Page 43 - The Toxicology of Fishes

Bioavailability of Chemical Contaminants in Aquatic Systems 23

+
EC TAMM = 110 pK a − pH
1 + 10 pK a − pH (2.5)
EC EC +
NH 3 NH 4
This describes a sigmoidal relationship of log EC TAMM to pH, with the value for EC TAMM approaching
EC at high pH and EC + at low pH.
NH 3 NH 4
Erickson (1985) reported a good fit of the above model to the data in Figure 2.5A, with LC
50 NH 3
estimated to be 0.63 mg N per liter and LC + estimated to be 170 mg N per liter. Substituting these
50 NH 4
values and pK = 9.6 (the value for 13.5°C, the average temperature in the study) into Equation 2.5
a
produces the solid line in Figure 2.5A, and multiplying these values by f produces the solid line in
NH 3
Figure 2.5B. The good fit of this model depends not only on these estimated values for LC and
50 NH 3
LC + but also on the adherence of the data to the model assumption that LC is a weighted
50 NH 4 50 TAMM
average of LC and LC + , with the weighting factors being the fractional species concentrations.
50 NH 3 50 NH 4
An important consequence of this assumption is that LC 50 TAMM closely parallels the fractional concen-
tration of the more bioavailable form (dotted lines in Figure 2.5) until this fraction is low enough that
the less bioavailable form is responsible for a significant portion of the toxicity.
Other datasets on ammonia toxicity to fish also indicate that pH effects can be primarily attributed to
joint toxicity of un-ionized ammonia and ammonium ion (Erickson, 1985); however, the relative bio-
availabilities of these two chemical species can differ among fish species. The ratio of LC to
50 NH 3
LC + varied from 0.001 to 0.006, mainly attributable to differences in the bioavailability of ammonium
50NH 4
ion for different fish. In particular, studies with channel catfish (Sheehan and Lewis, 1986; Tomasso et
al., 1980) have suggested little or no bioavailability of ammonium ion, with nearly constant LC values
50
on the basis of un-ionized ammonia. Such differences among fish species may be related to differences
in ionoregulatory and ammonia excretion mechanisms, but this has not been investigated.
Ammonia bioavailability to fish, however, is not just a matter of the above simple model. Several
additional factors that can make ammonia bioavailability relationships more complex may have to be
considered in specific assessment situations:
1. Excretion of respiratory carbon dioxide will depress the pH at the gill surface. This will cause
a shift in equilibrium of ammonia speciation, resulting in less un-ionized ammonia and more
ammonium ion at the gill surface than in the bulk exposure water. The degree of this effect
depends on the exposure pH, the pH-buffering strength of the exposure water, and the physi-
ology of the fish. Lloyd and Herbert (1962) and Szumski et al. (1982) suggested that this,
rather than some bioavailability of ammonium ion, could account for toxicity on a un-ionized
ammonia basis not being constant with pH. The analyses of Erickson (1985), however, indicated
that this mechanism is generally of secondary importance to the joint toxicity of un-ionized
and ammonium ion.
2. Ammonia is unique among toxicants of concern for fish because it is also an excretory product,
and this endogenous ammonia will contribute to some degree to the accumulation of ammonia
to toxic levels. If excretion rates are not constant with pH, this can affect the pH dependence
of the relationship of toxicity to exogenous ammonia. High pH can inhibit ammonia excretion
because more of the released ammonia remains in the un-ionized form, which reduces the
blood–water gradient of un-ionized ammonia and thus the net diffusive loss. Russo et al. (1988)
reported increased ammonia toxicity at pH values near and above 9 which might be related to
such effects. Exogenous ammonia might also affect excretory mechanisms, thereby altering
ammonia accumulation other than by direct uptake.
3. Sheehan and Lewis (1986) suggested that acute ammonia toxicity to channel catfish at low pH
is partly attributable to osmotic effects of high ammonium salt concentrations (>1000 mg N
per liter). To the extent their conclusions are true, the ammonium ion might be essentially
nonbioavailable and exert its effects simply based on its contribution to the external osmotic
pressure. Although this might be important for channel catfish, other fish species show apparent
effects of ammonium ion at much lower concentrations (ca. 100 mg N per liter) where such