Fish were fed four microdiets (A, B, C and D): microdiet A contained no phospholipid (PL), while microdiet B included 10 g kg⁻¹ Artemia nauplii PL (3.7 g kg⁻¹ PC). Microdiets C and D contained 10 g kg⁻¹ purified saturated PC dimyristoyl (C_14:0) and polyunsaturated PC dilinoleoyl (C_{18:2[cis]−9,12}), respectively.
Larvae from one or both of the PC microdiets demonstrated significantly higher ( P < 0.05) ingestion rates (μg diet larva⁻¹ h⁻¹) than the non-PL microdiet control in 15, 21, 22, 25 and 26-day-old larvae and the Artemia PL microdiet in 15, 22 and 26-day-old larvae. However, microdiet ingestion and fatty acid absorption rate appeared to be independent of the associated medium carbon chain saturated or polyunsaturated fatty acid moiety of the PC diets. Apparent absorption, as measured by the retention of radio-labelled [¹⁴C]16:0 following 8 h of non-labelled microdiet feeding, was possibly related to feeding.     

Histopathology, electron microscopy and isolation of channel catfish virus in experimentally infected European catfish, Silurus glanis L.

Abstract Two groups of European catfish, Silurus glanis L., fingerlings were infected with channel catfish virus (CCV) by either intraperitoneal injection with 10⁵ TCID₅₀ of CCV, or bathing in water containing 10⁵ TCID₅₀ of CCV per 1·0 ml. The virus was isolated from spleen, intestine and brain of CCV-injected fish at day 1 and the titres ranged from 10^2·1 to 10^3·3 TCID₅₀/g. However, the tissue distribution of CCV was irregular and no virus was isolated after day 3 post-exposure. In CCV-bathed fish, the virus was isolated only from the liver of one specimen at day 3 post-exposure. No clinical signs of CCV disease developed in any of the fish. Specimens in each regime from all sampling periods showed some minor histopathological changes, but there were no differences between treatments. Lesions included oedema and focal haemorrhage in the liver and the spleen was congested. Electron micrographs of tissue samples showed the presence of a few virus particles around the nuclei of kidney, spleen and intestinal cells, and in or around a myelinated nerve within the optic lobes of infected fish during the first 4 days of infection.     

Herpesviral haematopoietic necrosis virus (CyHV-2) infection: case studies from commercial goldfish farms

Herpesviral haematopoietic necrosis is a disease of goldfish, Carassius auratus , caused by Cyprinid herpesvirus-2 (CyHV-2) infection. Quantitative PCR was carried out on tissue homogenates from healthy goldfish fingerlings, broodfish, eggs and fry directly sampled from commercial farms, from moribund fish submitted to our laboratory for disease diagnosis, and on naturally-infected CyHV-2 carriers subjected to experimental stress treatments. Healthy fish from 14 of 18 farms were positive with copy numbers ranging from tens to 10⁷ copies μg⁻¹ DNA extracted from infected fish. Of 118 pools of broodfish tested, 42 were positive. The CyHV-2 was detected in one lot of fry produced from disinfected eggs. Testing of moribund goldfish, in which we could not detect any other pathogens, produced 12 of 30 cases with 10⁶–10⁸ copies of CyHV-2 μg⁻¹ DNA extracted. Subjecting healthy CyHV-2 carriers to cold shock (22–10 °C) but not heat, ammonia or high pH, increased viral copy numbers from mean copy number (±SE) of 7.3 ± 11 to 394 ± 55 μg⁻¹ DNA extracted after 24 h. CyHV-2 is widespread on commercial goldfish farms and outbreaks apparently occur when healthy carriers are subjected to a sharp temperature drop followed by holding at the permissive temperature for the disease.     

Utilization of terrestrial organic matter by the bivalve <Emphasis Type="Italic">Corbicula japonica</Emphasis> estimated from stable isotope analysis

ABSTRACT:   Carbon and nitrogen isotopic ratios in tissue of the bivalve corbicula ( Corbicula japonica ) and particulate organic matter (POM) were measured along a salinity gradient in the Kushida Estuary, Japan. The bivalve exhibited a gradual isotopic enrichment from the uppermost estuarine site (δ¹³C = −24.8‰ and δ¹⁵N = 8.6‰) to the marine site (δ¹³C = −16.1‰ and δ¹⁵N = 11.8‰). Using the concentration-weighted mixing model, the bivalves' food source is estimated from the isotope values for the bivalves and POM from terrestrial plants, marine phytoplankton and benthic microalgae. The results indicated that the contributions of benthic micro algae and phytoplankton were small, while terrestrial particulate matter is significantly important for the corbicula diet, although the contribution varies among sampling sites.