[Interview] Sentinels for detection of estrogenic compounds

After a PhD in cellular and molecular medicine, Daniel Gorelick continued to study how steroid hormones affect development and behavior. He already wrote a guest post about enhancer discovery few years ago, and some days ago I had occasion to speak with him about his last paper. In a recent Endocrinology issue, Daniel describes a transgenic ERE-Luc zebrafish developed to visualize estrogen receptor activation. This is of extreme interest to me because I previously used an ERE-Luc mouse to study some pharmacological and nutritional aspects of ER activity. Here, I summarise part of the conversation with Daniel in an interview.

5 day old ERE-Luc transgenic Danio rerio incubated in water containing 100 μg/L estradiol have fluorescent liver (arrowheads), indicating that ERs are activated in the liver. Scale bar = 200 μm.
  #1 - Daniel, what prompted you to study steroid hormones like estrogens?
Steroid hormones are fascinating molecules. They regulate diverse cellular and physiological processes in different tissues throughout development and in adulthood. I was drawn to estrogens in particular because of their actions in the central nervous system. For decades scientists have recognized that estrogens affect the development of neuronal circuitry that regulates behaviors such as aggression and mating. There is increasing evidence that estrogens regulate cell proliferation and death in the brain, as well as more disparate processes like inflammation.

#2 - As a model, what are the advantages of zebrafish compared to mouse?
Zebrafish and mice models are two of many tools in the scientist's tool box. Every tool has unique advantages and limitations. Zebrafish embryos are transparent and develop externally, allowing scientists to observe development in live embryos from the moment of fertilization. It's difficult to perform similar observations in mice. For example, we were able to monitor estrogen receptor activation in live zebrafish embryos and larvae through 5 days post fertilization and identify a novel site of potential estrogen receptor activity in developing heart valves. Doing the comparable experiment in mice would be impossible. Zebrafish are small and reproduce easily, producing several hundred embryos from a single mating, compared with the dozen or so progeny from a typical mouse mating. These qualities make performing high-throughput chemical and mutagenesis screens straightforward and economical in zebrafish.

#3 - Today's scientists are used to genome-wide scenarios but are still focusing on single cells, at single time-points. Don't you feel alone in following one (reporter) gene in different organs at multiple time points?
As a graduate student I received good advice from my mentor: Don't worry about what other people are doing. We have a lot to learn about how estrogens influence development, and a good way to do that is to identify the tissues and cell types that respond to estrogens during embryogenesis and organ formation. I don't worry that many publications focus on genome-wide estrogen receptor binding (for example) in a single cell type at a single time-point. My guess is that interesting discoveries will be made using both approaches.

#4 - Let's talk about reporter design. You have chosen to combine five tandem consensus sequences: older approaches would have used a target gene promoter. What are the advantages of your design? Do you had difficulties in having the referee accepting this strategy?
The limitations of using a gene promoter is that the reporter can only monitor gene expression in tissues where that promoter is expressed. For example, the vitellogenin gene promoter responds to estrogen. Several labs have taken fragments of this promoter, placed them upstream of a reporter gene and generated transgenic zebrafish that report estrogen receptor activation. But this vitellogenin gene is only expressed in the liver, so you can't monitor estrogen receptor activity in other tissues, such as the heart or brain. Using tandem estrogen response elements without a promoter allowed us to monitor estrogen receptor transcriptional activation in every tissue, not just the liver. Luckily, several groups had pioneered this strategy previously in cultured cells and in mice so the journal referees had no trouble accepting this strategy.

#5 - Estrogens are quite active in the brain. Do you think that your model could help in coupling behavioral responses to specific maps of neuronal activation by estrogens?
I hope so! This is exactly one of the reasons why I generated the estrogen reporter zebrafish. We've done some studies in adult zebrafish demonstrating that we can detect estrogen-responsive cells in the brain. Correlating estrogen-response with a behavioral response, and comparing the number and type of estrogen-responsive cells in males versus females is proving to be more challenging that I originally thought. But I am optimistic that we will make some interesting observations in the future.

Original publication
Gorelick and Halpern, Visualization of Estrogen Receptor Transcriptional Activation in Zebrafish. Endocrinology 2011 vol. 152 no. 7 2690-2703.