Toward a generalized computational workflow for exploiting transient pockets as new targets for small molecule stabilizers: Application to the homogentisate 1,2-dioxygenase mutants at the base of rare disease Alkaptonuria

Highlights

• A new approach to the treatment of Alkaptonuria is proposed by use of Pharmacological Chaperones (PCs).

• Transient pockets at the surface of mutated enzyme are exploited as targets for PC.

• A workflow employing ready-to-use tools is proposed.

Abstract

Alkaptonuria (AKU) is an inborn error of metabolism where mutation of homogentisate 1,2-dioxygenase (HGD) gene leads to a deleterious or misfolded product with subsequent loss of enzymatic degradation of homogentisic acid (HGA) whose accumulation in tissues causes ochronosis and degeneration. There is no licensed therapy for AKU. Many missense mutations have been individuated as responsible for quaternary structure disruption of the native hexameric HGD. A new approach to the treatment of AKU is here proposed aiming to totally or partially rescue enzyme activity by targeting of HGD with pharmacological chaperones, i.e. small molecules helping structural stability. Co-factor pockets from oligomeric proteins have already been successfully exploited as targets for such a strategy, but no similar sites are present at HGD surface; hence, transient pockets are here proposed as a target for pharmacological chaperones. Transient pockets are detected along the molecular dynamics trajectory of the protein and filtered down to a set of suitable sites for structural stabilization by mean of biochemical and pharmacological criteria. The result is a computational workflow relevant to other inborn errors of metabolism requiring rescue of oligomeric, misfolded enzymes.

Introduction

Pharmacological chaperones (PCs) are small molecules designed to facilitate the correct folding of a protein and to re-establish its functionality (Aymami et al., 2013). Their role is similar to physiological chaperones: instead of macromolecules they are small molecules with the task of stabilizing an already folded protein through binding (Petsko et al., 2009) and subsequent increase in non-covalent interactions among protein structural segments. PCs, obeying to the Lipinski rule of five (Lipinski, 2009), have a low molecular weight giving the advantage of being administered orally with a non-invasive treatment (Leeson and Springthorpe, 2007).

PCs should not be confused with chemical chaperones. The latter are non-specific molecules, that exert their stabilization action on protein structures without binding to a specific site; trehalose (Crowe, 2007) and glycerol (Ohnishi et al., 1999) are well known examples. The non-specific mechanism of action of chemical chaperones requires a high concentration for them be effective, preventing their use as therapeutics.

Most of the PCs currently in clinical trial or already registered as therapeutics are designed to target active (Germain and Fan, 2009) or co-factor binding sites (Santos-Sierra et al., 2012). They are analogues of physiological substrates or co-factors, therefore binding with high affinity and selectivity to the target binding site.

It has been shown that the binding of an inhibitor to an enzyme may stabilize the protein against denaturation, leading to a greater stability of the structure (Ringe and Petsko, 2009), as proven also by the improvement in the crystallization process of inhibitor-bound enzymes (Hassel et al., 2007). Thus, enzyme inhibitors are good candidates as PCs provided that they are reversible competitive, hence allowing for the presence of a small amount of free enzyme to be available for substrate binding. Again, especially in the case of severely destabilized structures, a very high concentration of such inhibitor acting as PC would be necessary, shifting the equilibrium toward the bound, inactive form of the enzyme, thus neutralizing the stabilization effect.

Currently, use of co-factor sites as targets for PC designed as analogues of natural binders has proven to be the most suitable strategy in small molecule-assisted stabilization of protein structures, as for the case of Tafamidis and Transthyretin (TTR) (Bulawa et al., 2012). Transthyretin-related hereditary amyloidosis (ATTR) is an autosomal neurodegenerative disease caused by mutations in the TTR gene leading to loss of three-dimensional structure at both tertiary and quaternary level, which in turn leads to protein aggregation causing cell death. The commercially available PC Tafamidis exploits the co-factor binding site for thyroxine (T4) located at the protomer–protomer interface, increasing the number of inter-chain interactions, which in turn stabilize the tetramer structure and prevent aggregation. Such example well illustrates why modulation of protein–protein interface by PCs is becoming a new important branch in drug discovery, especially for diseases characterized by quaternary structure instability (Bier et al., 2015, Makley and Gestwicki, 2013)

Unfortunately, co-factor sites are not always available in protein structures while PCs may, ideally, bind anywhere on the surface of a protein to give stabilization; hence, the search for alternative, non-biologically active binding sites is needed (Metz et al., 2012). A test case is possibly the hexameric homogentisate 1,2-dioxygenase (HGD), whose activity deficiency due to gene mutations leads to Alkaptonuria (AKU), an inborn error of metabolism resulting in the accumulation of homogentisic acid (HGA) and ochronosis (Mitri et al., 2017, Braconi et al., 2015, Braconi et al., 2017). AKU is a multisystemic disease (Bernardini et al., 2015) in which also secondary amyloidosis (Millucci et al., 2012, Millucci et al., 2014a, Millucci et al., 2014b) and angiogenesis occur (Millucci et al., 2016), with an easy diagnosis (Jacomelli et al., 2017, Millucci et al., 2014c) but no current licensed therapy (Ranganath et al., 2016). Several missense mutations have been individuated as responsible for HGD quaternary structure disruption (see Fig. 1) (Ranganath et al., 2016). Contrarily to TTR, no co-factor sites are present at HGD enzyme surface to be exploited for PC targeting, therefore alternative approaches must be sought.

A new approach to the treatment of AKU is here proposed by targeting the mutated HGD by PCs aiming to hexamer stabilization, seeking for alternative pockets showing along the molecular dynamics simulation of the protein in a transient fashion, i.e. transient pockets (Bernini et al., 2014).