A multi-step, dynamic allosteric model of testosterone's binding to sex hormone binding globulin

Highlights

• Accurate estimation of free testosterone (fT) is used widely in the diagnosis of hyogonadism.

• Complexity of direct measurements have led to development of models for calculation of fT.

• Current models used for fT estimation show systematic deviation from values measured by equilibrium dialysis.

• The SHBG exhibits complex behaviour in associating with testosterone.

• fT values from new mechanistic model incorporating allostery in SHBG:T match closely with those measured by equilibrium dialysis.

Abstract

Purpose

Circulating free testosterone (FT) levels have been used widely in the diagnosis and treatment of hypogonadism in men. Due to experimental complexities in FT measurements, the Endocrine Society has recommended the use of calculated FT (cFT) as an appropriate approach for estimating FT. We show here that the prevailing model of testosterone's binding to SHBG, which assumes that each SHBG dimer binds two testosterone molecules and that the two binding sites on SHBG have similar binding affinity is erroneous and provides FT values that differ substantially from those obtained using equilibrium dialysis.

Methods

We characterized testosterone's binding to SHBG using binding isotherms, ligand depletion curves, and isothermal titration calorimetry (ITC). We derived a new model of testosterone's binding to SHBG from these experimental data and used this model to determine FT concentrations and compare these values with those derived from equilibrium dialysis.

Results

Experimental data on testosterone's association with SHBG generated using binding isotherms including equilibrium binding, ligand depletion experiments, and ITC provide evidence of a multi-step dynamic process, encompassing at least two inter-converting microstates in unliganded SHBG, readjustment of equilibria between unliganded states upon binding of the first ligand molecule, and allosteric interaction between two binding sites of SHBG dimer. FT concentrations in men determined using the new multistep dynamic model with complex allostery did not differ from those measured using equilibrium dialysis. Systematic error in calculated FT vales in females using Vermeulen's model was also significantly reduced. In European Male Aging Study, the men deemed to have low FT (<2.5th percentile) by the new model were at increased risk of sexual symptoms and elevated LH.

Conclusion

Testosterone's binding to SHBG is a multi-step dynamic process that involves complex allostery within SHBG dimer. FT values obtained using the new model have close correspondence with those measured using equilibrium dialysis.

Introduction

Testosterone, the major androgen in humans, circulates in blood bound largely to sex hormone binding globulin (SHBG) and albumin (Bhasin et al, 2010, Hammond, Bocchinfuso, 1996, Mendel, 1989, Rosner, 1991, Rosner et al, 2007). Testosterone can also bind to orosomucoid and transcortin proteins, but the amount of testosterone bound to these proteins in human plasma is negligible. In many conditions that affect SHBG concentrations, such as obesity, diabetes, aging, hyperthyroidism, liver disease, and HIV-infection, total testosterone concentrations are altered because of changes in SHBG concentrations; in these conditions, expert panels have recommended the determination of free testosterone (FT) concentration to obtain an accurate assessment of androgen status (Bhasin et al, 2010, Hammond, Bocchinfuso, 1996, Mendel, 1989, Rosner, 1991, Rosner et al, 2007).

Reflecting the growing interest in men's health and the success of pharmaceutical advertising, testosterone sales have grown from 23 million dollars in 1993 to 70 million in 2000 to 1.7 billion dollars in 2012 (Spitzer et al., 2012). Testosterone is the second most frequently ordered test, next only to 25-hydroxyvitamin D. In 2012, nearly 4 million free testosterone tests were performed in the USA alone. A number of direct and indirect methods – equilibrium dialysis, ultrafiltration, tracer analog methods, and calculations based on simple law-of-mass action equations – have been developed for the determination of FT levels (Adachi et al, 1991, Mazer, 2009, Rosner, 1997, Rosner et al, 2007, Sinha-Hikim et al, 1998, Sodergard et al, 1982, Van Uytfanghe et al, 2004, Vermeulen et al, 1971, Vermeulen et al, 1999, Winters et al, 1998). Expert panels have expressed concern about the accuracy and methodological complexity of the available assays for FT (Rosner et al, 2007, Sodergard et al, 1982, Vermeulen et al, 1971). Recognizing these methodological difficulties in the measurement of free testosterone, the Endocrine Society's Expert Panel suggested that “the calculation of free testosterone from reliably measured total testosterone and SHBG using mass action equations provides the best approach for the estimation of free testosterone…” (Rosner et al., 2007). Therefore, algorithms for calculating FT from total testosterone, SHBG and albumin concentrations using the extant linear binding model (also referred to as law-of-mass-action equations) (Adachi et al, 1991, Bhasin et al, 2011, Ly, Handelsman, 2005, Ly et al, 2010, Mazer, 2009, Morales et al, 2012, Morley et al, 2002, Rosner et al, 2007, Sartorius et al, 2009, Sodergard et al, 1982, Vermeulen et al, 1999) or empirically-derived equations (Ly, Handelsman, 2005, Ly et al, 2010, Nanjee, Wheeler, 1985, Sartorius et al, 2009) have been published and used widely (Adachi et al, 1991, Bhasin et al, 2011, Ly, Handelsman, 2005, Ly et al, 2010, Mazer, 2009, Morales et al, 2012, Morley et al, 2002, Rosner et al, 2007, Sartorius et al, 2009).

The current model of testosterone's binding to SHBG assumes that each SHBG dimer binds two testosterone molecules, and that each of the two binding sites on SHBG dimer has similar binding affinity irrespective of the occupancy of the adjacent binding site (no allostery). Equations to determine FT, using this model, have been proposed by Vermeulen, Sodergard, and Mazer (Mazer, 2009, Rosner et al, 2007, Sodergard et al, 1982, Vermeulen et al, 1971). In present work we characterized testosterone's binding to SHBG using equilibrium dialysis (varying ligand and SHBG concentrations) and isothermal titration calorimetry (ITC) to characterize testosterone's binding to SHBG. We considered several possible mechanistic models of molecular interactions, including the prevailing model of homogeneous binding of testosterone to SHBG as envisioned by Vermeulen (Vermeulen et al., 1999), Sodergard (Sodergard et al., 1982) and implemented in a spreadsheet by Mazer (Mazer, 2009), and various allosteric mechanisms, including positive and negative cooperativity (Koshland et al, 1966, Monod et al, 1965), and ensemble allostery (Freiburger et al, 2011, Hilser, Thompson, 2007) (Fig. 1). Based on our analyses of the experimental data of testosterone's binding to SHBG, we constructed a novel multistep binding model with complex intra-dimer allostery for the calculation of FT, which provided the best fit to the totality of experimental data. This new model was then utilized to determine FT concentrations in samples derived from randomized testosterone trials in men and women, and to compare the results with those obtained using equilibrium dialysis. Finally, we used the algorithm to examine the distribution of FT levels in community-dwelling men in the Framingham Heart Study (FHS) and related the deviations from the mean to the risk of sexual symptoms and elevated LH levels in an independent sample of men in the European Male Aging Study (EMAS).