Keywords

Elwood Jensen’s discovery of the estrogen receptor (ER) made a paradigm shift toward our understanding of steroid hormone action. It launched the field of nuclear receptors, which has profoundly impacted the discipline of molecular medicine. A perfect example of “Bench to Bedside” translational research, his work has saved thousands of lives of breast cancer patients. Moreover, his work has led to the understanding of how ligand-dependent transcription factors mediate the cell-type-specific gene expression in amplifying hormonal actions. It is fitting that many of his friends and colleagues, who are leaders in the field, have contributed chapters in this book. This is a testament that Elwood Jensen was regarded in high esteem among scientific contributions and their impact on the field with a historical perspective.

1 The Elusive Mechanism of Estrogen Action

Adolf Butenandt and Edward Doisy independently purified estrogen in 1929. Over the following decades, observations accumulated that tiny amount of the hormone could cause profound target tissue growth (e.g., uterus). However, the mechanism of its action remained elusive. In the 1950s, being the era of enzymology, research community assumed that enzymes mediate the hormone-dependent tissue growth. The popular belief was that the mechanism of estrogen action entailed trans-hydrogenation in which the 17-hydroxyl group of estradiol is oxidized by one coenzyme and the resulting estrone reduced by another, thus using NADH to produce NADPH [1]. However, there was one caveat with this thought process: such a mechanism would not explain the uterotropic actions of diethylstilbestrol (DES), a synthetic estrogen that lacks any aliphatic hydroxyl group and thus could not undergo that reversible oxidation/reduction [1].

2 Fellowship in Zurich and the Matterhorn Experience

While doing a steroid chemistry fellowship in Zurich with the Nobel Laureate, Professor Leopold Ruzicka, Elwood was fascinated by the natural beauty of the surrounding areas. He was particularly attracted toward the towering Matterhorn. Although never climbed a mountain, physically, he was in good condition from his collegiate sports activities (Boxing/Judo/Tennis) and decided to scale the Matterhorn. He teamed up with a lab colleague with mountaineering experience and a guide to climb the Matterhorn (Fig. 1) from an alternate route (Swiss side), rather than from the seemingly simple but more hazardous Italian side. The latter approach was used by Edward Whymper to scale the Matterhorn peak but at the cost of many unsuccessful attempts and a few human lives. Matterhorn was the last European mountain to be climbed. The successful Matterhorn experience by a novice like Elwood Jensen instilled a lifelong passion of applying “alternative strategy ” approaches in his research pursuits [1].

Fig. 1
figure 1

At the base of the Matterhorn with his first wife Mary Jensen

Full size image

3 Faculty Position at the University of Chicago: From Chemist to Endocrinologist

When Charles Huggins, who won the Nobel Prize for his work on prostate cancer, recruited Elwood at the University of Chicago, a vexing question in the endocrinology field was “how does tiny amount of estrogen induce massive uterine growth” [1]? Elwood Jensen, himself a chemist, and his postdoctoral fellow, Herbert Jacobson, who earned his PhD degree with the famous chemist Morris Kharasch at the University of Chicago, embarked upon solving an agelong endocrinology problem: what is the mechanism of estrogen action? Elwood invoked the “alternative approach” and decided to understand the fate of the hormone itself rather than what hormone does to the tissue—the prevailing approach in the field at the time . Because estrogens are active at such low doses (in nanomolar range), they planned to label the hormone with tritium and follow the radioactivity in various rat tissues. However, their experimental strategy entailed using the hormone radioactively labeled to prohibitively high specific activity, normally not permitted by the university regulatory authorities. But as luck would have it, the “Fermi Lab”—an epicenter of the Second World War “Atomic Bomb Project”—was located in the nearby Argonne National Laboratory and was made accessible to the Jensen team [1]. They built an apparatus, tritiumator (Fig. 2), to measure the uptake of tritium by a catalytic reduction of a double bond in the precursor. They reasoned that one could radiolabel the sixth and seventh position of the hormone with carrier-free tritium gas. Thus, using the Fermi Lab facilities to handle carrier-free tritium (60 Ci/mmole), they succeeded in labeling high specific activity estradiol. When they injected the tritiated estradiol to immature rats or to castrated rats, to their surprise, they found that the hormone remained biochemically unchanged and the uterus showed the usual massive growth. Moreover, when they examined various rat tissues for the uptake of radioactive estradiol (Fig. 3), they found uptake and retention was 100-fold higher in uterus and vagina than in nonreproductive tissue such as blood [1, 2]. Some skeptics raised the concern that estrogen might have undergone oxidoreduction of its 17-beta hydroxyl group such that the hydrogen atom lost during oxidation is not the same one that replaces it during reduction. Jensen and colleagues addressed this by injecting a mixture of 6,7 tritium-labeled estrogen plus 17-tritium-labeled estrogen in their rat model and clearly demonstrated that there was no loss of tritium from position 17 of estrogen during the hormone -induced uterotropic growth [1].

Fig. 2
figure 2

Tritiation apparatus designed by Jensen and Jacobson

Full size image
Fig. 3
figure 3

Selective uptake and retention of tritiated estradiol by reproductive tissues (uterus and vagina)

Full size image

4 Birth of the Nuclear Receptor Family

When Elwood announced his groundbreaking findings at the International Congress in Vienna, five people came to listen to him—three of whom were speakers. Whereas, in a concurrent plenary session, hundreds went to hear the now debunked enzymatic theories of estrogen action! The “factor” that Jensen initially termed “estrophilin” is now known as “estrogen receptor” (ER). This momentous discovery shifted attention away from the involvement of enzymes in the mechanism of hormone action. Subsequently, Elwood’s contemporary, Jack Gorski at University of Illinois at Urbana/Champaign used state-of-the-art sedimentation gradient procedures to isolate and characterize a macromolecular component, which possessed the attributes of a specific receptor for estrogens [3]. The sedimentation density gradient would go on to play a critical role in the Jensen laboratory in research related to ER. These findings stimulated the search for other hormone receptors. The pioneering work by John Baxter, Pierre Chambon, Ron Evans, Jan-Ake Gustafsson, Bert O’Malley, and Keith Yamamoto led to the discoveries of the glucocorticoid receptor, progesterone receptor , retinoic acid receptor, and orphan receptors. In a remarkably short span of time, the 49 nuclear receptors described to date have become a family—“Nuclear Receptor Family” [4]. At the 2004 Lasker Award ceremony, Nobel Laureate, Joseph Goldstein, paid tribute to the discovery and called Elwood Jensen the patriarch and estrogen receptor the matriarch of the family [5]. The Lasker Foundation recognized these discoveries with Lasker Awards to Drs. Jensen, Chambon, and Evans; many in the field believe that it is deserving of recognition by the Nobel Committee as well because of its significant basic science and clinical impacts .

5 Estrogen Receptor and RNA Synthesis

After distinguishing two forms of the receptor [cytoplasmic and nuclear], Elwood Jensen, as well as Jack Gorski [3, 6], showed that the hormone-receptor complex becomes tightly bound in the nucleus and enhances RNA synthesis (transcription) in nuclei specifically isolated from hormone-dependent tissues [7]. Shortly thereafter, Bert O’Malley’s group used estrogen-stimulated chicken oviduct system and published landmark papers not only describing the receptor for progesterone but also showing that it also stimulated transcription of specific mRNAs [8, 9]. This phenomenon of hormone-induced receptor activation has since proved to be a key step in the actions of various classes of steroid hormones, and it identified a definitive biochemical role for the steroid .

6 Estrogen Receptor Domain Structure and Ligand-Dependent Receptor Dimerization

In a relatively short period of time after the discovery of ER, many labs contributed toward identifying the domain structures of the receptor protein and establishing that indeed the family of steroid receptors shares several common features in their domain structures, enabling them to utilize a similar mode of action. Based on sequence homology and other criteria, the estrogen receptor protein can be divided into six functionally and physically independent domains (A–F) . These domains are required for DNA binding (region C), nuclear localization (region D), and steroid binding (region E). The ER has two well-characterized transcriptional activation functions, which is located in the N-terminal A/B region and AF2 that is located in region E and whose activity is ligand dependent. The ER has been shown to form stable homodimers in solution [10], and several studies have provided evidence that a number of nuclear receptors including ER bind to their response elements as dimers [10,11,12,13]. However, Gorski et al. proposed a model where the ER protein bound to an ERE either as a monomer or with a heterodimeric partner [14, 15]. Most of the data for and against ER dimerization used in vitro experimental procedures such as gel mobility shift assays or complex assays requiring ER to bind DNA. It remained unclear whether estrogen was required [10] or not required [10,11,12] for high affinity ER/ERE interaction . More importantly, it remained unclear if the ER could form a dimer in vivo. Wang et al. approached this dilemma by using the yeast two-hybrid system, which is independent of the ER/ERE interaction, and showed that ER protein dimerization in vivo is ligand dependent [16].

7 Estrogen Receptor-Interacting Protein and Cloning of SRC-1

It was generally believed that accessory factors mediate hormone-dependent ER function in the nucleus, but their identity was elusive. Myles Brown used Southwestern blot analysis that indeed an ER coactivator protein, which he termed ER-associated protein (ERAP) , exists in estrogen-sensitive cells [17] and acts as coactivator. A major advancement in the field occurred when O’Malley’s group used the yeast two-hybrid system, using PR LBD as a bait, for the cloning of steroid receptor coactivator-1 (SRC-1). This discovery that has made a paradigm shift in our understanding of how ligand-induced activation of steroid receptors culminates into the assembly of multi-protein complexes on the target gene promoters and transcribe specific mRNAs. A majority of these proteins harbor enzymatic activities for chromatin modification . Several contributors in this book describe current status of their research that involves coactivators/corepressors. Such an intricate mechanism of steroid hormone action has broadened the scope of exploring new approaches to identify novel therapeutic approaches. One great example is that of AIB1/SRC3, which has been identified as an oncogene in breast cancer and is being explored for therapeutic values .

8 An Alternative Method to Detect Immune Complex Leads to the Production of Antibodies to Estrogen Receptor

Another major contribution of Elwood and his postdoctoral fellow, Geoffrey Greene, was the successful purification of the estrogen receptor , an achievement attributed to the first use of steroid affinity chromatography. Greene and Jensen then used the purified receptor to obtain monoclonal as well as polyclonal antibodies to ER. These were highly significant developments in the field, because, in collaboration with Pierre Chambon, they led to the cloning and structural determination of the cDNA for estrogen receptor and prompting an exponential increase in our understanding of how steroid receptors function as transcription factors [18]. Elwood credits the success in preparing the ER antibodies to the use of “alternative approach” to detect the immune complex of antigen and antibody. After many laboratories tried without success to prepare antibodies against estrogen or other steroid receptors, it had been considered that these proteins might be non-immunogenic “because they are so ubiquitous.” Greene and Jensen suspected that antibodies to the estrogen receptor form soluble immune complexes, not detectable by the conventional immunoprecipitation techniques [19, 20].

They again invoked the “alternative approach” and used sucrose gradient sedimentation to identify ER antibodies by their ability to shift the sedimentation peak of the receptor with tritiated estradiol as a marker [21]. They succeeded in producing the first polyclonal as well as monoclonal antibodies to any steroid receptor .

9 Estrogen Signaling and Breast Cancer

More than 100 years ago , Sir George Beatson, a Scottish surgeon, noticed that severity of breast cancer strictly correlated with the menstrual cycle in his patients [1]. He reasoned that the culprit is in the ovary and decided to remove the ovary to manage his breast cancer patients. He successfully tried oophorectomy to manage advanced breast cancer, although not knowing the reason for his successful approach. Around late 1940s, the use of cortisol replacement had made it feasible to remove adrenal gland. Although known for his pioneering contributions with antiandrogenic treatment of prostate cancer for which he was awarded the Nobel Prize, Elwood’s mentor , Dr. Charles Huggins at the University of Chicago, also revolutionized adrenalectomy to manage breast cancer in postmenopausal women [1]. However, there were no means to predict which breast cancer patients would respond. Jensen and colleagues exploited their success with the estrogen receptor research and described two types of breast cancers: ER-positive and ER-negative. His group later refined this finding, showing that estrogen receptor content of excised breast cancer tissue provides an indication of whether or not the tumor is hormone-dependent type, likely to be responsive to endocrine manipulation. Such predictive test had been the goal in the breast cancer field ever since the value of hormone therapy was recognized. Measurement of estrogen receptor in breast cancer specimens is now used routinely as a guide to prognosis and therapy selection. This development has been of immense value to breast cancer patients, not only to spare those with advanced disease from receiving unnecessary treatment that cannot help them but also in guiding the physician in the choice of adjuvant therapy following mastectomy . The following two immunoassays were developed for determining the estrogen receptor content of breast cancer specimens as a guide to therapy selection:

  1. 1.

    The enzyme immunoassay (EIA) measures the receptor in the cytosol fraction of a tumor homogenate by a “sandwich” technique. One monoclonal antibody (D547), bound to a Sepharose bead, absorbs the receptor from the diluted cytosol. The immobilized receptor is then treated with a second monoclonal antibody (H222), which binds to a different region of the receptor and is linked to the enzyme, horseradish peroxidase, which gives rise to a yellow color when exposed to a mixture of hydrogen peroxide and o-phenylenediamine. The color intensity is read in a colorimeter, and the receptor content of the cytosol calculated from a standard curve obtained with lyophilized cytosol from MCF-7 human breast cancer cells .

  2. 2.

    The immunocytochemical assay (ERICA) utilizes the peroxidase-antiperoxidase method in which frozen tumor sections, after gentle fixation, are treated first with an anti-receptor antibody (H222), then with a bridging antibody (goat anti-rat immuno-globulin), and finally with the peroxidase-antiperoxidase (PAP) reagent. Subsequent treatment with hydrogen peroxide and p-diaminobenzidine produces a brown stain in the cells where the receptor has retained the antibody and, thereby, the PAP reagent . Counterstaining is with hematoxylin to delineate cell nuclei (Fig. 4).

Fig. 4
figure 4

Immunocyto-chemical staining of breast cancer tissue (right) and control tissue (left). The ER antibody H222 made in Jensen Lab was used in these assays

Full size image

Because ER-positive tumors can be managed with antiestrogens, their studies revealed that ER+ tumors had a better prognosis than ER-negative tumors . These ER antibody-based assays were later developed as part of the diagnostic kit by Abbott Laboratories and remained as a gold standard on the market until 2000, when FDA put Abbott’s diagnostic division under consent decree for some violation, not connected with ER assay kit . Abbott had to remove all its diagnostic kits from the market, including the ER kits. When Elwood moved to University of Cincinnati in 2002, he often felt sorry for the breast cancer patients that they were deprived of the gold standard ER measurement kits. With a colleague of ours, Elwood and I decided to reintroduce the kit in the market and set up a company—Estrocept Diagnostics. Abbott cooperated with us and sublicensed the pair of cell lines, producing ER antibodies to Estrocept for developing the ER assay kits .

10 Antiestrogens as a “Selective Estrogen Receptor Modulator (SERM)”: A Novel Concept to Treat Breast Cancer

Elwood Jensen also left his marks in the development of antiestrogens as a therapeutic agent for breast cancer. For the first time, he “demonstrated that the antiuterotropic activity of MER-25, a non-steroidal antiestrogen, partly depends on its ability to prevent the incorporation and retention of administered estradiol in the rat uterus. Thus, a foundation for the molecular mechanism of action of antiestrogen was established.”

Due to its toxicity, MER-25 never qualified as a therapeutic agent. During the 1960s, antiestrogens were the focus of attention to develop them as female contraceptives, and the best-known antiestrogen for this purpose at the time was tamoxifen. When tamoxifen failed as contraceptive, Craig Jordan, a pioneer in the science of antiestrogens, repurposed tamoxifen as a therapeutic agent to treat breast cancer. In his initial career, Craig Jordan established a highly productive collaboration with Elwood Jensen, whose own discovery of estrogen receptor would play a pivotal role in advancing the concept of developing tamoxifen as a SERM and targeted therapeutic agent for managing breast cancer. The success of tamoxifen led to one of the largest clinical trials led by NIH (NASBP) to evaluate if tamoxifen could also be used as a prophylactic agent in women with high risk for breast cancer. Craig Jordan was among the first ones to raise concerns of developing endometrial cancer in those women taking tamoxifen as a prevention measure. Our own experiments using the rat model indicated that tamoxifen induced the expression of protooncogene c-fos and fosb [22]. However, Jordan showed that the second-generation antiestrogen raloxifene did not show uterotropic activity [23]. Jensen and Jordan have exemplified the concept of Bench to Bedside research and in recognition received several coveted awards: they shared the AACR ’s inaugural Dorothy Landen Award in 2003.

An avid Limerick composer, Elwood wrote:

A lady with growth neoplastic

thought castration was just a bit drastic

She preferred that her ill

could be cured with a pill

Today it’s no longer fantastic .

11 Estrogen Receptor Harbors Two Antiestrogen Binding Sites

For many years, type I antiestrogens were noted to display unusual pharmacological behavior compared with type ll “pure” antiestrogens [24]. Furthermore, ER was shown to bind nearly twice as much antiestrogen when compared with estradiol [25]. This provocative model relates to the agonist activity of type l antiestrogens to the occupancy of the cognate ligand-binding site, whereas antagonist activity results from an additional interaction with a secondary location. Depending on the particular species and tissue studied, tamoxifen exhibits properties of a pure agonist, a partial agonist/antagonist, or a pure antagonist [26]. For example, tamoxifen displays mixed agonist/antagonist behavior in human and rat tissues but under certain circumstances is a pure agonist in mouse and guinea pig [26, 27]. Tamoxifen also displays differential effects in human tissues, where it is an antagonist in breast and an agonist in uterine and bone tissues [28, 29]. By contrast, type II antiestrogens, including U11110A, LY 117018, and ICI 164384, display exclusively antagonist behavior. Type I and type II antiestrogens were shown to expose an ER epitope specific for the monoclonal ER antibody H222 [25, 30, 31]. These studies, which were carried out in the presence of a saturating amount of E2, led to the hypothesis that exposure of the H222-specific epitope results from the interaction of the antiestrogens at a distinct site from the cognate ligand-binding site. This alleged second ligand-binding site appeared to have a lower affinity for ligand, when compared with the cognate ligand-binding site for E2. In support, sucrose sedimentation and tritiated ligand experiments revealed that ER from MCF-7 cytosol binds approximately twice the amount of type I (HT) or type II (RU 58668) antiestrogens when compared with E2 and studied in the low nM range [25]. Interestingly, the concentration at which the second site-binding event occurs correlates to approximately the same concentration where agonism changes to antagonism [32] thus suggesting a two-site model for antiestrogen action [24].

12 Structural Validation of a Second ER Ligand-Binding Site

This second 4OHT ligand-binding site was recently corroborated in a report of the structure of human ERβ LBD bound to two molecules of 4OHT [33]. The overall fold of ERβ LBD structure (Fig. 5a) is similar to previously published NR LBD structures, namely, a three-layered α-helical sandwich. One 4OHT molecule is located in the cognate ligand-binding pocket, whereas the second 4OHT molecule is located in the hydrophobic groove of the AF-2 coactivator-binding surface, comprising helices 3, 4, 5, and 12—the surface that provides a binding site for coactivator proteins containing the LXXLL recognition motif. Thus, this structure suggests a direct means of antagonism: a ligand blocking the coactivator-binding surface of ERβ. The binding of the second 4OHT molecule in the AF-2 surface primarily involves hydrophobic and van der Waals interactions. The unsubstituted phenyl group of the second 4OHT molecule is buried deep into the hydrophobic cavity of the coactivator surface, which was noted to likely provide the main contribution to binding in the AF-2 surface. Consistent with the aforementioned biological data suggesting a second lower-affinity binding site, the second 4OHT molecule has higher crystallographic B-factors and is not fully buried when compared with the 4OHT molecule in the cognate ligand-binding site. Charge clamp residues K314 and E493, important for recognition of the LXXLL motif, do not interact with the second 4OHT. Similar second-site ligand interactions were reported for androgen receptor (AR) [34] (Fig. 5b) and thyroid receptor (TR) [35].

Fig. 5
figure 5

Structures of human ERα and ERβ, depicting the positioning of helix 12 and ligand-binding site locations in various NR protein structures. (a) 4OHT-ERβ (second 4OHT-binding site; PDB 2FSZ) (b) DHT/4HY-AR (PDB 2PIU). Display is similar to Fig. 2. Helix 12 from a neighboring molecule of 4OHT-ERβ is orange, and the AR C-terminal extension of helix 12 is circled

Full size image