Principal Investigators Meetings

Prostate Cancer and Androgens
Sandra Gaston

---

Table of Contents:

• Title and Author
• Abstract
• Prostate Cancer and Androgens
• Yeast Based AR Bioassays: Basic Elements
• AR Yeast BioAssay: Basic Protocol
• AR Bioassay vs Androgen Immunoassays
• Bioavailability of Serum Testosterone
• Human Serum Samples
• Titration of Bioavailable Androgen in Human Serum Samples
• Analysis of Nutritional Supplements for AR-active Compounds
• AR Bioassay: Biomarker for Soy Based Dietary Interventions for Prostate Cancer
• Tumor Response
• LNCaP Implant Tumor Response and Serum Biomarkers
• The Living Chip™ – A Nanotiter Plate
• Parallel Assay Initiation
• AR Biochip Screening System
• Yeast Cell Culture in Chip
• Implementation of AR Bioassays in Chip
• Tissue Sampling Device: Single Micro-Channel
• Tissue Sampling Device: Micro-Array Format
• New Biochips for the Analysis of Tissue Samples
• Acknowledgements

---

Prostate Cancer and Androgens

Sandra Gaston
Urological Research Laboratories

---

Abstract

The major objective of this project is the design and fabrication of a bettery of micro-scale yeast-based bioassays in a biochip format that canb e used to monitor androgen receptor (AR) ligands in sera and tissue from patients with prostate cancer. In contrast to the immunoassays currently used to measure steroid hormones, yeast-based AR micro-bioassays are designed to assess the receptor response to all of the available ligand in a complex biological sample. We have now formatted the assay so that individual differences in bioavailable androgen can be assessed across thy physiological range, with the goal of supporting individualized hormonally based bioassays using Living Chip™ technology. The Living Chip™ consists of a precisely constructed, high-density array of through-holes, or bottomless wells, in a plate. Each through-hole can hold approximately 30 nl of liquid via surface tension. Instrumentation is currently being developed at BioTrove to perfrom the yeast-based AR bioassays on human sera samples using the biochip technoogy.

In addition, to more completely characterize the endocrine microenvironment within solid tissues and tumors, we have begun development of a series of devices that will allow us to measure receptor expression in parallel with bioavailable ligand. Toward this goal, we have designed and fabricated a highly sensitive rt-PCR instrument to measure fluorescence polarization of sub-micro liter assay volumes utilizing theoretical models to predict and optimize components and performance. Ongoing characterization of the assembled instrument has indicated desired performance specifications are achievable given further optimization of the design. A micro-sampling device, currently in the early stages of development, will be utilized to permit simultaneous rt-PCR and bioassay analysis on each sub-microliter tissue sample. This integrated analytical strategy, on a scale compatible with needle biopsies, will provide both a means for direct assessment of critical variables in the tissue/tumor microenvironment and a strategy for monitoring those variable in response to hormonally based therapies.

---

Prostate Cancer and Androgens

---

• Prostate cancer is an androgen dependent malignancy (at least at first)
• When primary therapy fails, treatment for metastatic disease includes androgen ablation (surgical or chemical castration)
• Androgen independent prostate cancer emerges in the later stages of the disease

Changing Role of Hormonal Therapy in Prostate Cancer Management

---

• Many patients with asymptomatic relapse of their prostate cancer prefer an alternative to androgen ablation
• Men at high risk of prostate cancer (primary and recurrent) are keenly interested in endocrine-active dietary supplements
• Prevention and management of androgen insensitive relapse is an ongoing challenge

---

Yeast Based AR Bioassays: Basic Elements

The basic design of yeast-based steroid hormone bioassays follows from the molecular characteristics of steroid hormone receptors, including the androgen receptor. Steroid receptors are ligand modulated transcription factors, and in the presence of specific ligand they bind to DNA sequences (response elements) in the promoter region of hormonally responsive genes and activate transcription. In the transcriptional AR bioassay illustrated above, steroid hormones diffuse freely in and out of the yeast cells. In the presence of specific hormone, the androgen receptor binds it's ligand, forms dimers, binds to the androgen response elements in the reporter promoter and activates transcription of the reporter gene. b-galactosidase activity is a thus a function of the net concentration of AR ligand available to activate transcription.

AR bioassay response to Dihydrotestosterone (DHT) is illustrated above right (DHT is the primary androgen in the prostate gland).

---

AR Yeast BioAssay: Basic Protocol

---

AR Bioassay vs Androgen Immunoassays

In a clinical setting, serum concentrations of the major steroid hormones are determined by immunoassay, and these assays are designed to be chemically specific. However, the biological activity of a particular hormone is a function of it's binding to a specific steroid hormone receptor and of that receptor's response in a live cell. Yeast based steroid hormone assays, in which a specific mammalian steroid receptor and a reporter gene is introduced into a yeast host cell (Saccharomyces cerevisae), have proven to be useful for evaluating the biological activity of natural and synthetic steroid hormones in complex biological samples.

Unlike immunoassays, yeast-based bioassays are designed to assess the net response to all of the available AR ligands in the sample, and these bioassays can be formatted to detect both agonists and antagonists. In addition, yeast based AR bioassays can be designed to evaluate the impact of AR mutations and coactivators on the ligand response. In addition, in contrast to standard immunoassays and biochemical determinations, yeast based assays discriminate between biologically available and total steroid hormone. This is a critical distinction when evaluating potential ligands in blood or tissue extracts, because the fraction of the ligand that is tightly bound to protein may significantly alter it's effective concentration in vivo .

The use of yeast based bioassays for steroid receptor ligands in a clinical environment is currently limited by the availability of human expertise and labor, the cost of reagents and the amount of biological sample. What is needed, and what we are developing, is an automated, inexpensive system for performing a large number of micro-scale bioassays in a clinical setting.

---

Bioavailability of Serum Testosterone

There are three major testosterone fractions in human serum

• 30% bound to steroid hormone binding globulin (SHBG)
• essentially unavailable
• 60% bound to albumin
• partially available
• 2% not bound to protein ("free")
• available

NOTE: IN MICE AND RATS - NO SHBG IN THE SERUM! Extrapolation of serum androgen correlates from mouse models to human patients must incorporate this physiological difference

---

Human Serum Samples

---

Titration of Bioavailable Androgen in Human Serum Samples

Physiological range for immunoassay testosterone fractions:
Males >50 years old:	7.4-26 nM total T	37-85 pM free T
Males 20-49 years old:	9.4-60 nM total T	43-134 pM free T

---

Analysis of Nutritional Supplements for AR-active Compounds

Yeast based bioassays can be utilized to detect estrogenic and androgenic compounds in the diet and in the environment. PC-SPES is a complex nutritional supplement widely utilized by prostate cancer patients, with variable results. ER yeast bioassays demonstrated that PC-SPES extracts contain estrogenic activity (DiPaola et al. NEJM 339:785-791, 1998). Our AR yeast bioassay showed no AR ligand in these extracts, but detect an antagonist for AR-activation by serum androgens (see Sera A, B and C) and by testosterone.

---

AR Bioassay: Biomarker for Soy Based Dietary Interventions for Prostate Cancer

Endocrine active dietary supplements are of great interest to many patients and clinicians concerned with prostate cancer management. Effective use of such supplements depends upon a better understanding of their in vivo effects on androgen dependent tumor growth. We have utilized a well established orthotopic implant animal model to evaluate serum androgen and tumor responses to three dietary soy dietary supplements. Because total serum testosterone has proven to be a surprisingly poor marker of prostate tumor response to soy dietary supplements, we evaluated serum androgens with both bioassay and immunoassay techniques.

Methods:
Each of 30 male severe combined immunodeficiency (SCID) mice was inoculated intraprostatically with 2 million LNCaP cells and randomized into one of four experimental groups: control diet, soy protein supplement (isoflavone depleted), soy phytochemical supplement or genistin supplement. When the experiment was finished at ten weeks, each tumor was excised and weighed and serum samples were assayed for total testosterone and DHT using standard ELISA immunoassays. In addition, bioavailable serum androgen was evaluated with a microscale AR yeast bioassay that measures androgen receptor (AR) ligand as a function of androgen dependent transcription in an intact target cell. Statistical correlations between tumor mass and serum androgen measurements were evaluated by standard linear (Pearson) analysis.

Results:
Consistent with previous studies, each of the three soy dietary supplements showed anti-tumor activity in this SCID-LNCaP prostate cancer model. Interestingly, AR bioassay serum androgen is positively correlated with tumor mass across all of the experimental groups, while total serum testosterone and total serum DHT are not. This positive correlation between bioassay androgen and tumor response is strongest in the soy protein supplement group.

Conclusions:
In contrast to total serum testosterone, bioassay serum androgen may be useful in designing and monitoring soy dietary supplement protocols that inhibit prostate tumor growth in vivo . Our microscale bioassay is particularly useful in assessing soy dietary interventions in mouse models of prostate cancer, where available serum volumes are an important limitation.

---

Tumor Response

	Tumor response (weight in grams)
		mean (SD)		change (P value)
Control Diet		0.60 (0.40)		--
Soy protein		0.34 (0.15)		-0.26 (0.018)**
NIC		0.17 (0.23)		-0.43 (0.091)
Genistin		0.26 (0.27)		-0.34 (0.195)

---

LNCaP Implant Tumor Response and Serum Biomarkers

---

The Living Chip™ – A Nanotiter Plate

---

Parallel Assay Initiation

Parallel mixing of assay components takes place when plates are stacked such that the through-holes align.

---

AR Biochip Screening System

---

Yeast Cell Culture in Chip

---

Implementation of AR Bioassays in Chip

Chip 1: Sera spiked with testosterone + AR transactivation yeast

Chip 2: FDG fluorescent substrate.

---

Tissue Sampling Device: Single Micro-Channel

---

Tissue Sampling Device: Micro-Array Format

---

New Biochips for the Analysis of Tissue Samples:
Measuring AR mRNA Expression in Parallel with the AR Biossay

A: Schematic of microscale rt-PCR Assay

Tissue is homogenized by apposing a biochip containing homogenization buffer to the surface of the tissue sampling device and mixing with an array of plungers. A third biochip dip-loaded with reverse transcriptase, RT-PCR primers and reagents is apposed to the stacked chips. MRNA is reverse transcribed and teh mutant allele amplified using TaqMAMA-a novel assay that combines the 5’ fluorogenic TaqMan assay and the mismatch amplification mutation assay.

B: Real time microscale RT-PCR using fluorescence polarization

Fluorescence polarization in combination with real-time PCR analysis and allele-specific amplification (as in A.) can be used to quantitate mRNA expression. When fluorescent molecules are excited by plane-polarized light, they emit polarized fluorescence into a fixed plane (Perrin 1926). The molecules rotate under brownian motion during the excited lifetime of the fluorescence causing a net depolarization of the emission. The rotation rate and therefore the depolarization of the fluorescent emission are proportional to the relaxation time of the molecule which is a function of solvent viscosity, temperature, and the molecular volume. For measurements at constant viscosity and temperature, FP is proportional to molecular volume/weight. Therefore a large fluorescent molecule rotates and tumbles more slowly preserving FP while a small molecule rotates and tumbles faster resulting in depolarization (panel I.). Since FP is expressed as the ratio of fluorescence detected in two perpendicular planes, a detector can differentiate between labeled primer and product in the same well. Panel II is a depiction of the fluorescence polarization instrument. Laser light (488nm Argon Ion) is passed through a large pinhole and polarizer onto the surface of the stacked chips. The signal is collected perpendicular to the direction of excitation. Emitted light is collected by an objective, passed through a high pass filter (513nm cutoff), then separated by a polarizing beam splitter into two detection arms of the instrument. Each detection arm contains an additional polarizer and a CCD camera.

---

Acknowledgements

Massachusetts Institute of Technology
Leonard Lerman

Beth Israel Deaconess Medical Center/Harvard Medical School
Jin-Rong Zhou, William C. DeWolf, Glen Bubley, Rena Nassr, Hershey Foundation Prostate Cancer Serum and Tissue Bank

BioTrove, Inc.
Holly Allen, Colin Brenan, Linda Kiley, John Linton, Tom Morrison, Elen Ortenberg, Patrick Owens, Mahima Santhanam, Kristine Stone, Karl Yoder

National Cancer Institute (AR Biochip)
Award No. R21CA86365-01

CaPCURE Research Award (AR Biochip)

Advanced Technology Program (Antibody Phage Display)
Award No. 70NANB1H3003