Medical Physiology, 3rd Edition

Steroid and Thyroid Hormones

Cholesterol is the precursor for the steroid hormones: cortisol, aldosterone, estradiol, progesterone, and testosterone

Members of the family of hormones called steroids share a common biochemical parentage: all are synthesized from cholesterol. Only two tissues in the body possess the enzymatic apparatus to convert cholesterol to active hormones. The adrenal cortex makes cortisol (the main glucocorticoid hormone), aldosterone (the principal mineralocorticoid in humans), and androgens. The gonads make either estrogen and progesterone (ovary) or testosterone (testis). In each case, production of steroid hormones is regulated by trophic hormones released from the pituitary. For aldosterone, the renin-angiotensin system also plays an important regulatory role.

The pathways involved in steroid synthesis are summarized in Figure 47-6. Cells that produce steroid hormones can use, as a starting material for hormone synthesis, the cholesterol that is circulating in the blood in association with low-density lipoprotein (LDL; see p. 968). Alternatively, these cells can synthesize cholesterol de novo from acetate (see Fig. 46-16). In humans, LDL cholesterol appears to furnish ~80% of the cholesterol used for steroid synthesis (see Fig. 47-6). An LDL particle contains both free cholesterol and cholesteryl esters, in addition to phospholipids and protein. The cell takes up this LDL particle via the LDL receptor and receptor-mediated endocytosis (see p. 42) into clathrin-coated vesicles. Lysosomal hydrolases then act on the cholesteryl esters to release free cholesterol. The cholesterol nucleus, whether taken up or synthesized de novo, subsequently undergoes a series of reactions that culminate in the formation of pregnenolone, the common precursor of all steroid hormones. Via divergent pathways, pregnenolone is then further metabolized to the major steroid hormones: the mineralocorticoid aldosterone (see Fig. 50-2), the glucocorticoid cortisol (see Fig. 50-2), the androgen testosterone (see Fig. 54-6), and the estrogen estradiol (see Fig. 55-8).


FIGURE 47-6 Uptake of cholesterol and synthesis of steroid hormones from cholesterol. The cholesterol needed as the starting material in the synthesis of steroid hormones comes from two sources. Approximately 80% is taken up as LDL particles via receptor-mediated endocytosis. The cell synthesizes the remaining cholesterol de novo from acetyl coenzyme A (Acetyl CoA). Apo B-100, apolipoprotein B-100; VLDL, very-low-density lipoprotein.

Unlike the peptide and amine hormones considered above, steroid hormones are not stored in secretory vesicles before their secretion (Table 47-4). For these hormones, synthesis and secretion are very closely linked temporally. Steroid-secreting cells are capable of increasing the secretion of steroid hormones many-fold within several hours. The lack of a preformed storage pool of steroid hormones does not appear to limit the effectiveness of these cells as an endocrine regulatory system. Furthermore, steroid hormones, unlike peptide and amine hormones, mediate nearly all their actions on target tissues by regulating gene transcription. As a result, the response of target tissues to steroids typically occurs over hours to days.

TABLE 47-4

Differences Between Steroid and Peptide/Amine Hormones




Storage pools


Secretory vesicles

Interaction with cell membrane

Diffusion through cell membrane

Binding to receptor on cell membrane


In cytoplasm or nucleus

On cell membrane


Regulation of gene transcription (primarily)

Signal-transduction cascade(s) that affect a variety of cell processes

Response time

Hours to days (primarily)

Seconds to minutes

Like cholesterol itself, steroid hormones are poorly soluble in water. On their release into the circulation, some steroid hormones associate with specific binding proteins (e.g., cortisol-binding globulin) that transport the steroid hormones through the circulatory system to their target tissues. The presence of these binding proteins, whose concentration in the circulation can change in response to a variety of physiological conditions, can complicate efforts to measure the amount of active steroid hormone in the circulation.

Steroid hormones bind to intracellular receptors that regulate gene transcription

Steroid hormones appear to enter their target cell by simple diffusion across the plasma membrane (Fig. 47-7). Once within the cell, steroid hormones are bound with high affinity (KD in the range of 1 nM) to receptor proteins located in the cytosol or the nucleus. As detailed in Chapter 4, binding of steroid hormone to its receptor results in a change in the receptor conformation so that the “active” receptor-hormone complex now binds with high affinity to specific DNA sequences called hormone response elements (see p. 90) or steroid response elements (SREs), also called sterol regulatory elements. These sequences are within the 5′ region of target genes whose transcription is regulated by the specific steroid hormone–receptor complex. Termination of gene regulation by the steroid hormone–receptor complex is not as well understood as initiation of the signal. The receptor protein may be modified in a manner that permits dissociation of the hormone and DNA. The receptor itself could then be recycled and the steroid molecule metabolized or otherwise cleared from the cell.


FIGURE 47-7 Action of steroid hormones. The activated steroid hormone receptor binds to specific stretches of DNA called steroid response elements (SREs), which stimulates the transcription of appropriate genes. hsp, heat shock protein.

Steroid receptors are monomeric phosphoproteins with a molecular weight that is between 80 and 100 kDa. A remarkable similarity is seen among receptors for the glucocorticoids, sex steroids, retinoic acid, the steroid-like vitamin 1,25-dihydroxyvitamin D, and thyroid hormone. The genes encoding the receptors for these diverse hormones are considered part of a gene superfamily (see pp. 71–72). Each of these receptors has a similar modular construction with six domains (A through F). The homology among receptors is especially striking for the C domain, particularly the C1 subdomain, which is the part of the receptor molecule that is responsible for binding to DNA (see Fig. 3-14).

Steroid hormone receptors dimerize on binding to their target sites on DNA. Dimerization appears essential for the regulation of gene transcription. Within the C1 DNA-binding domain of the steroid receptor monomer are two zinc fingers that are involved in binding of the receptor to DNA (see p. 82). Even receptors with very different biological actions have a striking sequence similarity in this domain of the receptor. Because the specificity with which genes are regulated by a specific steroid receptor arises from the specificity of the DNA-binding domain, mutations in this region can greatly alter hormone function. For example, substitution of two amino acids in the glucocorticoid receptor causes the mutated glucocorticoid receptor to bind to DNA to which the estrogen receptor normally binds. In such a system, a glucocorticoid could have an estrogen-like effect.

The activated steroid receptor, binding as a dimer to SREs in the 5′ region of a gene, regulates the rate of transcription of that gene. Each response element is identifiable as a consensus sequence of nucleotides, or a region of regulatory DNA in which the nucleotide sequences are preserved through different cell types. The effect of gene regulation by activated steroid receptors binding to an SRE is dramatically illustrated by the chick ovalbumin gene. Chicks that are not exposed to estrogen have approximately four copies of the ovalbumin mRNA per cell in the oviduct. A 7-day course of estrogen treatment increases the number of copies of message 10,000-fold! This increase in message is principally the result of an increased rate of gene transcription. However, steroid hormones can also stabilize specific mRNA molecules and increase their half-life. imageN47-2


Stabilization of mRNA by Estrogen

Contributed by Gene Barrett

For example, in frogs, estrogen increases the half-life of the mRNA for vitellogen (which is formed by Xenopus liver) from <20 hours to ~500 hours.

The 5′ flanking region of the gene typically has one or more SREs upstream of the TATA box, a nucleotide sequence rich in adenine and thymine that is located near the starting point for transcription (see p. 78). The activated steroid hormone receptors recognize these SREs from their specific consensus sequences. For example, one particular consensus sequence designates a site as a glucocorticoid response element if the SRE is in a cell with a glucocorticoid receptor. This same consensus sequence in a cell of the endometrium would be recognized by the activated progesterone receptor or, in the renal distal tubule, by the activated mineralocorticoid receptor. The specificity of the response thus depends on the cell's expression of particular steroid receptors, not simply the consensus sequence. For example, the renal distal tubule cell expresses relatively more mineralocorticoid receptors than it does progesterone receptors when compared with the endometrium. As a result, changes in plasma aldosterone regulate Na+ reabsorption in the kidney with greater sensitivity than does circulating progesterone. However, very high levels of progesterone can, like aldosterone, promote salt reabsorption.

From the foregoing it should be apparent that the specificity of response of a tissue to steroid hormones depends on the abundance of specific steroid receptors expressed within a cell. Because all somatic cells have the full complement of DNA with genes possessing SREs, whether a cell responds to circulating estrogen (e.g., breast), androgen (e.g., prostate), or mineralocorticoid (e.g., renal collecting duct) depends on the receptors present in the cell. This specificity raises the obvious, but as yet unanswered, question of what regulates the expression of specific steroid receptors by specific tissues.

Within a given tissue, several factors control the concentration of steroid hormone receptors. In the cytosol of all steroid-responsive tissues, steroid receptor levels usually drop dramatically immediately after exposure of the tissue to the agonist hormone. This decrease in receptor level is the result of net movement of the agonist-receptor complex to the nucleus. Eventually, the cytosolic receptors are repopulated. Depending on the tissue, this repopulation may involve new synthesis of steroid hormone receptors or simply recycling of receptors from the nucleus after dissociation of the agonist from the receptor. In addition, some steroids reduce the synthesis of their own receptor in target tissues. For example, progesterone reduces the synthesis of progesterone receptor by the uterus, thus leading to an overall net reduction or downregulation of progesterone receptor concentration in a target tissue. An interesting observation in this regard is that the genes for steroid receptor proteins do not appear to have SREs in their 5′ flanking region. Thus, this regulation of receptor number probably involves trans-acting transcriptional factors other than the steroid hormones themselves.

Other factors that affect the concentration of steroid receptors in target tissues include the state of differentiation of the tissue, the presence of other hormones that affect steroid receptor synthesis, and whether the steroid hormone has previously stimulated the tissue. For example, estrogen receptor concentrations are low in an unstimulated uterus but rise dramatically in an estrogen-primed uterus (receptor upregulation). Regulation of steroid receptor number is clearly one factor that alters overall tissue sensitivity to these hormones (Box 47-3).

Box 47-3

Quantitation of Steroid Receptors in Patients with Cancer

The affinity of steroid molecules for their receptors can be studied in vitro in a manner analogous to that described for the radioimmunoassay of peptide hormones (see Fig. 47-1). In a typical immunoassay, an antibody with high affinity for a hormone or other compound binds to a radioactively labeled hormone or other molecule. A sample containing an unknown amount of the compound to be measured is added to the antibody-labeled hormone mixture and displaces the radioisotope from the antibody in proportion to the concentration of the unknown. The amount of unknown can be quantitated by comparison to the displacing activity of known standards.

For quantitating steroid receptors, cell extracts containing an unknown amount of steroid receptors are incubated with increasing concentrations of labeled steroid hormone. At each concentration, hormone that is bound by the receptor is separated from that remaining free in the extract. The result is a saturation curve (Fig. 47-8A), provided the tissue extract has a finite number of specific hormone receptors. This saturation curve can often be linearized by a simple arithmetic manipulation called a Scatchard plot (see Fig. 47-8B). This analysis allows quantitation of the affinity of the receptor for the hormone and provides an estimate of the number of receptors (actually, the concentration of receptors) for that particular hormone.


FIGURE 47-8 Quantitating receptor affinity and number of receptors. A, A plot of bound hormone (i.e., hormone-receptor complex) on the y-axis versus free steroid concentration on the x-axis. In this example, we have assumed that the KD for hormone binding is 3 nM and that the maximal bound-hormone concentration is 0.5 nM. B, This plot is a transformation of the data in A. The colored points in the plot match the points of like color in A. Plotted on the y-axis is the ratio of [bound hormone] to [free hormone]. Plotted on the x-axis is [bound hormone]. The slope of this relationship gives the −1/KD, where KD is the dissociation constant (3 nM). The x-axis intercept gives the total number of receptors (0.5 nM).

The technique of quantitating receptor number has found an important application in determining the number of estrogen and progesterone receptors present in breast cancer cells. The number of estrogen and progesterone receptors per milligram of breast cancer tissue (obtained by biopsy of the breast or involved lymph node) is quantitated with radiolabeled estrogen (or progesterone). For postmenopausal women with estrogen receptor–positive breast cancer (i.e., a tumor with a high level of estrogen receptors), treatment with an antiestrogen (e.g., tamoxifen) is effective therapy. For premenopausal women, an antiestrogen may be used as well, or the ovaries can be removed surgically. The woman might be given an aromatase inhibitor that blocks estradiol synthesis or a long-acting gonadotropin-releasing hormone (GnRH) agonist, which paradoxically blocks both LH and FSH production by the anterior pituitary and thereby reduces estradiol production and accomplishes a medical oophorectomy. The continuous (as opposed to the normally pulsatile) administration of GnRH downregulates GnRH receptors in the anterior pituitary (see Box 55-2).

These therapies are not effective in patients with cancers that do not express significant numbers of estrogen receptors. The presence of abundant estrogen and progesterone receptors in breast tumors correlates with a more favorable prognosis, possibly because of the relatively advanced state of differentiation of the tumor, as well as the tumor's responsiveness to manipulation by estrogen or progesterone therapy.

Thyroid hormones bind to intracellular receptors that regulate metabolic rate

In many respects, the thyroid gland and thyroid hormone are unique among the classic endocrine axes. In Chapter 49, we will see that this uniqueness begins with the structure of the thyroid gland, which is composed of follicles. Each follicle is an epithelial monolayer encircling a protein-rich fluid. The principal protein component of the follicular fluid is an extremely large protein, thyroglobulin. Neither of the two thyroid hormones—T4 nor T3—is free in the follicular fluid. Rather, these hormones are formed by the iodination of tyrosine residues within the primary structure of the thyroglobulin molecule.

T4 and T3 remain part of the thyroglobulin molecule in the follicle lumen until thyroid secretion is stimulated. The entire thyroglobulin molecule then undergoes endocytosis by the follicular cell and is degraded within the lysosomes of these cells. Finally, the follicular cell releases the free T4 and T3 into the circulation. Once secreted, T4 is tightly bound to one of several binding proteins. It is carried to its sites of action, which include nearly all the cells in the body. In the process of this transport, the liver and other tissues take up some of the T4 and partially deiodinate it to T3; this T3 can then re-enter the circulation.

Both T3 and T4 enter target cells and bind to cytosolic and nuclear receptors. These receptors are similar to those for steroid hormones (see pp. 71–72). T3 has higher affinity than T4 for the thyroid hormone receptor. Even though it accounts for only ~5% of the circulating thyroid hormone, T3 is probably the main effector of thyroid hormone signaling. The activated thyroid hormone receptor binds to thyroid hormone response elements in the 5′ region of responsive genes and regulates the transcription of multiple target genes.

Thyroid hormone receptors are present in many tissues, including the heart, vascular smooth muscle, skeletal muscle, liver, kidney, skin, and CNS. A major role for thyroid hormone is overall regulation of metabolic rate. Because T4 affects multiple tissues, individuals affected by disorders involving oversecretion or undersecretion of thyroid hormone manifest a host of varied symptoms that reflect the involvement of multiple organ systems.

Steroid and thyroid hormones can also have nongenomic actions

A central dogma has been that all the diverse actions of steroid and thyroid hormones are due to genomic regulation. However, the very rapid onset of some effects (occurring within 2 to 15 minutes) appear incompatible with a mechanism requiring new protein synthesis. Such accumulating evidence suggests that steroid and thyroid hormones can bind to receptors that modulate the activity of cytosolic proteins and thereby regulate their activity or behavior via a nongenomic action. An example is the binding of extracellular estrogen or aldosterone to the GPCR called GPR30, also known as the G-protein estrogen receptor (GPER). Receptor occupancy rapidly activates Gαs and then adenylyl cyclase (see p. 53), thereby raising [cAMP]i and stimulating PKA (see p. 57). In addition, GPR30 indirectly activates the epidermal growth factor receptor (EGFR), a receptor tyrosine kinase (see pp. 68–70), and some of its downstream effectors, such as phosphatidylinositol 3-kinase (PI3K; see pp. 69–70) and the mitogen-activated protein kinases (MAPKs; see pp. 68–69). Besides estrogen and aldosterone, other ligands of nuclear receptors have nongenomic actions: thyroid hormones, testosterone, and glucocorticoids.