An Evaluation of Models to Predict the Activity of Environmental Estrogens Candice M. Johnson and Rominder Suri, Ph.D.,P.E. NSF Water and Environmental Technology (WET) Center, Department of Civil and Environmental Engineering, Temple University,

1
An Evaluation of Models to Predict the Activity
of Environmental EstrogensCandice M. Johnson
and Rominder Suri, Ph.D.,P.E.NSF Water and
Environmental Technology (WET) Center, Department
of Civil and Environmental Engineering, Temple
University, Philadelphia Pennsylvania 19122 
2
(No Transcript)
3
Endocrine Disrupting effects observed in the
environment
4
(No Transcript)
5
Routes of entry for endocrine disruptors in the
environment
6
Endocrine disrupting activity is related to
wastewater treatment
Relative distance from water treatment plant
7
Removal of EDCs

• End-of-Pipe Technologies
• Ozonation, Ultrasound, Adsorbents
• Source Control Strategies
• Risk assessment hazard characterization
• Development of policy and laws
• Research and development of safer products
• Replacement of endocrine active ingredients

8
Methods for detecting potential endocrine
disruptors

• Chemical Analysis
• - target analysis
• - low limits of detection
• - rapid analysis methods
• - high throughput
• - no indication of biological activity
• Biological Analysis (Bioassays)
• - detects the activity of mixtures and
  unknowns
• detects interactions, measures net biological
  activity
• does not indicate the identity or concentrations
  of specific contaminants

9
Approaches to testing EDCs

• Chemical-by- chemical approach
• May be too simplistic and may underestimate the
  risks of chemicals
• Test mixture toxicity on a case by case basis
• Chemical mixtures vary with respect to
  constituents and to concentrations of those
  constituents, Provides site specific data
• Band-Aid but not a cure to the characterization
  of chemical mixtures (LeBlanc Olmstead, 2004)
• Component-based approach (estimating the total
  toxicity from information on identified
  components)
• A step towards a generalized understanding and
  assessment of mixture toxicity

10
Effect Directed Analysis (EDA) scheme
extraction and pre-concentration
Antagonistic activity?
Mathematical models are used to estimate the
biological effects from the concentration of
target compounds
11
Additive models

• Concentration addition (CA) model
• Independent action (IA) model (probabilistic
  model)

RP Relative Potential Cn Concentration of
Component n in the mixture IEQ Induction
equivalents in terms of a standard
Emix Predicted effect of the mixture Emax
Maximum effect Fi,(ci) activating effects
determined from the regression of the
concentration response relationships
12
CA versus IA

• Concentration Addition (CA)
• Applied to chemicals with a similar mode of
  action
• EC50 of a mixture can be predicted based on the
  EC50 values of the individual components
• Independent action (IA)
• Applied to chemicals with diverse modes of action
  
• Mixture effects predicted from precise effects of
  each individual component and at the
  concentration found in the mixture. This
  information is not readily available
• Assumes strictly independent events, may not be
  relevant in biological systems due to converging
  signalling pathways and inter-linked subsystems

13
Objective To assess the ability of additive
models to predict estrogenic activity

• Approach
• Extract hormones from wastewater influent and
  effluent samples
• Measure the estrogenic activity of the extracts
  using the Yeast Estrogen Screen (YES) Assay
• Quantify the concentrations of suspected
  estrogens using LC-MS/MS
• Estimate the estrogenicity of the extracts using
  additive models

14
Assessment of additive models
Table 1 Concentrations of hormones detected in
wastewater extracts
  Estriol 17ß-estradiol Estrone 17a-dihydroequilin
Influent (ngL-1) 8.66 5.07 0.15 679.18
Effluent (ngL-1) 6.55 ltLOD 0.65 311.61
LOD 17ß-estradiol 0.15 ngL-1
Table 2 Total estrogenic activity of the
wastewater extracts measured in the YES
Effect Level () Influent EEQ, µg/L Effluent EEQ, µg/L Reduction
50 0.0193 0.00751 61.1
15
Assessment of additive models
Predicted and observed concentration response
curves in the YES
Antagonistic- like activity is evident in both
the wastewater influent and effluent samples
16
Assessment of additive models in clean water
Predictions based on simulated samples do not
suggest that the mixture should be
interactive Clear contribution from the
wastewater matrix
Comparison of predicted and observed mixture
responses for 17ß-estradiol, estriol, estrone,
and 17a-dihydroequilin in simulated sample
17
Assessment of additive models for estimating
estrogenicity and androgenicity
18
Conclusions and Recommendations

• Incomplete degradation of estrogen hormones
  during wastewater treatment
• - 24 - gt 99 removal of
  steroid hormones from this wastewater treatment
  plant. Similar results were reported by
  Chimchirian et al., 2007
• Residual estrogenicity after water treatment may
  lead to endocrine disrupting effects in fish
• Suggested no effect concentration for
  17ß-estradiol is 2ngL-1 (Caldwell et al., 2012)
• Estrogenicity of effluent in our study is 7ngL-1
  EEQ
• No synergism or antagonism between estrogen
  hormones in clean water

19
Conclusions and Recommendations

• Other unknown components in the wastewater matrix
  may cause antagonistic responses
• Additive models are applicable to clean water
  but may be limited in their use with complex
  mixtures
• More advanced models that can capture
  interactions or antagonistic effects are needed

20
Cn- concentration of nth mixture component ? -
interaction index RP - relative potential IEQ -
Induction equivalent concentrations
Testosterone aRBPA bRDBP aRPBPA aRPDBP TEQ(µg/L) TEQ(µg/L) TEQ(µg/L)
(µg/L) aRBPA bRDBP aRPBPA aRPDBP CA Model Interaction Observed
  aRBPA bRDBP aRPBPA aRPDBP ( Error) Model  
  aRBPA bRDBP aRPBPA aRPDBP   ( Error)  
2 0 0 -2.61E-04 -5.37E-05 2 (0.1) 2.000 (0.1) 1.999
2 40 5000 -2.58E-04 -5.37E-05 2 (39) 1.437 (0.4) 1.443
2 80 5000 -2.55E-04 -5.37E-05 2 (35) 1.411(5.1) 1.487
2 160 5000 -2.50E-04 -5.37E-05 2 (46) 1.362 (0.3) 1.367
2 320 5000 -2.40E-04 -5.37E-05 2 (53) 1.270 (3.0) 1.31
2 640 5000 -2.22E-04 -5.37E-05 2 (50) 1.109 (17) 1.333
2 1280 5000 -1.88E-04 -5.37E-05 2 (53) 0.869 (33) 1.305
2 1875 10000 -1.62E-04 -5.37E-05 2 (433) 0.186 (50) 0.375
2 2560 5000 -1.36E-04 -5.37E-05 2 (96) 0.625 (39) 1.024
2 3750 10000 -1.01E-04 -5.37E-05 2 (748) 0.039 (83) 0.236
2 5120 5000 -7.13E-05 -5.37E-05 2 (133) 0.630 (27) 0.858
2 7500 10000 -3.91E-05 -5.37E-05 2 (534) 0.289 (8) 0.417
plt0.01 (These predictions are significantly
different from the observed values) a
concentration ratio of BPA to testosterone b
concentration ratio of DBP to testosterone TEQ
Testosterone equivalents
Johnson, C.M., et al., Environmental Science and
Technology. 2013
21
17ß-E2 (µgL-1) E3 17a- EQN aDBP EEQ (µg/L) EEQ (µg/L) EEQ (µg/L)
17ß-E2 (µgL-1) (µgL-1) (µgL-1) (µgL-1) CA Model Interaction Observed
17ß-E2 (µgL-1)       ( Error) Model  
17ß-E2 (µgL-1)         ( Error)  
0.0625 6.25 0.5 1200 0.1219 (123) 0.0473 (13) 0.0546
0.0625 6.25 0.5 600 0.1219 (59) 0.0878 (14) 0.0765
0.0625 6.25 0.5 300 0.1219 (34) 0.1080 (18) 0.0913
0.0625 6.25 0.5 150 0.1219 (9) 0.1181 (6) 0.1118
0.0625 6.25 0.5 75 0.1219 (6) 0.1231 (5) 0.1292
0 6.25 0.5 600 0.0594 (77) 0.0253 (25) 0.0335
0 6.25 0.5 300 0.0594 (50) 0.0455 (15) 0.0396
0 6.25 0 300 0.0362 (68) 0.0200 (7) 0.0216
0 0 4 1200 0.1858 (38) 0.1229 (9) 0.1349
0 0 4 600 0.1858 (27) 0.1633 (12) 0.1459
0.0313 6.25 0.5 600 0.0907 (50) 0.0565 (6.8) 0.0607
0.0313 3.125 0.5 600 0.0726 (51) 0.0364 (24) 0.0481
0.0313 1.563 0.5 600 0.0635 (44) 0.0263 (40) 0.0441
0.0313 0.781 0.5 600 0.0590 (42) 0.0213 (49) 0.0416
0.0313 0.391 0.5 600 0.0567 (54) 0.0188 (49) 0.0368
0.0313 6.25 0.5 1200 0.0907 (136) 0.0161 (58) 0.0384
0.0313 6.25 0.25 600 0.0791 (92) 0.0438 (6) 0.0413
0.0313 6.25 0.125 300 0.0733 (67) 0.0576 (31) 0.0439
0.0313 6.25 0.0625 150 0.0703 (64) 0.0645 (50) 0.043
0.0313 6.25 0.0313 75 0.0689 (48) 0.0680 (46) 0.0467
22
Thank you!Questions?