Comparative analysis of the dual EGFR-DNA targeting and growth inhibitory properties of 6-mono-alkylamino- and 6,6-dialkylaminoquinazoline-based type II combi-molecules

Highlights

• Development of the first synthetic strategy to access 6-N, N-disubstituted quinazolines.

• Comparison between di-substituted and mono-substituted EGFR-DNA combimolecules

• Hemi-mustard combi-molecules induce less DNA damage than full mustards.

• In contrast to mono-, di-substitution adversely affects EGFR binding affinity.

• In silico and X-Ray studies confirmed that steric bulk hinder EGFR binding.

Abstract

Over the past decade, we described a novel tumour targeted approach that sought to design “combi-molecules” to hit two distinct targets in tumour cells. Here, to generate small combi-molecules with strong DNA damaging potential while retaining EGFR inhibitory potency, we developed the first synthetic strategy to access the 6-N, N-disubstituted quinazoline scaffold and designed JS61 to possess a nitrogen mustard function directly attached to the 6-position of the quinazoline ring. We compared its biological activity with that of structures containing either a hemi mustard or a non-alkylating substituent. Surprisingly, the results showed that JS61, while capable of inducing strong DNA damage, exhibited moderate EGFR inhibitory potency. In contrast, “combi-molecules” with no bulky substituent at the N-6 position (e.g. ZR2002 and JS84) showed stronger EGFR and growth inhibitory potency than JS61 in a panel of lung cancer cells. To rationalize these results, X-ray crystallography and molecular modeling studies were undertaken, and the data obtained indicated that bulkiness of the 6-N,N-disubstituted moieties hinder its binding to the ATP site and affects binding reversibility.

Introduction

The epidermal growth factor receptor (EGFR) is known to be overexpressed in a wide variety of human cancers including non-small cell lung carcinoma (NSCLC) for which this overexpression has been associated with poor prognosis [[1], [2], [3], [4]]. Significant advances in the biological understanding of EGFR expression and its mutational status have led to the design of several inhibitors. A large variety of first-class EGFR inhibitors have been developed such as gefitinib (1), which has been used in advanced NSCLC patients (Fig. 1) [[5], [6], [7]]. However, acquired resistance to these inhibitors is frequently developed, leading to the design of second and third generation of EGFR inhibitors. These include afatinib (2) (2nd generation) and osimertinib (3) (3rd generation) [[8], [9], [10]]. However, acquired resistance still arises rapidly, generally over a period of 9–13 months [[11], [12], [13]].

Another approach to enhance the potency of EGFR inhibitor-based treatment consists of its combination with cytotoxic agents. Indeed, several studies showed that the combination of an inhibitor of EGFR and a DNA damaging agent induced either additive or synergistic growth inhibitory effects [14]. However, one of the major challenges of combination therapy is the risk of additive toxicity associated with the deleterious effects of each individual drug. To circumvent these problems, we developed a novel drug design approach termed “combi-targeting” that seeks to reduce the pharmacotoxicity of multiple antitumour mechanisms to that of a single drug. The proof-of-concept of this approach has already been confirmed in our laboratory [[15], [16], [17], [18], [19], [20], [21], [22], [23], [24]]. We classified these “combi-molecules” in different types according to their mechanism of action. As depicted in Fig. 2, type I combi-molecule (EGFR-DNA) is designed as a single agent that, intact, can act exclusively on EGFR. Hydrolysis under physiological conditions allows the activation of the second targeting arm and the subsequent release of both bioactive species, whereupon they can act independently on both their targets EGFR and DNA. Type II combi-molecules are systems that cannot undergo hydrolysis and the single molecule can act only on one of its respective targets. Finally, a type III combi-molecule; AL776, has been designed which acts as a hybrid between type I and type II, but is not described here [25]. Regardless of the combi-molecule type, these combi-molecules have shown superior potency and/or reduced toxicity when compared with traditional combinations that compose these molecules, indicating the potential therapeutic benefit of the whole being greater than the sum of its parts. However, the synthesis of multifunctional molecular systems capable of reacting with or binding to the target, and ultimately overcoming drug resistance is a real challenge. Indeed, these systems must maintain strong binding affinity for their desired targets, have sufficient solubility for biological applications and be of appropriate size for diffusion through the cell membrane.

ZR2002 (5) was the first type II combi-molecule designed to inhibit EGFR while inducing DNA damage through its 2-chloroethylamino moiety (Fig. 1, compound 5) [23]. Previous work on this molecule has shown that it can target either EGFR with its inhibitory scaffold or DNA with its 2-chloroethyl function, however a single molecule cannot effect both at the same time (Fig. 2). We have recently achieved a key step in its pre-clinical development by demonstrating that ZR2002 was able to increase overall survival of mice intracranially injected with glioblastoma mesenchymal stem cells harboring temozolomide resistance [26].

In order to enhance the lipophilicity of the drug as well as DNA alkylating potential, we designed and synthesized JS61 (Chart 1), a type II combi-molecule carrying an additional chloroethyl group on the nitrogen, leading to a highly reactive aromatic nitrogen mustard moiety. We and others have already studied N-mustard-quinazoline molecules (Fig. 1, compound 6) [21,[27], [28], [29]]. However, these molecules were only achieved through coupling with a linker between the nitrogen mustard and the quinazoline moieties. Furthermore, the addition of a lipophilic chloroethyl group was expected to enhance its lipophilicity without significantly increasing its molecular weight (<500 MW), a property that may be attractive for penetration through the blood brain barrier and potential indication for the therapy of advanced brain tumours.

Herein, for the first time, the synthesis and the biological activity of the first prototype of 6,6-dialkylaminoquinazoline JS61, a combi-molecule with a nitrogen mustard directly grafted on the quinazoline scaffold, is described. Further, we challenged its activity by replacing the chloroethyl group(s) with various non alkylating moieties such as methyl (JS84), hydroxyethyl and acetoxyethyl (Chart 1) and compared their EGFR binding affinity and growth inhibitory potencies with those of ZR2002. To rationalize the results, we performed molecular modeling and X-ray diffraction analyses of the N-disubstituted combi-molecules JS61, JS84 and the N-monosubstituted ZR2002.