D.R. Livingstone

The fate of organic xenobiotics in aquatic ecosystems: quantitative and qualitative differences in biotransformation by invertebrates and fish.

Biotransformation of natural and man-made foreign compounds (xenobiotics) proceeds via introduction of a functional group (phase I metabolism) and subsequent attachment of a polar moiety to the group (phase II metabolism). The biotransformation fate of xenobiotics depends on the activities, complement and inducibility of the biotransformation enzymes. Previous analysis of the dependence of in vivo rates of biotransformation on tissue parent compound concentration for marine invertebrates revealed that hydrocarbons are metabolised more slowly than xenobiotics already containing functional groups, and crustaceans metabolise both types of xenobiotics faster than molluscs (Livingstone D.R., Persistent pollutants in marine ecosystems, pp. 3-34, Pergamon, Oxford). Use of the same approach showed that fish metabolise pentachlorophenol (PCP) and benzo[a]pyrene (BaP) faster than certain aquatic invertebrates, viz. rates of biotransformation to total metabolites (pmol min-1 g-1 wet wt.) at a tissue parent compound concentration of 10 nmol g-1 were, respectively, 19.2 +/- 3.7 (Carassius auratus) and 4.8 +/- 6.6 (molluscan species) (PCP), and 19.1 +/- 6.3 (fish species) and 2.1 +/- 0.2 (crustacean species) (BaP). The higher rate of biotransformation of BaP in fish is consistent with higher levels of total cytochrome P450 and inducible cytochrome P4501A (CYP1A) activity. The similar rate of metabolism of a hydrocarbon (BaP) (requires initial metabolism by cytochrome P450) and a functional group compound (PCP) by fish may also be due to the high levels of cytochrome P450, compared with the situation in invertebrates where rate-limiting cytochrome P450 may be responsible for the lower rates of hydrocarbon compared with functional group compound metabolism.

Effects of the “Aegean Sea” oil spill on biotransformation enzymes, oxidative stress and DNA-adducts in digestive gland of the mussel (Mytilus edulus L.)

Abstract Possible molecular biomarkers of impact by organic pollution on mussels were applied to samples from five sites along the Galician Coast, Spain, taken 6 months after the oil spill from the tanker “Aegean Sea.” Whole body aliphatic hydrocarbon concentrations were similar at all sites, but specific chemical ratios (resolved/unresolved hydrocarbons; carbon preference index; pristane/phytane) indicated a predominance of degraded petrogenic hydrocarbons nearer the oil spill. Levels of whole body polycyclic aromatic hydrocarbons (sum of 13 PAHs) increased steadily towards the oil spill, and were paralleled by increases in digestive gland levels of total cytochrome P-450, CYP1A-like protein and lipid peroxidation (corr. coeffs. with PAHs of 0.64–0.67). Differences were more marked in CYP1A-like protein than total cytochrome P450, indicating induction of specific P450 isoenzyme(s). No differences between sites were seen for benzo[a]pyrene hydroxylase, glutathione S-transferase, Superoxide dismutase and DT-diaphorase activities. Bulky, hydrophobic DNA-adducts were detected in digestive gland of mussels from industrial and urban sites, but not from the site nearest to the oil spill which had the highest tissue levels of PAHs. Overall the results indicate induction of cytochrome P450(s) and oxidative damage in mussel with oil exposure.

Cytochrome P450 and dependent activities in unexposed and PAH-exposed terrestrial annelids

The cytochrome P450 system of the oligochaetes Eisenia f. fetida (tiger worm) and Enchytraeus crypticus (pot worm) was analysed using ethoxy-, pentoxy- and benzoxyresorufin as substrates for monooxygenase activity. Whole body microsomes of the earthworm E.f. fetida displayed PentROD activity in the range from 0.26 to 1.05 pmol mg protein-1 min-1 and BenzROD activity in the range from 0.14 to 0.30 pmol mg protein-1 min-1. Exposure of the animals for up to four weeks to 100 mg fluoranthene or benzo[a]pyrene kg-1 soil (dry weight) did not induce significant changes in the activity of these monooxygenases. In E. crypticus EROD activity was in the range from 2.10 to 6.18 pmol mg protein-1 min-1 and PentROD activity in the range from 1.75 to 4.78 pmol mg protein-1 min-1. Short-term exposure to BaP by feeding reduced the EROD activity significantly by 45%, but did not effect PentROD activity. After long-term (8 weeks) exposure to BaP in the agar-agar medium EROD activity was not changed but PentROD had decreased to zero. In both species cytochrome P420 and NADPH-cytochrome C reductase activity were present. In E.f. fetida microsomes are associated with the giant haemoglobin. Both can be separated by gel filtration on a Sepharose B2 column or by hydrophobic interaction chromatography after solubilisation with cholate. NADPH-cytochrome C reductase elutes together with haemoglobin. Cytochrome P420 is eluted with Emulgen 911 and can be further purified by ion exchange chromatography using HA-Ultrogel. By SDS-PAGE of the purified microsomal proteins three protein bands are visualised in the range of cytochrome P450 displaying an apparent molecular mass of 54, 56 and 58 kDa. Only the 54-kDa protein interacts weakly with perch (Perca fluviatilis) CYP1A antibodies, while two proteins with an apparent molecular mass of 65 and 71 kDa give a strong antibody signal.

Aromatic Hydrocarbon Quinone-mediated Reactive Oxygen Species Production in Hepatic Microsomes of the Flounder (Platichthys flesus L.)

The NAD(P)H-dependent redox cycling of a range of eight 1 ring to 5 ring aromatic hydrocarbon (AH) quinones by hepatic microsomes of flounder (Platichthys flesus) was studied in terms of oxygen consumption (Clark electrode) and reactive oxygen species (ROS) production (detection of hydroxyl radical by iron/EDTA-mediated oxidation of 2-keto-4-methiolbutyric acid). Stimulated oxygen consumption was detectable for only five AH-quinones (duroquinone, 1,2- and 1,4-naphthoquinones, menadione, 9,10-phenanthrenequinone), whereas stimulated ROS production was seen, or is known, for all eight (others plus 1,4-benzoquinone, anthraquinone, benzo[a]pyrene-3,6-dione), indicating that the former measurement is a more sensitive assay of redox cycling. Both processes showed Michaelis-Menten kinetics with respect to AH-quinone concentrations, with values for Vmax and apparent Km being, respectively, 146- to 9895-fold and 3- to 344-fold higher for stimulated oxygen consumption than ROS production. Marked correlation in values for both Vmax and apparent Km was seen between stimulated oxygen consumption and ROS production for 1,2-naphthoquinone, 1,4-naphthoquinone and 9,10-phenanthrenequinone, indicative of redox cycling and the univalent reduction of O2 to superoxide anion radical. Rates of stimulated oxygen consumption and ROS production were up to 10-fold higher for NADH- than for NADPH-dependent reactions and were highest for the naphthoquinones and 9,10-phenanthrenequinone. Comparison of the results for different AH-quinones indicates that enzyme substrate specificity is an important factor in determining redox cycling potential. Under the assay conditions used (0.1-2.0 mM AH-quinone), mutagenicity of the AH-quinone mediated processes could not be demonstrated using the Salmonella typhimurium umu assay. Overall, the results indicate a widespread potential for AH-quinone stimulated ROS production.

Characterization of the presence and seasonal variation of a CYP1A-like enzyme in digestive gland of the common mussel, Mytilus edulis

The presence and putative catalytic properties of a CYP1A-like enzyme in the digestive gland of Mytilus edulis L. were investigated by molecular biological and seasonal studies. Reverse-transcriptase PCR using oligonucleotide primers to amplify a sequence around the conserved haem binding cysteine region of hepatic CYP1A1 of trout (Oncorhynchus mykiss) produced several cDNA bands resolved by electrophoresis, including major bands of about 220 and 280 bp compared to the predicted size for O. mykiss of 208 bp. Following Southern blotting and probing with a cDNA probe to O. mykiss CYP1A1 (pfP1450-3′ probe modified by removal of 3′-non-coding region by digestion by Cla1), a single band (280 bp) only was detected using moderate stringency conditions of sequence recognition (i.e. hybridization at 42 °C followed by washing at 55 °C in 1 × SSC containing 0.1% SDS), providing evidence for the presence of a CYP1A orthologous gene sequence. The seasonal variation in levels of putative CYP1A mRNA (Northern analysis using the modified cDNA probe for O. mykiss CYP1A1) over 1 year showed some similarity to seasonal patterns of change in microsomal metabolism of 3H-benzo[a]pyrene (BaP) to polar metabolites (dials, diones and phenols resolved by HPLC). Maxima for putative CYP1A mRNA and BaP metabolism levels were in late spring-early summer. However, differences were also apparent, possibly indicative of other P450s contributing to BaP metabolism. Overall the results indicate the existence of a CYP1A-like enzyme which is, at least, partly responsible for the mono-oxygenase activity of BaP metabolism.

Seasonal variation in mussel Mytilus edulis digestive gland cytochrome P4501A- and 2E-immunoidentified protein levels and DNA strand breaks (Comet assay)

Mytilus edulis digestive gland microsomes were prepared from indigenous populations sampled from a clean reference site (Port Quin) and an urban-industrial contaminated site (Blackpool) in the UK. Samples were collected in March/April, May, August and December 1998. Western blot analysis was performed using polyclonal antibodies to fish CYP1A and rat CYP2E using partially purified M. edulis CYP as a positive control, to aid identification. CYP1A- and CYP2E-immunopositive protein levels showed different site-specific seasonal variation with higher levels of CYP2E determined in May (P < 0.05). At both sites, lower levels of CYP1A-immunopositive protein but not CYP2E-immunopositive protein were observed in the samples collected in December (P < 0.05). This correlated with lower levels of nuclear DNA damage (Comet assay expressed as per cent tail DNA) observed in December compared to August (P < 0.05).

Characterisation of biotransformation enzyme activities and DNA integrity in isolated cells of the digestive gland of the common mussel, Mytilus edulis L.

Characterisation of biotransformation and antioxidant enzyme activities and DNA integrity was carried out on isolated cells of the major biotransformation organ (digestive gland) of Mytilus edulis as a first step in developing a cell culture system for use in toxicology. Digestive gland cells were isolated by trypsin or non-protease tissue dissociation procedures, followed by filtration and differential centrifugation. Both dissociation methods produced a mixture of smaller cell types and large digestive cells at a viability of over 90% (EOSIN Y exclusion). The specific activities (per milligramme of protein) of 10 000×g supernatants from freshly isolated digestive gland mixed-cell populations produced by either dissociation method were similar to those from whole digestive gland for the antioxidant enzymes superoxide dismutase (EC 1.15.1.1) and catalase (EC 1.11.1.6), but from 3- to 10-fold lower for the biotransformation enzyme activities benzo[a]pyrene hydroxylase, glutathione-S-transferase (EC 2.5.1.18; substrate: 1-chloro-2,4-dinitrobenzene) and NADPH-dependent DT-diaphorase (EC 1.6.99.2). Cells produced by trypsin dissociation showed significant increased DNA strand breakage as measured by the single cell electrophoretic ‘comet’ assay, viz.% comet tail DNA increased from 10.3±2.3 (control) to 19.8±4.1 (0.01% w/v trypsin) to 23.7±3.2 (0.1%) (mean±S.D.). Cell yields from digestive gland were only slightly lower for non-enzymic compared to trypsin dissociation for the same time period of dissociation, indicating the former as a preferred method of cell culture preparation.

Increased potential for NAD(P)H-dependent reactive oxygen species production of hepatic subcellular fractions of fish species with in vivo exposure to contaminants

The present study investigated the proposed involvement of contaminant-stimulated reactive oxygen species (ROS) production in disease processes in fish. NAD(P)H-dependent ROS production of subcellular fractions was determined by the iron/EDTA-mediated oxidation of 2-keto-4-methiolbutyric acid. Hepatic cytosolic NADPH-dependent and microsomal NAD(P)H-dependent ROS production were increased 51-160% (P < 0.05) in rainbow trout (Oncorhynchus mykiss) 15 weeks after a single i.p. injection of polychlorobiphenyl (PCB) (100 mg Clophen A50 kg-1 wet wt.). Hepatic microsomal NADH-dependent ROS production was 114% higher in perch (Perca fluviatilis) from PCB-contaminated Lake Järnsjön compared to clean Lake Vänern, Sweden. Hepatic mitochondrial NADH-dependent, cytosolic NADH-dependent and microsomal NADPH-dependent ROS production were variously elevated up to 160% in flounder (Platichthys flesus) at various sites along two pollution transects near to the ports of Rotterdam and Amsterdam, Netherlands. Overall the data indicate increased potential for ROS production in liver of fish exposed to field pollution, and support the hypothesis of oxidative stress as a mechanism of contaminant-mediated disease in fish.

Elevation of cytochrome P450-immunopositive protein and DNA damage in mussels (Mytilus edulis) transplanted to a contaminated site

Mytilus edulis were collected from a reference site (Port Quin) and an urban/industrial contaminated site (New Brighton) in the UK during June 1999. Levels of PCBs (sigma7 congeners) and CB-138 were determined to be, respectively, 21 fold and 16 fold higher in the mussel digestive glands from New Brighton. Levels of CYPIA-immunopositive protein were 1.5 fold higher (P < 0.05) at the polluted site but the levels of DNA strand breaks were 1.3 fold higher (P<0.05) at the reference site. Mussels from Port Quin were placed in cages at both sites and both transplanted and indigenous populations sampled in September (13 weeks). Mussels transplanted from the reference site to the industrial site, reported elevated levels of CYP1A-immunopositive protein (1.4 fold; P < 0.05) and higher levels of DNA damage (1.2 fold; P < 0.05) compared to caged populations at the reference site and a PCB loading similar to the populations from the polluted site. Moreover, transplanted mussels had DNA damage 1.8 fold greater (P < 0.05) than indigenous mussels at the transplant site. These changes were small but significant when compared to the observed temporal changes in the indigenous populations.

Variation in levels of cytochrome P4501A, 2B, 2E, 3A and 4A-immunopositive proteins in digestive gland of indigenous and transplanted mussel Mytilus galloprovincialis in Venice Lagoon, Italy

Mytilus galloprovincialis digestive gland microsomes were prepared from (i) indigenous populations sampled from a clean reference (Lido) and an urban-contaminated site (Salute) and (ii) mussels transplanted for up to 3 weeks from Lido to an industrial-contaminated site (CVE) in Venice Lagoon, Italy. Western blot analysis was performed using antibodies to five mammalian or fish CYP forms (1A, 2B, 2E, 3A, 4A). Simultaneously run M. edulis digestive gland partially purified CYP aided identification of immunopositive bands. Levels of CYP1A, CYP2E, and CYP4A-immunopositive proteins were 50 to 300% higher in indigenous M. galloprovincialis from Salute compared to Lido (p < 0.05). Three weeks after transplantation to CVE, levels of only the CYP1A-immunopositive protein were determined to be higher (63%) than levels for Lido (p < 0.05), indicating that anti-CYP1A shows greater specificity for a contaminant-inducible CYP form than the other antibodies.