Screening Assays for Estrogen and Androgen Related End Points

Paul Foster

foster2@niehs.nih.gov

OR: Investigator, Know thy Assay!

Acknowledgements

• Earl Gray – EPA

• Vickie Wilson – EPA

• Tina Teng – Biomolecular Screening Branch, NTP

Overview

• A brief introduction to the USEPA Tier 1 screening battery for estrogen and androgen activity.

• Are there tiers within the Tier 1? How does the battery get used?

• Examples from some of our work (with a focus on androgen signaling).

• The move to HTS for estrogen and androgen activity – what are the issues and compromises with these data?

• Can we use this information to accurately predict chemical hazard?

EPA’s Tier 1 EDC mammalian screens

• ER binding and transactivation

• AR binding (and transactivation)

• Aromatase activity

• Steroidogenesis in H295R cells

• Uterotrophic assay

• Hershberger assay

• Male pubertal assay

• Female pubertal assay

• Focus on steroid hormones (especially E & A) and disruption of thyroid action. http://www.epa.gov/oppts/pubs/frs/publications/Test_Guidelines/series890.htm

About the Battery

• Little information on how the battery will be used.

– Does every chemical require every assay?

– Do we use the in vitro data to tailor the assays of in vivo pharmacological activity?

– Does an in vivo response “trump” an in vitro response?

• No in utero exposure in Tier 1

• What comes next? What will Tier 2 look like?

• Will we base regulatory decisions only on Tier 1 screening data?

ER and AR receptor binding and transactivation assays

• Basic methodology for receptor binding is robust and has been used by the scientific community for decades.

– Source of receptor (rat uterine and prostate cytosol)

– Full length receptor plus various co-factors present

– Displacement of (radioactive) ligand

– Cannot directly determine agonist from antagonist

PMD Foster 7

Androgen Receptor Binding

0

20

40

60

80

100

120

DHT

FLUT-OHDDE

DBP

0 .005 .05 .5 5 50(uM)

% 3

H-R

1881

Bin

ding

Receptor - Transcription ComplexReceptor - Transcription Complex

EstrogenEstrogenReceptorReceptor

EstrogenEstrogenReceptorReceptorERE

ERE

Repressor

ActivatorActivator

CoactivatorsCore PromoterTATA box

CodingRegion

L

L

1

2

3

ER and AR receptor binding and transactivation assays

• Basic methodology for receptor binding is robust and has been used by the scientific community for decades.

– Source of receptor (rat uterine and prostate cytosol)

– Full length receptor plus various co-factors present

– Displacement of (radioactive) ligand

– Cannot directly determine agonist from antagonist

• Use of animal material can be problematic

– Human recombinant protein

– Chimeric constructs (LBD [rat and human] with yeast reporter) - NCGC

– LNCaP (mutated hAR) - in ToxCast

Human recombinant ER binding

ER and AR receptor binding and transactivation assays

• Basic methodology for receptor binding is robust and has been used by the scientific community for decades.

– Source of receptor (rat uterine and prostate cytosol)

– Full length receptor plus various co-factors present

– Displacement of (radioactive) ligand

– Cannot directly determine agonist from antagonist

• Use of animal material can be problematic

– Human recombinant protein

– Chimeric constructs (LBD with reporter)

• Patent issues with human AR

• No metabolism

Cell based assays

• Transactivation

– Transient vs stable transfection (EPA uses HeLa; ERE driven by MT promotor)

– Issues with construct employed

• Cell type, endogenous vs transfected receptor, choice of reporter

– Can easily look at antagonist activity

– Issues of cytotoxicity (particularly important with antagonists)

– NB there is no protocol for AR transactivation in the EPA battery

PMD Foster 13

NHCCHCH3

CH3

O

NO2

CF3

NHCNCH3

OCH3

O

Cl

Cl

Flutamide

Linuron

Chemical Structures of Flutamide and Linuron

PMD Foster 14

-10 -9 -8 -7 -60

250

500

750

1000

DHTDHT + 10 -6 M LinuronDHT + 3x10 -6 M LinuronDHT + 10 -5 M Linuron

Log Dose DHT (M)

Nor

mal

ized

Res

pons

e

1.07Flutamide75.8Linuron

Kb (10-8M)AR antagonist

Linuron is a Competitive Antagonist of Androgen Receptor-Mediated Gene Transcriptional Activity

McIntyre et al TAP 167: 87-99 (2000)

Linuron-Induced Testicular and Epididymal Lesions (PND 100)

McIntyre B.S., et al. (2000) Toxicol. Appl. Pharmacol. 167:87l

PMD Foster 16

-12-11 -10 -9 -8 -7 -6 -5 -40

25

50

75

100

125

Log Dose (M)

E2HPTEHPTE + 10-7 M E2

-12 -11 -10 -9 -8 -7 -6 -5 -40

25

50

75

100

Log Dose (M)

Perc

ent R

espo

nse

ER-α ER-ß

HPTE is an ERα Agonist and ERß Antagonist in HepG2 Cells

PMD Foster 17

Activity of HPTE in HepG2 AR Assay

-12 -11 -10 -9 -8 -7 -6 -5 -40

25

50

75

100

125

150

Dihydrotestosterone

M DHT-7HPTE +10

Log Dose (M)

Perc

ent R

espo

nse

Other Tier 1 in vitro assays

• Aromatase.

– Conversion of androgens to estrogens (eg testosterone to estradiol).

– CYP 19

– Use of human recombinant aromatase for activity measurement

– Fenarimol would be one of the classic environmental agents that inhibit aromatase activity. A number of triazole fungicides also inhibit CYP 19.

• Steroidogenesis

– Original assays used isolated testicular Leydig cells or testis tissue

– Now using H295R cells (human adrenal tumor cell line) and measures the production of testosterone and estradiol (ELISA), plus an evaluation of cytotoxicity.

– A number of fungicides are excellent inhibitors of CYP enzymes involved in steroidogenesis.

PMD Foster 19

Effect of Triazole fungicide on Leydig cell steroidogenesis

0.3 1 3 10 300

10

20

30

0

2

4

6

8

10

Triazole µM

Test

oste

rone

ng/

mill

ion

cells

Prog

esta

gens

ng/

mill

ion

cells

And

rost

ened

ione

ng/

mill

ion

cells

Prog

17 ProgTesto

A4

0

30

20

10

0

hCG stimulated 0-24h

PMD Foster 20

Effect of a Triazole & KTZ on rat Leydig cell Testosterone production

concentration µM

0.01 0.1 1 10 100

200

150

100

50

0

Test

oste

rone

ng/m

illio

n ce

lls

TriazoleKTZ

0-24h hCG stimulated

PMD Foster 21

3000

2500

2000

1500

1000

500

0

Effects of a Triazole fungicide and Ketoconazole on human Leydig cell steroidogenesis

triazole

KTZ

hCG stimulatedconcentration µM (log)0 0.1 1 10 1000.01

Test

oste

rone

ng/m

illio

n ce

lls

In vivo Screens

• Based on pharmacological assays that have been used for decades

• Rat Uterotrophic assay– Short term 3 days;

– sc route to maximize exposure

– Non-functional HPG axis (OVX adult female preferred over pre-puberty immature female)

– Metabolism included (TK different?)

– Measure uterine weight

– Can be adapted for antiestrogens (but not part of EPA or OECD protocols)

– Only from signals with ER?

Hershberger assay

• Short term bioassay to determine androgen or antiandrogen action.

• 10 days exposure via sc or oral route (relevant to human exposure?)

• Non-functional HPG axis (use of pubertal castrate males)

• Metabolism included

• Measure growth of androgen dependent organs (VP, SV, LABC, Cowper’s glands, glans penis)

• Can be adapted for antiandrogens by injecting castrate animals with a reference androgen (eg testosterone proprionate) and then looking for reductions in androgen dependent organ weights.

• Oral administration studies of anti-androgens show very good predictivity for malformation induction after in utero exposure.

• Only for signals from AR?

PMD Foster 24

Hershberger Assay

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

Sem

inal

Ves

icle

Wt (

g)

Corn Oil TP(50ug) Flut+TP DDE+TP DBP+TP

Antagonist Activity

Rat Pubertal Assays

• Longer term assays (up to 20 - 30 days treatment in life)

• Oral exposure

• Intact HPG axis

• Metabolism included

• Can measure agonist and antagonist activity, plus interference with steroid action, effects on pituitary control etc.

• Includes organ weights, pathology plus hormone assays etc

• Advances and delays in puberty could be considered adverse effects. Only 2-dose levels required. Care in normalization of data.

• Dose range finder for Tier 2 studies.

• Directed from previous Tiers?

– AR + Hershberger = male pubertal? ↓steroidogenesis = pubertal?

PMD Foster 26

Assign to Treatments based

upon body weight. 15/group

Dose Female

Necropsy

Immature (21 - 43 Days of Age) Pubertal Female Rat Protocol to Evaluate Pubertal Development and Thyroid Function. Detects inhibition of steroidogenesis, antithyroid and (anti)estrogenic

activities and altered HPG maturation.

Wean

Days of age

Required endpoints: Growth

Age and weight at vaginal opening (VO) and Vaginal cytology Serum thyroxine and thyroid-stimulating hormones Ovarian and uterine weights and histology Liver, kidney, pituitary and adrenal weights

Optional endpoints:Serum hormone levels, ex vivo ovarian steroidogenesisOther organ histology and function

Dose dailyDaily examination for VO and lavage

PMD Foster 27

Assign to Treatments based

upon body weight. 15/group

Dose Male

Necropsy

Immature (21 - 52 Days of Age) Intact Male Rat Protocol to Evaluate Pubertal Development and Thyroid Function. Detects inhibition of

steroidogenesis, antithyroid and (anti)androgenic activities and altered HPG maturation.

Wean

Days of age

Required endpoints: Growth

Age and weight at preputial separation (PPS)Serum thyroxine and thyroid-stimulating hormones

Thyroid HistologySeminal vesicle plus coagulating gland weight (with fluid)

Ventral prostate weight, Levator ani/bulbocavernosus muscle weightTestis and epididymal weights and histology

Liver, kidney, adrenal and pituitary weights Optional endpoints

Serum hormones and ex vivo testis and pituitary hormone production

Dose daily

Daily examination for PPS

Battery data

• AR binding and TA -ve

• ER binding -ve

• ER TA ±

• LC Steroidogenesis -ve

• Uterotrophic assay -ve

• Hershberger assay -ve

• Female pubertal -ve

• Di-n-butyl phthalate

Some issues noted with HTS measurements of estrogen and androgen activity in Tox 21

• Shift to need for human receptors or assays in human cells.

• Need to scale assay to use in HT format.

• No assay for steroidogenesis or aromatase activity.

• No metabolism included.

• Use of rat or hLBD Gal-4 chimeric receptor (NCGC) – no HRE. Better at agonists than antagonists. Rat and human differences.

• Use of AR from LnCaP cells (Toxcast). Mutated human receptor (binds progestins better than androgens).

• Dose levels and cytotoxicity still complicates interpretation. Do we want an EC50 for cytotoxicity?

• Curve fitting methods to generate AC50’s.

• Other points to consider:

OCH 3N C

OC

H2 ON

CF3 CH 3R

R=OH, HydroxyflutamideR=H, Flutamide

CH3

Vinclozolin, Metabolite

CH=CH 2

OH

CH3

HC

OCN

Cl

Cl

Cl

Cl

Linuron

CH 3H

OCN N

O PS

O-CH 3O-CH 3

CH 3

2ON

Fenitrothion

NH

C X

Substituents O

: X= C, N

Essential structural moiety

Fenitrothion

-11 -10 -9 -8 -7 -6 -5 -40

25

50

75

100

125 DHTDHT + 10-7M FenDHT + 3x10-7M FenDHT + 10-6M Fen

Log Dose Dihydrotestosterone (M)

Fenitrothion is an AR competitive antagonist in HepG2 cells

Kb = 2.18 x 10-8 M

PMD Foster 32

In Utero Fenitrothion Exposure Induces Developmental Toxicity

Fenitrothion (mg/kg/day)0 5 10 15 20 25

# Li

ve P

ups

Per L

itter

0

2

4

6

8

10

12

14

16

**

# of Live Pups

Fenitrothion (mg/kg/day)0 5 10 15 20 25

Prop

ortio

n of

Pup

s B

orn

Aliv

e0.0

0.2

0.4

0.6

0.8

1.0

1.2

****

Mean ± SE, * p<0.05, ** p<0.01, *** p<0.001

Pups born Alive

Turner K.J. et al. (2002) Tox. Sci. 68: 174

PMD Foster 33

Effects of Prenatal Fenitrothion Exposure on Anogenital distance (AGD) and the Retention of

Areolae

Fenitrothion (mg/kg/day)0 5 10 15 20 25

Mea

n A

GD

on

PND

1 (m

m)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

***

FemaleAGD

p=0.06

Fenitrothion (mg/kg/day)0 5 10 15 20 25

Mea

n nu

mbe

r of a

reol

ae p

er ra

t

0

2

4

6

8

10

12

***

Maximum # of Areolae

Turner K.J. et al. (2002) Tox. Sci. 68: 174

Male Offspring PND 100

No dose-dependent effects on:

• Nipple retention

• AGD

• Incidence of gross malformations

• Androgen-dependent organ weights

• Histopathology

• The OP toxicity limits the ability to characterize anti-androgenic activity

Conclusions

• There are a number of useful screens for the detection of estrogen and androgen activity,

• The EPA battery provides some internal consistency and redundancy for increasing the levels of organization and complexity of assays.

• How to use the battery?

– All chemicals get all tests?

– Use the in vitro signals to define the next in vivo assays?

• Some data on ER and AR from the HTS efforts needs very careful examination and should be placed in context with the Tier 1 battery data.

– Significant issues with false positive and negatives from HTS and even binding versus TA.

• Currently, the short term in vivo studies are much better predictors of hazard than in vitro assays.

PMD Foster 36

“Where observation is concerned, chance favors only the prepared

mind.”

- Louis Pasteur (1854)