Distribution and Effects of Chemicals in Communities and Ecosystems - Organic Pollutants: An Ecotoxicological Perspective - page 78

BOX 4.2 mOdelS fOR BIOCOnCentRatIOn
and BIOaCCumulatIOn
As indicated in Table 4.1, aquatic mollusks present a relatively simple picture
because they have little capacity for biotransformation of organic pollutants,
the principal mechanism of both uptake loss being diffusion. It is not sur-
prising, therefore, that bioconcentration factors (BCFs) for diverse lipophilic
compounds, measured at the steady state, are related linearly to log K ow val-
ues (Figure 4.1). Thus, the more hydrophobic a compound is, the greater the
tendency to partition from water into the lipids of the mollusk. The relation-
ship shown in Figure 4.1 has been demonstrated in several species of aquatic
mollusks, including the edible mussel ( Mytilus edulis) , the oyster ( Crassostrea
virginica ), and soft clams ( Mya arenaria ) (Ernst 1977). A similar relationship
has also been found with rainbow trout and other fish for some pollutants. On
the other hand, some organic pollutants do not fit the model well (Connor 1983).
It seems probable that some compounds that are metabolized relatively rapidly
by fish will be eliminated faster than would be expected if diffusion were the
only process involved (Walker 1987). Such compounds would not be expected
to follow closely a model for BCF based on K ow alone. This point aside, K ow
values can give a useful prediction of BCF values at the steady state for lipo-
philic pollutants in aquatic invertebrates. A great virtue of the approach is that
K ow values are easy and inexpensive to measure or predict (Connell 1994).
Other more complex and sophisticated models have been developed for fish
(see, for example, Norstrom et al. 1976) but are too time-consuming/expensive
to be used widely in environmental risk assessment where cost-effectiveness
is critically important. Modeling for bioaccumulation by terrestrial animals
presents greater problems, and BAFs cannot be reliably predicted from K ow
values (Walker 1987). For example, benzo[ a ]pyrene and dieldrin have log K ow
values of 6.50 and 5.48, respectively, but their biological half-lives range from
a few hours in the case of the former to 10 -369 days for the latter. Endrin
is a stereoisomer of dieldrin with a similar K ow , but has a half-life of only 1
day in humans, compared with 369 days in the case of dieldrin. These large
differences in persistence have been attributed to differences in the rate of
metabolism by P450-based monooxygenases (Walker 1981). Effective predic-
tive models for bioaccumulation of strongly lipophilic compounds by terrestrial
animals need to take account of rates of metabolic degradation. This is not a
straightforward task and would require the sophisticated use of enzyme kinet-
ics to be successful. In one model, it has been suggested that Lineweaver-Burke
plots for microsomal metabolism might be used to predict BAF values in the
steady state (Walker 1987) (Figure 4.2). In principle, when an animal ingests a
lipophilic compound at a constant rate in its food, a steady state will eventually
be reached where the rate of intake of the compound is balanced by the rate of
its metabolism. It is assumed that the rate of loss of the unchanged compound
by direct excretion is negligible. Primary metabolic attack upon many highly