et al. examined the ability of several estrogen mimics to
act alone and then in combination with one another, in an experimental
system that examined the strength of an estrogen response measured
in yeast cells in culture that had been genetically modified to
contain human estrogen receptors. This assay is now well-established
as a means to measure estrogenicity of compounds.
carefully characterizing the dose-response curves of compounds by
themselves, they performed experiments with combinations of the
compounds. They then tested to see whether the impacts of combinations
of compounds could be predicted on the basis of adding the observed
impacts of the compounds when studied separately.
found that the additions accurately predicted the impacts of the
compounds together in mixtures.
is important because it addresses a more general and vital question
about the potential hazards of endocrine disrupters. Without exception,
in the real world contaminants come in mixtures. Virtually all of
regulatory toxicology, however, has focused on the actions of contaminants
one-by-one. Do these one-by-one experiments help in anticipating
how contaminants behave in mixtures? Do compounds cancel one another
or multiply each others' impacts. [More...]
did they do?
Payne et al. used four well-established estrogen mimics,
o,pī-DDT, genistein, 4-nonylphenol, and 4-n-octylphenol,
in the yeast estrogen screen (YES) system. These four compounds
are all known to bind with the estrogen receptor. Payne et al.
established dose response curves for each compound, describing the
strength of the estrogen response for different concentrations of
the compounds singly. They then combined the estrogen mimics in
different mixtures and measured their total impact.
crucial test was then to determine whether the total impact was
predictable on the basis of the sum of the independent effects.
So, for example, if a particular test mixture of DDT, genistein,
nonylphenol and octylphenol contained concentrations of those compounds
that separately had provoked estrogen responses (on some arbitrary
scale) of 0.1, 3, 2.2 and 5, respectively, the question was whether
the mixture provoked a response of 10.3 (the sum of the separate
need not have been the case. Compounds could have interfered with
one another, decreasing the total response, even negating it. Or
they might have interacted synergistically, causing a response in
mixture significantly greater than a simple additive effect.
did they find?
They found that the compounds interacted additively. The results
of mixtures could be predicted by adding up their individual impacts.
noted above, this is an important demonstration. It has several
limitations, however, which limit its applicability to the real
world. These limitations, discussed below, suggest that these results
imply that interactions of these four compounds in live animals
at a minimum would be additive. They might be much more.
this pioneering work tells us that some compounds interact additively,
but work of this nature is too rare, as yet, to know whether this
is a general pattern to expect with other compounds.
the experimental system was purposefully made as simple as possible:
compounds known to bind with the estrogen receptor with the assay
being a direct measure of the result of receptor binding in an in
vitro cell culture. They designed the experiment that way to
limit the mechnisms of action to the one they could measure directly.
Many compounds have more than one mechanism of action. Other mechanisms
of action by the same compound might provoke other responses simultaneously
that could alter the pattern of interaction.
to simplify these experiments (and make interpretation as feasible
as possible, they were done with a single type of cell in a monoculture.
Mixtures in real animals interact with many different cells, organs
and tissues simultaneously, with the potential for myriad positive
and negative feedback loops.