In Vitro Aromatase Assay: Prevalidation Studies Susan Laws, Ph.D. Endocrinology Branch Reproductive Toxicology Division NHEERL Office of Research and Development U.S. EPA

In Vitro Aromatase Assay: A cytochrome P450 enzyme complex bound in endoplasmic reticulum Catalyzes the conversion of androgens to estrogens Androstendione Estrone Testosterone Estradiol Present in ovary, placenta, testis, brain, bone, vasculature and adipose tissue Present in all vertebrates Known to be inhibited by EDCs Entire vertebrate phylum including mammals, birds, reptiles, amplhibians, teleost and elasmobranch fish , and agnatha (hagfish and lampreys). In most vertebrates, aromatse occurs in gonads and brain. (sex-related behavior such as mating responses, frequently a marked sexually dimorphic difference. Human and higher primates (more extensive tissue distribution….placenta and liver of developing fetus, adipose tissue of adult. Placenta = ungulate species such as cows, pigs, horses and sheep….note: not in placenta of rats.

In Vitro Aromatase Assay: Prevalidation Studies Historical Perspective EDSTAC recommended as alternative assay EDSP Detailed Review Paper Radiometric method Human placental microsomes Initial prevalidation studies DRP protocol Compared tissue sources for enzyme

Prevalidation Studies: Goals Optimize protocols Enzyme, substrate and cofactor concentrations Linear time course response Positive control Performance Criteria Intra- and inter-assay variation Technician variation Compare placental and recombinant microsomes (11 test chemicals) Protocol for multi-laboratory studies

Androstenedione to Estrone Reaction Mechanism: Androstenedione to Estrone Enzyme complex bound to endoplasmic reticulum of cell and comprised of 2 proteins Cytochrome P450 (hemoporteins that converts C19 steroids (androgen) to C18 (estrogens) containing a phenolic A ring NADPH-cytochrome P450 reductase, transfers the reducing equivalents to cytochrome P450 aromas. 3 moles of NADPH and 2 moles of Oxygen are used in the conversion of one mole of substrate into one mole of estrogen product. Androstenedione (perferred substrate) proceed via 3 sucessive oxidation steps, with the first two being hydroxylations of the angular c-19 methyl group. Final oxidation step, whose mechanism not fully understood, proceeds with the aromatization of the A ring and loss of the c-19 carbon atom as formic acid. This is the basis for the classical 3H-water methods used for measuring aromatase acativity, quantifies the relase of tritium from the 1beta-position of androsteindion into the aqueous phase. Km and Vmas in human placental microsomes was 68 nM and 57 pmol/min/mg protein, respectively. Positive control: 2-hydroxyandrostendione (1 uM) inhibits reactions. The heme protein is responsible for binding of the C19 androgenic steroid substrate and catalyzine the series of reactions leading to formation of the phenolic A ring characteristic of estrogen. ) (reducing equivalents are supp;ied from NADPH via a ubiquitosu microsomal flavoprotein, NADPH-cytochrome P450 reductase. Humans: ovaries and testes, placenta and fetal (not adult) liver, adipose tissue, chondrocytes and osteoblasts of bone, vasculature smooth muscle, brain (hypothalamus, limbic system, cerebral cortex). - Cytochrome P450arom and NADPH-cytochrome P450 reductase

In Vitro Aromatase Assay: Radiometric (3H20) Method

Indicators of Optimized Protocol Small fraction (10-15%) substrate converted to estrone Estrone production linear with time and enzyme concentration Estrone production dependent upon presence of enzyme and NADPH Estrone formation can be inhibited

Placental Microsomes: Product Formation versus Protein (38%) (18%) (9%) (%) conversion substrate to product Inhibited: 4OH-androstenedione (100nM) Intra-assay (triplicates) CV=3%; Inter-assay (3 exp.) CV=7.4% Aromatase Optimization Supplementary Studies (pages 11,28,29)

Placental Microsomes: Product Formation versus Time 2.7, 5.1, 10, 14, 17.5% Substrate conversion over time Inhibited= 4OH-androstenedione (100 nM) Estrogen Production: Human Placental Assay Results Quick Response Task 2 (Table 1, pages 2-3)

Placental Microsomes: Inhibition of Aromatase Activity Estrogen Production: Human Placental Assay Results Quick Response Task 3 (Figure 3)

Placental Microsomes: Intra- and Inter-Assay Variance Intra-assay (triplicates) CV 1.4-7.5% Inter-assay (3 days) CV 1.7 – 11.5% Estrogen Production: Human Placental Assay Results Quick Response Task 3 (Table 3, Figure 4)

Human Recombinant: Protocol Optimization Experiments 7, 12, 25, 33, 42% Substrate conversion over time Inhibited= 4OH-androstenedione (100 nM) Intra-assay (triplicates): CV = 1-3% Inter-assay (2 days): CV = 5 – 20% Estrogen Production: Human Placental Assay Results Quick Response Task 4 and 5 (Tables 4, 5, Figures 5-7)

In Vitro Aromatase Assay: Optimized Assay Conditions Assay Factor Assay Type Human Placenta Human Recombinant Protein (mg/mL) 0.0125 0.004 NADPH (mM) 0.3 [3H]ASDN (nM) 100 Incubation time (min) 15 Activity (nmol/mg/min) 0.053 +/-0.001 (3) 0.283 +/- 0.0005 (2) Estrogen Production: Human Placental Assay Results Quick Response Task 4

Optimized Protocols: Variability Between Assay Day and Technicians Experiment design Three technicians conducted each assay independently over 3 days Triplicate assay tubes Maximum aromatase activity determined Comparison of coefficient of variations Estrogen Production: Human Placental Assay Results Quick Response Task 4 Tables 7 and 8

Coefficient of Variation: Intra-assay, Assay Day, and Technician Variability Parameter Placenta Recombinant Triplicates 4% 3.7% Tech 1 22% 17% Tech 2 49% (12%)* 50% (11%)* Tech 3 12% 19% Day 3 36% Day 4 29% 30% Day 5 47% (10%)* 53% (15%)* (%)* CV after Tech 2, Day 5 data deleted Estrogen Production: Human Placental Assay Results Quick Response Task 4, Tables 7 and 8

In Vitro Aromatase Assay: Comparison of Test Chemicals Experiment Design Optimized protocols using placental and recombinant microsomes Test chemicals (11, positives and negatives) Complete concentration curve for each chemical ran on 4 separate days Two technicians (one ran placental, the other recombinant) Single set of test chemical concentrations shared by 2 tech. each day

Test Chemicals: Inhibitors Negative for Inhibition 4-OH-androstenedione Chrysin Ketoconazole Aminoglutethimide Econazole Genistein (?) Negative for Inhibition Nonylphenol Atrazine Bis-(2-ethylhexyl)phthlate Lindane Dibenz(a,h)anthracene

In Vitro Aromatase Activity: Comparison of Inhibition CV (20%) CV (54%) Figure 10-Placenta aromatase response curves Figure 11-Recombinant aromatase response curves

In Vitro Aromatase Activity: Comparison of Inhibition CV (47%) CV (14%) Figure 10-Placenta aromatase response curves Figure 11-Recombinant aromatase response curves

In Vitro Aromatase Activity: Examples of Data Figure 10-Placenta aromatase response curves Figure 11-Recombinant aromatase response curves

Conclusions: Test Chemical Experiment Variability between reps. is greater than expected for both assays IC50s for inhibitors (CVs ranged 7 – 49%) Technician error rather than inadequate protocol method is likely cause of variability Despite variability, both protocols correctly identified inhibitors

In Vitro Aromatase Assay: Next Steps Identify source of variability Substrate concentration Technician training Conduct additional experiment to evaluate day-to-day and technician variability (e.g, better estimate of performance criteria) 2 Tech., 3 test chemicals (8-9 concentrations in triplicates), 4 days, both protocols

In Vitro Aromatase Assay: Next Steps Rerun assays for test chemicals with incomplete curves econazole, ketoconazole Evaluate the usefulness of estrone measurement rather than 3H20 for recombinant protocol Prepare updated protocols for validation Broader concentration range for test chemicals Guidelines for data analysis and interpretation

In Vitro Aromatase Assay: Summary Protocols optimized for placenta and recombinant assays Assays produce similar data Assays differ in advantages/disadvantages High throughput assays KGN cell line CYP19/Fluorescent substrate (HTP) kit available

Acknowledgements: Endocrinology Branch Battelle Memorial Institute Columbus, OH David Houchens Paul Feder Terri Pollock Chemical and Life Sciences Research Triangle Institute RTP, NC Sherry Black RTI Technical Staff James Mathews Marcia Phillips Rochelle Tyl Endocrinology Branch RTD, NHEERL, ORD U.S. EPA RTP, NC Ralph Cooper Earl Gray Tammy Stoker Vickie Wilson Jerome Goldman OSCP, U.S. EPA Washington, DC Gary Timm Jim Kariya Jane Scott-Smith