A recent U.S. Geological Survey study has demonstrated that surface waters in the state of Arizona have
widespread contamination of endocrine disrupting compounds (EDCs) and including various estrogenous
compounds. Little is known about the fate of EDCs in drinking water treatment processes. Most of
the water treatment plants around the nation use chlorine as a disinfectant. The ability of HOCl to
oxidize organic compounds with amine functional groups and/or cyclic nitrogen is well
established. Many of the EDC/PPCP compounds contain benzene rings, which impact their
reactivity with chlorine and the transformed product can also show varying degree of estrogenicity.
The issue becomes especially important in light of the fact that these chemicals occur at trace
concentration levels. The objective of this study was to characterize the type of estrogenicity found in
source water and its response to advanced water treatment processes, which can help in
understanding the removal and transformation of EDCs from drinking water.
The water samples from the Verde River, the Salt River, the Central Arizona Project
(CAP) canal, and two local water treatment plants were collected in amber glass bottles on a
monthly basis. The water samples were split into two sets before concentration. One set of samples
was chlorinated and the other set was kept untreated. Both the sets were concentrated by liquid-liquid
extraction using Dichloromethane. The concentrated samples were assayed for the presence of
Endocrine Disrupting Chemicals (estrogenic) by using MCF cell line grown in 12 well plates. The
17b-estradiol was used as standard in all cell proliferation assays.
The Relative Proliferation Effect (RPE) for source and finished waters were calculated. The
estrogenicity in water from the Salt River, the Verde River, CAP canal, raw water from water treatment
plant 1 (WTP1) and water treatment plant 2 (WTP2) averaged at 62.46%, 66.69%, 63.06%,
35.87%, and 29.87%. These levels suggest the presence of partial to fully agonistic chemicals in these
source waters. The relative proliferation effect for finished water from WTP2 and WTP1 averaged
at 24.31% and 23.58%, suggesting that treatment processes were able to reduce the levels of
estrogenic chemicals during water treatment processes. The relative proliferation effect for
chlorinated raw waters from all sources were found to be always less than respective unchlorinated
raw waters but greater than finished drinking water. This approach shows a promise to
comprehensively evaluate the effectiveness of advanced treatment processes to mitigate
cumulative estrogenicity in drinking water. Includes figures.