Biomedical Engineering Reference
In-Depth Information
Figure3.1 Glucuronidation
including minnows, trout and flounders. The source of environmental oestrogens
is not confined to outfall from sewage treatment plants, however, the fate of
endocrine disrupters, examples of which are given in Figure 3.2, in sewage treat-
ment plants is the subject of much research (Byrns, 2001). Many other chemicals,
including PAHs, dichlorodiphenyltrichloroethane (DDT), alkyl phenols and some
detergents may also mimic the activity of oestrogen. There is general concern
as to the ability of some organisms to accumulate these endocrine disrupters in
addition to the alarm being raised as to the accumulative effects on humans of
oestrogen like activity from a number of xenobiotic sources.
To date there is no absolute evidence of risk to human health but the Environ-
mental Agency and Water UK are recommending the monitoring of environmen-
tal oestrogens in sewage treatment outfall. Assays are being developed further to
make these assessments (Gutendorf and Westendorf, 2001) and to predict poten-
tial endocrine disrupter activity of suspected compounds (Takeyoshi et al ., 2002).
Oestrogen and progesterone are both heat labile. In addition, oestrogen appears
to be susceptible to treatment with ultra violet light, the effects of which are aug-
mented by titanium dioxide (Coleman et al ., 2004). The oestrogen is degraded
completely to carbon dioxide and water thus presenting a plausible method for
water polishing prior to consumption.
Another method for the removal of oestrogens from water, in this case
involving Aspergillus , has also been proposed (Ridgeway and Wiseman, 1998).
Sulphation of the molecule by isolated mammalian enzymes, as a means of
hormone inactivation is also being investigated (Suiko et al ., 2000). Taken
overall, it seems unlikely that elevated levels of oestrogen in the waterways
Search WWH ::




Custom Search