Biomolecules at Kenyon XX HHMI at Kenyon xx Jmol Home xx Biology Dept xx COMMENTS and CORRECTIONS

The Ancestral Protein of the Glucocorticoid Receptor (GR) and Mineralcorticoid Receptor (MR)

Haley Adcox '11 and Cami Odio '11


I. Introduction

The glucocorticoid receptor (GR) and its sister protein, mineralcorticoid receptor (MR), are transcription factors present in all jawed vertebrates (1). GR is activated by adrenal steroid cortisol and regulates stress response, glucose homeostasis, and other functions (2).  MR, on the other hand, is activated by aldosterone in tetrapods and by deoxycorticosterone (DOC) in teleosts.  It then acts as a hormone signal regulating sodium and water homeostasis as well as cardiovascular function, neuronal fate, and adipocyte differentiation (3).  Similarities between the two receptors and their ligands suggests that they descended from a single ancestor.

II. AncCR (Ancestrol Corticoid Receptor): The ancestor of the sister proteins MR and GR

A  phylogenetic analysis of existing GR and MR sequences revealed that the receptors descend from the duplication of a single ancient gene, the ancestrol corticoid receptor (AncCR), in the vertebrate lineage  ~450 million years ago . This phylogenetic analysis was also used to infer a genetic sequence  for the AncCR that was biochemically synthesized, expressed in cultured cells, and the resulting protein was crystallized. Overall, the structure of AncCR consists of nine helices and an activation-function helix(AF-H). The study of its structure was then used to understand the evolution of MR and GR. Importantly, the “resurrected” AncCR binds aldosterone and DOC with highest affinity, while it’s affinity to cortisol is much lower. Further, AncCR’s network of hydrogen bonds for aldosterone binding is conserved in the modern MR and GR. Thus, GR's hormone specificity evolved from the AncCR through a specific set of historical mutations without altering these three residues.  

III. Conformational Epistasis: The evolutionary switch in binding preference from aldosterone to cortisol (mutations S106P and L111Q)

Conformational epistasis is a mode of structural evolution where one substitution changes the position and thus, the functional effect of a second substitution. Two mutations of the ancestor protein, S106P and L111Q, are characterized as epistasic changes and are conserved in modern GR. In combination, the two substitutions destabilize binding with aldosterone or DOC and re-establish stability in a cortisol-specific fashion. Thus they switch the binding preference from aldosterone to cortisol. 

Specifically, these mutations occur on helices 6 and 7 of the ancestor protein. The position of helix 7 of the ancestor protein is stabilized by a hydrogen bond between Ser106 and the carbonyl backbone of Met103.  The S106P mutation  breaks the H-bond and creates a sharp kink in the backbone. This kink repositions the loop downward,  and partially unwinds helix 7 (see 'mutations' link above). Thus, through this destabilization of a crucial region of the receptor, S106P impairs activation by all ligands. However, the movement of helix 7 brings site 111 into close proximity with the binding ligand. Further, the mutation of this site, L111Q, creates a hydrogen bond with cortisol’s C17-hydroxyl (see 'mutations' link above).  Because aldosterone and DOC lack this hydroxyl, the new bond is cortisol specific. Alone, the L111Q mutant has no effect on the binding of any of the ligands because it is too far from the ligand binding site. However, the repositioning provided by S106P, brings position 111 close to the ligand and increases the effect of the L111Q mutation. These mutations can be characterized as conformational epistasis because both mutations are required for the switch in binding preference from aldosterone to cortisol.

IV.  Disruption of the activation-function helix (AF-H): Three mutations (L29M, F98I, and deletion S212∆) that complete the functional switch

Although S106P and L111Q mutations result in stabilization of cortisol binding, receptors with these mutations still retain some affinity to aldosterone and DOC. Thus, further mutations are required to decrease affinity for MR’s ligands, and complete the switch to the singular cortisol binding that characterizes GR. Specifically, the transcriptional activity of all steroid receptors depends on the stability of the activation-function helix (AF-H), whose repositioning due to ligand binding creates an interface for transcriptional co-activators. The stability of the AF-H depends on interactions between S212, the loop preceding the AF-H, the ligand, and helix 3. Three mutations (L29M, F98I, and deletion S212∆ ) decrease the receptor’s affinity for aldosterone and DOC by disrupting these interactions. In particular, by removing a serine (that contains an H-bonding hydroxyl), S212∆ eliminates a hydrogen bond that stabilizes the AF-H loop. Second, L29M on helix 3 creates a steric clash and unfavorable interactions between the inserted methionine and the D-ring of aldosterone. This mutation also enhances cortisol specificity by forming an H-bond between the Met29 and cortisol’s unique C-17 hydroxyl. The last mutation, F98I opens up space between the ligand, helix 3 and helix 7 through the elimination of base stacking by Phe98 and the insertion of the longer side chain Ile29. This indirectly de-stabilizes the AF-H thus impairing activation by all ligands. However, the space created by F98I does stabilize the crucial H-bond between Q111 and cortisol's C -17 hydroxyl by relieving a steric clash between the repositioned loop and Met108 .

V.  Permissive mutations (N26T, Q105L, and Y27R): Stabilizing receptors to tolerate the functional switch

The three mutations (L29M, F98I, and deletion S212∆ ) that decrease binding affinity for MR’s ligands and thus complete the functional switch to singular cortisol binding, actually result in nonfunctional receptors that cannot activate transcription. Thus, more mutations are required to stabilize specific elements of the protein allowing it to tolerate the function switching changes. These mutations, which have no consequence in the ancestor receptor but stabilize the functional-switches of the new form, are called permissive mutations. The first of these, N26T, creates a new hydrogen bond on the loop between helix 3 and the AF-H . The next permissive mutation, Q105L, allows helix 7 to pack more tightly against helix 3 by increasing Van-der-waals interactions. As a result, this mutation stabilizes helix 3 and indirectly stabilizes AF-H. These two permissive substitutions restore transcriptional activity to the receptor in a fully GR-like fashion. A final permissive mutation that may have evolved even earlier than those previously noted is Y27R. The permissive character of this mutation is supported by the finding that it has a negligible effect on AncCR, but it’s reversal in receptors containing cortisol binding preference abolishes ligand activation. Specifically, the inserted arginine stabilizes helix 3 and the ligand pocket by forming a cation-π; interaction with Tyr17 .

VI. The Overall Evolution by Epistasis: from AncCR to MR and GR  

The overall evolution from the AncCR to MR and GR resulted from a gene duplication of the ancestor ~450 million years ago. One of these duplicates, which resulted in MR, has been conserved for ~400 million years. Consequently, it binds preferentially to aldosterone and DOC in a similar fashion as AncCR. The other gene duplicate evolved to GR through multiple mutations that resulted in high affinity binding for cortisol and very low activation by aldosterone and DOC.  Because some of these mutations (namely those that complete the functional switch: L29M, F98I, and deletion S212∆) abolish ligand binding all together, evolution by epistasis was crucial to maintaining an active receptor. Specifically, the permissive mutations (N26T, Q105L, and Y27R) had to occur on the same receptors as the evolutionary switch mutations (S106P and L111Q) and before the AF-H disrupting mutations could occur. If the  AF-H disrupting mutations occurred before the other two groups of mutations could be coupled on a receptor a non-functional intermediate would have resulted. In general, non-functional intermediates are less likely to support evolutionary trajectories. Thus, a specified order of the three mutations in which permissive substitutions occurred before the destabilizing mutations that confer new function was crucial to GR development. This order of general mutations along with a high incidence of conformational epistasis indicates two factors that may be crucial to all genetic evolution and should be considered in studies that involve evolutionary trajectories.


(1) Eric A. Ortlund, Jamie T. Bridgham, Matthew R. Redinbo, and Joseph W. Thornton (2007). "Crystal structure of an ancient protein: evolution by conformational epistasis." Science  317: 1544-1548

(2)  Lombes, M., L. Pascual-Le Tallec.  (2005). "The mineralocorticoid receptor: a journey exploring its diversity and specificity of action."  Molecular Endrocrinology. 9:2211-21

(3) Lu NZ, Wardell SE, Burnstein KL, Defranco D, Fuller PJ, Giguere V, Hochberg RB, McKay L, Renoir JM, Weigel NL, Wilson EM, McDonnell DP, Cidlowski JA (2006). "International Union of Pharmacology. LXV. The pharmacology and classification of the nuclear receptor superfamily: glucocorticoid, mineralocorticoid, progesterone, and androgen receptors". Pharmacol Revl 58 (4):782–97.

Back to Top