Research article
Structural and energetic insight into the interactions between the benzolactam inhibitors and tumor marker HSP90α

https://doi.org/10.1016/j.compbiolchem.2015.07.013Get rights and content

Highlights

  • Computational methods were used to reveal the interactions of benzolactam inhibitors to HSP90α.

  • Hydrophobic interactions contributed the most to the binding affinity.

  • A good linear correlation was obtained between the calculated and the experimental binding free energies.

  • Ala55, Ile96, and Leu107 are responsible for the different binding affinities of compounds.

Abstract

The heat shock protein 90α (HSP90α) provides a promising molecular target for cancer therapy. A series of novel benzolactam inhibitors exhibited distinct inhibitory activity for HSP90α. However, the structural basis for the impact of distinct R1 substituent groups of nine benzolactam inhibitors on HSP90α binding affinities remains unknown. In this study, we carried out molecular docking, molecular dynamics (MD) simulations, and molecular mechanics and generalized Born/surface area (MM–GBSA) binding free energy calculations to address the differences. Molecular docking studies indicated that all nine compounds presented one conformation in the ATP-binding site of HSP90α N-terminal domain. MD simulations and subsequent MM–GBSA calculations revealed that the hydrophobic interactions between all compounds and HSP90α contributed the most to the binding affinity and a good linear correlation was obtained between the calculated and the experimental binding free energies (R = 0.88). The per residue decomposition revealed that the most remarkable differences of residue contributions were found in the residues Ala55, Ile96, and Leu107 defining a hydrophobic pocket for the R1 group, consistent with the analysis of binding modes. This study may be helpful for the future design of novel HSP90α inhibitors.

Introduction

As ubiquitous proteins in both prokaryotic and eukaryotic, Heat shock proteins (HSPs) are a family of proteins induced by heat shock. The quintessential members of this group are associated with the folding and unfolding of a wide variety of other client proteins (Taipale et al., 2010). According to their molecular size, mammalian HSPs can be divided into six families, such as HSP27, HSP40, HSP60, HSP70, HSP90, and HSP100 (Khalil et al., 2011). Mammalian HSP90, a highly conserved molecular chaperone, has two major isoforms in the cytoplasm, namely HSP90α (inducible form) and HSP90β (constitutive form). To date, over 200HSP90 client proteins have been identified, including signaling protein kinases (e.g. epidermal growth factor receptor, Akt kinase, Ras), transcription factors (e.g. p53), and steroid hormone receptors (Khalil et al., 2011). These client proteins are well involved in the cellular signal transduction pathways, such as PI3K-Akt-MAPK/ERK and cAMP-dependent pathways. Therefore, in addition to the involvement in protein folding, HSP90 also plays fundamental roles in a plethora of cellular processes, encompassing intracellular transport, signal transduction, protein degradation, gene regulation, and apoptosis (Zhao et al., 2005).

Various studies demonstrate that overexpression of HSP90 has been implicated in various human carcinomas, including both solid tumors and hematological malignancies, marking HSP90 a promising molecular target for cancer therapy (Li et al., 2007, Wang et al., 2009, Shiau et al., 2006). It is also well-documented that the secretion of HSP90α from the cytosol to the extracellular space can be processed under the stress conditions such as hypoxia and oxidative stress (Li et al., 2007). Recently, Wang et al. (2009) have ascertained that the secreted HSP90α is responsible for tumor invasiveness and blockade of the secreted HSP90α inhibits tumor invasiveness. Moreover, the level of plasma HSP90α is correlated with tumor malignancy in clinical cancer patients. These data not only imply the effective target of HSP90α in cancer therapeutics, but also reveal the feasibility of using HSP90α as a viable marker for diagnosis of malignant tumors. Indeed, HSP90α has recently been proved to be a new tumor marker (Khalil et al., 2011).

Structurally, HSP90 is a homodimer and each monomer is composed of three flexible domains, including an N-terminal domain, a middle domain, and a C-terminal dimerization domain. The N-terminal domain is the site of ATP binding and is related to the weak ATPase activity. Association of ATP molecule to the N-terminal domain allosterically induces conformational changes in the arrangement of the N-terminal and middle domains and then is hydrolyzed to ADP when the allosteric regulations are complete (Shiau et al., 2006, Lu et al., 2014a, Lu et al., 2014b). This allosteric activation mechanism is greatly conserved in the HSP family (Lu et al., 2014c). From the therapeutic standpoint, it is feasible to design potent inhibitors to compete with ATP, thereby resulting in the inhibition of HSP90 function which contributes to an effective cancer therapy. Therefore, the functional role of HSP90 in the pathogenesis of cancer is gaining increasingly attention in developing HSP90 inhibitors.

Inhibitors targeting the ATP-binding site of HSP90α are currently in various stages of developments. As of yet, none of HSP90α inhibitors are available in the market. In fact, some promising HSP90α inhibitors display excellent antitumor activities and have entered clinical phase II and/or III trials in non-small-cell lung cancer, melanoma, and acute myeloid leukemia (Bhat et al., 2014), such as AUY922, AT13387, STA9090, KW-2478, and SNX5422 (Fig. 1). Very recently, Ernst et al. (2014) reported a novel benzolactam series of HSP90α inhibitors. Of them, the potent compound 33 showed an IC50 of 3 nM for HSP90α. Remarkably, the compounds with the different carbon linked substitution ortho to the lactam, namely the R1 substituent groups (Table 1), may reduce the HSP90α binding affinity compared to the compound 33 with the R1 substituent group of the 4-tetrahydro-2H-pyran. For instance, the compound 36 with the R1 group of tert-butyl showed 60-fold lower potency against the compound 33. However, the structural basis for the impact of distinct R1 substituent groups on HSP90α binding affinities remains unknown.

Computational modeling, with its ability to assess the protein–ligand interactions at atomic level, has been increasingly employed to insight into the structural and energetic properties of the formation of biomolecular complexes. In this study, we performed molecular docking, molecular dynamics (MD) simulations, and the binding free energy calculations (molecular mechanics and generalized Born/surface area, MM–GBSA) to illuminate the detailed interactions between the benzolactam inhibitors and HSP90α. Based on the obtained results, we would unravel the structural basis for the effect of different R1 substitutes of benzolactam inhibitors on the inhibitory capability of HSP90α, which may be the benefit of developing the next-generation of potent HSP90α inhibitors.

Section snippets

Models setup

The 1.93 Å X-ray crystal structure of the HSP90α N-terminal domain in complex with the compound 31 was extracted from the RCSB Protein Data Bank (PDB ID: 4O0B) (Ernst et al., 2014). The chemical structures for the remaining eight compounds were constructed using the ACD/ChemSketch software and subsequently energy-minimized using the B3LYP/6-31G* as implemented in the Gaussian03 program.

Molecular docking

The compound 31 was deleted from the ATP-binding site of HSP90α and the resulting unbound structure was used in

Results and discussion

HSP90 function is associated with ATP that binds to the N-terminal domain and an intrinsic ATPase activity is essential for the operation of a functional “chaperone” cycle, which contributes to stabilize client proteins (Picard, 2002). Fig. 2 shows the HSP90α N-terminal domain of (residues: 16-224) in complex with the compound 31 in the ATP-binding site. The N-terminal domain of HSP90α consists of eight anti-parallel β-sheets and seven α-helices. The inhibitor binding site is formed by helices

Conclusions

In the present study, molecular docking, MD simulations, MM–GBSA binding free energy calculations, and decomposition of binding energy were performed to unravel the binding modes and the different binding abilities of the nine benzolactam inhibitors to the ATP-binding site of HSP90α N-terminal domain. The MM–GBSA calculations showed that the hydrophobic interactions between the nine compounds and HSP90α contributed the most to the binding affinity. Moreover, a good linear correlation was

Acknowledgement

The authors thank Prof. Y Jiang at Zhengzhou University for proving the computational resources.

References (33)

  • P.G. Blachly et al.

    Utilizing a dynamical description of IspH to aid in the development of novel antimicrobial drugs

    PLoS Comput. Biol.

    (2013)
  • D.A. Case et al.

    AMBER 11

    (2010)
  • T. Darden et al.

    Particle mesh Ewald: an N log (N) method for Ewald sums in large systems

    J. Chem. Phys.

    (1993)
  • Y. Duan et al.

    A point-charge force field for molecular mechanics simulations of proteins

    J. Comput. Chem.

    (2003)
  • J.T. Ernst et al.

    Identification of novel HSP90α/β isoform selective inhibitors using structure-based drug design. Demonstration of potential utility in treating CNS disorders such as huntington’s disease

    J. Med. Chem.

    (2014)
  • W.L. Jorgensen et al.

    Comparison of single potential function for simulating liquid water

    J. Chem. Phys.

    (1983)
  • Cited by (3)

    • Molecular dynamics simulations reveal the allosteric effect of F1174C resistance mutation to ceritinib in ALK-associated lung cancer

      2016, Computational Biology and Chemistry
      Citation Excerpt :

      The propagation of allosteric signals from allosteric to active sites leads to alteration of dynamics of active sites. In this context, all-atom molecular dynamics (MD) simulations with an explicit solvent can provide important details pertaining to the internal motions and resulting local and large-scale conformational changes (Sun et al., 2014; Guo et al., 2015a; Omotuyi and Ueda, 2015), which have proved capable of capturing allosteric communications between allosteric and active sites (Doyle et al., 2014; Neale et al., 2015; Lu et al., 2015a,b; Köhler et al., 2015; Guo et al., 2015b). In addition, the Molecular Mechanics/Generalized Born Surface Area (MM/GBSA) technology has been developed to predict the binding affinities between proteins and ligands (Yang et al., 2015; You et al., 2016; Frieg et al., 2016; Brender and Zhang, 2015).

    View full text