Protein-Protein Interactions in Transcription: A Fertile Ground for Helix Mimetics

Abstract

Designed ligands that inhibit protein-protein interactions involved in gene expression are valuable as reagents for genomics research and as leads for drug discovery efforts. Selective modulation of protein-protein interactions has proven to be a daunting task for synthetic ligands; however, the last decade has seen significant advances in inhibitor design, especially for helical protein interfaces. This review discusses examples of transcriptional complexes targeted by designer helices.

Introduction

Transcription factors are involved in an intricate web of interactions with partner proteins and promoter DNA, which result in recruitment of chromatin-remodeling enzymes and assembly of the preinitiation complex.¹ Because of the essential role gene expression plays in the progression of diseases,² synthetic agents that modulate transcription in a defined manner are attractive candidates for drug design (Figure 1).^3-8 Artificial DNA binding ligands, such as pyrrole-imidazole polyamides,^9-11 triplex forming oligos,^12,13 peptide nucleic acids,^14-16 and zinc-finger proteins^17,18 have been shown to efficiently inhibit the transcription factor–DNA interface and control transcription of targeted genes. In contrast to the success in targeting DNA with specific ligands, selective inhibition of protein-protein contacts remains a formidable challenge.^19,20

Figure 1.

The DNA-protein and protein-protein interactions provide attractive targets for the design of inhibitors and activators of gene expression. (i) Programmable sequence-specific DNA binding ligands, such as pyrrole-imidazole polyamides, represent a successful class of synthetic modulators of transcription. (ii) Emerging strategies for targeting protein-protein interactions are offering new approaches for the design of transcription regulators.

A broad effort for the construction of specific ligands for protein targets has focused on a rational-design approach that seeks to adapt protein recognition principles utilized by nature. These efforts recognize that protein secondary structures play key roles in the interaction of proteins with other biomolecules, and that stable mimics of these secondary structures may potentially provide new classes of ligands.^21-27 The distinguishing element of this approach is that it offers medium-sized molecules with greater number of contacts to target protein surfaces with higher specificity, than traditional small molecules. α-Helices constitute the largest class of protein secondary structures and play a major role in mediating protein-protein interactions. Over the last decade, several classes of synthetic helix mimetics have shown selective inhibition of chosen protein-protein interactions.

A recent analysis of multiprotein complexes in the Protein Data Bank (PDB) indicates that roughly 62% of all complexes feature helical interfaces and, of these, 20% participate in gene regulation.^28,29 These statistics suggest that inhibitors of helical protein interactions can potentially become a broadly useful class of transcription regulators. Table 1 shows several examples of helical protein-protein interactions involved in transcriptional cascades that have been targeted by small molecules and helix mimetics. Entry 1 depicts the interaction of the activation domain of p53 with Murine Double Minute (MDM2). Tumor suppressor protein p53, generally considered the guardian of the genome, plays a fundamental role in apoptotic signaling and cell cycle arrest.³⁰ In response to DNA damage or cellular stress, phosphorylation of p53 signals for the expression of genes that activate apoptosis and prevent proliferation. MDM2 and the human homolog (HDM2) have been shown to bind the activation domain (AD) of p53 and repress its activity; accordingly, the p53/MDM2 interaction has become a target for drug discovery.³¹ The p53 AD adopts an α-helical conformation when bound to MDM2,³² and several classes of stabilized helices and helix mimetics, discussed below, have been shown to target this interaction.^33-40

Table 1.

Helical protein-protein interactions that mediate transcription.

Entry	Complex	PDB code	Description
1		1ycr	p53 activation domain (orange) is a transcriptional activator and tumor suppressor.
			MDM2 (cyan) is an E3 ubiquitin ligase and cellular inhibitor of p53
2		1l8c	Hif-1α (orange) is a transcriptional activator induced by hypoxia.
			CBP (cyan) is a coactivator that regulates gene expression of many cellular functions such as growth and differentiation, modifys chromatin through acetyltransferase activity.
3		1kdx	CREB pKID domain (orange) is a transcriptional activator.
			CBP KIX domain (cyan) is a coactivator that regulates expression of many genes involved in cellular functions such as growth and differentiation.
4		3erd	Peptide from the NR box II region of coactivator GRIP1 (orange) is essential for regulation of estrogen receptor mediated genes.
			Human estrogen receptor α-ligand binding domain (cyan) is a transcription factor that mediates estrogen related gene expression
5		2f8x	Ankyrin domain of Notch1 (green) is a transcription activator that transduces signals involved in development.
			MAML-1 (orange) is a polypeptide coactivator important for the transcriptional machinery.
			CSL (cyan) is a transcription factor involved in spatial and temporal activation of transcription.

Entry 2 highlights the structure of the C-terminal activation domain of hypoxia-induced transcription factor, HIF-1α, and the cysteine-histidine rich 1 (CH1) region of CBP/p300. In response to a low level of tissue oxygen, HIF-1 activates the expression of genes involved in glucose metabolism, cell proliferation, and vascularization. Particularly, the Hif-1α subunit contains the oxygen-sensing and transactivating domains and subsequently is found overexpressed in most human cancers. CREB binding protein (CBP), and the related protein p300, is a transcriptional coactivator that integrates multiple signal transduction pathways and serves as a scaffold for the recruitment of other transcription factors by modifying chromatin. Essential for basic cellular functions, CBP/p300 is implicated in cell growth and differentiation and plays a critical role in embryonic development. This interaction has been targeted by small molecules^4,5,41-43 and stabilized helices.⁴⁴ The interaction of Hif-1α with its cognate promoter, and transcription of hypoxia inducible genes, has also been successfully controlled by polyamides developed by Dervan and coworkers.^45,46

Entry 3 shows the complex between the phosphorylated kinase inducible activation (pKID) domain of cAMP-regulated transcription factor CREB and the KIX region of CBP.⁴⁷ CREB proteins are involved in neuronal function and plasticity and have been implicated in both Huntington's Disease and cancer. pKID adopts a helical conformation when bound to CBP.⁴⁸ Entry 4 is an example of ligand-activated transcription of estrogen-activated genes by the conserved estrogen receptor (ER).⁴⁹ Binding of the GRIP1 coactivator peptide through a conserved LXXLL, where L is leucine and X is any amino acid residue, motif to the ER ligand binding domain leads to specific interactions of the receptor with DNA. The resulting transcription activation and expression of estrogen-activated genes is associated with a variety of diseases including breast cancer, osteoporosis, and cardiovascular disease.

Entry 5 depicts the complex of Notch with CSL and coactivator protein of the mastermind-like family, MAML1.⁵⁰ Notch is a transmembrane receptor that is displayed at cell surfaces and binds to protein ligands on the surfaces of adjacent cells. Ligand binding to the extracellular domain of Notch triggers intramembrane cleavage of the receptor which enters the cell nucleus, docks with the DNA-bound transcription factor CSL and recruits MAML1 to initiate expression of Notch-targeted genes. Inappropriate expression of the Notch-targeted genes has been implicated in several human diseases, including cancers of the lung, ovary, pancreas and T-cell acute lymphoblastic leukemia.⁵¹

Recent reviews have extensively examined the mechanisms of transcription initiation and the strategies employed in the regulation of gene expression by synthetic ligands.^3-8,52,53 In this report we focus on helix design strategies to illustrate their exciting potential as transcription regulators and as leads for drug design.

α-Helix Mimetics

The overall aim of stabilizing or mimicking α-helical conformation is to endow peptidic and nonpeptidic oligomers with conformational rigidity, proteolytic stability, and the desired array of protein-like functionality. Several approaches have bee described to realize these aims. These approaches can be divided into three general categories: helix stabilization, helical foldamers and helical surface mimetics. Helix stabilizing methods based on side chain crosslinks and hydrogen-bond surrogates preorganize amino acid residues and initiate helix formation; miniproteins that display helical domains also fall under this category. Helical foldamers, such as beta-peptides and peptoids, are composed of amino acid analogs and are capable of adopting conformations similar to those found in natural proteins. Helical surface mimetics utilize conformationally restricted scaffolds with attached functional groups that resemble the i, i+4, i+7 pattern of side chain positioning along the face of an α-helix.

Stabilized α-Helices

Side chain crosslinked helices

The α-helix features 3.6 residues per complete turn, which places the i, i+4, i+7, and i+11 side chains on the same face of the folded structure. The classical strategy to stabilize the α-helical conformation in peptides employs covalent bonds between the i and i+4 or i and i+7 side chain groups (Figure 2). Earliest side chain crosslinks utilized lactam, disulfide and metal-mediated bridges.^54-63 Helices containing lactam-bridges and disulfide links have succesfully targeted their intended protein receptors.^59,64-69 Grubbs and Verdine have described hydrocarbon bridged side chain crosslinked helices obtained from olefin metathesis reactions.^70,71 Verdine and coworkers utilized hydrocarbon stapled helices to target the interaction between the activation domain of p53 and its negative regulator HDM2, demonstrating upregulation of p53 expression and activation of the apoptotic signal.³⁴ Recently, this group also demonstrated that stapled helices obtained from the wild-type sequence of the coactivator competitively inhibits MAML1 binding to transcription factors CSL–Notch^ICN, and represses Notch-mediated gene expression.⁷²

Figure 2.

Side chain crosslinks and hydrogen bond surrogates provide two successful methods for stabilizing the helical conformation in peptides. Crosslinking of residues on the same face of the helix enhances helix stability. Hydrogen bond surrogate (HBS) helices feature a covalent bond in place of the intramolecular hydrogen bond to initiate helix formation. Green circles represent amino acid side chain functionality.

HBS helices

An alternative strategy for the stabilization of α-helices involves replacement of one of the main chain intramolecular hydrogen bonds with a covalent linkage (Figure 2).⁷³ The hydrogen bond surrogate (HBS) approach is based on the helix-coil transition theory in peptides, which suggests that the energetically demanding organization of three consecutive amino acids into the helical orientation inherently limits the stability of short α-helices. The HBS strategy affords preorganized α-turns to overcome this intrinsic nucleation barrier and initiate helix formation.⁷⁴ The preorganized α-turns are obtained by replacing the N-terminal main chain hydrogen bond between the C=O of the i^th amino acid residue and the NH of the i+4^th amino acid residue with a carbon-carbon bond through a ring-closing metathesis reaction.⁷⁵ HBS helices have been shown to target their expected protein partners with high affinity in cell-free and cell culture assays.^44,76,77 In a recent study Olenyuk, Arora and coworkers showed that HBS helices that mimic a helical segment in the C-terminal activation domain of HIF-1α can bind to the CH1 region of CBP/p300 and inhibit transcription of hypoxia inducible genes in cell culture.⁴⁴ Hypoxia inducible genes encode vascular endothelial growth factor (VEGF) and its receptor VEGFR2, which are involved in the induction of new blood vessels (angiogenesis) in solid tumors.⁵ This work suggests that designed ligands that inhibit hypoxia-inducible gene expression could aid drug discovery efforts for the treatment of neovascularization in cancers.

Together the studies with side chain crosslinked and HBS helices highlight the promise of stabilized helices to target gene-specific transcription factors.

Miniproteins

Well-folded miniature proteins that present solvent-exposed helices can serve as templates for the development of novel ligands for transcription factors and cofactor protein interactions.²⁶ Schepartz and coworkers have engineered avian pancreatic polypeptide (aPP, Figure 3) scaffolds with the recognition epitope from CREB KID transcription activation domain to target the KIX domain of coactivator CBP with high affinity.^78,79 Fusion of the CREB miniature protein with a heterologous DNA-binding domain provided artificial ligands that activated transcription through the CBP/p300 pathway.⁷⁹ An attractive feature of miniprotein scaffolds is that they may be diversified using common protein evolution strategies, such as phage display. Several miniproteins including aPP, and α-helical scaffolds derived from scorpion toxin and apamin, have been designed to target the interaction between the activation domain of p53 and HDM2.^80-83

Figure 3.

Miniature proteins that display stable helical folds and regulate transcription: (a) avian pancreatic protein (PDB code: 1ppt), (b) scorpion toxin miniprotein (PDB code: 1r1g), and (c) apamin (PDB code: 3iux).

β-Peptide Foldamers

β-Peptide foldamers are composed of β-amino acid residues and are capable of adopting conformations reminiscent of α-helices (Figure 4).^21,22,84-86 The salient feature of these nonpeptidic oligomers is that they display protein like functionality to target biomolecular receptors while resisting proteolytic degradation.⁸⁵ Recent studies have shown that foldamers composed of heterogeneous backbones consisting of α- and β-amino acid residues are also capable of adopting diverse sets of stable conformations.⁸⁷ Pioneering studies from the Gellman, Schepartz and Seebach groups have demonstrated the vast potential of these nonnatural oligomers to pursue a variety of protein targets including those involved in transcription.^88-90 Cell-penetrating β-peptides that mimic p53 activation domain were shown to upregulate cellular levels of p53, p21 and HDM2.⁹¹ The Gellman and Schepartz groups have utilized combinatorial chemistry strategies to generate diverse sets β-peptide ligands for protein targets.^92,93

Figure 4.

α-Helix and β-peptide foldamers array side chain functionality in similar fashion: (a) primary and (b) secondary structures of α- and β-peptide helices (Cambridge Structural Database accession code 633286).

Helical Surface Mimetics

In many proteins complexes, one face of the α-helix featuring the i, i+4, and i+7 residues participates in protein-protein interactions. Helical surface mimetics take advantage of this feature by capturing the functionality of the primary face of the helix on a nonpeptidic scaffold. Hamilton and coworkers pioneered the development of helical surface mimetics with terphenyl and related scaffolds (Figure 5).^94,95 Molecular modeling and crystal structures suggest that these scaffolds project protein-like functionality in a manner reminiscent of the i, i+4 (or i+3), and i+7 positions of a canonical α-helix. Terphenyl derivatives displaying key p53 binding residues were able to selectively inhibit p53/HDM2 interaction in vitro with high affinity.⁹⁶ The same group also demonstrated that pyridylpyridone derivatives can effectively mimic the conserved nuclear receptor box motif, LXXLL, and target the interaction of estrogen receptor and its coactivator responsible for the expression of estrogen-activated genes.^49,97

Figure 5.

Nonpeptidic helix mimetics such as (b) terphenyls and (c) pyridylpyridone derivatives array protein like functionality to mimic their arrangement on an α-helix.

Conclusion

The past decade has seen noteworthy progress in the design of helix mimetics and their application in targeting protein-protein interactions, which are difficult targets for traditional small molecules. Transcription factors present unique opportunities and challenges for inhibitor design. The inappropriate activation of these key regulators of cell fate is linked to the etiology of numerous genetic diseases, making them attractive drug targets. However, with the exception of ligand-activated nuclear hormone receptors, transcription factors have been difficult targets for synthetic ligands. The success of helix mimetics in targeting these recalcitrant proteins suggests that an improved understanding of the recognition principles underlying protein–protein interactions may allow rational design of target-specific protein–interface mimetics.