• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molbiolevolLink to Publisher's site
Mol Biol Evol. Dec 2008; 25(12): 2643–2652.
Published online Sep 17, 2008. doi:  10.1093/molbev/msn204
PMCID: PMC2721552

Evolution of Hormone Signaling in Elasmobranchs by Exploitation of Promiscuous Receptors


Specific interactions among proteins, nucleic acids, and metabolites drive virtually all cellular functions and underlie phenotypic complexity and diversity. Despite the fundamental importance of interactions, the mechanisms and dynamics by which they evolve are poorly understood. Here we describe novel interactions between a lineage-specific hormone and its receptors in elasmobranchs, a subclass of cartilaginous fishes, and infer how these associations evolved using phylogenetic and protein structural analyses. The hormone 1α-hydroxycorticosterone (1α-B) is a physiologically important steroid synthesized only in elasmobranchs. We show that 1α-B modulates gene expression in vitro by activating two paralogous intracellular transcription factors, the mineralocorticoid receptor (MR) and glucocorticoid receptor (GR), in the little skate Leucoraja erinacea; MR serves as a high-sensitivity and GR as a low-sensitivity receptor. Using functional analysis of extant and resurrected ancestral proteins, we show that receptor sensitivity to 1α-B evolved millions of years before the hormone itself evolved. The 1α-B differs from more ancient corticosteroids only by the addition of a hydroxyl group; the three-dimensional structure of the ancestral receptor shows that the ligand pocket contained ample unoccupied space to accommodate this moiety. Our findings indicate that the interactions between 1α-B and elasmobranch GR and MR proteins evolved by molecular exploitation: a novel hormone recruited into new functional partnerships two ancient receptors that had previously interacted with other ligands. The ancestral receptor's promiscuous capacity to fortuitously bind compounds that are slight structural variants of its original ligands set the stage for the evolution of this new interaction.

Keywords: evolution of protein function, evolution of gene networks, hormone receptors, steroid hormones, elasmobranchs, molecular interactions


Virtually, all cellular functions are driven by specific interactions among biomolecules, such as enzymes and their substrates, transcription factors and their DNA-binding sites, and receptors and their ligands. Despite extensive “top-down” work on the global structure of molecular interaction networks (Barabasi and Oltvai 2004; Cork and Purugganan 2004; Teichmann and Babu 2004; Wilkins 2005), only limited knowledge is available concerning the specific mechanisms and dynamics by which the molecular interactions that constitute these networks evolve (Zhu et al. 2005; Bridgham et al. 2006; Prud'homme et al. 2006; Hittinger and Carroll 2007; McGregor et al. 2007; Ortlund et al. 2007).

Corticosteroid hormones and their receptors provide an elegant model of molecular interactions, and the existence of a lineage-specific corticosteroid in the taxon Elasmobranchii offers the opportunity to investigate how new hormone–receptor interactions evolve. Corticosteroids are produced in the adrenal/interrenal gland through an enzyme-mediated biosynthetic pathway and secreted into the blood. The classic actions of corticosteroids are mediated by intracellular corticosteroid receptors (CRs), members of the steroid hormone receptor family that also includes receptors for androgens, progestins, and estrogens (Thornton 2001). These proteins are ligand-activated transcription factors: upon binding to their hormone partners with high specificity and affinity, the receptor changes conformation, dimerizes, binds to specific DNA response elements, and attracts coregulator proteins, resulting in increased expression of nearby target genes (Beato et al. 1995). Like other steroid receptors, CRs have a modular domain structure, consisting of functionally autonomous conserved domains for DNA-binding and ligand-activated transcription as well as a nonconserved hinge and an N-terminal domain with additional ligand-independent transcriptional regulatory functions (Kumar and Thompson 2005). Most vertebrates possess two paralogous CRs, a glucocorticoid receptor (GR) and mineralocorticoid receptor (MR), which arose from a single ancestral CR in a large-scale gene duplication some 470–440 MYA (Thornton 2001; Bridgham et al. 2006). In bony vertebrates, GR is activated by cortisol and regulates metabolism, immunity, and the stress response; MR controls electrolyte homeostasis and blood pressure and is activated primarily by aldosterone in tetrapods and 11-deoxycorticosterone (DOC) in teleosts (Bentley 1998; Sturm et al. 2005). In addition to their classic receptor-mediated effects, some steroids also trigger rapid “nonclassical” effects through other mechanisms (Thomas et al. 2004, 2005; Prossnitz et al. 2007; Harvey et al. 2008).

A better understanding of steroid hormone–receptor interactions in basal vertebrates would help illuminate how GR, MR, and the physiological processes they regulate have evolved (Baker et al. 2007; Bury and Sturm 2007). Some 40 years ago, it was discovered that the plasma of elasmobranchs—sharks, skates, and rays—contains very high levels of the corticosteroid 1α-hydroxycorticosterone (1α-B), which is not known to be produced in any other taxon (Idler and Truscott 1966). 1α-B is synthesized in the interrenal gland in copious amounts relative to other endogenous hormones (Idler et al. 1967; Truscott and Idler 1972). It differs from other corticosteroids by the addition of a hydroxyl group at the C1 position of the steroid backbone. The enzyme activity that drives this unusual 1α-hydroxylation reaction is present and functional in in vitro preparations of elasmobranch steroidogenic tissues (Truscott and Idler 1968), but the underlying gene has not been identified. The molecular basis for 1α-B action in elasmobranchs, particularly its potential interactions with MR and GR, remains uncharacterized. It has been speculated that 1α-B functions as a dual ligand in both glucocorticoid and mineralocorticoid signaling pathways (Gelsleichter and Musick 1999; Nunez and Trant 1999; Nunez et al. 2006; Manire et al. 2007). 1α-B clearly has a mineralocorticoid-like role in the regulation of salt and osmolyte balance (Hazon and Henderson 1984; Armour et al. 1993), but the evidence for its glucocorticoid-like effects is more speculative. Levels of 1α-B have been shown to increase under stress and reduced osmolarity (Hazon and Henderson 1984; Armour et al. 1993; Manire et al. 2007). Administration of glucocorticoids to elasmobranchs stimulates the regulation of carbohydrate metabolism (Patent 1970), cartilage growth (Gelsleichter and Musick 1999), and immune responses (Walsh et al. 2002). Like the glucocorticoids of other vertebrates, synthesis of 1α-B is regulated by adrenocorticotropic hormone and angiotensin II from the hypothalamic–pituitary–adrenal axis (Honn and Chavin 1978; Hazon and Henderson 1985; O'Toole et al. 1990; Nunez and Vedeckis 2002). Although cortisol and other recognized glucocorticoids are absent or present at very low concentrations in elasmobranchs, 1α-B circulates at extremely high levels in elasmobranchs (Truscott and Idler 1972; Kime 1977; Armour et al. 1993), just as glucocorticoids do in most other vertebrates.

Here we characterize the functional interactions of 1α-B with the GR and MR of an elasmobranch, the little skate (Leucoraja erinacea), and reconstruct how these receptor–hormone interactions evolved. We combine molecular functional assays, ancestral gene resurrection (Thornton 2004), and analysis of protein structure to determine the functions of GR and MR with respect to this hormone and to characterize how the lineage-specific partnership of 1α-B with its receptors evolved.


Isolation and Reconstruction of CRs

Skate (L. erinacea) and hagfish (Myxine glutinosa) receptor ligand–binding domains (LBDs) were amplified using degenerate polymerase chain reaction (PCR) and rapid amplification of cDNA ends from liver cDNA (Bridgham et al. 2006). Lamprey (Petromyzon marinus) CR cDNA was amplified similarly from a cDNA library (Thornton 2001). Teleost (Astatotilapia burtoni) GR2a and MR were provided by R. Fernald, the human MR by R. Evans, and the human GR by B. Darimont.

Reconstruction and synthesis of ancestral receptors were performed as described in (Bridgham et al. 2006; Ortlund et al. 2007). Briefly, ancestral protein sequences were inferred using maximum likelihood (ML) phylogenetic reconstruction and a large data set of steroid and related receptors. Nucleic acid sequences coding for the LBD were synthesized de novo. Ambiguously reconstructed sites were defined as having an alternate amino acid state with a posterior probability greater than 0.20 or as having a different ML state when reconstructed on any tree in the 95% credible set collected by Bayesian Markov chain Monte Carlo analysis. Alternative states were introduced singly into the ML sequence using site-directed mutagenesis (QuikChange II, Stratagene, La Jolla, CA).

Receptor Activation

LBDs (with hinge and carboxy-terminal extension [CTE]) of extant receptors were amplified using high-fidelity PCR and cloned into pSG5-GAL4DBD (gift of D. Furlow). Ancestral receptor LBDs (with CTE) were cloned into pSG5-GAL4DBD with a human GR hinge region. CHO-K1 cells were grown in 96-well plates in phenol red-free α-MEM plus 10% dextran–charcoal-stripped fetal bovine serum (Hyclone, Logan, UT), then transfected with 1 ng of receptor LBD, 100 ng of pFRluc reporter, and 0.1 ng of normalization vector phRLtk using Lipofectamine and Plus Reagents (Invitrogen, Carlsbad, CA). After 4-h incubation, cells were treated with fresh medium, allowed to recover overnight, and then treated in triplicate with hormone or vehicle control (ethanol) for 24 h. Reporter expression was measured assayed using Dual-Glo (Promega, Madison, WI) and expressed as the ratio of firefly luciferase to Renilla luciferase. Dose–response relationships were analyzed using Prism4 software (GraphPad, La Jolla, CA). Receptors were considered unresponsive if they displayed <2-fold activation at >1 uM hormone. The 1α-B was synthesized and provided by J. Rimoldi, University of Mississippi.

Quantitative PCR

Expression of MR and GR was measured in various organs of L. erinacea provided by J. St. George, Boston University. Total RNA was isolated using RNeasy (Qiagen, Valencia, CA) and cDNA prepared using QuantiTect Reverse Transcription (Qiagen). Skeletal and cardiac muscle samples were digested with proteinase K (Qiagen) to improve yield. Primers were designed to amplify the ligand-dependent activation (AF-2) domain of MR and GR LBDs. Housekeeping gene glyceraldehyde 3-phosphodehydrogenase (GAPDH GenBank DQ382343) was isolated using degenerate PCR and used as an internal reference for normalization. Quantitative PCR (Q-PCR) was performed on an ABI Prism 7900 HT with 1× QuantiTect SYBR Green PCR Master Mix (Qiagen), 0.3 μM each primer, and skate cDNA. Cycling was as follows: 95°/15′, (94°/15″, 54°/30″, 72°/30″) × 39 cycles, followed by melting curve analysis from 65° to 95°. Reactions were run in triplicate and fluorescence detected during extension. Primer efficiencies (EMR = 1.96, EGR = 1.98, and EGapdh = 1.97) were determined by standard curve analyses (Simon 2003) of serially diluted and linearized skate MR and GR in GAL4-DBD-pSG5 and skate Gapdh gene in pCR2.1 (Invitrogen).

In Silico Structural Analysis

A model of 1α-B was constructed and energy minimized using ChemOffice Ultra (CambridgeSoft, Cambridge, MA) and imported into MacPyMOL (Delano Scientific, Palo Alto, CA) along with the crystal structure of AncCR/DOC (PDB 2Q3Y). 1α-B was aligned to DOC at the C10 carbon with the steroid backbones oriented in the same plane.


High- and Low-Sensitivity Receptors for 1α-B

To determine the intrinsic hormone sensitivity of skate GR and MR, we expressed fusion constructs of receptor LBDs with the Gal4 DNA-binding domain. We evaluated transcription of a UAS-driven luciferase reporter gene in response to the predominant elasmobranch hormone 1α-B and several other steroid hormones found at lesser concentrations in elasmobranchs, including 11-deoxycorticosterone (DOC), corticosterone (B), and 11-dehydrocorticosterone (11-DHC) (Truscott and Idler 1972). We also examined several mammalian corticosteroids not present in elasmobranchs, including aldosterone (Aldo), cortisol (F), and 11-deoxycortisol (DOF).

We found that both skate MR and GR LBDs are activated by 1α-B and other corticosteroid hormones. MR is a high-sensitivity receptor, activating transcription in response to low nanomolar concentrations of all hormones examined, including 1α-B. GR, in contrast, is activated by the same hormones but requires two to four orders of magnitude higher concentrations to achieve half-maximal activation (fig. 1). Based on these results, endogenous concentrations of 1α-B and possibly other steroids in elasmobranchs are expected to activate the MR but only concentrations of 1α-B are likely to be high enough to activate the GR (see Discussion). The difference in quantitative sensitivity but not hormone specificity between GR and MR in elasmobranchs contrasts with the situation in humans and other bony vertebrates, in which GR and MR have distinct hormone preferences.

FIG. 1.
Skate MR and GR are high- and low-sensitivity receptors, respectively, for 1α-B and other corticosteroids. (A) Ligand-dependent transcriptional activity of receptor LBDs was determined in the presence of increasing concentrations of various hormones ...

MR and GR are Ubiquitously Coexpressed

To determine whether differences in gene expression could be important for generating distinct tissue-specific gluco- or mineralocorticoid responses, we used Q-PCR to measure GR and MR transcripts across skate tissues. We found that both MR and GR are ubiquitously expressed, with little variation in the relative quantities of the two transcripts. GR and MR transcript levels, normalized to expression of the housekeeping gene Gapdh, varied by less than an order of magnitude among tissues, except for in skeletal muscle, where Gapdh levels were very high (fig. 2A); a similar pattern was observed when transcripts were not normalized to Gapdh (supplementary fig. S1, Supplementary Material online). The ratio of MR to GR expression, which is not affected by differences in Gapdh levels, varied by a factor of less than two among tissues (fig. 2B; supplementary fig S1, Supplementary Material online).

FIG. 2.
MR and GR are widely coexpressed in skate. Abundance of MR, GR, and Gapdh mRNAs were determined in various tissues of adult skate using quantitative PCR. (A) The mean expression level and standard error of the mean (bars) were calculated for MR and GR ...

1α-B Activates CRs from Other Vertebrates

We next sought to determine how the lineage-specific partnership of 1α-B with CRs was assembled during evolution. We began by determining whether CRs from other species are also sensitive to 1α-B. MRs from bony vertebrates—both tetrapods (Homo sapiens) and teleosts (A. burtoni)—activated reporter gene transcription in the presence of submicromolar concentrations of 1α-B (fig. 3 and table 1), despite the hormone's absence from lineages other than elasmobranchs. The CR of an agnathan—the Atlantic hagfish (M. glutinosa), which possesses a single unduplicated gene orthologous to both GR and MR—had similar 1α-B sensitivity. Only the GRs of bony vertebrates—which are activated only by 17-hydroxylated corticosteroids like cortisol (Bridgham et al. 2006)—and the CR of the sea lamprey (P. marinus) were insensitive to 1α-B.

Table 1
Sensitivity of Extant and Ancestral Receptors to 1α-B
FIG. 3.
GR and MR sensitivity to 1α-B is taxonomically widespread. Sensitivity of CRs from a variety of vertebrates to 1α-B was assessed using a luciferase reporter assay. Black and gray branches show receptors sensitive and insensitive to 1α-B, ...

Ancient Receptor Sensitivity to 1α-B

The widespread sensitivity of vertebrate CRs to 1α-B suggests an ancient origin of 1α-B responsiveness. To test the hypothesis that receptor sensitivity to 1α-B predates the elasmobranch-specific emergence of the hormone, we resurrected and functionally characterized three ancient CRs as they existed in ancestral vertebrates. Specifically, we used a large database of extant receptor sequences and used phylogenetic techniques to infer the ML amino acid sequences of three ancient receptors: the ancestral corticoid receptor (AncCR, the unduplicated ancestral gene from which extant MRs and GRs descend by gene duplication), GR in the last common ancestor of all jawed vertebrates (AncGR1), and GR from the last common ancestor of all bony vertebrates (AncGR2), after their split from cartilaginous fishes (fig. 4). We then synthesized DNAs coding for the LBDs of these reconstructed proteins, expressed them, and characterized their functions using a reporter transcription assay in cell culture (Bridgham et al. 2006; Ortlund et al. 2007).

FIG. 4.
CR sensitivity to 1α-B predates the evolution of 1α-B synthesis. Three ancestral receptors from early vertebrates (circled nodes) were resurrected by ancestral sequence reconstruction and gene synthesis. Graphs depict luciferase reporter ...

As predicted, we found that the most ancient receptors—AncCR and AncGR1—are extremely sensitive to 1α-B, activating transcription with EC50s of ~20 nM. AncGR2, in contrast, was unresponsive, as expected based on the lack of sensitivity to 1α-B in its descendants, the GRs of tetrapods and teleosts (fig. 4 and table 1). To determine whether the 1α-B sensitivity of AncCR might be an artifact of uncertainty in the inference of the ancestral sequence, we identified sites that were ambiguously reconstructed (defined as having an alternative amino acid state with posterior probability >0.20). In all cases but five, the alternate state is found in other 1α-B–activated receptors and is therefore not sufficient to abolish sensitivity to that hormone. Introducing each of these five alternate states into the AncCR by site-directed mutagenesis had no effect on ligand activation (table 2). Among sites that make contact with the ligand in the AncCR crystal structure (Ortlund et al. 2007), only one was ambiguously reconstructed; introducing this alternate state into AncCR had no effect on sensitivity to 1α-B (table 2). We conclude that AncCR's response to 1α-B is not an artifact of uncertainty in the ancestral reconstruction.

Table 2
AncCR's Sensitivity to 1α-B Is Robust to Statistical and Phylogenetic Uncertainty

To determine whether AncCR's sensitivity to 1α-B may be due to error in the phylogeny on which the ancestral reconstruction is based, we inferred the ML sequence of AncCR on each of the 467 trees in the 95% credible set from a large Bayesian analysis. At only one sequence site did the ancestral reconstructions differ among trees. We introduced the alternate state at this site (A7V) into AncCR by mutagenesis and found it had no effect on sensitivity to 1α-B (table 2).

We conclude that AncCR and AncGR1 were sensitive to 1α-B, and this ancient sensitivity was retained in most of the lineages descending from those ancestors, including the MR and GR of elasmobranchs. After the divergence of bony from cartilaginous fishes, the GRs of bony vertebrates subsequently lost 1α-B sensitivity, during the same period in which the receptor became cortisol specific (Bridgham et al. 2006; Ortlund et al. 2007). This result indicates that CRs were capable of being activated by 1α-B many millions of years before synthesis of the hormone itself evolved in the elasmobranch lineage.

Structural Analysis of 1α-B Docked in the AncCR

To determine why ancient receptors were activated by 1α-B, we examined the previously solved crystal structure of AncCR (Ortlund et al. 2007). We hypothesized that AncCR was structurally “preadapted” to bind 1α-B because of that hormone's similarity to DOC, an ancient hormone that is the putative ancestral ligand for AncCR (Ortlund et al. 2007). We generated a structural model of 1α-B docked into the LBD of the AncCR with DOC. The 1α-B differs from DOC only by the presence of hydroxyl groups at the C1 and C11 positions. The structure shows that AncCR's ligand pocket contains ample room to accommodate the 11-hydroxyl of 1α-B; this conclusion is supported by the receptor's previously identified ability to bind cortisol and aldosterone, which also carry the 11-hydroxyl, without conformational adjustment (Ortlund et al. 2007). The ligand pocket also contains unoccupied space in the alpha plane above the C1 carbon (fig. 5A, 5C); in the model of AncCR with 1α-B, this space is occupied by and adequate to accommodate the 1α-hydroxyl (fig. 5B, 5D). A slight hydrophobic clash of this hydroxyl with AncCR's Leu32 and Phe92 is the likely cause of the receptor's slightly reduced sensitivity to 1α-B compared with other ligands that lack the 1α-hydroxyl. That the receptor retains nanomolar sensitivity to 1α-B, however, indicates that minor adjustments of the receptor backbone or side chain rotamers are sufficient to relieve the clash.

FIG. 5.
The structural basis for receptor promiscuity. The crystal structure of AncCR's ligand pocket in complex with DOC (green, A and C) and an in silico model of AncCR with 1α-B (yellow, B and D) are shown from two perspectives (A, B and C, D). Electron ...


Our analysis indicates that the lineage-specific partnership of 1α-B with MR and GR in elasmobranchs evolved due to molecular exploitation—recruitment of older molecules, which previously had different functions, into new functional relationships (Thornton 2001; Bridgham et al. 2006). Both the widespread taxonomic distribution of 1α-B sensitivity and experimental analysis of resurrected ancestral receptors indicate that the ancestral gene from which both GR and MR evolved was already sensitive to 1α-B, millions of years before synthesis of the hormone itself evolved (fig. 6). Sensitivity to 1α-B has been retained to the present in numerous descendants of AncCR, including those in numerous species that lack 1α-B, such as mammals, in which administration of exogenous 1α-B elicits a strong mineralocorticoid response (Idler et al. 1967). Our results indicate that the partnership of 1α-B with its receptors in elasmobranchs evolved when a newly synthesized hormone-recruited preexisting receptors, which previously had different ligands, into new functional partnerships. This evolutionary dynamic is similar to that previously observed for the interaction between aldosterone and the MR of tetrapods (Bridgham et al. 2006). Other steroid hormone–receptor interactions evolved when intermediates in the biosynthesis of ligands for more ancient receptors were recruited into partnerships with newly duplicated receptors (Thornton 2001; Thornton et al. 2003). Several other apparent examples of evolutionary recruitment of ancient molecules into new partnerships have recently been described (Krasowski et al. 2005; Cai et al. 2007; Cardoso et al. 2007), suggesting that the evolution of specific molecular interactions by molecular exploitation may be a dominant theme in the emergence of biological systems.

FIG. 6.
Receptor exploitation in the evolution of corticosteroid signaling. The phylogeny traces the evolution of receptor genes and their ligand sensitivity (black and gray lines, sensitive and insensitive to 1α-B, respectively) within the divergence ...

Molecular exploitation is greatly facilitated by untapped promiscuity in ancient proteins, which allows them to accommodate new partners that are minor variants of their original binding partners. 1α-B is identical to the likely ancestral ligand DOC, except for additional oxygen atoms at the C1 and C11 positions. Our structural analysis of AncCR indicates that the ancestral ligand pocket had ample room to accommodate these modifications when the hormone evolved later in the elasmobranchs. AncCR's promiscuous response to steroids that had not yet evolved contrasts with its unresponsiveness to the more ancient androgens, estrogens, or progestins, which differ from the corticosteroids at other key locations on the steroid backbone (Bridgham et al. 2006). AncCR therefore appears to have been only as specific as it needed to be: it excluded other endogenously produced hormones that would inappropriately activate the receptor but not other potential ligands whose synthesis had not yet evolved. When steroids that fortuitously fit the pocket appeared later, this promiscuity set the stage for the evolution of novel receptor–ligand partnerships with physiological or developmental functions. Structural promiscuity—a ligand pocket that contains extra space or flexibility, allowing it to accommodate a range of ligands—is a common protein feature and is likely to have facilitated the evolution of new functions in both receptors and enzymes (Brzozowski et al. 1997; James and Tawfik 2001; Bledsoe et al. 2002; Copley 2003; Khersonsky et al. 2006; Pereira de Jesus-Tran et al. 2006; Bloom et al. 2007; Nettles et al. 2007; Suino-Powell et al. 2008).

Our results shed light on the functional roles of corticosteroids and their receptors in elasmobranchs. The LBDs of skate MR and GR differ little in hormone specificity but radically in sensitivity: MR is activated by much lower corticosteroid concentrations than are required to activate GR. Analysis of serum steroid concentrations in a wide variety of elasmobranch species have indicated that 1α-B levels are very high—up to several hundred nanomolar (Truscott and Idler 1968; Kime 1977)—and that corticosterone, DOC, and 11-DHC are present at orders of magnitude lower concentrations (Idler et al. 1967; Truscott and Idler 1972; Kime 1977; Hazon and Henderson 1984; Armour et al. 1993; Rasmussen and Crow 1993; Snelson et al. 1997). Our results indicate that skate MR has corticosteroid sensitivity similar to that of tetrapod and teleost MRs (Bridgham et al. 2006), whereas the skate GR's sensitivity is orders of magnitude lower. Although in vitro reporter assays using fusion proteins are not precise predictors of in vivo dose–response relationships, physiological 1α-B levels, at least under some conditions, are likely to be well beyond those necessary to activate skate MR. It is possible that DOC and corticosterone may also serve as MR ligands. Skate GR, in contrast, is likely to be activated only by elevated concentrations of 1α-B (and possibly corticosterone) such as those that occur under conditions of stress (Manire et al. 2007). Our findings suggest the hypothesis that low corticosteroid levels regulate MR to control osmolarity, whereas stress-induced increases in the same hormones regulate GR to control the stress response. Like the elasmobranch GR, the GR of bony vertebrates is a low-sensitivity receptor that activates the stress response only when cortisol levels are very high (Bentley 1998). Unlike the elasmobranch GRs, however, GRs in bony vertebrates evolved novel specificity as well, making them insensitive to even high doses of mineralocorticoids (Bridgham et al. 2006). Other factors, such as posttranslational modification and the availability of coregulators, may also contribute to differences in GR and MR function.

MR and GR are ubiquitously coexpressed, at roughly equal proportions, in a wide variety of skate tissues. This is largely consistent with patterns of expression found in teleosts (Bury et al. 2003; Greenwood et al. 2003; Sturm et al. 2005) and tetrapods (de Castro et al. 1996; Le Menuet et al. 2000). These data provide no evidence for subfunctionalization of expression domains as important in the maintenance of the GR and MR after duplication. We cannot rule out the possibility, however, that GR and MR may be differentially expressed on a finer scale, in specific cell types or groups of cells or at specific stages.

The interaction of CRs, their ligands, and the DNA response elements they regulate represent a relatively simple but physiologically essential molecular network. Gene duplication provides raw material for the genesis and expansion of such networks (Wagner 1994; Teichmann and Babu 2004), but duplication alone is not enough to make a network more elaborate or generate a new one. To increase complexity, functional interactions must diversify, either by the evolution of new interactions or by the partitioning of ancestral functions among new network nodes generated by duplication. We have shown that the corticosteroid hormone/receptor network was elaborated when receptor gene duplication and divergence, together with the extension of a biosynthetic pathway, produced new receptors and hormones that were slight structural and functional modifications of their ancestral forms. The receptors’ promiscuity set the stage for these new hormones to be integrated into the existing receptor signaling network. Subsequently, divergence of the two sister receptors—leading to reduced sensitivity in the GR—resulted in markedly different quantitative sensitivities to corticosteroid hormones and the possibility of distinct network responses to varying levels of hormone. These observations indicate that the complex molecular networks that drive physiological process can evolve by mechanisms as simple as molecular exploitation and the partial degradation of function after gene duplication.

Supplementary Material

Supplementary figure S1 is available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).

[Supplementary Data]


We thank S. Nunez and members of the Thornton laboratory for advice and comments, E. Ortlund for assistance with protein structural analyses, J. Rimoldi for providing 1α-B, and J. St George and I. Callard for skate tissue samples. Supported by National Institutes of Health (NIH) R01-GM081592, NSF-IOB-0546906, and a Sloan fellowship (J.W.T.); NIH F32-GM074398 (J.T.B.); and National Science Foundation IOB-0546906.


  • Armour KJ, O'Toole LB, Hazon N. The effect of dietary protein restriction on the secretory dynamics of 1 alpha-hydroxycorticosterone and urea in the dogfish, Scyliorhinus canicula: a possible role for 1 alpha-hydroxycorticosterone in sodium retention. J Endocrinol. 1993;138(2):275–282. [PubMed]
  • Baker ME, Chandsawangbhuwana C, Ollikainen N. Structural analysis of the evolution of steroid specificity in the mineralocorticoid and glucocorticoid receptors. BMC Evol Biol. 2007;7(1):24. [PMC free article] [PubMed]
  • Barabasi AL, Oltvai ZN. Network biology: understanding the cell's functional organization. Nat Rev Genet. 2004;5(2):101–113. [PubMed]
  • Beato M, Herrlich P, Schutz G. Steroid hormone receptors: many actors in search of a plot. Cell. 1995;83(6):851–857. [PubMed]
  • Bentley PJ. Comparative vertebrate endocrinology. New York: Cambridge University Press; 1998. p. 526.
  • Bledsoe RK, Montana VG, Stanley TB, et al. (14 co-authors) Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition. Cell. 2002;110(1):93–105. [PubMed]
  • Bloom JD, Romero PA, Lu Z, Arnold FH. Neutral genetic drift can alter promiscuous protein functions, potentially aiding functional evolution. Biol Direct. 2007;2:17. [PMC free article] [PubMed]
  • Bridgham JT, Carroll SM, Thornton JW. Evolution of hormone-receptor complexity by molecular exploitation. Science. 2006;312(5770):97–101. [PubMed]
  • Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M. Molecular basis of agonism and antagonism in the oestrogen receptor. Nature. 1997;389(6652):753–758. [PubMed]
  • Bury NR, Sturm A. Evolution of the corticosteroid receptor signalling pathway in fish. Gen Comp Endocrinol. 2007;153(1–3):47–56. [PubMed]
  • Bury NR, Sturm A, Le Rouzic P, Lethimonier C, Ducouret B, Guiguen Y, Robinson-Rechavi M, Laudet V, Rafestin-Oblin ME, Prunet P. Evidence for two distinct functional glucocorticoid receptors in teleost fish. J Mol Endocrinol. 2003;31(1):141–156. [PubMed]
  • Cai SY, Xiong L, Wray CG, Ballatori N, Boyer JL. The farnesoid X receptor FXRalpha/NR1H4 acquired ligand specificity for bile salts late in vertebrate evolution. Am J Physiol Regul Integr Comp Physiol. 2007;293(3):R1400–R1409. [PubMed]
  • Cardoso JC, Vieira FA, Gomes AS, Power DM. PACAP, VIP and their receptors in the metazoa: insights about the origin and evolution of the ligand-receptor pair. Peptides. 2007;28(9):1902–1919. [PubMed]
  • Copley SD. Enzymes with extra talents: moonlighting functions and catalytic promiscuity. Curr Opin Chem Biol. 2003;7(2):265–272. [PubMed]
  • Cork JM, Purugganan MD. The evolution of molecular genetic pathways and networks. Bioessays. 2004;26(5):479–484. [PubMed]
  • de Castro M, Elliot S, Kino T, Bamberger C, Karl M, Webster E, Chrousos GP. The non-ligand binding beta-isoform of the human glucocorticoid receptor (hGR beta): tissue levels, mechanism of action, and potential physiologic role. Mol Med. 1996;2(5):597–607. [PMC free article] [PubMed]
  • Gelsleichter J, Musick JA. Effects of insulin-like growth factor-I, corticosterone, and 3,3′, 5-tri-iodo-L-thyronine on glycosaminoglycan synthesis in vertebral cartilage of the clearnose skate, Raja eglanteria. J Exp Zool. 1999;284(5):549–556. [PubMed]
  • Greenwood AK, Butler PC, White RB, DeMarco U, Pearce D, Fernald RD. Multiple corticosteroid receptors in a teleost fish: distinct sequences, expression patterns, and transcriptional activities. Endocrinology. 2003;144(10):4226–4236. [PubMed]
  • Harvey BJ, Alzamora R, Stubbs AK, Irnaten M, McEneaney V, Thomas W. Rapid responses to aldosterone in the kidney and colon. J Steroid Biochem Mol Biol. 2008;108(3–5):310–317. [PubMed]
  • Hazon N, Henderson IW. Secretory dynamics of 1 alpha-hydroxycorticosterone in the elasmobranch fish, Scyliorhinus canicula. J Endocrinol. 1984;103(2):205–211. [PubMed]
  • Hazon N, Henderson IW. Factors affecting the secretory dynamics of 1 alpha-hydroxycorticosterone in the dogfish, Scyliorhinus canicula. Gen Comp Endocrinol. 1985;59(1):50–55. [PubMed]
  • Hittinger CT, Carroll SB. Gene duplication and the adaptive evolution of a classic genetic switch. Nature. 2007;449(7163):677–681. [PubMed]
  • Honn KV, Chavin W. In vitro tropic action of ACTH upon adrenocortical cyclic nucleotide (cAMP, cGMP) production and corticosterone output in blue (Prionace glauca Linnaeus) and in mako (Isurus oxyrinchus Rafinesque) sharks. Gen Comp Endocrinol. 1978;36(1):161–169. [PubMed]
  • Idler DR, Freeman HC, Truscott B. Biological activity and protein-binding of 1 alpha-hydroxycorticosterone: an interrenal steroid in elasmobranch fish. Gen Comp Endocrinol. 1967;9(2):207–213. [PubMed]
  • Idler DR, Truscott B. 1-alpha-hydroxycorticosterone from cartilaginous fish: a new adrenal steroid in blood. J Fish Res Board Can. 1966;23:615–619.
  • James LC, Tawfik DS. Catalytic and binding poly-reactivities shared by two unrelated proteins: the potential role of promiscuity in enzyme evolution. Protein Sci. 2001;10(12):2600–2607. [PMC free article] [PubMed]
  • Khersonsky O, Roodveldt C, Tawfik DS. Enzyme promiscuity: evolutionary and mechanistic aspects. Curr Opin Chem Biol. 2006;10(5):498–508. [PubMed]
  • Kime DE. Measurement of 1alpha-hydroxycorticosterone and other corticosteroids in elasmobranch plasma by radioimmunoassay. Gen Comp Endocrinol. 1977;33(3):344–351. [PubMed]
  • Krasowski MD, Yasuda K, Hagey LR, Schuetz EG. Evolution of the pregnane x receptor: adaptation to cross-species differences in biliary bile salts. Mol Endocrinol. 2005;19(7):1720–1739. [PMC free article] [PubMed]
  • Kumar R, Thompson EB. Gene regulation by the glucocorticoid receptor: structure: function relationship. J Steroid Biochem Mol Biol. 2005;94(5):383–394. [PubMed]
  • Le Menuet D, Viengchareun S, Penfornis P, Walker F, Zennaro MC, Lombes M. Targeted oncogenesis reveals a distinct tissue-specific utilization of alternative promoters of the human mineralocorticoid receptor gene in transgenic mice. J Biol Chem. 2000;275(11):7878–7886. [PubMed]
  • Manire CA, Rasmussen LE, Maruska KP, Tricas TC. Sex, seasonal, and stress-related variations in elasmobranch corticosterone concentrations. Comp Biochem Physiol A Mol Integr Physiol. 2007;148(4):926–935. [PubMed]
  • McGregor AP, Orgogozo V, Delon I, Zanet J, Srinivasan DG, Payre F, Stern DL. Morphological evolution through multiple cis-regulatory mutations at a single gene. Nature. 2007;448(7153):587–590. [PubMed]
  • Nettles KW, Bruning JB, Gil G, et al. (12 co-authors) Structural plasticity in the oestrogen receptor ligand-binding domain. EMBO Rep. 2007;8(6):563–568. [PMC free article] [PubMed]
  • Nunez BS, Evans AN, Simpson MA, Wong WP, Ip YK. Characterization of cDNAs encoding cholesterol side chain cleavage and 3beta-hydroxysteroid dehydrogenase in the freshwater stingray Potamotrygon motoro. Comp Biochem Physiol B Biochem Mol Biol. 2006;145(3–4):306–317. [PubMed]
  • Nunez BS, Vedeckis WV. Characterization of promoter 1B in the human glucocorticoid receptor gene. Mol Cell Endocrinol. 2002;189(1–2):191–199. [PubMed]
  • Nunez S, Trant JM. Regulation of interrenal gland steroidogenesis in the Atlantic stingray (Dasyatis sabina) J Exp Zool. 1999;284(5):517–525. [PubMed]
  • Ortlund EA, Bridgham JT, Redinbo MR, Thornton JW. Crystal structure of an ancient protein: evolution by conformational epistasis. Science. 2007 [PMC free article] [PubMed]
  • O'Toole LB, Armour KJ, Decourt C, Hazon N, Lahlou B, Henderson IW. Secretory patterns of 1 alpha-hydroxycorticosterone in the isolated perifused interrenal gland of the dogfish, Scyliorhinus canicula. J Mol Endocrinol. 1990;5(1):55–60. [PubMed]
  • Patent GJ. Comparison of some hormonal effects on carbohydrate metabolism in an elasmobranch (Squalus acanthias) and a holocephalan (Hydrolagus colliei) Gen Comp Endocrinol. 1970;14(2):215–242. [PubMed]
  • Pereira de Jesus-Tran K, Cote PL, Cantin L, Blanchet J, Labrie F, Breton R. Comparison of crystal structures of human androgen receptor ligand-binding domain complexed with various agonists reveals molecular determinants responsible for binding affinity. Protein Sci. 2006;15(5):987–999. [PMC free article] [PubMed]
  • Prossnitz ER, Arterburn JB, Sklar LA. GPR30: a G protein-coupled receptor for estrogen. Mol Cell Endocrinol. 2007;265–266:138–142. [PMC free article] [PubMed]
  • Prud'homme B, Gompel N, Rokas A, Kassner VA, Williams TM, Yeh SD, True JR, Carroll SB. Repeated morphological evolution through cis-regulatory changes in a pleiotropic gene. Nature. 2006;440(7087):1050–1053. [PubMed]
  • Rasmussen LEL, Crow GL. Serum corticosterone concentrations in immature captive whitetip reef sharks, Triaenodon obesus. J Exp Zool. 1993;267(3):283–287. [PubMed]
  • Simon P. Q-Gene: processing quantitative real-time RT-PCR data. Bioinformatics. 2003;19(11):1439–1440. [PubMed]
  • Snelson FF, Jr, Rasmussen LE, Johnson MR, Hess DL. Serum concentrations of steroid hormones during reproduction in the Atlantic stingray, Dasyatis sabina. Gen Comp Endocrinol. 1997;108(1):67–79. [PubMed]
  • Sturm A, Bury N, Dengreville L, Fagart J, Flouriot G, Rafestin-Oblin ME, Prunet P. 11-deoxycorticosterone is a potent agonist of the rainbow trout (Oncorhynchus mykiss) mineralocorticoid receptor. Endocrinology. 2005;146(1):47–55. [PubMed]
  • Suino-Powell K, Xu Y, Zhang C, Tao YG, Tolbert WD, Simons SS, Jr, Xu HE. Doubling the size of the glucocorticoid receptor ligand binding pocket by deacylcortivazol. Mol Cell Biol. 2008;28(6):1915–1923. [PMC free article] [PubMed]
  • Teichmann SA, Babu MM. Gene regulatory networks growth by duplication. Nat Genet. 2004;36(5):492–496. [PubMed]
  • Thomas P, Pang Y, Filardo EJ, Dong J. Identity of an estrogen membrane receptor coupled to a G protein in human breast cancer cells. Endocrinology. 2005;146(2):624–632. [PubMed]
  • Thomas P, Pang Y, Zhu Y, Detweiler C, Doughty K. Multiple rapid progestin actions and progestin membrane receptor subtypes in fish. Steroids. 2004;69(8–9):567–573. [PubMed]
  • Thornton JW. Evolution of vertebrate steroid receptors from an ancestral estrogen receptor by ligand exploitation and serial genome expansions. Proc Natl Acad Sci USA. 2001;98(10):5671–5676. [PMC free article] [PubMed]
  • Thornton JW. Resurrecting ancient genes: experimental analysis of extinct molecules. Nat Rev Genet. 2004;5(5):366–375. [PubMed]
  • Thornton JW, Need E, Crews D. Resurrecting the ancestral steroid receptor: ancient origin of estrogen signaling. Science. 2003;301(5640):1714–1717. [PubMed]
  • Truscott B, Idler DR. The widespread occurrence of a corticosteroid 1-alpha-hydroxylase in the interrenals of elasmobranchii. J Endocrinol. 1968;40(4):515–526. [PubMed]
  • Truscott B, Idler DR. Corticosteroids in plasma of elasmobranchs. Comp Biochem Physiol A. 1972;42(1):41–50. [PubMed]
  • Wagner A. Evolution of gene networks by gene duplications: a mathematical model and its implications on genome organization. Proc Natl Acad Sci USA. 1994;91(10):4387–4391. [PMC free article] [PubMed]
  • Walsh CJ, Wyffels JT, Bodine AB, Luer CA. Dexamethasone-induced apoptosis in immune cells from peripheral circulation and lymphomyeloid tissues of juvenile clearnose skates, Raja eglanteria. Dev Comp Immunol. 2002;26(7):623–633. [PubMed]
  • Wilkins AS. Recasting developmental evolution in terms of genetic pathway and network evolution and the implications for comparative biology. Brain Res Bull. 2005;66(4–6):495–509. [PubMed]
  • Zhu G, Golding GB, Dean AM. The selective cause of an ancient adaptation. Science. 2005;307(5713):1279–1282. [PubMed]

Articles from Molecular Biology and Evolution are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...