Quinoline-Based Molecules Targeting c-Met, EGF, and VEGF Receptors and the Proteins Involved in Related Carcinogenic Pathways.

The quinoline ring system has long been known as a versatile nucleus in the design and synthesis of biologically active compounds. Currently, more than one hundred quinoline compounds have been approved in therapy as antimicrobial, local anaesthetic, antipsychotic, and anticancer drugs. In drug discovery, indeed, over the last few years, an increase in the publication of papers and patents about quinoline derivatives possessing antiproliferative properties has been observed. This trend can be justified by the versatility and accessibility of the quinoline scaffold, from which new derivatives can be easily designed and synthesized. Within the numerous quinoline small molecules developed as antiproliferative drugs, this review is focused on compounds effective on c-Met, VEGF (vascular endothelial growth factor), and EGF (epidermal growth factor) receptors, pivotal targets for the activation of important carcinogenic pathways (Ras/Raf/MEK and PI3K/AkT/mTOR). These signalling cascades are closely connected and regulate the survival processes in the cell, such as proliferation, apoptosis, differentiation, and angiogenesis. The antiproliferative biological data of remarkable quinoline compounds have been analysed, confirming the pivotal importance of this ring system in the efficacy of several approved drugs. Furthermore, in view of an SAR (structure-activity relationship) study, the most recurrent ligand-protein interactions of the reviewed molecules are summarized.

1. Introduction

pan class="Chemical">Quinoline 1 is an imclass="Chemical">portant and versatile nucleus recurrent in several natural and class="Chemical">pharmacologicalclass="Chemical">pan class="Chemical">ly active molecules. Due to the presence of nitrogen, which withdraws electrons by resonance, this fused aromatic scaffold is an electron-deficient ring system (Figure 1), displaying weak tertiary base properties. Quinoline reacts with similar behaviour of pyridine and allows both nucleophilic and electrophilic substitution reactions. From a biological point of view, the nontoxic effect to humans, via inhalation and oral absorption, makes the 1-aza-naphthalene an interesting scaffold to be studied for the development of more selective drugs [1].

Figure 1

(a) Quinoline 1 or 1-aza-naphthalene and benzo[b]pyridine; (b) quinoline 3D structure; (c) quinoline electron density; (d) electron density encoded with the electrostatic potential.

pan class="Chemical">Quinoline and its related derivatives exhibit a broad sclass="Chemical">pectrum of class="Chemical">pharmacological activities from antibacterial, antifungal, antimalarial, anthelmintic, local anaesthetic, anticlass="Chemical">psychotic, and anticlass="Chemical">pan class="Disease">cancer [2]. Especially in the research of new anticancer agents, quinoline is one of the most important scaffolds in drug discovery. Indeed, over the last few years, an increase in the publication of papers about quinoline derivatives possessing antiproliferative properties has been observed. This trend can be justified by the “druggability”, versatility, and accessibility of the quinoline nucleus [3].

Furthermore, different pan class="Chemical">quinoline small molecules, acting class="Chemical">pan class="Chemical">as protein kinases inhibitors, have been approved by the Food and Drug Administration (FDA) for clinical uses [4,5]. In this contest, bosutinib, a potent inhibitor of Abl, was approved in 2012 for the treatment of Philadelphia chromosome-positive chronic myelogenous leukaemia (CML) [6]; lenvatinib, an inhibitor of RET and VEGFR has been employed, since 2015, for the treatment of differentiated thyroid cancers [7]; and neratinib and cabozantinib were authorized for the clinical treatment respectively in 2017 and 2012 [8,9].

Due to the importance of pan class="Chemical">quinoline nucleus in the medicinal chemistry field, the main class="Chemical">purclass="Chemical">pose of this review is to class="Chemical">provide an overview of the most relevant anticlass="Chemical">pan class="Disease">cancer quinoline molecules described in the literature, in the last decade. In particular, the attention has been focused on quinoline drugs and experimental active derivatives, able to interfere with c-Met (mesenchymal-epithelial transition factor), VEGF, and EGF receptors and the correlated proteins of the intracellular signalling pathways. The interest on these targets rises from their crucial involvement in several carcinogenic degenerations. In fact, closely connected by each other, they regulate the survival mechanisms in the cell, such as proliferation, apoptosis, differentiation, and angiogenesis.

In this light, the evaluation of the antiprolifepan class="Species">rative data and the anaclass="Chemical">pan class="Chemical">lysis of most frequent ligand–protein interactions of quinoline-based compounds could support, in a drug discovery process, the search for new and more efficacious multi-targets drugs.

EGFR, c-MET, and VEGFR Signal Pathways in Cancer Cell

Aberrant pan class="Gene">receptor tyrosine kinases (RTKs) are deeclass="Chemical">pclass="Chemical">pan class="Chemical">ly involved in cancer progression, and for this reason represent the key oncology therapeutic targets. Among the most studied, in this review the attention has been focused on three crucial growth factor receptors: c-Met, EGFR, and VEGFR [10].

pan class="Gene">C-Met is a class="Chemical">pan class="Gene">tyrosine kinase receptor belonging to the MET (MNNG HOS transforming gene) family, and is normally expressed on the surfaces of liver, pancreas, prostate, kidney, muscle, and bone marrow cells, where mediates tissue regeneration and wound healing [11,12]. The endogenous ligand of c-Met is HGF/SF (ligand hepatocyte growth factor/scatter factor), that induces signal transduction from the extracellular to the intracellular domain [13]. The kinase catalytic activity triggers the transphosphorylation of the tyrosines Tyr1234 and Tyr1235, turning on a whole signalling cascade of biological pathways correlated to the cellular growth and replication [14,15].

pan class="Gene">C-Met is strictclass="Chemical">pan class="Chemical">ly connected with the phosphoinositide 3-kinase (PI3K) a cytosolic family of kinases responsively to phosphorylate the 3’-hydroxyl group of phosphoinositides [16], (Figure 2). After activation by growth factors and/or cytokines, PI3K is able to turn on downstream proteins, such as AkT, a serine-threonine kinase, that consequently controls stimulation of mTOR complex 1 (mTORC1). AkT is responsible for the handling of the cell cycle, through genes transcription activation. Furthermore, another component of these pathways is mTOR complex 2 (mTORC2), that determinates activation of AkT by its phosphorylation. It has been studied that this signalling pathway is upregulated in many cancers and apparently drives the resistance to the typical cytotoxic anticancer drugs. For these reasons PI3K/AkT/mTOR appear attractive targets in cancer targeted therapy [17,18,19].

Figure 2

Cross-talk between EGFR, VEGFR, and c-Met signalling pathways; in boxes are reported some quinoline based inhibitors developed up to date.

pan class="Gene">C-Met intracellular signalling cclass="Chemical">pan class="Chemical">ascade is also correlated to the carcinogenic pathways stimulated by EGFR and VEGFR (Figure 2). The equilibrium in the activation of these three growth factor receptors is at the basis of normal cell survival processes, such as proliferation, apoptosis, differentiation, and angiogenesis [11,20].

pan class="Gene">EGFR belong to a famiclass="Chemical">pan class="Chemical">ly of tyrosine kinases receptors and include four members: EGFR (ErbB1), ErbB2, ErbB3, and ErbB4. These receptors are widely distributed in mammalian epithelial cell membranes and normally modulate different processes such as cell proliferation, death, and differentiation. It has been detected that an anomalous over-expression of these proteins has an important role in the formation and development of many types of solid tumours, principally Non-Small Cell Lung (NSCLC), breast, colorectal, head, and neck cancers types [21,22,23].

pan class="Gene">VEGFR can be secclass="Chemical">pan class="Gene">reted by tumour cells and, through its interaction with one of the three tyrosine kinase receptors (VEGFR1-3), it is able to promotes vascular growth and permeability, with an antiapoptotic effect on newly formed cells [24,25].

pan class="Gene">EGFR and/or class="Chemical">pan class="Gene">VEGFR recruits Rat sarcoma (RAS) protein family, with consequent activation of Raf/MEK/ERK and of PI3K/AkT/mTOR signalling cascade, directly activated also by c-Met [16,20].

With respect to the aforementioned intracellular effectors, Rpan class="Chemical">as, a membrane-class="Chemical">pan class="Chemical">associated guanine nucleotide-binding protein, plays a central role in regulating the cytoplasmatic transduction cascades, especially those of Raf/MEK and PI3K, leading to cell growth (Figure 2) [26,27,28].

2. Quinolines as Inhibitors of Carcinogenic Pathways

2.1. Quinoline Based Inhibitors of c-Met Receptor

An altepan class="Species">ration of the class="Chemical">pan class="Gene">HGF/c-Met signalling has been detected in liver, pancreatic, breast, ovarian, gastric, and prostate cancers and it contributes to the development, progression, dissociation of cells from its primary site, and distant colonization. Therefore, c-Met is considered an attractive target for cancer therapy [11,12]. Currently, a lot of small molecules with a quinoline nucleus, classified as c-Met inhibitors, have been approved for therapeutic treatments. Among these, we could mention cabozantinib, a multikinase inhibitor administered in the treatment of advanced medullary thyroid cancers [8,29], and foretinib, a drug candidate under experimental evaluation (Figure 3) [3].

Figure 3

C-Met quinoline inhibitors: cabozantinib and foretinib (in boxes structure analogies are highlighted).

From a molecular point of view, all the pan class="Chemical">quinoline-bclass="Chemical">pan class="Chemical">ased inhibitors of c-Met interact with the kinase domain of the receptor in the cytosolic side, involving amino acids residues from 1078 to 1345 [13,30]. As an example, in Figure 4a, the crystal structure of c-Met kinase domain in complex with foretinib (PDB id: 3LQ8) is reported: the ligand binds to c-Met in an extended conformation occupying the ATP-binding site and the adjacent deep hydrophobic pocket. The cyclopropane-1,1-dicarboxiamide framework interferes with the Phe1223 of the DFG motif, determining the disruption of the catalytic conformation (from DFG-in to DFG out) and the reorganization of the activation loop, which almost entirely envelops the ligand (Figure 4a). In the cavity the reoriented Phe1223 forms a π-π stacking with the central fluorophenyl ring, the quinoline nitrogen links with a hydrogen bond the backbone amide of Met1160 and also the dicarboxamide moiety forms additional hydrogen bonds with Lys1110 and Asp1222 (Figure 4b) [30,31,32].

Figure 4

(a) crystal structure of kinase domain of c-Met in complex with foretinib (PDB ID: 3LQ8). In the box the binding cavity of the receptor occupied by the quinoline small molecule is illustrated (the activation loop is represented with transparent surface) [31]; (b) interactions of foretinib with amino acid residues of the c-Met active site [32].

Considering pan class="Chemical">cabozantinib and class="Chemical">pan class="Chemical">foretinib as lead compounds, for the design of new quinoline derivatives selectively acting on c-Met, SAR studies highlight the 4-phenoxyquinoline and an aromatic moiety, connected each other by a linker, as the fundamental pharmacophoric portions in the modulation of c-Met tyrosine kinase activity (Figure 5). Generally, the linker should have two important features: (1) the distance of five atoms (the so-called “five atoms regulation”) between the 4-phenoxyquinoline moiety and the aromatic one; (2) the presence of donor and/or acceptor hydrogen-bond groups and at least one amide. These general features are also shown in the structure of cabozantinib and foretinib, in which there is a cyclopropane-1,1-dicarboxiamide framework as linker [33].

Figure 5

General structure for quinoline c-Met inhibitors as analogues of cabozantinib and foretinib.

In the lpan class="Chemical">ast few years, a lot of more active and selective class="Chemical">pan class="Chemical">quinoline derivatives have been synthesized and biologically evaluated by modifying lead compounds (Table 1). The most representative modifications are: (1) at position C-7 of quinoline ring system, generally with the introduction of appropriate substituents to improve solubility, (2) at the linker portion [34], that has been modified by different cyclic/acyclic 5-atoms-analogues with similar electronic features: pyridazinone-3-carboxyamide 3 [35], 3-oxo-3,4-dihydroquinoxaline 4 [36], 1H-imidazole-4-carboxamide or (E)-3-hydrosulfonylacrylamide 5,6 [37], 1,2,3-triazole-4-carboxamide 7,8 [38,39], 2-imidazolone-4-carboxamide 9 [40], acylhydrazone moiety 10 [41], pyridine/pyrimidine-2-carboxyamide 11 [42], 2-phenylquinoline-4-carboxamide 12 [33], 4-oxo-1,4-dihydrocinnoline-3-carboxamide moiety 13 [43], acyclic semicarbazones 14 [44,45], acylthiourea moiety [46], l,2,4-triazine-3,5-dione 15 [47], 1,8-naphthyridin-2-one 16 [48], 2-oxo-1,2-dihydropiridine-3-carboxamide 17 [49], 4-oxo-1,4-dihydroquinoline-3-carboxamide 18 [50], 5-(aminomethylene)pyrimidine-2,4,6-trione moiety 19 [51], 4-oxo-3,4-dihydrophthalazine-1-carboxamide 20 [52], pyrazolone 21 [53], and (thio)semicarbazone 22 [54].

Table 1

Examples of some interesting quinoline based c-Met inhibitors and their corresponding biological activities.

Quinoline Compound	Linker	IC50 (μM) Sensitive Cell Lines	IC50 on c-Met (nM)	Ref.
	pyridazinone-3- carboxamide	A549 (IC50 = 0.003 ± 0.001)HepG2 (IC50 = 0.49 ± 0.003)MCF-7 (IC50 = 0.006 ± 0.001)	0.6	[35]
	3-oxo-3,4-dihydroquinoxaline	HT-29 (IC50 = 0.06 ± 0.001)A549 (IC50 = 0.050 ± 0.006)H460 (IC50 = 0.18 ± 0.01)MKN-45 (IC50 = 0.023 ± 0.005)U87 MG (IC50 = 0.66 ± 0.09)	0.9	[36]
	1H-imidazole-4-carboxamide	HT-29 (IC50 = 0.08 ± 0.02)MKN-45 (IC50 = 0.22 ± 0.03)A549 (IC50 = 0.07 ± 0.01)	1.1 ± 0.21	[37]
	(E)-3-hydrosulfonylacrylamide	HT-29 (IC50 = 0.15 ± 0.003)MKN-45 (IC50 = 0.28 ± 0.003)A549 (IC50 = 0.15 ± 0.05)	17 ± 0.20	[37]
	1,2,3-triazole-4-carboxamide	HT-29 (IC50 = 0.10 ± 0.003)H460 (IC50 = 0.18 ± 0.03)A549 (IC50 = 0.07 ± 0.01)MKN-45 (IC50 = 0.03 ± 0.01)	2.27	[38]
	1,2,3-triazole-4-carboxamide	A549 (IC50 = 0.08 ± 0.004)H460 (IC50 = 0.15 ± 0.03)HT-29 (IC50 = 0.12 ± 0.02)MKN-45 (IC50 = 0.021 ± 0.003)U87MG (IC50 = 0.85 ± 0.12)	1.04 ± 0.09	[39]
	2-imidazolone-4-carboxamide	A549 (IC50 = 0.25 ± 0.02)H460 (IC50 = 0.10 ± 0.008)HT-29 (IC50 = 0.086 ± 0.005)MKN-45 (IC50 = 0.014 ± 0.004)	1.42 ± 0.14	[40]
	acylhydrazone	HT-29 (IC50 = 0.15 ± 0.03)H460 (IC50 = 0.031 ± 0.008)MKN-45 (IC50 = 0.37 ± 0.08)A549 (IC50 = 0.080 ± 0.01)U87MG (IC50 = 0.29 ± 0.08)	1.86	[41]
	pyridine -2-carboxyamide	HT-29 (IC50 = 0.026 ± 0.003)H460 (IC50 = 0.037 ± 0.01)MKN-45 (IC50 = 0.073 ± 0.01)A549 (IC50 = 0.10 ± 0.02)U87MG (IC50 = 0.81 ± 0.19)	1.39	[42]
	2-phenylquinoline-4-carboxamide	H460 (IC50 = 0.011 ± 0.002)HT-29 (IC50 = 0.062 ± 0.01)MKN-45 (IC50 = 0.011 ± 0.003)U87MG (IC50 = 0.15 ± 0.03)SMMC-7721 (IC50 = 0.03 ± 0.004)	1.32	[33]
	4-oxo-1,4-dihydrocinnoline-3-carboxamide	A549 (IC50 = 0.035 ± 0.0042)H460(IC50 = 0.055 ± 0.010)HT29 (IC50 = 0.11 ± 0.016)MKN-45 (IC50 = 0.022 ± 0.0035)U87MG (IC50 = 0.35 ± 0.060)SMMC-7721 (IC50 = 0.25 ± 0.072)	0.59 ± 0.07	[43]
	acyclic semicarbazone	HT-29 (IC50 = 0.059 ± 0.0039)MKN-45 (IC50 = 0.016 ± 0.0012)A549 (IC50 = 0.0090 ± 0.0012)MDA-MB-231 (IC50 = 0.77 ± 0.064)	4.3	[44]
	l,2,4-triazine-3,5-dione	H460 (IC50 = 0.14)HT-29 (IC50 = 0.21)MKN-45 (IC50 = 0.066)SMMC-7721 (IC50 = 0.73)	6.3	[47]
	1,8-naphthyridin-2-one	HepG2 (IC50 = 0.23 ± 0.01)MCF-7 (IC50 = 0.42 ± 0.06)A549 (IC50 = 0.21 ± 0.02)	2.36	[48]
	2-oxo-1,2-dihydropiridine-3-carboxamide	MKN45 (IC50 = 0.039 ± 0.009)BaF3/TRP-Met ((IC50 = 0.019 ± 0.0088)	0.9 ± 0.1	[49]
	4-oxo-1,4-dihydroquinoline-3-carboxamide	H460 (IC50 = 0.075)HT-29 (IC50 = 0.051)MKN-45 (IC50 = 0.010)U87MG (IC50 = 0.53)SMMC-7721 (IC50 = 0.20)	1.35	[50]
	5-(aminomethylene) pyrimidine-2,4,6-trione moiety	HT-29 (IC50 = 0.13 ± 0.060)H460 (IC50 = 0.051 ± 0.0080)MKN-45 (IC50 = 0.057 ± 0.0040)A549 (IC50 = 0.072 ± 0.0060)U87MG (IC50 = 0.64 ± 0.045)	1.15	[51]
	4-oxo-3,4-dihydrophthalazine-1-carboxamide	H460 (IC50 = 0.055 ± 0.010)MKN-45 (IC50 = 0.071 ± 0.011)HT-29 (IC50 = 0.13 ± 0.016)MDA-MB-231 (IC50 = 0.43 ± 0.050)	1.63	[52]
	pyrazolone	HT-29 (IC50 = 0.14 ± 0.02)H460 (IC50 = 0.18 ± 0.03)A549 (IC50 = 0.09 ± 0.02)MKN-45 (IC50 = 0.03 ± 0.001)U87MG (IC50 = 1.06 ± 0.05)	2.20	[53]
	(thio)semicarbazones	A549 (IC50 = 0.33 ± 0.02)HT-29 (IC50 = 0.21 ± 0.02)MKN-45 (IC50 = 0.71 ± 0.08)MDA-MB-231 (IC50 = 1.2 ± 0.17)	8.92	[54]

In this regard, after treatment with interesting pan class="Chemical">quinoline comclass="Chemical">pounds, in Tclass="Chemical">pan class="Gene">able 1 are reported the IC50 values (concentration of the compound producing 50% of cell growth inhibition) evaluated against the more responsive human cancer cell lines and the IC50 values (concentration of the compound producing 50% of inhibition activity) against the c-Met receptor, (bold data show the values of target compounds lower than that of the positive control in the corresponding experiment).

In order to extrapolate important general information about the interactions between ligands and amino acid residues in the active site, the binding mode of the previous pan class="Chemical">cabozantinib/class="Chemical">pan class="Chemical">foretinib-like derivatives has been explored through molecular docking simulations, (Figure 6). These studies pointed out π-π stacking interactions between the quinoline ring and Tyr1159 and the formation of a hydrogen bond between the lone pair of the quinoline nitrogen and Met1160, playing a pivotal role in the stabilization of the ligand-protein binding. Other important interactions occur through hydrogen bonds between the O and N atoms of the different linker regions and amino acids residues as Asp1222 and Lys1110 [35,36,38,40,41,48,51].

Figure 6

Binding models with c-Met kinase domain of quinoline- based compounds (a) 3 [35]; (b) 4 [36]; (c) 7 [38]; (d) 9 [40]; (e) 10 [41]; (f) 16 [48]; (g) 19 [51].

In 2011, Wang and co-workers designed a relativepan class="Chemical">ly unexclass="Chemical">plored class="Chemical">pan class="Chemical">quinoline chemotype, identifying the lead compound 3,5,7-trisubstituted derivative (zgwatinib) and the quinoline 23 as novel anticancer agents with a potent and selective inhibition activity on c-Met (IC50 of 0.93 and 0.95 nM, respectively) (Figure 7a). In particular, compound 23 showed significant antiproliferative effects (IC50 in the range 1–5 μM) especially on cancer cell lines characterized by c-Met overexpression (ICMKN45, SNU-5 and H1993), exhibiting a promising PK profile and a remarkable in vivo tumour growth inhibition in human glioblastoma xenograft models [55]. The crystal structure of zgwatinib in complex with the kinase domain of c-Met was determined in a further study, to elucidate the binding mode of this class of compounds. In detail, quinoline nitrogen forms an H-bond with Met1160 in the hinge region of the kinase, the trifluoromethyl and the benzylamino groups occupy the hydrophobic pocket, and the nitro substituent forms H-bond with Asp1222 (Figure 7b) [56].

Figure 7

(a) structures of a new cluster of 3,5,7-trisubstituted quinoline compounds with promising anticancer activity on c-Met [55,57]; (b) crystal structure of zgwatinib in complex with the kinase domain of c-Met (residues coloured in cyan form hydrophobic interactions with the ligand, whereas those coloured in yellow formed H-bonds, indicated with black dash lines; PDB id: 4GG5), [56].

Nevertheless, further toxicology studies of both derivatives demonstpan class="Species">rated considerclass="Chemical">pan class="Gene">able cardiovascular safety risk in normal rats, arising from significant inhibition of hERG K+ channels (IC50 in the range 37–463 nM). For this reason, in the attempt to reduce the off-target effect, the same researcher group developed a new set of quinoline molecules, among which the compounds 24 and 25 were identified as the most active, (Figure 7a). The modification was directed towards the C-3 piperazinyl portion with the aim of improve the PK profile and reduce the hERG activity without influencing the key interactions. In detail, bearing a Michael acceptor moiety to the N-4 piperazinyl ring, these new quinoline compounds showed c-Met IC50 values comparable to the parent molecules (2.3 and 2.6 nM respectively for 24 and 25), and an appreciable lower activity against hERG [57].

Nishii and co-workers explored modifications at the C-3 and C-6 positions of pan class="Chemical">quinoline scaffold. class="Chemical">pan class="Chemical">3,6-disubstituted quinoline 26 (Figure 8), shows an inhibition effect selectivity against c-Met kinase among more than 20 kinases with IC50 = 9.3 nM, and a potent antiproliferative activity against MKN45 cancer cell line (IC50 = 0.093 μM), [58].

Figure 8

4,6,7-substitued quinoline derivatives selectively active on c-Met [58,59].

In 2019, focusing the attention on C-6 substitution, Lien and co-workers designed and evaluated several pan class="Chemical">4,6,7-substituted quinolines, analogues of class="Chemical">pan class="Chemical">cabozantinib (Figure 8) [59]. In vitro antiproliferative assays highlight that derivatives 27 and 28 are more active than the parent compound on leukaemia, CNS, and breast cancer cell lines. The c-Met kinase inhibition assay denotes higher, or comparable activity than cabozantinib, with IC50 values of 19 and 64 nM respectively for 27 and 28 (IC50 cabozantinib = 40 nM). On this basis, from the point of view of structure–activity relationships, these derivatives demonstrated well fittings in the binding pocket of c-Met. Indeed, docking studies (Figure 9) demonstrated that, besides the interactions with crucial amino acids residues, above observed, these compounds could engage, through the p-aminophenyl moiety at C-6, additional hydrogen bonding with Ala1226 [59].

Figure 9

Main interactions of compound 28 with the amino acid residues in c-Met active site [59].

Different studies on hybrid derivatives, with a pan class="Chemical">quinoline scaffold linked at the C-6 class="Chemical">position to another heterocyclic system, class="Chemical">pan class="Chemical">as triazolo-pyrazine or imidazo-pyrazine, demonstrate inhibition activity on c-Met and antiproliferative effects unless the 4-phenoxy substitution typical of the cabozantinib/foretinib-like compounds. In detail, Cui and co-workers described anticancer properties of 2-(4-(1-(quinolin-6-ylmethyl)-1H-[1,2,3]triazolo[4–b]pyrazin-6-yl)-1H-pyrazol-1-yl)-ethanol methanesulfonate (PF-04217903), (Figure 10), an extremely potent and selective c-Met inhibitor with good oral bioavailability and an acceptable safety profile in preclinical trials [1,2,3]triazolo[4,5-b]pyrazin-6-yl)-1H-pyrazol-1 -yl)ethanol (PF-04217903) for the treatment of cancer. J. Med. Chem.. 2012 ">60].

Figure 10

Structures of some hybrid quinoline derivatives as c-Met inhibitors.

After SGX523 and pan class="Chemical">JNJ-38877605 clinical failure [61,62,63], Zhang and co-authors synthetized class="Chemical">pan class="Chemical">3-N-substituted quinoline triazolopyridine analogues to understand the electron-donating and steric hindrance effects on the metabolic activity of the aldehyde oxidase [64].

Furthermore, Zhao et al. have conducted some studies on both pan class="Chemical">triazolo[4,5-b]pyrazine and class="Chemical">pan class="Chemical">imidazo[4,5-b]pyrazine quinoline derivatives 29, 30 (Figure 10), showing significant inhibitory effects on both enzymatic and cellular assay, as well as in vivo selectivity and metabolic stability [65,66].

2.2. Quinoline as Inhibitor of PI3k/AkT/mTOR Pathway

A lot of pan class="Chemical">quinoline derivatives, bclass="Chemical">pan class="Chemical">asically substituted or condensed with other ring systems, exhibit inhibitory activity on PI3K/AkT/mTOR pathway and several molecules are currently in clinical trials. For example, omipalisib (GSK2126458) is a quinoline derivative, substituted at C-6 with a 2-(methoxy)-3-benzenesulfonamide pyridinyl moiety, at present under investigation in the first-in-human phase I study. This molecule shows the high inhibition effect of both PI3K and mTOR with an acceptable oral bioavailability [67,68]. Several omipalisib-like derivatives, characterized by a N-{5-[6-quinolinyl]-3-pyridinyl}benzenesulfonamide scaffold, have been synthesized and biologically evaluated for their PI3K/mTOR inhibition capability (biological data of omipalisib and compounds 31–34 are reported in Table 2) [69,70,71,72].

Table 2

Omipalisib and analogues 31–34 selective active on PI3Kα and mTOR.

Quinoline Compound	Inhibition Data (IC50 or GI50) In Antiproliferative Assays	Enzymatic Inhibition Data (IC50 or Ki)	Ref.
	T47D (IC50 = 3 nM)BT474 (IC50 = 2.4 nM)	PI3Kα IC50 = 0.04 nM;mTORC1 Ki = 0.18 nM;mTORC2 Ki = 0.3 nM	[67]
	U-87 MG (IC50 = 16 ± 5.8 nM)	PI3Kα (IC50 = 4.6 ± 3 nM; Ki = 0.6 ± 0.5 nM);mTOR IC50 = 3.9 ± 1 nM	[69]
	PC-3 (IC50 = 0.37 μM);HCT-116 (IC50 = 2.47 μM)	PI3Kα IC50 = 1.63 nMmTOR IC50 = 3.6 nM	[70]
	PC-3 (IC50 = 0.08 μM));HCT-116 (IC50 = 0.47 μM)	PI3Kα IC50 = 1.7 nMmTOR IC50 = 10.0 nM	[71]
	PC-3 (GI50 = 0.40 μM);HCT-116 (GI50 = 0.47 μM)	PI3Kα IC50 = 0.50 nMmTOR IC50 = 1.3 nM	[72]

Crystallography and in silico studies highlight some recurrent and crucial interactions between these ligands and the pan class="Gene">ATP-binding site of PI3Kγ: the backbone NH of class="Chemical">pan class="Chemical">Val882 (Val851 in PI3Kα) forms a hydrogen bond with quinoline nitrogen; Lys883 interacts, through a charged bond, with the sulfonamide group; and the C-4 quinoline substituent frequently stabilizes the ligand-protein complex, accommodating itself in the ribose pocket physiologically occupied by ATP. In Figure 11 it is reported the co-crystal structure of omipalisib in complex with the catalytic subunit of PI3Kγ, [67,69,70,71,72].

Figure 11

Crystal structure of omipalisib in complex with the catalytic subunit of PI3Kγ (PDB id: 3L08) and detailed interactions [67,73].

Ma and co-workers reported the synthesis and the biological activities of a series of pan class="Chemical">quinoline derivatives with a substituted class="Chemical">pan class="Chemical">aniline at C-4 position, and linked to another quinoline moiety at the C-6 one (compound 35–37 in Table 3). These quinoline compounds showed interesting inhibitory activities on both mTORC1 and mTORC2, with an adequate stability in simulated gastric, intestinal fluids, and liver microsome. The docking studies for these derivatives into the mTOR catalytic cleft showed the importance of some recurrent interactions, that stabilize the complex: hydrogen bond between the nitrogen of central quinoline scaffold and the critical residue Val2240, π– π interaction between indole moiety of Trp2239 and quinoline scaffold, and hydrogen bond between quinoline substituent at C-6 and Tyr2225 [74,75].

Table 3

Quinoline derivatives 35–37 with interesting activity on both mTORC1 and mTORC2.

Compound	Antiproliferative Assays (IC50, μM)	mTOR Inhibition Data (IC50, nM)	Ref.
	HCT-116 (IC50 = 0.46 μM)PC-3 (IC50 = 0.61 μM)MCF-7 (IC50 = 0.24 μM)	14	[74]
	HCT-116 (IC50 = 0.36 μM)PC-3 (IC50 = 0.50 μM)MCF-7 (IC50 = 0.11 μM)	22	[75]
	HCT-116 (IC50 = 0.11 μM)PC-3 (IC50 = 0.17 μM)MCF-7 (IC50 = 0.05 μM)	30	[75]

Zhao et al. published the synthesis and the biological activity of pan class="Chemical">quinoline derivatives with dual PI3K/class="Chemical">pan class="Gene">mTOR inhibitory effects. This new set of 4-aniline quinoline compounds containing phenylsulfonylurea are a combination of the benzensulfonamido and the anilino moieties, which are generally present in most of the quinoline active molecules reported in this review. Quinoline 38 (Figure 12) is the best inhibitor of the series, with IC50 values against PI3K and mTOR, respectively of 0.72 μM and 2.62 μM, and with an IC50 (half-maximal inhibitory concentration) against MCF-7 cell line comparable to the positive control [76].

Figure 12

The most active derivative of 3-phenylsulfonylurea-4-aniline quinoline series [76].

The PI3K/pan class="Gene">mTOR inhibition activity is also conserved in numerous derivatives in which the class="Chemical">pan class="Chemical">quinoline scaffold is condensed with other heterocycles. Among these, imidazo[4,5-c]quinolines are the most important, such as investigational drugs like dactolisib (BEZ-235), panulisib (P-7170), LY-3023414, SHR8443, and BGT-226 (Figure 13) [3].

Figure 13

Imidazo[4,5-c]quinoline derivatives as promising anticancer PI3K/mTOR inhibitors.

pan class="Chemical">Dactolisib,2-class="Chemical">pan class="Gene">methyl-2-[4-(3-methyl-2-oxo-8-quinolin-3-yl-2,3-dihydro-imidazo[4,5-c]quinolin-1- -yl)-phenyl]-propionitrile, was patented by Novartis in 2006 [77]. It demonstrated high selectivity and potency in the nanomolar concentration range against PI3K/mTOR kinases, acceptable oral bioavailability, potent antiproliferative activity on acute lymphoblastic leukaemia T-cell and also in paclitaxel-resistant gastric cancer models. Moreover, it has been tested in clinical trials, alone or in combination with cytotoxic agents, in patients with advanced solid tumours, as for example nasopharyngeal carcinoma [78,79,80,81,82,83]. From a molecular point of view, in Figure it 14 is shown the binding mode of dactolisib in the catalytic site of PI3Kα as an example for all the imidazo[4,5-c]quinolines: the nitrogen of the quinoline core forms an hydrogen bond with the backbone of the conserved Val851, interaction observed also for the ATP; other two hydrogen bonds involved the -CN substituent and the nitrogen of the lateral quinoline scaffold respectively with the amino acids Ser774 and Asp933 (Figure 14) [78,84].

Figure 14

Model for the interactions of dactolisib in the ATP-binding cleft of PI3Kα [78,84].

pan class="Chemical">Panulisib (P-7170) wclass="Chemical">pan class="Chemical">as patented in 2012 [85]. This quinoline compounds demonstrated potent inhibition activity on mTORC1/mTORC2/ALK1, cell cycle arrest, induction of apoptosis, antiangiogenic activity. Moreover, it showed antitumor efficacy on both in vitro and in vivo studies against the erlotinib–sensitive and –insensitive models NSCLC and anti-estrogen models of ER+ breast cancer [86,87,88], reaching the clinical trials evaluation for patients with advanced refractory solid tumours (ClinicalTrials.gov number NCT01762410) [89].

pan class="Chemical">LY-3023414, develoclass="Chemical">ped class="Chemical">pan class="Chemical">as a selective PI3K/mTOR dual inhibitor, was patented in 2012, with an absolute mTOR IC50 and PI3Kα IC50 values respectively of 0.165 μM and 0.00607 μM. In cancer cell human line panels, it showed broad antiproliferative activity [90,91]. In a first-in-human phase I study it demonstrated a tolerable safety profile and single-agent activity in patients with advanced tumours [92].

pan class="Chemical">SHR8443 hclass="Chemical">pan class="Chemical">as been evaluated in vitro for the enzymatic inhibition assay of mTOR, showing IC50 value of 1 nM [93]. Preclinical studies evidenced pharmaceutical properties favourable for the clinical use in antitumor treatments [94].

BGT-226, 8-(6-pan class="Gene">Methoxy-class="Chemical">pyridin-3-yl)-3-class="Chemical">pan class="Gene">methyl-1-(4-piperazin-1-yl-3-trifluoromethyl-phenyl)- 1,3-dihydro-imidazo[4,5-c]quinolin-2-one, developed by Novartis [77], demonstrated effectiveness in reducing in vitro and in vivo the growth of several human cancer types: head and neck [95]; human pancreatic [96]; human acute leukaemia and myeloma cancer cell lines [97,98]; NSCLC cancer (in combination with gefitinib) [99]; normoxic and hypoxic hepatocarcinoma cells [100]. BGT-226 reached the phase I of clinical trials (ClinicalTrials.gov Identifier: NCT00742105) where it demonstrated good tolerability, but exhibited insufficient results for a clinically meaningful PI3K pathway inhibition [101].

Recentpan class="Chemical">ly, new series of class="Chemical">pan class="Chemical">imidazo[4,5-c]quinoline derivatives have been reported in the literature as PI3K/mTOR inhibitors. Xiao et al. proved anticancer properties of a set of quinoline molecules, from which compound 39 (Figure 13) is the most active, with a mTOR IC50 value of 1.4 μM and PI3Kα IC50 of 0.9 μM [102].

Bahekar et al. synthesized pan class="Chemical">quinoline 40, an oralclass="Chemical">pan class="Chemical">ly available, selective, and effective inhibitor of PI3Kδ, with IC50 value of 1.9 nM [103].

Besides pan class="Chemical">imidazo[4,5-c]quinolines, also some interesting class="Chemical">pan class="Chemical">thieno[3,2-c]quinoline derivatives show PI3K inhibitory activity; compound 41 (Figure 15), for example, exhibits a PI3K IC50 = 1 μM and IC50 values against K562 and DU145 cancer cell lines of 0.15 μM and 2.5 μM, respectively [104].

Figure 15

Thieno[3,2-c]quinoline compound active as PI3K inhibitor [104].

Furthermore, recentpan class="Chemical">ly, Guo et al. through high-throughclass="Chemical">put screening (HTS), selected a novel hit comclass="Chemical">pound 42 (Figure 16a), containing class="Chemical">pan class="Chemical">pyrazino[2,3-c]quinolin-2(1H)-one scaffold, as mTOR inhibitor. This molecule demonstrated an mTOR IC50 = 31 nM, but weak antiproliferative activity in vitro. Further structural optimization led to compound 43 (Figure 16), that showed a better mTOR inhibitory activity than the parent compound (IC50 = 7 nM), good selectivity profile, potent antiproliferative activities against breast and cervical cancer lines, and significant tumor regression in the T-47D xenograft model after an oral once-daily dose [105].

Figure 16

(a) mTOR inhibitors containing pyrazino[2,3-c]quinolin-2(1H)-one scaffold; (b) structures of torin 1 and torin 2, mTOR inhibitors containing benzo[h]-1,6-naphthyridin-2(1H)-one scaffold [105,106,107].

Remarkpan class="Gene">able interesting, are torin 1 and torin 2, two class="Chemical">pan class="Chemical">benzo[h][1,6]-naphthyridin-2(1H)-one compounds, containing a quinoline moiety and acting as selective mTOR inhibitors (Figure 16b). Developed by Liu and co-workers, these tricyclic pyridoquinolines derivatives are the result of two consecutive generations of quinoline-based compounds. They are active on the phosphorylation process of mTORC1/ mTORC2 substrates in the cell, exhibiting EC50 values in the range 0.25–7 nM. Furthermore, with respect to torin 1, second generation compound torin 2 demonstrates 800-fold selectivity over PI3K, improving the bioavailability (54%), metabolic stability, and plasma exposure [106,107].

2.3. Quinoline Derivatives as Inhibitors of Epidermal Growth Factor Receptors

Various pan class="Chemical">quinoline derivatives, with class="Chemical">pan class="Gene">EGFR inhibitory potential, have been studied and the 4-anilinoquinoline-3-carbonitrile class is one of the first group developed. These compounds have been designed by molecular modelling studies, considering the approved EGFR 4-anilinoquinazoline inhibitors (gefitinib, erlotinib, and afatinib) as lead compounds (Figure 17).

Figure 17

4-anilinoquinoline-3-carbonitrile 44 and of the approved 4-anilinoquinazoline EGFR inhibitors gefitinib, erlotinib, afatinib.

Weissner et al. demonstpan class="Species">rated that the N-3 of class="Chemical">pan class="Chemical">quinazoline ring system can be replaced by a C-X, where X is an electron withdrawing group (i.e., CN). It was found that 4-anilinoquinoline-3-carbonitriles are effective inhibitors of EGFR kinase with activity comparable to the 4-anilinoquinazoline-based inhibitors. Compound 44 (Figure 17) showed the interesting IC50 value of 7.5 nM [108].

The same authors developed a second genepan class="Species">ration of class="Chemical">pan class="Gene">EGFR inhibitors, namely 4-anilinoquinoline-3-carbonitrile derivatives bearing Michael acceptor groups (such as 4-(dimethylamino)crotonamide) at the position C-6 (Figure 18).

Figure 18

Structures of the irreversible EGFR inhibitors pelitinib (EKB-569), neratinib (HKI-272), and pyrotinib (SHR-1258).

These molecules act pan class="Chemical">as class="Chemical">pan class="Gene">EGFR irreversible inhibitors, forming a covalent bond with a conserved cysteine residue located in the ATP binding pocket of EGFR (Cys773 or Cys797) and HER-2 (Cys805) [109]. In particular, pelitinib (EKB-569; EGFR IC50 = 0.083 μM), currently under investigation in clinical trials, showed excellent oral in vivo activity [3,110].

Subsequentpan class="Chemical">ly, Tsou et al., introducing a liclass="Chemical">poclass="Chemical">philic substituent at the class="Chemical">para class="Chemical">position of the 4-arylamino ring, isolated the molecule class="Chemical">pan class="Chemical">HKI-272, more active than pelitinib, and approved in 2017 with the name neratinib (Figure 18) for the extended adjuvant therapy against early-stage HER2/ErbB2-amplified/overexpressed breast cancer [3,4,109,111,112]. Crystal structure of EGFR/T290M mutant kinase domain in complex with neratinib showed the main interactions recurrent in the binding pocket (Figure 19): non-covalent interactions (H-bond of quinoline nitrogen with residues of the hinge region; hydrophobic interactions between the 2-pyridinyl group and the Met766, Phe856 and Met790) and a covalent bond (Cys797 at the edge of the active site cleft attacked the crotonamide group of the ligand) [113].

Figure 19

EGFR/T290M mutant kinase domain (PDB id: 2JIV) in complex with neratinib [113].

Starting from the lead compound pan class="Chemical">neratinib, SHR1258 hclass="Chemical">pan class="Chemical">as been developed as irreversible EGFR/HER2 tyrosine kinase inhibitor, and afterwards named pyrotinib (Figure 18). Figure 20 shows a comparison between binding modes of neratinib and pyrotinib in the catalytic region of HER2 kinase. In this model the two molecules are completely superimposed (Figure 20b), except for the substituents on the Michael acceptor group. Both display the same stabilizing interactions: hydrogen bond between N-1 of the quinoline ring and the hinge region of Met801, a covalent bond between Cys805 and Michael acceptor group (double bond) [114].

Figure 20

(a) binding mode of pyrotinib in the catalytic region of HER2 kinase; (b) overlap binding mode of neratinib (colored blue) and pyrotinib (colored yellow) [114].

In vitro preclinical studies show high potency of pan class="Chemical">pyrotinib, comclass="Chemical">parclass="Chemical">pan class="Gene">able to neratinib, against HER2 dependent BT474 (breast cancer) and SK-OV-3 (ovarian cancer) cell lines and also in vivo efficacy in HER2-dependent mouse xenograft models. Moreover, pyrotinib displays high selectivity when tested against a panel of different kinases [114]. Clinical trials demonstrate a safety profiles, desirable pharmacokinetic properties, tolerability and promising antitumor activity in HER2-positive patients with metastatic breast cancer, especially in combination with capecitabine [115,116]. Currently, pyrotinib is also under trial evaluation for the use in HER2-positive gastric cancer and NSCLC [117].

Pannala et al. described the synthesis and the evaluation of pan class="Gene">EGFR inhibition of some class="Chemical">pan class="Chemical">4-(2-aryl-cyclopropylamino)-quinoline-3-carbonitriles. These compounds possess an arylcyclopropylamino group at the C-4 position of the quinoline-3-carbonitrile core instead of 4-aniline one. Quinoline 45 (Figure 21) shows an interesting inhibition effect on EGFR with IC50 value of 5 nM [118].

Figure 21

Structure of 4-(2-aryl-cyclopropylamino)-quinoline-3-carbonitrile as selective inhibitor of EGFR [118].

Apan class="Chemical">ly et al. class="Chemical">planned and synthesized a set of class="Chemical">pan class="Chemical">4-anilino-3-carboxyamide derivatives. They based their studies on the high similarities between the 4-anilinoquinazoline, 4-anilinoquinoline-3-carbonitrile, and 4-anilinoquinoline-3-carboxamide nuclei. In a previous work, quinoline 46 (EGFR IC50 = 5.283 μM) was selected as a lead structure for the design of other more active derivatives [119]. In particular, compound 47 with a substituted-thiophene moiety at C-6 (Figure 22) exhibited selective activity on EGFR with IC50 value of 0.49 μM [120].

Figure 22

4-anilino-3-carboxyamide derivatives as EGFR inhibitors [119,120].

In addition, pan class="Chemical">2-styrylquinolines 48 and 49, without a class="Chemical">pan class="Chemical">4-anilino substitution (Figure 23), have been synthesized by El-Sayed et al., showing moderate inhibition effects towards EGFR with IC50 values of 2.23 and 1.11 μM respectively [121].

Figure 23

2-styrylquinolines 48 and 49 as promising EGFR inhibitors [121].

In lpan class="Chemical">asts years, several hybrids derivatives containing class="Chemical">pan class="Chemical">quinoline scaffold linked with other heterocyclic ring systems have been studied as EGFR inhibitors with interesting activity. Makawana et al. reported numerous Schiff’s base derivatives bearing nitroimidazole and quinoline nuclei as potential EGFR tyrosine kinase inhibitors. Compound 50 (Figure 24) showed an IC50 value on EGFR receptor of 0.12 ± 0.05 μM [122].

Figure 24

Schiff’s base quinoline derivative as EGFR tyrosine kinase inhibitor [122].

George et al. described some new pan class="Chemical">quinoline hybrid derivatives bearing either class="Chemical">pan class="Chemical">pyrazoline or pirazolinylthiazole heterocycles. Most of the tested compounds revealed promising anticancer activity especially against DLD1 cancer cell line (colorectal). Furthermore, compounds 51–53 (Figure 25) revealed inhibitory effect on EGFR at nanomolar concentration, with IC50 values of 31.80, 37.07, and 42.52 nM, respectively [123].

Figure 25

Pyrazoline/pirazolinylthiazole quinoline hybrids [123].

The pan class="Chemical">quinoline scaffold is frequentclass="Chemical">pan class="Chemical">ly present in numerous polycondensed ring system compounds with proved selective inhibition activity on EGFR.

Abdelbpan class="Chemical">aset et al., for examclass="Chemical">ple, designed several class="Chemical">pan class="Chemical">thieno[2,3-b]quinoline-2-carboxamide-chalcone derivatives, as molecule 54 (Figure 26a) that showed significant antiproliferative effects against tested cancer cell lines (IC50 values in the range 0.9–1.2 μM) and an EGFR IC50 = 0.5 μM. Molecular docking studies showed that the thienoquinoline moiety occupied the ATP-binding site of the receptor, whereas chalcone moiety was located in an allosteric pocket of the enzyme (Figure 26b) [124].

Figure 26

(a) Thieno[2,3-b]quinoline-2-carboxamide-chalcone derivative 54; (b) orientation and interactions of 54 within the EGFR active site [124].

Afterwards, Abdelsalam and co-workers designed some pan class="Chemical">benzo[h]quinoline derivatives, class="Chemical">pan class="Chemical">as molecule 55 (Figure 27), with good anticancer activity against MCF-7 cancer cell line (IC50 = 7.21 ± 0.43 μM) and also an EGFR IC50 = 0.14 μM [125].

Figure 27

Structure of benzo[h]quinoline derivative [125].

2.4. Quinoline Derivatives as Inhibitors of Vascular Endothelial Growth Factor Receptor

pan class="Gene">VEGF-class="Chemical">pan class="Gene">VEGFR abnormal signals play central roles in angiogenic processes in a variety of diseases, especially in cancer. Some studies demonstrated, that, although the VEGF binds more effectively the VEGFR-1, its major mitogenic and angiogenic effects seem to be mediated through the interaction with VEGFR of type 2. In recent years, a lot of selective angiogenesis and multi-receptor kinase inhibitors have been synthesized, and some of these have been approved for clinical use [24,25].

pan class="Chemical">Lenvatinib, (Figure 28), aclass="Chemical">pclass="Chemical">proved in 2015 for the treatment of differentiated thyroid, class="Chemical">pan class="Disease">hepatocellular cancers, and the second-line treatment of renal carcinoma, is a multikinase quinoline inhibitor with a selective effect against VEGFR [4,7].

Figure 28

Structures of VEGFR inhibitors with quinoline-urea moiety lenvatinib, Ki8751, tivozanib.

Many other pan class="Chemical">quinoline derivatives are reclass="Chemical">ported to class="Chemical">possess selective class="Chemical">pan class="Gene">VEGFR inhibitor activity, especially against VEGFR-2. Kubo et al. reported the synthesis of some N-Phenyl-N’-{4-(4-quinolyloxy)phenyl}ureas, such as compound Ki8751 (Figure 28), which exhibited the best inhibition effect on VEGFR with IC50 value of 0.9 nM [126].

Nakamura and co-workers firstpan class="Chemical">ly reclass="Chemical">ported the anticlass="Chemical">pan class="Disease">tumor and anti-angiogenic activity of KEN951, after called tivozanib, (Figure 28). With a very similar structure to the previous studied compounds, this quinoline-urea inhibits VEGFR-1, VEGFR-2, and VEGFR-3 tyrosine kinases (IC50 = 30, 6.5, and 15 nM, respectively). Tivozanib potently inhibited VEGF-induced VEGFR-2 phosphorylation in endothelial cells at in vitro sub-nanomolar concentration (IC50 = 0.16 nM). In xenografts models, tivozanib demonstrated significant inhibition of tumor-induced angiogenesis and tumor vessel normalization and in phase I of clinical trial it exhibited safety and tolerability when orally administered [127,128,129].

Anapan class="Chemical">lysing the structural analogies between class="Chemical">pan class="Chemical">lenvatinib, Ki8751, and tivozanib, it is possible to underline the recurrent and important presence of the C-4 oxyphenyl urea substituted moieties and the C-6, C-7 substitutions on the quinoline scaffold. These structure features play a pivotal role for the stabilization of ligand-protein complex, as it is shown by the X-ray crystal structure of VEGFR-2 in complex with lenvatinib (Figure 29, PDB id: 3WZD): the ligand binds the receptor kinase domain in its DFG-in conformation, occupying the ATP-binding site through its quinoline moiety and the neighboring region via the cyclopropane ring. In particular, the key interactions are three H-bonds (quinoline nitrogen and Cys919; urea oxygen and Asp1046; urea nitrogen and Glu885), two hydrophilic interactions (the C-6 carboxamide with Asn923, bridged by water molecules) and several π-interactions (quinoline and C-4 phenoxy moiety with Leu840, Phe818 and Lys868) [130]. Quinoline core and urea moiety in tivozanib showed the same key interactions observed for the parent compound (PDB id: 4ASE) [131,132].

Figure 29

(a) crystal structure of lenvatinib in complex with VEGFR-2 (PDB id: 3WZD); (b) binding pocket of VEGFR-2 occupied by lenvatinib; (c) scheme for interactions of lenvatinib in the VEGFR-2 binding pocket [130].

Yang et al. reported the synthesis of some pan class="Chemical">quinoline amide derivatives with class="Chemical">pan class="Gene">VEGFR-2 inhibition activity. Compound 56 (Figure 30) resulted to be the most active to inhibit VEGFR kinase (IC50 = 3.8 nM) and proliferation of HUEVEC cancer cells (IC50 = 5.5 nM); the docking analysis confirmed that this compound is suitable bonded to VEGFR-2 [133].

Figure 30

Structure of the quinoline VEGFR-2 inhibitor 56 [133].

Through a combination of some structural elements present in molecules with antiprolifepan class="Species">rative activity, such class="Chemical">pan class="Chemical">as 4-piperazinoquinoline scaffold and aminoacyl chain, Aboul-Enein et al. planned and synthesized 7-Chloro-4-(piperazin-1-yl)quinoline derivatives as VEGFR-2 inhibitors. The most promising compound 57 (Figure 31), exhibited cytotoxicity higher than that of reference drug (doxorubicin) against MCF-7 cell line (6.502 μM vs. 6.774 μM) and a VEGFR-2 IC50 = 1.38 μM, but less potent than the reference drug sorafenib (IC50 of 0.33 μM). Docking analysis in the ATP-binding site of VEGFR-2 proved that molecule 57 shows a binding mode similar to that of other VEGFR-2 inhibitors as lenvatinib: both displayed H-bonds between quinoline nitrogen and Cys919, and between the carbonyl group (in the aminoacyl and in urea moieties in derivative 57 and tivozanib, respectively) and Asp1046. However, 57 showed some additional π-interaction with Ile888 and Phe918 [134].

Figure 31

Structure of compound 57, a 7-Chloro-4-(piperazin-1-yl)quinoline derivative with VEGFR-2 inhibition activity [134].

In 2008, Westman et al. published a series of pan class="Chemical">quinoline-3-carboxylic acid derivatives (Figure 32), acting class="Chemical">pan class="Chemical">as VEGFR-2 and -3 inhibitors and with good potency and high selectivity for the treatment and prevention of cell proliferative and/or differentiation disorders. Compounds 58 and 59 showed IC50 values against VEGFR-2 of 1.2 μM and 2.5 μM, and against VEGFR-3 of 0.53 μM and 0.35 μM respectively [135].

Figure 32

Quinoline-3-carboxylic acid derivatives with VEGFR-2 and VEGFR-3 inhibition activity [135].

Different pan class="Chemical">quinoline derivatives are reclass="Chemical">ported to act class="Chemical">pan class="Chemical">as dual inhibitors of VEGFR and other targets involved in carcinogenic processes. For example, Li et al. discovered some 3-aryl-quinoline derivatives 60 and 61 as dual inhibitors of both VEGFR-2 and ERα. Two molecules (Figure 33) showed interesting IC50 against ERα and VEGFR compared to the reference compounds, tamoxifen and sunitinib, respectively (for compound 60, ERα IC50 = 2.33 μM and VEGFR-2 IC50 = 104 nM; for 61, ERα IC50 = 1.78 μM and VEGFR-2 IC50 = 86 nM) [136].

Figure 33

Structure of 3-aryl-quinoline derivatives 60 and 61, dual VEGFR-2/ERα inhibitors [136].

2.5. Quinoline Derivatives as Inhibitors of Ras/Raf/MEK Pathway

pan class="Gene">EGFR and class="Chemical">pan class="Gene">VEGFR extracellular activations trigger the intracellular cascade of signalling connected to Ras/Raf/MEK/ERK. This deregulated pathway, in neoplastic cells, improve the tumour genesis, acting on the survival processes of cellular proliferation; such as differentiation, apoptosis, angiogenesis.

Several pan class="Chemical">quinoline derivatives, tested class="Chemical">pan class="Chemical">as inhibitors of Ras/Raf/MEK cascade, demonstrate promising results both in antiproliferative and enzymatic inhibition assays. In detail, Feng and co-workers developed quinoline KAL-21404358 (Figure 34), the first ligand of the K-Ras P110 allosteric pocket able to disrupt downstream pathways (Raf/MEK/ERK and PI3K/AkT/mTOR). Through a combination of computational methods and with the validation of biochemical assays, the authors proposed a mechanism of action for KAL-21404358. The quinoline compound could interfere with the protein-protein interactions, binding and stabilizing K-Ras in its inactive GDP-bound state, then, it halts the nucleotide exchange process and the subsequent activation of Ras [137].

Figure 34

Quinoline-based molecules with inhibition activity on Ras.

In 2016, Li et al. filled a patent of a new set of fused-tricyclic compounds containing pan class="Chemical">quinoline scaffold with a class="Chemical">promising inhibition activity on the mutated form of Gclass="Chemical">pan class="Chemical">ly12Cys K-Ras (Figure 34). The analysis of the catalytic binding site, highlights that Cys12 forms a covalent bond with the electrophilic acryloyl moiety of quinolines 62 and 63 [.

El-Gamal et al. synthesized two clpan class="Chemical">ass of class="Chemical">pan class="Gene">Raf inhibitors possessing quinoline scaffold. In the first one, a diarylamide moiety, through a S or an O atom, was linked at the C-3 of a dimethoxy/dihydroxy-quinoline scaffold. Biological assays realized by National Cancer Institute demonstrated that the most active compounds were dimethoxyquinolines with an oxygen as linker and with electron-withdrawing groups (-Cl or CF3) on the terminal ring. One of the most active compounds, 64 (Figure 35a), exhibited GI50 values in the micromolar range against the full panel of NCI cancer cell lines and inhibited C-Raf kinase activity by 76.65% at 10 μM [139].

Figure 35

(a) Structures of diarylurea derivative sorafenib and of quinoline inhibitors 64,65 as selective inhibitors of Raf.; (b) The best docked pose of 64 within the C-Raf active site (two H-bond stabilize the ligand-protein complex: amidic carbonyl oxygen-Thr421, quinolinyl nitrogen-Asn472); (c) best docked pose for 65 in C-Raf active site (three H-bond stabilize the interaction: urea carbonyl oxygen-catalytic Lys470, urea terminal nitrogen-Asp486, quinolinyl nitrogen-Lys431) [139,140].

In the second clpan class="Chemical">ass of class="Chemical">pan class="Chemical">quinoline compounds, the same authors substituted the amide linker with urea one to obtain diarylurea derivatives. In NCI five-dose screening protocol, molecule 65, with the insertion of 4-chloro-3-(trifluoromethyl)phenylurea moiety, was the most promising compound, with a relevant antiproliferative effect compared to the reference diarylurea drug sorafenib (Figure 35a). Inhibition assays of C-Raf kinase evidenced quinoline 65 as more active than the lead compound 64 (for 65 %inhibition at 10 μM = 99.67% and IC50 = 0.10 μM). Docking studies of the active site of C-Raf kinase (PDB ID: 3OMV) further explained the high potency of 65 with respect to 64. Indeed, compound 65, due to the greater flexibility of the urea linker, formed additional H-bonds with Asp486 and Lys470 (Figure 35b,c) [140].

El-Dampan class="Chemical">asy et al. designed and synthetized several class="Chemical">pan class="Chemical">quinoline analogues of sorafenib, substituting the central phenoxy nucleus with a quinoline one and inserting the two recurrent moieties of the lead compound, the arylurea/arylamide and the N-methyl picolinamide, respectively at the C-2 and C-5 positions of the quinoline ring system.

NCI antiprolifepan class="Species">rative class="Chemical">pan class="Chemical">assays of urea derivatives with fluorinated phenyl ring, as 66 and 67 (Figure 36a), showed GI50 values in the low-submicromolar range for the majority of the tested cell lines. Furthermore, compound 67 obtained remarkable results in the kinase inhibition assays, with high selectivity against Raf family and IC50 of 316 nM and 61 nM against BRAFV600E and C-Raf, respectively (for the reference sorafenib BRAFV600E IC50 = 38 nM and C-Raf IC50 = 6 nM); instead of quinoline derivative 66, that exerted its inhibitory effect only against C-Raf. Docking studies showed a similar binding mode between 67 and sorafenib in the catalytic kinase domain of BRAFV600E, indeed both molecules formed interactions through the picolinamide, urea, and trifluoromethylphenyl moieties (Figure 36b); on the other hand, the absence of activity of 66 against BRAFV600E was justified by the unfit orientation of the 2,4-difluorophenyl group, far from the hydrophobic pocket of the allosteric site (Figure 36c) [141].

Figure 36

(a) Quinoline inhibitors of Raf compared with the reference compound sorafenib; (b) overlay of the docked pose of compound 67 (magenta) and co-crystalized sorafenib (black) in the catalytic kinase domain of BRAFV600E (PDB id: 1UVJ); (c) superimposition of the docked pose of compound 67 (magenta) and 66 (orange) in the catalytic kinase domain of BRAFV600E (PDB id: 1UVJ) [141,142].

Considering 67 pan class="Chemical">as lead comclass="Chemical">pound, the same authors, by the substitution of the class="Chemical">pan class="Chemical">urea linker with an amine one, developed a new series of 2-anilinoquinolines bearing the N-methylpicolinamide group at the C-5 position of the quinoline scaffold. The derivative 4-chloro-3-(trifluoromethyl)aniline 68 (Figure 36a) manifested more remarkable antiproliferative effects than compound 67 and sorafenib, especially against melanoma and breast cancer panels. However, the replacement of the urea linker with an amine in compound 68 has a negative effect on Raf kinase inhibition, that resulted modestly. Docking studies in the catalytic kinase domain of BRAFV600E, in fact, underline the capability of quinoline 67 to form additional interactions right through its urea moiety; furthermore, the short amine spacer inhibited the insertion of the 4-chloro-3-(trifluoromethyl)phenyl group within the hydrophobic allosteric site adjacent the ATP binding site, precluding some stabilizing hydrophobic interactions [142].

Li et al. developed pan class="Chemical">quinoline derivative 69 (Figure 37) with class="Chemical">pan class="Chemical">urea and a 4-fluoro-3-(trifluoromethyl)phenyl moieties, as sorafenib. It exhibited higher antiproliferative activity than the reference compound, against Hep G2, A549 and KCC-853 cancer cell lines and also a higher inhibitory activity against C-Raf (for 69 C-Raf IC50 = 8.7 nM; for sorafenib C-Raf IC50 = 28.5 nM) [143].

Figure 37

3,3-dimethyl-1H-pyrrolo[3,2-g]quinolin-2(3H)-one derivative as potent C-Raf inhibitor [143].

In the Rpan class="Chemical">as/class="Chemical">pan class="Gene">Raf/MEK signalling cascade, MEK has been also identified as an interesting target for quinoline-based agents able to inhibit its catalytic activity.

Zhang et al. developed pan class="Chemical">4-(phenoxyanilino)quinoline 70 (Figure 38), that showed high selectivity against class="Chemical">pan class="Gene">MEK enzyme with an IC50 value of 25 nM. SAR studies demonstrated that the cyanoquinoline core is crucial in the inhibition of the enzyme if compared with the less active quinazoline analogues [144]. The same authors also explored the effects of different substitutions on the 4-anilino group in some 4-anilino-3-cyano-6-methoxy-7-(3-morpholino-4-yl-propoxy)quinolines; compounds 71 and 72 (Figure 38), with a phenoxy and phenylsulfonyl substitution respectively, were the most potent (MEK IC50 of 7 and 5,7 nM, respectively) [145].

Figure 38

Structures of some MEK inhibitors with 4-anilino-quinoline-3-carbonitrile core [144,145,146,147].

Berger and coworkers discovered some pan class="Gene">ATP comclass="Chemical">petitive inhibitors of class="Chemical">pan class="Gene">MEK1 kinase with a 6,7-dialkoaxy-3-quinolinecarbonitrile core, as 73 and 74 (Figure 38), in which the 4-anilino moiety is substituted with the 1-methyl-1H-imidazol-2-ylsulfonyl one. Compound 73 exhibited a MEK IC50 = 3 nM, exceptional in vitro, (IC50 = 7 nM), and in vivo activity against LoVo xenografts in nude mice [146].

In order to improve the solubility and the oral bioavailability, the same authors further developed molecule 74, with an alkenyl group at C-7 of the pan class="Chemical">quinoline core; this comclass="Chemical">pound demonstclass="Chemical">pan class="Species">rated good in vitro potency (MEK1 IC50 = 12 nM) and high plasma levels after oral dosing in H358 xenograft models [147].

Popan class="Chemical">lycondensed class="Chemical">pan class="Chemical">quinoline systems, as the 1H-imidazo[4,5-c]quinoline 75 (Figure 39) show interesting results in MEK kinase inhibition assays, with IC50 values lower than 100 μM; it was demonstrated that the high selectivity for MEK over other kinases depends on the –F substituent at the C-7 position of the quinoline nucleus [148].

Figure 39

Structure of MEK inhibitor 1H-imidazo[4,5-c]quinoline 75.

An interesting derivative of pan class="Chemical">ursolic acid class="Chemical">possessing a condensed class="Chemical">pan class="Chemical">chloro-substituted quinoline nucleus and a hydrazide moiety was developed recently by Jin et al. (76 in Figure 40). Compound 76 showed remarkable results as antiproliferative agent, capability to inhibit MEK1 kinase activity (MEK1 IC50 = 64 nM) and activation of Ras/Raf/MEK/ERK pathway. Molecular docking analysis into the MEK1 binding site proved that the ursolic acid skeleton ensured the correct orientation of the hydrazide side chain and of the quinoline ring into the pocket, while the latter were involved in stabilizing interactions with the amino acids [149].

Figure 40

(a) structure of the quinoline derivative of ursolic acid 76; (b) binding pose of 76 within MEK1 kinase domain (PDB id: 3EQF); (c) ligand interactions of 76 docked into MEK1 active site [149].

3. Conclusions

The pan class="Chemical">quinoline ring system is a scaffold highclass="Chemical">pan class="Chemical">ly frequent in druggable molecules. Currently, a great amount of quinoline derivatives exhibit pharmaceutical activity as antibacterial, antifungal, antimalarial, anthelmintic, local anaesthetic, antipsychotic, and anticancer drugs.

In the antipan class="Disease">cancer field, the class="Chemical">pan class="Chemical">quinoline nucleus has been observed as crucial molecular moiety recurrent in several inhibitors of kinases, for the treatments of a wide range of tumours in the targeted therapeutic approach.

In particular, this review summarizes the biological data of several pan class="Chemical">quinoline comclass="Chemical">pounds, which are active on class="Chemical">pan class="Gene">c-Met, VEGF, and EGF receptors, and on the related proteins involved in the intracellular signalling cascades. Overexpressed in tumour cells, these three protein kinases receptors trigger carcinogenic pathways closely connected with each other, regulating the survival processes in the cell, such as proliferation, apoptosis, differentiation, and angiogenesis.

The druggpan class="Gene">able attitude of the class="Chemical">pan class="Chemical">quinoline scaffold probably lies on its proved biocompatibility, versatility, and chemical accessibility, from which new derivatives can be easily designed and synthesized.

The pan class="Chemical">quinoline nucleus is an electron-deficient ring system with tertiary bclass="Chemical">pan class="Chemical">ase properties. The presence of nitrogen withdraws electrons by resonance, interfering with the equal distribution of the π-electron density and suggesting a chemical behaviour similar to the pyridine.

In view of an pan class="Species">SAR study, the anaclass="Chemical">pan class="Chemical">lysis of the interactions of the quinoline molecules in the protein binding sites, under investigation, highlights recurrent hydrogen bonds with the nitrogen of the quinoline ring and π-π stacking complexes with complementary amino acid residues. The presence of flexible moiety, especially at C-4, C-6, and C-7 positions, frequently consolidates the force of the ligand–protein complexes.

All the biological data of the pan class="Chemical">quinoline comclass="Chemical">pounds, anaclass="Chemical">pan class="Chemical">lysed in the present review, are reported in detail, confirming the antiproliferative activity and the pivotal importance of this ring system in the efficacy of several approved drugs. In drug discovery, the information recovered in this manuscript could help in the design and rationale optimization studies for the development of new quinoline multitargets molecules.