From steroid receptors to cytokines: the thermodynamics of self-associating systems

Abstract

Since 1987, the Gibbs Conference on Biothermodynamics has maintained a focus on understanding the quantitative aspects of gene regulatory systems. These studies coupled rigorous techniques with exact theory to dissect the linked reactions associated with bacterial and lower eukaryotic gene regulation. However, only in the last ten years has it become possible to apply this approach to clinically relevant, human gene regulatory systems. Here we summarize our work on the thermodynamics of human steroid receptors and their interactions with multi-site promoter sequences, highlighting results not available from more traditional biochemical and structural approaches. Noting that the Gibbs Conference has also served as a vehicle to promote the broader use of thermodynamics in understanding biology, we then discuss collaborative work on the hydrodynamics of a cytokine implicated in tumor suppression, prostate derived factor (PDF).

Figure 1. Schematic representation of steroid receptor structures

Receptors are as indicated on the left; number of amino acids is shown on right. Functional regions are as indicated: DBD, DNA binding domain; HBD, hormone binding domain; activation function, AF.

Figure 2. Schematic representation of receptor:HRE₂ assembly states

Circles represent hormone-bound receptor monomers, squares represent dimers. Dimerization (k_dimer) is coupled to response element binding (k_int); complete occupancy is coupled to an inter-site cooperative interaction (k_c). Arrow refers to the direction of transcriptional start site.

Figure 3. Sedimentation equilibrium analysis of PR-A

Panels represent each PR-A loading concentration: (A) 1.0 μM, (B) 0.5 μM and (C) 0.25 μM. Symbols represent PR-A absorbance at 15,000 rpm (squares), 18,000 rpm (circles), and 21,000 rpm (triangles). Solid lines represent the best fit model (monomer-dimer) from simultaneous analysis of all nine data sets using the program NONLIN [42]. Square root of the variance was 0.010 absorbance units. Buffer conditions were 20 mM Hepes, pH 8.0, 300 mM NaCl, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. Identical results were obtained for NaCl concentrations ranging from 50 mM to 1 M. Figure reproduced with permission from [16].

Figure 4. Quantitative footprint titration and individual-site isotherms for PR-A and PR-B assembly at the HRE₂ promoter

(A) Representative titration of the HRE₂ promoter by PR-A. Positions of the two HREs (site 1, filled rectangle; site 2, open rectangle) are indicated to right. (B) Individual-site isotherms for PR-A binding to site 1(filled circles) and site 2 (open circles) of the HRE₂ promoter, and binding to site 2 (open squares) of the HRE_1- promoter. Lines represent best fit to all data using equations 2–3. Buffer conditions were 20 mM Hepes, pH 8.0, 50 mM NaCl, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. (C) Same as B, but PR-B individual-site isotherms are displayed.

Figure 4. Quantitative footprint titration and individual-site isotherms for PR-A and PR-B assembly at the HRE₂ promoter

(A) Representative titration of the HRE₂ promoter by PR-A. Positions of the two HREs (site 1, filled rectangle; site 2, open rectangle) are indicated to right. (B) Individual-site isotherms for PR-A binding to site 1(filled circles) and site 2 (open circles) of the HRE₂ promoter, and binding to site 2 (open squares) of the HRE_1- promoter. Lines represent best fit to all data using equations 2–3. Buffer conditions were 20 mM Hepes, pH 8.0, 50 mM NaCl, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. (C) Same as B, but PR-B individual-site isotherms are displayed.

Figure 4. Quantitative footprint titration and individual-site isotherms for PR-A and PR-B assembly at the HRE₂ promoter

(A) Representative titration of the HRE₂ promoter by PR-A. Positions of the two HREs (site 1, filled rectangle; site 2, open rectangle) are indicated to right. (B) Individual-site isotherms for PR-A binding to site 1(filled circles) and site 2 (open circles) of the HRE₂ promoter, and binding to site 2 (open squares) of the HRE_1- promoter. Lines represent best fit to all data using equations 2–3. Buffer conditions were 20 mM Hepes, pH 8.0, 50 mM NaCl, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. (C) Same as B, but PR-B individual-site isotherms are displayed.

Figure 5. Predicted distribution of each macroscopic PR-A:HRE₂ and PR-B:HRE₂ ligation state

(A) Distribution of PR-A ligation states as predicted from the experimentally determined energetics (Table 1). (B) Same as A except that PR-B ligation states are displayed. Unligated HRE₂ is represented by dashed line; singly ligated promoter is represented by solid line; fully ligated promoter is represented by dotted line, and proportion of free dimer is represented by “+”. The shaded box represents the estimated intracellular PR concentration [43]. Figure reproduced with permission from [21] Copyright 2007 National Academy of Sciences, USA.

Figure 6. Quantitative footprint titration, schematic representation of assembly states, and individual-site isotherms for PR-A binding to the MMTV promoter

(A) Representative titration of the MMTV_wt promoter by PR-A. Positions of the HREs are shown to right. (B) Schematic of PR-A:MMTV promoter assembly states. Monomers dimerize in solution (k_dimer) and bind at a half-sites (k₁); dimers bind at the palindromic site (k₂). Monomers bound at half-sites cooperatively interact in a pair-wise fashion (k_c1). Non-additive cooperativity induced by addition of a third monomer to the half-sites is accounted for by k_c2. Saturation of the promoter is linked to a third cooperative interaction between the palindrome and the three half-sites (k_c3). (C) Individual-site isotherms for PR-A binding to the reduced-valency MMTV_1-,3- and MMTV_1-,4- promoters (see figure inset). Solid line represents a simultaneous fit to a pair-wise cooperative model by fixing ΔG₁ at −8.1 kcal/mol and allowing ΔG_c1 to float. Dotted line represents a global fit to all data using a non-cooperative two-site model. Monomer binding to site 2 (open circles); binding to site 3 (open diamonds); binding to site 4 (open triangles).

Figure 6. Quantitative footprint titration, schematic representation of assembly states, and individual-site isotherms for PR-A binding to the MMTV promoter

(A) Representative titration of the MMTV_wt promoter by PR-A. Positions of the HREs are shown to right. (B) Schematic of PR-A:MMTV promoter assembly states. Monomers dimerize in solution (k_dimer) and bind at a half-sites (k₁); dimers bind at the palindromic site (k₂). Monomers bound at half-sites cooperatively interact in a pair-wise fashion (k_c1). Non-additive cooperativity induced by addition of a third monomer to the half-sites is accounted for by k_c2. Saturation of the promoter is linked to a third cooperative interaction between the palindrome and the three half-sites (k_c3). (C) Individual-site isotherms for PR-A binding to the reduced-valency MMTV_1-,3- and MMTV_1-,4- promoters (see figure inset). Solid line represents a simultaneous fit to a pair-wise cooperative model by fixing ΔG₁ at −8.1 kcal/mol and allowing ΔG_c1 to float. Dotted line represents a global fit to all data using a non-cooperative two-site model. Monomer binding to site 2 (open circles); binding to site 3 (open diamonds); binding to site 4 (open triangles).

Figure 6. Quantitative footprint titration, schematic representation of assembly states, and individual-site isotherms for PR-A binding to the MMTV promoter

(A) Representative titration of the MMTV_wt promoter by PR-A. Positions of the HREs are shown to right. (B) Schematic of PR-A:MMTV promoter assembly states. Monomers dimerize in solution (k_dimer) and bind at a half-sites (k₁); dimers bind at the palindromic site (k₂). Monomers bound at half-sites cooperatively interact in a pair-wise fashion (k_c1). Non-additive cooperativity induced by addition of a third monomer to the half-sites is accounted for by k_c2. Saturation of the promoter is linked to a third cooperative interaction between the palindrome and the three half-sites (k_c3). (C) Individual-site isotherms for PR-A binding to the reduced-valency MMTV_1-,3- and MMTV_1-,4- promoters (see figure inset). Solid line represents a simultaneous fit to a pair-wise cooperative model by fixing ΔG₁ at −8.1 kcal/mol and allowing ΔG_c1 to float. Dotted line represents a global fit to all data using a non-cooperative two-site model. Monomer binding to site 2 (open circles); binding to site 3 (open diamonds); binding to site 4 (open triangles).

Figure 7. Individual-site binding isotherms for PR-B binding to the HRE₂ promoter in KCl, and weight-average s plots for PR-B and PR-A as a function of monovalent cation type

(A) Individual-site binding isotherms for PR-B binding to site 1 (filled circles) and 2 (open circles) of the HRE₂ promoter, and site 2 (open square) of the HRE_1- promoter in 50 mM KCl. Lines (HRE₂, solid; HRE_1- dotted) represent best fit model using Eq. 2–3. (B) Initial PR-B loading concentration was 1.0 μ M. The weight average sedimentation coefficient was determined as implemented in the program DCDT+ [44, 45]. Error estimates represent one standard deviation as reported by DCDT+. Solution conditions minus MCl were 20 mM Tris, pH 8.0, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. The concentration of MCl was 10 mM (squares, solid line), 50 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot. (C) As described for B except the results for PR-A are displayed. Concentration of MCl was 100 mM (squares, solid line), 200 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot.

Figure 7. Individual-site binding isotherms for PR-B binding to the HRE₂ promoter in KCl, and weight-average s plots for PR-B and PR-A as a function of monovalent cation type

(A) Individual-site binding isotherms for PR-B binding to site 1 (filled circles) and 2 (open circles) of the HRE₂ promoter, and site 2 (open square) of the HRE_1- promoter in 50 mM KCl. Lines (HRE₂, solid; HRE_1- dotted) represent best fit model using Eq. 2–3. (B) Initial PR-B loading concentration was 1.0 μ M. The weight average sedimentation coefficient was determined as implemented in the program DCDT+ [44, 45]. Error estimates represent one standard deviation as reported by DCDT+. Solution conditions minus MCl were 20 mM Tris, pH 8.0, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. The concentration of MCl was 10 mM (squares, solid line), 50 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot. (C) As described for B except the results for PR-A are displayed. Concentration of MCl was 100 mM (squares, solid line), 200 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot.

Figure 7. Individual-site binding isotherms for PR-B binding to the HRE₂ promoter in KCl, and weight-average s plots for PR-B and PR-A as a function of monovalent cation type

(A) Individual-site binding isotherms for PR-B binding to site 1 (filled circles) and 2 (open circles) of the HRE₂ promoter, and site 2 (open square) of the HRE_1- promoter in 50 mM KCl. Lines (HRE₂, solid; HRE_1- dotted) represent best fit model using Eq. 2–3. (B) Initial PR-B loading concentration was 1.0 μ M. The weight average sedimentation coefficient was determined as implemented in the program DCDT+ [44, 45]. Error estimates represent one standard deviation as reported by DCDT+. Solution conditions minus MCl were 20 mM Tris, pH 8.0, 2.5 mM MgCl₂, 1 mM CaCl₂, 1 mM DTT, and 10⁻⁵ M progesterone at 4°C. The concentration of MCl was 10 mM (squares, solid line), 50 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot. (C) As described for B except the results for PR-A are displayed. Concentration of MCl was 100 mM (squares, solid line), 200 mM (open circles, dashed line) or 300 mM (triangles, dotted line). M⁺ is as indicated on the plot.

Figure 8. Resolved binding parameters for receptor assembly at the HRE₂ promoter

Results are plotted as the natural log of each microstate constant, k_i. Symbols represent k_int (open circles), k_dimer (closed triangles), and k_c (closed squares). Each set of constants was fit to a straight line to emphasize the trend in the data.

Figure 9. Purification and sedimentation velocity analysis of PDF

(A) 1 μg PDF was resolved by SDS-PAGE and imaged by silver staining. (B) Representative sedimentation velocity data collected at 50,000 rpm. Only every thirtieth scan is shown for clarity. Buffer conditions were 5 mM HCl, 4°C. (C) c(s) distributions for PDF at 40 μM (black line), 24 μM (dark grey) and 5 μM (light grey). Distributions were determined using SedFit [38].

Figure 9. Purification and sedimentation velocity analysis of PDF

(A) 1 μg PDF was resolved by SDS-PAGE and imaged by silver staining. (B) Representative sedimentation velocity data collected at 50,000 rpm. Only every thirtieth scan is shown for clarity. Buffer conditions were 5 mM HCl, 4°C. (C) c(s) distributions for PDF at 40 μM (black line), 24 μM (dark grey) and 5 μM (light grey). Distributions were determined using SedFit [38].

Figure 9. Purification and sedimentation velocity analysis of PDF

(A) 1 μg PDF was resolved by SDS-PAGE and imaged by silver staining. (B) Representative sedimentation velocity data collected at 50,000 rpm. Only every thirtieth scan is shown for clarity. Buffer conditions were 5 mM HCl, 4°C. (C) c(s) distributions for PDF at 40 μM (black line), 24 μM (dark grey) and 5 μM (light grey). Distributions were determined using SedFit [38].