Polychlorinated biphenyl tissue-concentration thresholds for survival, growth, and reproduction in fish

Polychlorinated biphenyls (PCBs) have left a legacy of environmental contamination. Even though they were banned from production and active use in the 1970s, they persist in the environment and still have the potential to impact aquatic life. Our objective was to identify data from controlled laboratory studies of PCB-related adverse effects in fish and to conduct a meta-analysis on mortality, growth, and reproductive (MGR) threshold responses. For each endpoint type, we compiled data on the lowest-observed-adverse effect concentration (LOAEC) and the degree of effect at the LOAEC as a percentage of control. The LOAECs were expressed as tissue concentrations, so the term lowest-observed-adverse-effect residue concentration (LOAER) was used to represent PCB exposures. The lower limit of applicability was set at 0.1 μg/g total PCB tissue concentration, below which adverse MGR effects in fish were not supported by the data. Sensitivity distributions identifying the probability of adverse effects in fish populations or communities predicted that 25% of fish species would be impacted between 0.1 and 7.5 μg/g. Concentration-response threshold regressions were developed from the MGR datasets. For example, a 1 μg/g total PCB tissue concentration would predict effects of 17% mortality, 15% growth, and 39% reproductive. The analysis determined the degree of adverse response, with uncertainty estimates, expected across a broad range of PCB tissue exposure concentrations in fish. Data generated from MGR endpoints were combined to determine an approach for overall effect thresholds for PCB-related injury in fish. The MGR datasets included only laboratory data; however, responses were compared with field-observed effects. The present review provides a comprehensive assessment of PCB-induced injury in fish utilizing a data-inclusive approach. Environ Toxicol Chem 2019;38:712-736. Published 2018 Wiley Periodicals Inc. on behalf of SETAC. This article is a US government work and, as such, is in the public domain in the United States of America.

Patterns in stream fish assemblage structure and function associated with a PCB gradient

Stream fish assemblage structure and function were examined for significant response along a polychlorinated biphenyl (PCB) gradient from two PCB-contaminated streams (Clear Creek and Richland Creek watershed) at three locations and a control stream (Little Indian Creek), Indiana, USA. Fish were sampled in the summer months of 1995 and from 1999 to 2002. 51 fish assemblage attributes-including structure (i.e., fish composition) and function (i.e., trophic, reproductive, condition guilds), biomass, and index of biotic integrity (IBI) metric scores-were evaluated for significance according to an increasing PCB gradient. Eight biomass attributes of fish assemblages decreased with increasing PCB concentration: number of species biomass, number of sunfish biomass, percent sunfish biomass, number of sucker biomass, percent sucker biomass, biomass of sensitive species, percent sensitive species biomass, and percent carnivore biomass. Three biomass attributes increased with PCB concentration: percent minnow biomass, percent pioneer species biomass, and percent tolerant species biomass. Seven species composition and relative abundance characters decreased with increasing PCB concentration: number of species; number of darter, madtom, and sculpin; number of darter; number of sunfish; number of sucker; number of sensitive species; and percent individuals as carnivores. Percent individuals as pioneer species increased with increasing PCB concentration. Two IBI metrics, percent individuals as headwater species and number of minnow species, increased as PCB concentrations increased, whereas number of sucker species and percent individuals as pioneer species decreased with increasing PCB concentration class. We observed a direct response between decreased relative abundance and biomass of carnivores and increased relative abundance minnows as the PCB gradient increased. Total IBI score did not detect subtle changes to the fish community that were observed along a PCB gradient, whereas diagnostic analysis of the individual metrics did.

Evaluation of PCB bioaccumulation by Lumbriculus variegatus in field-collected sediments

Review of data from several contaminated sediment sites suggested that biota-sediment accumulation factors (BSAFs) declined with increasing contaminant concentrations in the sediment. To evaluate the consistency and possible causes of this behavior, polychlorinated biphenyl (PCB)-contaminated sediment samples from the Hudson, Grasse, and Fox River Superfund sites were used in sediment bioaccumulation tests with the freshwater oligochaete, Lumbriculus variegatus, with PCB concentrations in interstitial water (IW) quantified using polyoxymethylene passive samplers. Measured BSAFs tended to decrease with increasing PCB concentration in sediment, especially for the more highly chlorinated congeners. Measures of partitioning between sediment, IW, and oligochaetes showed that measured sediment-IW partition coefficients (KTOC ) tended to increase slightly with increasing sediment contamination, whereas the ratio of tissue PCB to IW PCB tended to decrease with increasing concentration in IW. Variation in accumulation among sediments was clearly influenced by bioavailability, as reflected by IW measurements, although the specific cause of varying KTOC was not clear. Calculated partitioning between IW and organism lipid (Klipid ) indicated that accumulation was generally 5 to 10-fold higher than would be predicted if Klipid was approximately equal to the n-octanol-water partition coefficient (KOW ). While affirming previous observations of decreasing BSAFs with increasing PCB contamination, the relatively shallow slope of the observed relationship in the current data may suggest that this concentration dependence is not a major uncertainty in sediment risk assessment, particularly if measurements of PCBs in IW are incorporated.

Early life-stage toxicity in zebrafish (Danio rerio) following embryonal exposure to selected polychlorinated biphenyls

Polychlorinated biphenyls are a widespread aquatic contaminant. In this article, specific polychlorinated biphenyl congeners were examined for embryo and early life-stage toxicity in zebrafish (Danio rerio). A set of three polychlorinated biphenyl congeners (non-ortho polychlorinated biphenyl 126, mono-ortho polychlorinated biphenyl 28 and di-ortho polychlorinated biphenyl 153) were tested. The typical lesions observed were yolk sac edema, vertebra defect, craniofacial malformations (double head, triple retina), anaxial body and inhibition of swim bladder inflation. Moreover, embryo and larval mortality increased and hatching success decreased. The severity of abnormalities and mortalities were concentration- and congener-dependent. Of the compounds tested, polychlorinated biphenyl congener 126 was found to be highly toxic to the fish embryos following exposure. The Lethal Concentration 50 values for polychlorinated biphenyl 28, polychlorinated biphenyl 126, polychlorinated biphenyl 153 calculated by probit analysis were 3.270, 1.298 and 5.375 ppm, respectively. The inhibition of swim bladder inflation was the most sensitive endpoint measured, and it is suggested that the inhibition of swim bladder inflation may be mediated by mechanism with an aryl hydrocarbon receptor activation.

Environmental contaminants and the reproductive success of lake trout in the Great Lakes: an epidemiological approach

Epidemiological criteria were used to examine the influence of environmental contamination on reproductive success of lake trout (Salvelinus namaycush) in the Laurentian Great Lakes. Most of the information was obtained from lake trout eggs collected in southeastern Lake Michigan and reared in the laboratory. Two separate end points that measure reproductive success--egg hatchability and fry survival--were used in the evaluation. Strong evidence for maternally derived polychlorinated biphenyls causing reduced egg hatchability were observed for the time order, strength of association, and coherence criteria. Equally strong evidence for organic environmental contaminants, also of maternal origin, causing a swim-up fry mortality syndrome were presented for the strength of association, specificity, replication, and coherence criteria. The epidemiological approach for demonstrating cause-and-effect relations was useful because of the difficulty in demonstrating definite proof of causality between specific environmental contaminants and reproductive dysfunction in feral fish.

Reproductive impairment in a fish inhabiting a contaminated coastal environment off Southern California

White croaker (Genyonemus lineatus), collected from a highly contaminated site in San Pedro Bay and from a reference site 80 km away (Dana Point), were induced to spawn in the laboratory. Forty-one per cent of San Pedro Bay females and 54% of Dana Point females spawned. Examination of the ovaries of non-spwaning females revealed that spawning was imminent in the remainder of Dana Point fish but only in 16% of the San Pedro Bay fish. The remainder of the San Pedro Bay fish (43%) contained only immature, yolky oocytes. No croakers containing more than 3.8 ppm ovarian total DDT could be induced to spawn whereas 36% of a contemporaneous San Pedro Bay sample had ovarian total DDT residues in excess of 4 ppm. This suggests that the inability to induce spawning in white croaker may be associated with an ovarian total DDT threshold of about 4 ppm. These data, coupled with observed decreases in fecundity (32%), fertility (14%), and early oocyte loss (30%) relative to reference fish, could partially explain the population declines observed for many southern California fishes since the 1940s.

Toxic substances in rivers and streams

Many of the toxic substances entering freshwaters today are those which were present several decades ago, but others have become significant recently. The effects of toxicants in flowing waters are modified by unidirectional transport and dispersion which afford the potential for a degree of 'self-purification'. The chemical quality of the receiving water also affects toxicity. Biological factors also contribute to the ultimate effect of pollutants. The potential for accumulation of toxic substances within tissues increases the significance of certain pollutants which may be present in water even though ambient concentrations are very low. The biota of flowing waters may be restored, following catastrophic entry of pollutants, by drift from unaffected regions upstream. The range of potential toxic substances is very extensive and includes inorganic poisons, organic poisons, heavy metals, pesticides and PCBs. Metals, pesticides and PCBs have the greatest potential for bioaccumulation. Few generalisations can be made regarding the effects of toxic substances on the biota. Each species tends to respond to different toxicants in different ways and even at different stages in its life-history. Toxicity tests conducted under controlled laboratory conditions sometimes produce conflicting results: it is not then to be unexpected that field observations should sometimes vary widely. Determinations of toxicity in laboratory tests must be applied with caution to field conditions and it is not wise to extrapolate findings to other species or environments.

Dose determinations for waterborne 2,5,2',5'-[14C]tetrachlorobiphenyl and related pharmacokinetics in two species of trout (Salmo gairdneri and Salvelinus fontinalis): a mass-balance approach

A mass-balance study was undertaken to evaluate the accuracy of dose determinations of waterborne 2,5,2',5'-[14C]tetrachlorobiphenyl (TCB) made on transected brook trout (Salvelinus fontinalis) and rainbow trout (Salmo gairdneri) and to determine any pharmacokinetic differences between the two species. The total calculated [14C]TCB absorbed by brook (17.8 micrograms) and rainbow (24.5 micrograms) trout was compared to the actual body burden measurements of [14C]TCB for brook (17.4 micrograms) and rainbow (25.6 micrograms) trout; the latter measurements also included excretory losses through the urine, feces, and across the gill surface. Approximately 1% of the total dose was excreted of which 75% was in the urine and 25% in the feces. The agreement between the whole body burden measurements of [14C]TCB and the total calculated micrograms of [14C]TCB absorbed was within 10% in both species. Mass-balance measurements were converted to dose by dividing by fish weight. Mean calculated and measured doses were 31.4 and 30.4 micrograms/kg/48 hr for brook trout and 32.3 and 33.6 micrograms/kg/48 hr for rainbow trout. No species differences were seen in either calculated or measured doses or in total radioactivity excreted. Respiratory function between the two species was similar except for a significantly higher mean ventilation rate for brook trout (84 +/- 14/min) than for rainbow trout (65 +/- 4/min). There were also no fluctuations noted in either respiratory function of [14C]TCB uptake efficiency across the gills over the 48-hr exposure period. These studies demonstrated the ability to accurately calculate a water dose in micrograms per kilogram per hour for individual fish that could be directly compared to other fish species or to mammals.