key: cord-0022286-nku6dezl
authors: Recio, Rocío; Lerena, Patricia; Pozo, Esther; Calderón-Montaño, José Manuel; Burgos-Morón, Estefanía; López-Lázaro, Miguel; Valdivia, Victoria; Pernia Leal, Manuel; Mouillac, Bernard; Organero, Juan Ángel; Khiar, Noureddine; Fernández, Inmaculada
title: Carbohydrate-Based NK1R Antagonists with Broad-Spectrum Anticancer Activity
date: 2021-07-08
journal: J Med Chem
DOI: 10.1021/acs.jmedchem.1c00793
sha: aa5b1238d0db00850b0711efbfb3c021e8afdb6f
doc_id: 22286
cord_uid: nku6dezl

[Image: see text] NK1R antagonists, investigated for the treatment of several pathologies, have shown encouraging results in the treatment of several cancers. In the present study, we report on the synthesis of carbohydrate-based NK1R antagonists and their evaluation as anticancer agents against a wide range of cancer cells. All of the prepared compounds, derived from either d-galactose or l-arabinose, have shown high affinity and NK1R antagonistic activity with a broad-spectrum anticancer activity and an important selectivity, comparable to Cisplatin. This strategy has allowed us to identify the galactosyl derivative 14α, as an interesting hit exhibiting significant NK1R antagonist effect (k(inact) 0.209 ± 0.103 μM) and high binding affinity for NK1R (IC(50) = 50.4 nM, K(i) = 22.4 nM by measuring the displacement of [(125)I] SP from NK1R). Interestingly, this galactosyl derivative has shown marked selective cytotoxic activity against 12 different types of cancer cell lines.

The NK1 receptor (NK1R), also known as tachykinin receptor 1 (TACR1), 1 belongs to the superfamily of G-protein coupled receptors, which constitute ∼35% of the therapeutic targets of all of the pharmaceutical products on the market. 2 The preferred endogenous agonist of NK1R is the undecapeptide substance P (SP), which acts as a neurotransmitter and neuromodulator. 3 NK1R is present in the central and peripheral nervous systems, smooth muscle, endothelial cells, and also on cells that participate in immune response. 4 Over the past four decades, intensive research has linked the SP-NK1R system to broad pathophysiological processes including nausea, 5 analgesia, 6 inflammation, 7 and depression. 8 In addition, NK1R is overexpressed in several cancers, 9 including melanoma, 10 astrocytoma, 11 pancreatic ductal carcinomas, 12 bone marrow, 13 and gastric cancer, 14 highlighting the potential therapeutic value of NK1R antagonists. This potential has recently been accentuated following several studies demonstrating the beneficial effect of NK1R antagonists on the health of patients infected with the SARS-CoV-2 virus, responsible for the current COVID19 pandemic. 15 This perspective has boosted the search not only in academia but also in industry, with almost all important pharmaceutical companies investing in this field of selective and potent NK1R antagonists. 16 A turning point in this race was the discovery of the first nonpeptide NK1R antagonist CP-96,345, 17 which has been instrumental in the development of a number of antagonists with improved pharmacological properties; Figure 1 . 16, 18 Structural optimizations around the central skeleton ultimately led to the development of Aprepitant, 19 which became the first oral drug approved to enter the clinic, specifically targeting NK1R for the treatment of chemotherapy-induced nausea and vomiting. 20 During the last 5 years, two other molecules, namely, Netupitant and Rolapitant, have been approved for clinical use for the same indication. 21 It is worth mentioning that the discovery of effective NK1R antagonists is challenging due to the complexity of the NK1 transmembrane receptor, the crystal structure of which has only recently been determined. 22 In a project directed toward the asymmetric synthesis of new NK1R antagonists, we have recently reported the asymmetric synthesis of 5-arylsulfinyl-2-amino-4H-pyrans and their application as antitumoral compounds. 23 In the present work, we report on the stereoselective synthesis of carbohydrate-based NK1R antagonists (CarbNK1RAnt) and the determination of their selective cytotoxic activities against different types of cancer cell lines. 24 ■ RESULTS AND DISCUSSION Chemistry. Preliminary structure−activity relationship studies carried out on a large number of NK1R antagonists developed after the discovery of CP-96,345 allowed the proposition of a pharmacophore model, which consists of a heterocyclic scaffold substituted with at least two aromatic rings in a cis orientation; Figure 1A . 25 In most cases, the fixed orientation between the two aromatic groups aforementioned can be a parallel face-to-face ( Figure 1A(a) ), 26 a perpendicular T ( Figure 1A(b) ), or an edge-on L arrangement ( Figure  1A(c) ). 27 The scaffolds used have evolved from the quinuclidine in CP-96,345, to a simpler piperidine in CP-99,994, Casopitant and Ralopitant; a morpholine ring in Aprepitant; a pyridine ring in Netupitant; or a simple acyclic chain in L-733,060; Figure 1B ,C. The use of a heterocyclic saturated scaffold with two substituted carbons implies that the molecule is chiral with at least two stereogenic centers. Indeed, both Aprepitant and Ralopitant have three chiral centers, and of the eight possible diastereoisomers, only one has the desired activity. Accessing the desired compound as a single enantiomer is challenging, time consuming, and highly expensive. As an illustrative example, the Merck process developed for the synthesis of Aprepitant consists of a catalytic asymmetric (transfer) hydrogenation process coupled with a successful crystallization-induced diastereoselective transformation and a diastereoselective imine hydrogenation for the creation of three stereocenters. 28 A simple and economical alternative to access heterocyclic compounds with multiple chiral centers in close proximity is to use compounds belonging to the chiral pool. A family of compounds well suited for this comprises the carbohydrates, stereochemically rich compounds with hydroxyl groups in virtually all arrangements, allowing the easy tuning of their steric, electronic, and three-dimensional (3D) structures. Moreover, as abundant and renewable biomolecules, they are accessible on a large scale at low cost. Indeed, some monosaccharides are even less expensive than the most common solvents used in synthetic laboratories. As part of a large program aimed at the utilization of carbohydrates in the asymmetric synthesis of synthetically and pharmacologically relevant molecules as well as in the synthesis of biomaterials, 29 we decided to develop new NK1R antagonists using sugars as raw materials. As functional groups, we planned to incorporate at the anomeric position the 3,5-bis(trifluoromethyl)benzyl fragment, present in many NK1R antagonists including all currently marketed ones. As a second aromatic fragment, we opted for a pfluorophenyl function, present in Aprepitant, Fosaprepitant, and other selective NK1R antagonists, as O-benzyl ether at the C-2 position of the carbohydrate (Figure 2 ). When both aromatic groups are at the α-face of the pyranose ring, an intramolecular π−π stacking interaction can be established, which, as stated before, has been proposed to play a significant role in other active analogues profiled in structure−activity relationship studies.

Synthesis of D-Galactose-Derived NK1R Antagonists (GalNKR1Ant). Considering the planned sugar-based 1,2-cis substituted NK1R antagonist analogues, the choice of the starting carbohydrate as well as the sequence followed for implementing the substituents is crucial. From a synthetic point of view, it is a question of using the right sugar and a synthetic sequence that allows the selective functionalization of the C2-OH and to introduce the aglycon group in the carbohydrate α-face. Taking these considerations into account, among all of the hexopyranoses, D-galactose is the sugar of choice due to the different reactivity of its five hydroxyl groups, mainly due to the cis arrangement of C3-OH and C4-OH. Moreover, the α position of the aglyconic group requires the use of a nonparticipating group at C2 during the glycosylation step. Therefore, using commercially available D-galactose pentaacetate 1(α,β), we first introduced the p-fluorophenyl group at position 2 of the pyranose ring as the corresponding O-benzyl ether, before introducing the (R)-1-[3,5-bis-(trifluoromethyl) phenyl]ethan-1-ol at the anomeric position; Scheme 1.

Condensation of thiophenol with D-galactose pentaacetate 1(α,β) in the presence of boron etherate trifluoride afforded the corresponding thioglycoside 2β in high chemical yield (90%, Scheme 1), as a single anomer. A Zemplen deacetylation, followed by acid-catalyzed acetalation with 2,2- dimethoxypropane (DMP), afforded the 3,4-acetal 4β in 85% yield. A regioselective silylation of the primary alcohol with tert-butyldiphenylsilyl chloride in DMF at rt afforded the mono hydroxylated derivative 5β in high yield (96%). Finally, the installation of the p-fluorophenyl fragment was carried out in THF, using NaH as the base and p-fluorobenzyl chloride 6. Thus, in only five high-yielding steps, the fully O-protected derivative 7β was obtained on a multigram scale (Scheme 1).

The introduction of the (R)-1-[3,5-bis(trifluoromethyl)phenyl]ethan-1-ol fragment in the anomeric position was accomplished as indicated in Scheme 2, following two different approaches. The first one, based on the use of the wellestablished trichloroacetimidate glycosylation reaction, consisted of treating 7β with N-bromosuccinimide (NBS) in wet acetone and subsequent base-catalyzed addition of the obtained lactol 8(α,β) to trichloroacetonitrile, to give the trichloroacetimidate donor as a 3:1 mixture of both anomers 9α and 9β (Scheme 2). Lewis-acid-catalyzed glycosylation of the chiral alcohol acceptor, (R)-1-[3,5-bis(trifluoromethyl)phenyl]ethan-1-ol 10R, with the mixture of the trichloroacetimidate donors gave the fully protected O-glycosyl derivative 11(α,β), as a 3:1 mixture of both diastereomers, which were easily separated by column chromatography to give the 11α and 11β anomers, in 60 and 25% chemical yields, respectively. As an alternative route, the alcohol 10R was directly Oglycosylated, using thioglycoside 7β as the glycosyl donor. For this, the mixture NIS/trimethylsilyl trifluoromethanesulfonate was used as an activator in the presence of MS (4 Å), affording Scheme 2. Syntheses of 11α and 11β by Glycosylation of p-Fluorobenzyl D-Galactose Derivative 7β a a an equimolecular mixture of both anomers, 11(α,β), in a high 81% chemical yield.

Considering the rigid conformations of the new Dgalactoderivatives, as a fused bicyclic compound with a pyranose ring and a cyclic acetal, we were interested in studying the difference in bioactivity between both anomers, despite the fact that in the case of most NK1R antagonist analogues the cis isomer is the most active one. Moreover, in both diastereomers, 11α and 11β, deprotection of the hydroxylic groups at 3, 4, and 6 positions and modification of the protecting groups, as indicated in Scheme 3, give us the opportunity to modulate the lipophilicity of the carbohydrate derivatives and study, at the same time, the structure−activity relationship. Desilylation of the O-silyl ether in position 6 with TBAF, followed by acid hydrolysis of 3,4-dimethyl acetal and subsequent formation of the 4,6-O-benzylidene acetal, allows us to obtain the corresponding monoalcohols (12 and 14) or the more hydrophilic trihydroxylated analogues (13) , with α or β configurations, starting from 11α and 11β, respectively (Scheme 3). The presence of the phenyl ring in the benzylidene moiety of 14α and 14β favors their hydrophobic and/or π−π stabilizing interactions with some NK1R amino acids, as we have determined by docking studies (vide infra).

Upon regioselective protection of the trihydroxylated epimers 13 (Scheme 3), a single benzylidene acetal diastereoisomer was formed. Although expected, we, however, conducted selective NOESY1D experiments to confirm the stereochemical outcome of the process. 30 For both 14α and 14β, the registered NOESY1D spectra with selective excitation of the benzylidene acetal protons (see the Supporting Information) display three sets of signals, for an aromatic proton, H4, and H6 of the sugar. While the NOE contacts observed with the aromatic proton and H6 may be seen for the two diastereomers, the NOE contact observed with H4 is clearly indicative that the absolute configuration of the benzylidene acetal carbon center is indeed R.

Synthesis of L-Arabinose-Derived NK1R Antagonists (AraNK1Ant). The pentapyranose L-arabinose is structurally related to D-galactose with the sole, and important, difference of lacking the 6-hydroxymethyl group. Consequently, starting from L-arabinose tetraacetate 15α, a synthetic approach similar to that developed for D-galactose has allowed us to obtain a series of analogues in only five or six steps and in the multigram scale; Scheme 4. Additionally, the absence of the 6hydroxymethyl group in the obtained analogues ( Figure 2 ) will affect both their lipophilicity and conformational behavior.

Thus, the 2-O-p-fluorobenzyl derivative 19α was obtained as a single diastereomer, starting from per-O-acetylated α-Larabinose 15α, in only four high-yielding steps, with a 77% overall yield; Scheme 4. The chiral alcohol 10R was introduced in the anomeric position using the trichloroacetimidate method to give a 1:2 mixture of both anomeric O-glycosyl derivatives 22α and 22β. They were also obtained directly from the phenylthioglycoside 19α, as a 1:1 mixture of anomers by activation with trifluoromethylsilyl triflate as the Lewis acid, in the presence of NIS and 4 Å MS, at 0°C. The two diastereoisomers 22α and 22β showed a very different separation factor, which, after column chromatography, allowed them to be obtained in the pure form with an overall 66% yield. Finally, acid hydrolysis of the 3,4-dimethyl acetal using CSA in methanol yielded the corresponding dihydroxy derivatives 23α and 23β, in quantitative yields; Scheme 5.

Anticancer Activity. First, we evaluated the cytotoxicity of our carbohydrate derivatives against MRC-5 human nonmalignant lung cells and A549 human lung cancer cells using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay under the same experimental conditions. Both cell lines were exposed to different concentrations of compounds 12α, 12β, 13α, 13β, 14α, 14β, 22α, 22β, 23α, and 23β during 48 h before quantifying the cell viability. Aprepitant was used as a reference control to assess the activity of the new derivatives with a known NK1R antagonist, and Cisplatin, a well-known anticancer standard drug, was used as a positive control to study the possible selective cytotoxicity. The obtained results are collected in Table 1 (see Figures S2 and   S3 ). All of the carbohydrate derivatives showed some selective cytotoxicity against the cancer cell line, i.e., for a given concentration, A549 cancer cells were more sensitive than MRC-5 normal cells. 12α, 12β, and Aprepitant showed the lowest selective activity, with selectivity index values ∼1.5 (Table 1) . 13α, 13β, 23α, and 23β showed a modest selectivity, with IC 50 values in A549 ∼2-fold lower than in MRC-5. Interestingly, both α and β diastereomers of 14 and 22 showed the highest selective cytotoxicity against cancer cells, even higher than that of the anticancer drug Cisplatin. 14α and 22β were the most selective compounds. A549 cancer cells were over 10 times more sensitive than MRC-5 cells to these derivatives.

It should be noted that, in general, the carbohydrate derivatives showed higher selectivity than Aprepitant. D-Galactosyl derivatives were the most active compounds, with similar activity to Aprepitant and Cisplatin against A549 cancer cells. Deprotection of the hydroxyl groups at 3, 4, and 6 positions increased the selective activity in 13α and 13β, but it also decreased their cytotoxic activity. The introduction of the 4,6-O-benzylidene acetal in diastereomers 14 increased the selective activity without compromising the cytotoxicity against cancer cells. Indeed, 14α showed similar cytotoxicity as Cisplatin against A549 cancer cells, being less cytotoxic against MRC-5 normal cells. L-Arabinose derivatives 22α and 22β also showed high selective anticancer activity; however, they were less cytotoxic than the galactose derivatives 14 against cancer cells. 14α was ∼2.5-times more cytotoxic against A549 than 22β, being both the most selective anticancer derivatives. For that reason, 14α was selected to delve into its anticancer activity.

Next, we used MCF7 breast cancer cells, MCF 10 normal breast epithelial cells, UACC-62 melanoma cells, and VH10 skin nonmalignant cells to explore whether the new carbohydrate derivatives were also selective against other types of cancer. These cells were exposed to several concentrations of the compound 14α for 48 h, and cell viability was measured by the MTT assay ( Figure 3 ). 14α was ∼10 times more cytotoxic against MCF7 breast cancer cells than against MCF 10 normal cells. The IC 50 values (mean ± The selectivity index is the mean of the selectivity indices calculated in each individual experiment. The selectivity index is calculated by dividing the IC 50 standard error of mean (SEM); μM) in MCF7 and MCF 10 cells were, respectively, 23.9 ± 5.0 and 291.3 ± 52.6. 14α also showed selective cytotoxic activity against melanoma cells. UACC-62 melanoma cells were 4.3 times more sensitive than VH10 skin normal cells to the cytotoxic effect of 14α. The IC 50 values (mean ± SEM; μM) in UACC-62 and VH10 were, respectively, 31.9 ± 6.0 and 117.9 ± 12.3. It is worth mentioning that 14α showed similar cytotoxicity against the three cancer cell lines, with IC 50 values between 25 and 30 μM.

As 14α displayed cytotoxicity against three cancer cell lines (lung cancer, breast cancer, and melanoma), we decided to assess whether 14α was also active against other types of cancer. Thirteen human cancer cells lines and one human nonmalignant cell line were treated with 14α for 96 h. We used longer exposure times because some of these cell lines have long cell cycles (24−48 h); this longer exposure time allows the cells to pass several cycles in the presence of 14α. Cell viability was estimated with the resazurin assay. HaCaT human skin nonmalignant cells were used to study selectivity; these cells are derived from normal adult tissue and have a division rate similar to that of cancer cells. One key limitation of most anticancer drugs is that, in addition to targeting cancer cells, they also target nonmalignant cells with similar division rates. Indeed, results in Table 2 show that our positive control, Cisplatin, did not spare normal HaCaT cells from its cytotoxicity, with 14α being more selective than Cisplatin. Compound 14α showed a modest selectivity against HT29 (colorectal cancer) and A64-CLS (submaxillary gland adenoma) cells and a very low selectivity against T24 (bladder cancer) and MDA-MB-231 (triple-negative breast cancer). Interestingly, 14α showed a marked selective cytotoxic activity against GAMG (glioblastoma), HNO97 (tongue cancer), MeWo (melanoma), PC-3 (prostate), Sk-Br-3 (HER2-positive breast cancer), An3Ca (endometrial cancer), and Sk-OV-3 (ovarian cancer). These cell lines were at least 3 times more sensitive than the normal cell line to 14α treatment. HepG2 (hepatocarcinoma) and KATO III (gastric cancer) were at least 9-fold more vulnerable to 14α than HaCaT nonmalignant cells. These data show that 14α induces selective anticancer activity against a variety of cancer cell lines and suggest that our D-galactose derivative has anticancer potential. The key role of NK1R in cell proliferation and the elevated expression of NK1R identified in several cancer types 31 could explain the higher sensitivity of cancer cell lines (A549, MCF7, UACC-62, GAMG, HNO97, MeWO, PC-3, HepG2, KATO III) than normal cell lines (MRC-5, MCF 10, VH10, HaCat) to compound 14α.

Study of the Mechanisms of Anticancer Activity. Our next aim was to study the possible mechanisms involved in the selective anticancer activity of 14α. Because tumor cells rely on glycolysis for survival more than normal cells, 32 we evaluated if our compound behaved as a glycolysis inhibitor. Since glycolysis consumes glucose and produces lactate, we measured the concentrations of glucose and lactate in untreated A549 cells and in cells exposed for 8 h to 14α and to the known glycolysis inhibitor dichloroacetate. Unlike dichloroacetate, 14α did not reduce glucose consumption or lactate production, therefore indicating that 14α does not inhibit glycolysis.

It is known that cancer cells have higher basal levels of reactive oxygen species (ROS) than normal cells, making them more sensitive to exogenous induction of ROS. 33 Several studies have shown that Aprepitant and other NK1R antagonists as SR140333 increased mitochondrial ROS production, leading to apoptosis. 34 Therefore, we decided to study whether the generation of ROS was involved in the selective cytotoxic effect of 14α. A549 cancer cells were treated with 14α for 48 h in the presence or absence of three antioxidants: N-acetylcysteine (antioxidant activity through the glutathione system), catalase (hydrogen peroxide-degrading enzyme), and Mn(III) tetrakis(1-methyl-4-pyridyl)porphyrin pentachloride (MnTMPyP; superoxide anion scavenger). None of these three antioxidants reduced the cytotoxicity of 14α on A549 cells. These results suggest that the anticancer effect of 14α is not mediated by ROS production.

Several anticancer drugs (e.g., antimetabolites) inhibit DNA synthesis, block replication fork progression, and generate DNA damage that ultimately leads to cell death. In normal cells, this type of DNA damage is usually repaired by homologous recombination (HR). 35 However, some cancers are HR-deficient and, therefore, they are more sensitive to these drugs. 36 Because data suggest that NK1R blockade decreases the synthesis of DNA through the MAPK pathway, 31a,37 we tested whether HR-deficient cells were more sensitive to 14α. The HR-deficient VC8 cell line (V79 Chinese hamster lung cells mutated in BRCA2) and the HRproficient VC8B2 cell line (VC8 cells complemented with human BRCA2) were treated with 14α for 24 h. After a recovery period of 48 h, cell viability was estimated with the MTT assay. The HR-deficient VC8 cells were slightly more sensitive to 14α than the HR-proficient VC8B2 cells. IC 50 values (mean ± SEM; μM) in VC8 and VC8B2 were, respectively, 36.5 ± 4.9 and 63.7 ± 11.1. These data suggest that the possible generation of DNA damage by 14α, which could be repaired by HR, might play a minor role in the cytotoxicity of this compound.

Compounds. Finally, tests were carried out to, first, confirm the NK1R antagonist activity of the new carbohydrate derivatives and, then, its correlation with the observed anticancer effect. For this, we have chosen three representative derivatives for their chemical structure and their antitumor activity including the D-galactosyl derivative 14α, which provided the best effect, its trihydroxylated analogue 13α, exhibiting a lower anticancer activity, and an arabinose derivative 23β, exhibiting an intermediate activity.

The antagonist activity of these compounds was determined by their ability to inhibit NK1R using the IPone test. This test quantifies the inositol monophosphate (IP1) accumulated inside the cell by time homogeneous fluorescence (HTRF) technology. The accumulation of IP1 is an indicator of the activation of NK1R, so that the NK1R agonist ligands cause an increase in the levels of IP1 in the absence of SP while, on the contrary, the antagonist ligands produce a decrease of these levels in the presence of the endogenous ligand SP. As a control compound, we used the N-acetyl-L-typtophan 3,5(bistrifluorometil)benzyl ester derivative (L732,138), whose NK1R antagonist activity is well-known at the molecular level.

As shown in Figure 4 , both D-galactosyl derivatives, 13α and 14α, as well as the arabinose derivative 23β, exhibit significant inhibitory effect on the SP activity and can therefore be considered as NK1R antagonists. Specifically, the k inact values for the galactosyl derivatives 13α and 14α are 0.651 ± 0.239 and 0.209 ± 0.103 μM, respectively, and 0.494 ± 0.047 μM for the arabinosyl derivative 23β; Figure 4 . Interestingly, by comparing the potency of the NK1R inhibitory activity of the three synthetic derivatives (Table 3 , entries 2−4), as well as that known of Aprepitant (Table 3 , entry 1), with their anticancer activity against the lung cancer cell line A549, determined previously, we note that there is a clear correlation between both activities; Table 3 .

Indeed, Aprepitant (Table 3 , entry 1) > 14α (Table 3 , entry 3) > 23β (Table 3 , entry 4) > 13α (Table 3 , entry 2) for both activities. Although the exact mechanism(s) of the antitumoral activity of NK1R antagonists is poorly understood, and is currently the subject of intense research, 39 the results described indicate that the anticancer activity obtained with the . Concentration−response curves for the agonist and antagonist effects of SP (blue), L732,138 (olive green), α-D-galactosyl derivatives (13α green; 14α pink), and β-D-arabinosyl derivative 23β (black) using the IPone assay. The IP 1 accumulation (pmoles) was measured as described in the Experimental Section. Data are illustrated from a representative experiment performed at least three times.

carbohydrate analogues is mediated, at least in part, by the NK1 receptor.

Next, a binding affinity assay conducted by measuring the displacement of [ 125 I]SP from the hNK1R from U-373MG cells 40 reveals that 14α has an excellent affinity for NK1R with an IC 50 = 50.4 nM and K i = 22.4 nM. The improved NK1R antagonist activity of 14α over the trihydroxylated analogue 13α is likely due to the presence of the benzylidene acetal moiety, which can establish additional stabilizing interactions with the NK1 receptor, as has been confirmed in modeling calculations.

Docking Calculations. To gain knowledge about the interactions involved in the molecular recognition among ligands 13α, 13β, 14α, 14β, Aprepitant, and NK1R, we performed docking calculations. The obtained docking scores were −9.5, −9.6, −12.0, −12.4, and −11.2 kcal/mol for 13α, 13β, 14β, 14α, and Aprepitant, respectively. The capability of docking calculations has been widely recognized for precise predictions of the optimal ligand binding geometries as well as binding interactions. 41 Thus, using the obtained docking poses and analyzing the hydrophobicity of the binding site, we characterize the intermolecular interactions between the studied ligands and NK1R.

The binding site of NK1R has been previously described, 22b and it consists of a deep concave pocket with a large hydrophobic region that can maximize favorable protein− ligand contacts. For this reason, we used the octanol−water partition coefficient (or log P) of the different fragments of the ligands to obtain information about the interaction patterns of the ligands in the binding site. 42 Docking results revealed that all of the carbohydrate derivatives show a noticeable spatial overlap of the hydrophobic fragments 3,5-bis(trifluoromethyl)phenyl and 4-fluorophenyl, with root-mean-square deviation (RMSD) values from 0.115 to 0.207 Å for the former and from 0.120 to 0.950 Å for the latter ( Figure 5 ). The hydrophobicity surface of the binding pocket was performed because according to complementary ligand−protein binding interactions, strong hydrophobic regions of a binding site are usually occupied by hydrophobic fragments of ligands. Figure 6 reveals that most of the amino acids located at the bottom of the binding site have hydrophobic or amphipathic character (Ile113, Ile116, Ile204, Ala294, Met81, Met295, Met291, Phe110, Phe111, Phe264, Trp261, Tyr196). On the other hand, Figure 6A shows a top view of the middle-region binding site, which is dominated by charged/hydrophilic amino acids with large H-bonding donor/acceptor capacities (Gln165, His108, Asn109). On the contrary, Figure 6B indicates that hydrophobic amino acids are located at the down view of the middle region (Phe264 and Phe268). In addition to this, a strong hydrogen-bonding acceptor residue, Glu193, is found on the left side of the top view of the binding site outer region, which is accompanied by other amino acids with noticeable H-bonding donor/acceptor capacities (Trp184, Tyr196, Hys197). By contrast, the opposite side of this outer region is mainly surrounded by Ala93, Phe287, Ile283, and Tyr287 residues.

The analysis of the ligand−protein interactions of the docked complexes (Figure 7) shows that the hydrophobic fluorinated aromatic fragments, 3,5-bis(trifluoromethyl)phenyl (log P = 3.70) and p-fluorophenyl (log P = 2.10), in all of the studied carbohydrate derivatives are stabilized at the bottom side of the binding site through a large number of hydrophobic interactions, π−π stacking interactions with Phe268, His197, and Phe264, π−alkyl interaction with Pro112, and alkyl−alkyl interactions with Met295, Phe264, Met291, Ile113, Ile204, and Trp261 ( Figure 7) . Besides, two types of electrostatically driven interactions were observed (values in parenthesis reflect the interaction distances range found): (i) weak hydrogen bonds between fluoromethyl fragments and amino groups of Asn89 (2.51−2.65 Å) and Thr201 (2.36−3.22 Å), and (ii) halogen bonding between some fluorine atoms of the 3,5bis(trifluoromethyl)phenyl fragment and the carbonyl group of Pro112 (3.36−3.48 Å). These latter interactions are present in many protein−ligand complexes. 43 The nonhydrophobic residues Gln165, Asn109, and His108 are in the central part of the binding pocket, where the Osubstituted groups and central moieties of the ligands are placed. Oxygen atoms O-1 and O-2 and the amino group of Gln165 established two hydrogen bonds with lengths of 2.52 and 2.67 Å for 14α and 2.49 and 2.91 Å for 13α (Figure 7) , which largely contributes to the stabilization of the central moieties of these ligands in the binding site. Such a hydrogen bond was also found for 14β (2.62 Å) and none for 13β ( Figure 7) .

The calculations revealed that the pyranosyl fragment in 14α and 14β is a moderate hydrophilic moiety (log P = −0.42), which can make close contacts with the polar amino acids Asn109 and His108 ( Figure 6A ). On the other hand, Figure 6B reveals that the large hydrophobic character of the opposite side of the middle binding region is mainly caused by the presence of Phe268 and Phe264 residues, so a small ligand hydrophobicity would tend to have a weaker binding ability with these residues, but the moderate hydrophilicity of the pyranose fragment in 14α and 14β will result in low docking score penalties due to its interaction with these residues. In addition to this, the oxygen atom O-6 participates in the formation of a strong hydrogen bond with Tyr287 in 14α (2.16 Å) but not in 14β, suggesting that the former shows better structural requirements for the docking.

Regarding compounds 13α and 13β, the different spatial orientations of their hydroxylated pyranosyl fragments lead to different intermolecular interactions. The amino group of the Glu165 residue establishes a hydrogen bond (2.82 Å) with the oxygen atom of the pyranose fragment of ligand 13α. Moreover, the carbonyl group of Asn109 (2.52 Å) and the basic nitrogen of His108 (2.16 Å) form hydrogen bonds with the hydroxyl group at the 6 position. In the case of 13β, the hydroxyl groups at 3 and 4 positions and the oxygen atom of Tyr268 act as a H-bonding donor (2.17 Å) and acceptor (2.00 Å), respectively. The log P value of this pyranosyl fragment (−1.29) is significantly more negative than that obtained for the galactosyl derivative (see above), which suggests that the observed poor docking score in these ligands in part may be related to the extremely low hydrophobicity of the pyranosyl fragment, which is partially surrounded by a hydrophobic environment ( Figure 6B ).

Concerning Aprepitant, two hydrogen bonds are formed between the amino group of Glu165 and an oxygen (2.29 Å) and nitrogen (2.67 Å) atom, located on opposite sides of the morpholine moiety. This provokes an optimal anchoring of this fragment within the central region of the binding site. Besides, the log P value of this fragment (−0.41) is similar to that of the pyranosyl fragment of galactosyl derivatives, which allows a low scoring penalty due to the interactions with hydrophobic residues of the middle binding region.

Finally, together with Aprepitant, the only ligands that have fragments on the binding site outer region are 14α and 14β.

The first one has the triazolone group, while the other two have the phenyl group of the benzylidene acetal fragment. These fragments are docked in different binding regions ( Figure 6 ). Thus, the triazolone group (log P = −0.64) is located in a more polar region of this binding site region and it interacts through strong H-bonds with Glu193 (2.04 Å) and Trp184 (1.67 Å) and establishes π−π staking interactions and π−alkyl interactions with His197 and Ile182, respectively. By contrast, the benzylidene groups (log P = 1.94) of 14α and 14β are surrounded by apolar residues like Ala93 and Phe267 and they form π−π staking interactions with Tyr287.

The obtained results show that hydrophobic and π−π stacking interactions play an important role in the binding affinity at the bottom of the binding site. By contrast, hydrogen-bond interactions seem to be the key factor for the docking in the middle binding region, for which the Gln165 residue plays an important role; however, a low hydrophobicity value of the ligand moiety docked in this region seems to provide an unfavorable binding factor. Finally, the presence of molecular fragments in the outer binding region is needed to increase the docking score. The different nature of residues in separate locations of the outer region binding site allows either a hydrophilic fragment, through hydrogen-bonding interactions, or a hydrophobic one, through strong hydrophobic interactions, to increase the binding affinity.

In summary, we have reported the synthesis of a family of compounds designed as NK1R antagonists, using carbohydrate as the central scaffold. The use of sugars as starting substrates greatly facilitates the obtaining of final products as single isomers, important both for their biological activity and for their possible translation to the market. In addition, the multiple hydroxyl groups allow the regioselective anchoring of different substituents, which facilitates the modulation of the lipophilic/hydrophilic balance and optimizes the bioavailability of the final products. The products synthesized showed a strong affinity and antagonist activity against NK1R and were furthermore shown to be broad-spectrum anticancer agents with high selectivity, comparable to Cisplatin. Among all of the analogues tested, compound 14α derived from galactose and A clear correlation between antagonist and anticancer activities was observed by comparing the potency of the NK1R inhibitory activity of this galactosyl derivative 14α with two other synthetic derivatives (13α and 23β) and that of Aprepitant.

Interestingly, this galactosyl derivative has shown marked cytotoxic activity against 12 different types of cancer cell lines. Even more interesting is the selectivity observed, with cancer lines being up to 20 times more sensitive than nonmalignant cell lines to the treatment with 14α. Docking studies on selected carbohydrate derivatives have provided new information on the key role of the 4,6-O-benzylidene acetal fragment of 14α as the most likely responsible for its higher NK1R affinity. Taken together, these results strongly support the possibility of using carbohydrate-based NK1R antagonists as selective anticancer drugs, with the product 14α being an interesting compound toward this end.

Chemistry: General Procedures. For the reactions that were run under an atmosphere of dry argon, oven-dried glassware and dried solvents were used. Chemicals were obtained from commercial sources and were used without further purification. Thin-layer chromatography (TLC) was carried out on silica gel GF254 (Merck), and compounds were detected by charring with phosphomolybdic acid/EtOH or sulfuric acid/EtOH. For flash chromatography, Merck 230−400 mesh silica gel was used. Chromatographic columns were eluted with a positive pressure of air, and eluents are given as volume-to-volume ratios (v/v). Nuclear magnetic resonance (NMR) spectra were recorded with Bruker Avance 500 MHz spectrometers. Chemical shifts are reported in ppm, and coupling constants are reported in Hz. High-resolution mass spectra (HRMS) were recorded in the Centro de Investigacioń, Tecnología e Innovacioń in the University of Seville with a Kratos MS-80RFA 241-MC apparatus. Different ionization methods than chemical ionization (CI) are indicated. Optical rotations were determined with a Perkin-Elmer 341 polarimeter. Melting points were measured with a Stuart SMP3 apparatus in open-ended capillary tubes. A Waters Alliance 2690 HPLC instrument with an ACE Excel C18 (4.6 mm × 100 mm, 2 μm particle size) column was used for analytical high-performance liquid chromatography (HPLC) analyses. The elution conditions were as follows: CH 3 CN/H 2 O, 70% (v/v) CH 3 CN gradient with 0.1% formic acid in 7 min, flow rate 1.0 mL/ min, calculation of the relative purity of each compound at 254 nm. For compound 22β, 80% (v/v) CH 3 CN gradient with 0.1% formic acid in 7 min was used. The purity of all tested compounds was above 95%.

Phenyl 2,3,4,6-Tetra-O-acetyl-1-thio-β-D-galactopyranoside (2β). 24 To a solution of 3.90 g of D-galactose pentaacetate (10.00 mmol) in dry dichloromethane (40 mL) at 0°C under an argon atmosphere, 4.93 mL of boron trifluoride etherate (40.00 mmol) was added dropwise. After 15 min of stirring at room temperature, 1.07 mL of thiophenol (10.50 mmol) was added. After stirring overnight, the starting material was consumed. The reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The aqueous phase was extracted with dichloromethane (2 × 40 mL), and the combined organic phases were washed with saturated NaCl aqueous solution and dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:4) to obtain 3.96 g of 2β (9.00 mmol, 90% yield) as a white solid; R f = 0.79 (EtOAc/ hexane Phenyl 1-Thio-β-D-galactopyranoside (3β). To a solution of 3.96 g of 2β (9.00 mmol) in methanol, at 0°C under an argon atmosphere, 36.00 mL of a 1 M sodium methoxide methanolic solution (36.00 mmol) was added dropwise. After stirring for 35 min, the starting material was consumed. The reaction mixture was neutralized with acid resin and filtered to obtain 2.40 g of 3β (8.80 mmol, quantitative yield) as a white solid, which was used in the next reaction without further purification; Phenyl 3,4-O-Isopropylidene-1-thio-β-D-galactopyranoside (4β). To a suspension of 2.21 g of 3β (8.10 mmol) in 60.00 mL of 2,2dimethoxypropane (2,2-DMP) at room temperature under an argon atmosphere, 7.00 mg of 10-camphorsulfonic acid (CSA) (0.03 mmol) was added. After stirring for 48 h, the reaction mixture was neutralized with triethylamine and filtered to remove the ammonium salts formed. The solvent was evaporated under reduced pressure, and the residue obtained was dissolved in the minimum possible amount of toluene and evaporated to dryness. This process was repeated twice to obtain the mixed acetal with a small amount of the desired diol. Then, the crude obtained was dissolved in the minimum possible amount of methanol and a catalytic amount of CSA at 0°C was added. After 5 min of stirring at room temperature, the reaction mixture was neutralized with triethylamine, the ammonium salts were filtered, and the solvent was evaporated under reduced pressure. The residue obtained was dissolved in toluene and evaporated to dryness. This process was repeated twice, and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:2) to obtain 2.16 g of 4β 

To a solution of 4.00 g of 5β (6.07 mmol) in THF (80.00 mL) at room temperature under an argon atmosphere, a solution of 0.73 g of sodium hydride (18.25 mmol) in THF (10.00 mL) was added. After 1 h, 0.90 g of Bu 4 NI (2.43 mmol) was added and the reaction mixture was stirred for 30 min. Then, a solution of 1.10 mL of p-fluorobenzyl chloride 6 (9.10 mmol) in THF (5.00 mL) was added. After 48 h, the reaction mixture was quenched with a saturated NH 4 Cl aqueous solution. The aqueous phase was extracted with ethyl acetate (3 × 40 mL), and the combined organic phases were washed with a saturated NaCl aqueous solution and dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:8) to obtain 3. 8(α,β) ). To a solution of 3.02 g of 7β (4.59 mmol) in acetone/water 99:1 (120.00 mL) in the darkness at −15°C, 1.03 g of N-bromosuccinimide (NBS) (5.78 mmol) was added. After 30 min, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The aqueous phase was extracted with dichloromethane (3 × 40.00 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:8) to obtain 2.37 g of a mixture of both anomers 8α:8β in a 2:1 ratio (4.18 mmol, 91% yield) as a yellow syrup; R f = 0.72 (EtOAc/hexane, 1:4). 

To a solution of 2.12 g of 8(α,β) (3.74 mmol) in 50 mL of cyclohexane/dichloromethane 4:1 at room temperature under an argon atmosphere, a catalytic amount of DBU (0.23 mL, 1.50 mmol) and 0.96 mL of a 98% 2,2,2trichloroacetonitrile solution was added dropwise. After stirring overnight, the reaction mixture was quenched with water, and the organic phase was washed with a saturated NaCl aqueous solution and dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure to obtain 2.58 g of trichloroacetimidate tertbutyldiphenylsilyl-2-O-(p-fluorobenzyl)-3,4-O-isopropylidene-6-Oα,β-D-galactopyranoside, 9(α,β) (3.63 mmol, 97% yield), as a mixture of both anomers α:β in a 1:1 ratio, as a yellow syrup, which was used directly in the next reaction without further purification; R f = 0.69 (EtOAc/hexane, 1:6). To a solution of 0.16 g of the obtained tricloroacetimidate 9(α,β) (0.22 mmol) and 0.17 g of (R)-1-[3,5-bis(trifluoromethyl)phenyl]ethanol (0.66 mmol) with 200.00 mg of a molecular sieve (4 Å) in 7.00 mL of ether at 0°C under an argon atmosphere, 2.80 mL of trimethylsilyl trifluoromethanesulfonate (0.02 mmol) was added dropwise. After stirring for 1 h at room temperature, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The resulting suspension was filtered through a pad of Celite. The aqueous phase was extracted with dichloromethane (3 × 40 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure to obtain 0.15 mg (0.19 mmol, 85% yield) of a mixture of both anomers 11α:11β in a 3:1 ratio, as a yellow syrup. After purification by flash chromatography (EtOAc/hexane, 1:15), 104.00 mg of 11α (0.13 mmol, 60% yield) and 44 mg of 11β (0.06 mmol, 25% yield) were obtained.

Method B. To a solution of 0.20 g of thioglycoside 7β (0.30 mmol) and 0.23 g of (1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethanol (0.91 mmol) with 200.00 mg of a molecular sieve (4 Å) in 8.00 mL of dichloromethane at 0°C under an argon atmosphere, 3.90 mL of trimethylsilyl trifluoromethanesulfonate (0.02 mmol) and 0.30 g of NIS (1.50 mmol) were added. After stirring for 1 h at room temperature, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The resulting suspension was filtered through a pad of Celite. The aqueous phase was extracted with dichloromethane (3 × 40 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure to obtain 193.0 mg (0.24 mmol, 81% yield) of the mixture of both anomers 11α:11β in a 1:1 ratio, as a yellow syrup; R f = 0.81 (EtOAc/hexane, 1:6). After purification by flash chromatography (EtOAc/hexane, 1:15), 97.00 mg of 11α (0.12 mmol, 40% yield) and 97.00 mg of 11β (0.12 mmol, 40% yield) were obtained. To a solution of 100.00 mg of 13β (0.19 mmol) in DMF (15.00 mL) and 0.32 mL of dimethoxymethyl benzene (0.21 mmol) under an argon atmosphere, a catalytic amount of CSA was added. After stirring for 1 h in vacuo at 40°C, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The aqueous phase was extracted with dichloromethane (3 × 40 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:4) to obtain 100.00 mg of 14β (0.18 mmol, 95% yield) as a white solid; 1.1 ppm) .

Phenyl 3,4-O-Isopropylidene-1-thio-α-L-arabinopyranoside (18α). To a suspension of 3.91 g of 17α (16.20 mmol) in 120.00 mL of 2,2-dimethoxypropane (2,2-DMP) at room temperature under an argon atmosphere, 7.00 mg of 10-camforsulfonic acid (CSA) (0.03 mmol) was added. After stirring for 30 min, the reaction mixture was neutralized with triethylamine and filtered to remove the ammonium salts formed. The solvent was evaporated under reduced pressure, and the residue obtained was dissolved in the minimum possible amount of toluene and evaporated to dryness. This process was repeated twice to obtain the mixed acetal with a small amount of the desired diol. Then, the crude obtained was dissolved in the minimum possible amount of methanol and a catalytic amount of CSA at 0°C was added. After 5 min of stirring at room temperature, the reaction mixture was neutralized with triethylamine, the ammonium salts were filtered, and the solvent was evaporated under reduced pressure. The residue obtained was dissolved in toluene and evaporated to dryness. This process was repeated twice, and the reaction crude was purified by flash chromatography (EtOAc/hexane, 1:3) to obtain 4. To a solution of 2.80 g of the obtained tricloroacetimidate 21(α,β) (4.70 mmol) and 1.21 g of (1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethanol (4.70 mmol) with 200.00 mg of a molecular sieve (4 Å) in 95.00 mL of ether at 0°C under an argon atmosphere, 0.11 mL of trimethylsilyl trifluoromethanesulfonate (0.01 mmol) was added dropwise. After stirring for 24 h at room temperature, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The resulting suspension was filtered through a pad of Celite. The aqueous phase was extracted with dichloromethane (3 × 40 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure to obtain 1.68 g (3.12 mmol, 66% yield) of a mixture of both anomers 22α:22β in a 1:2 ratio, as a yellow syrup. After purification by flash chromatography (EtOAc/hexane, 1:20; and acetone/hexane, 1:50), 420.00 mg of 22α (0.75 mmol, 16% yield) and 1.26 mg of 22β (2.31 mmol, 50% yield) were obtained.

Method B. To a solution of 4.85 g of thioglycoside 19α (12.42 mmol) and 9.62 g of (1R)-1-[3,5-bis(trifluoromethyl)phenyl]ethanol (37.26 mmol) with 300.00 mg of a molecular sieve (4 Å) in 150.00 mL of dichloromethane at 0°C under an argon atmosphere, 14.00 g of NIS (62.10 mmol) was added. After stirring for 6 h at room temperature, the reaction mixture was quenched with a saturated NaHCO 3 aqueous solution. The resulting suspension was filtered through a pad of Celite. The aqueous phase was extracted with dichloromethane (3 × 40 mL), and the combined organic phases were dried with anhydrous Na 2 SO 4 . The solvent was evaporated under reduced pressure to give 4.41 g (8.20 mmol, 66% yield) of the mixture of both anomers 22α:22β in a 1:1 ratio, as a yellow syrup. After purification by flash chromatography (EtOAc/hexane, 1:4; and acetone/hexane, 1:50), 2.21 g of 22α (4.10 mmol, 33% yield), R f = 0.64 (EtOAc/hexane, 1:2), and 2.21 g of 22β (4.10 mmol, 33% yield), R f = 0.72 (EtOAc/hexane, 1:2), were obtained. at 37°C in a humidified atmosphere of 95% air and 5% CO 2 . Nonessential amino acids (Invitrogen) were also added to the media.

Transient transfection of the cell lines was performed using electroporation in a 300 μL volume with a total of 10 μg of DNA (pRK5 Neo-NK1 wild type) plasmid up to 500 ng plus pRK5 as carrier DNA to reach 10 μg containing 107 cells in an electroporation buffer (50 mM K 2 HPO 4 , 20 mM CH 3 COOK, 20 mM KOH, and 26 mM MgSO 4 , pH 7.4). After electroporation (280 V, 1 mF, Gene Zapper 450/2500; IBI, New Haven, CT), cells were suspended in a complete medium and seeded into 96-well culture plates at a density of 105 cells per well. First, 96-well culture plates were coated with polyornithine diluted in phosphate-buffered saline (PBS), incubated at 37°C for 30 min, and then rinsed with PBS before seeding.

Enzyme-Linked Immunosorbent Assay (ELISA). To measure the expression of the transfected receptors, cells were transfected with pRK5-NK1-6His. After 24 h of electroporation, cells were fixed with 4% paraformaldehyde in PBS for 5 min and rinsed three times with PBS. A blocking step of 30 min with PBS + 1% decomplemented FBS was performed before incubation with an anti-6 His primary antibody (0.5 μg/mL) for 30 min. The cells were then rinsed four times for 5 min in PBS + 1% FBS and incubated for 30 min with an antimouse antibody conjugated with horseradish peroxidase (1/1000; Amersham, Orsay, France). The cells were rinsed three times with PBS + 1% FBS and three times with PBS. Afterward, 60 μL of PBS and 20 μL of Supersignal ELISA Femto (Perbio-Pierce, Brebieres, France) were added to the wells. The luminescence was read using a Wallac Victor2 (PerkinElmer Life and Analytical Sciences, Courtaboeuf, France)

Second Messenger (IP1) Accumulation. Activation/inhibition of the IP pathway by NK1R agonists or antagonists, respectively, was determined using the IP-One dynamic kit (Cisbio Bioassays, Bagnolssur-Ceze, France). In brief, after transfection, 105 cells were distributed in 100 μL of complete medium into a 96-well assay plate (Greiner Bio-One, Courtaboeuf, France). Twenty-four hours later, the medium was removed and replaced with 40 μL of incubation medium containing the agonist and/or antagonist at the appropriate concentrations. The IP-One assay is based on the accumulation of IP 1 , a downstream metabolite of the IP pathway that is produced by phospholipase C activated by the Gq/11 protein; IP 1 is stable in the presence of LiCl. The homogeneous time-resolved fluorescence− fluorescence resonance energy transfer (HTRF−FRET) assay was performed as described previously. This assay involves the transfer of energy from a Lumi4TM-Terbium cryptate donor fluorophore to a d 2 acceptor fluorophore. The assay is an immunoassay that measures competition between native IP1 produced by the cells and IP1 labeled with the d 2 acceptor, as revealed by a monoclonal antibody against IP 1 labeled with Lumi4TM-Terbium cryptate. Fifteen microliters of antibody and 15 μL of competitor diluted in lysis buffer provided in the kits were added to the wells after a 30 min incubation at 37°C with the agonist. As a negative control, some wells only received the donor fluorophore-labeled antibody. After 1 h of incubation at room temperature, fluorescence emissions were measured at both 620 and 665 nm on a RubyStar fluorometer (BMG Labtechnologies, Offenburg, Germany) equipped with a nitrogen laser as the excitation source (337 nm). A 400 μs reading was recorded after a 50 μs delay to eliminate the short-lived fluorescence background from the acceptor fluorophore-labeled antibody. The fluorescence intensities measured at 620 and 665 nm correspond to the total europium cryptate emission and to the FRET signal, respectively. The specific FRET signal was calculated using the following equation

with R pos being the fluorescence ratio (665/620 nm) calculated in wells incubated with both donor-and acceptor-labeled antibodies and R neg being the same ratio for the negative control incubated only with the donor fluorophore-labeled antibody. The FRET signal (ΔF%), which is inversely proportional to the concentration of IP 1 in the cells, was then transformed into IP 1 accumulation using a calibration curve prepared on the same plate. It is worth noting that all comparisons of agonist or antagonist effects were done on the same day, on the same culture and plate, and were made against the SP effect. The experiments were repeated at least three times on different cultures. Values corresponding to the low basal activities, determined in unstimulated cells, were first subtracted. Activation/inhibition curves were plotted to the log of agonist or antagonist concentrations and fitted to the Hill equation to extract the EC 50 , the Hill coefficient, and minimal/maximal values. The inhibitory effect of the specific nonpeptidic NK1 antagonist on IP 1 accumulations induced by SP was studied according to Arunlakshana and Schild. 44 Preincubation for 10 min with the antagonist was followed by a 30 min incubation with the antagonist and SP. IP 1 accumulation was then measured as described above.

Cell Lines. MRC-5 (human fetal lung fibroblastic cells) and A549 (human nonsmall cell lung cancer cells) were purchased from the European Collection of Cell Cultures. HaCaT cells (human The IC 50 value (concentration causing a half-maximal inhibition of control specific binding) and the Hill coefficient (nH) were determined by nonlinear regression analysis of the competition curve generated with mean replicate values using the Hill equation curve fitting (Y = D + [(A − D)/(1 + (C/C 50 ) nH )], where Y is the specific binding, D is the minimum specific binding, A is the maximum specific binding, C is the compound concentration, C 50 is IC 50 , and nH is the slope factor). This analysis was performed using a software developed at Cerep (Hill software) and validated by comparison with data generated by commercial software SigmaPlot 4.0 for Windows (©1997 by SPSS Inc.).

The inhibition constant (K i ) was calculated using the Cheng Prusoff equation (K i = IC 50 /(1 + (L/K D )), where L is the concentration of the radioligand in the assay and K D is the affinity of the radioligand for the receptor). A scatchard plot is used to determine the K D .

Cell Viability Assay. Cell viability was estimated with the 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay or the resazurin assay. Both assays are redox-based colorimetric techniques based on the capability of viable cells to reduce the yellow product MTT or the blue reagent resazurin into a purple formazan dye or a pink-colored product, respectively. The number of live cells is directly proportional to the amount of the final product created. Exponentially growing cells were seeded in 96-well plates and were allowed to grow during 24 h. The cells were then treated with several concentrations of the tested compounds for 48−96 h (the incubation periods are specified in the figures and table legends) before measuring the cell viability using the MTT assay or the resazurin assay.

For the MTT assay, after the incubation period, the growth medium was removed and 125 μL of MTT solution (1 mg/mL in medium) was added to each well for 3−4 h. Then, 80 μL of 20% sodium dodecyl sulfate (SDS) in 20 mM HCl was added to dissolve the insoluble purple formazan product and plates were incubated for 15 h at 37°C. The optical density (OD) of each well was measured at 540 nm with a multiwell plate spectrophotometer reader to quantify cell survivals.

For the resazurin assay, after treatment, the medium was removed and 150 μL of resazurin solution (20 μg/mL in medium) was added to each well for 3−6 h. The OD of each well was measured at 540 and 620 nm on a multiwell plate spectrophotometer reader.

In both assays, results were expressed as the percentage of cell viability in relation to untreated cells (controls). All data were averaged from two to five independent experiments and were expressed as the means ± standard error of the mean (SEM). For statistical analysis, the t-test (paired, two-tailed) was used. A p value >0.05 is not considered statistically significant and is not represented by any symbol, a p value <0.05 is considered statistically significant and is indicated with an asterisk (*), a p value <0.01 is indicated with a double asterisk (**), and a p value <0.001 is indicated with a triple asterisk (***).

Selectivity indices (SIs) are useful to evaluate the anticancer potential in vitro. 45 SIs were calculated as the mean of the IC 50 value in the normal cell line divided by the IC 50 in the cancer cell line obtained in each independent experiment. Glycolysis inhibition was assessed by measuring concentrations of glucose (initial product of glycolysis) and lactate (final product of glycolysis) in control and treated cells. Briefly, 4 × 10 5 cells were seeded into 24-well plates. After 10 h, the medium of cells was renewed and drugs were added. Cells were exposed to the tested compounds for 8 h, and glucose and lactate concentrations were determined in cell supernatants using the Accutrend Plus analyzer together with Accutrend glucose strips and BM-Lactate strips (Roche Diagnostics). After calibrating the instrument with glucose and lactate calibration strips, test strips were used to determine glucose and lactate levels via colorimetric-oxidase mediator reactions according to the manufacturer's instructions. 46 Results are expressed as a percentage of lactate production and percentage of glucose consumption in relation to untreated cells and are shown as the means ± SEM of two independent experiments. Molecular Modeling. To validate the docking method used with AutoDock Vina, we redocked three cocrystallized ligands existing with the NKR1 proteinCP-99,994 (PDB ID: 6HLL), Aprepitant (PDB ID: 6HLO), and Netupitant (PDB ID: 6HLP)and then compared the obtained Cartesian coordinates of the docked ligand atoms with those of the native ones, using root-mean-square deviation (RMSD) values. All of the predicted docking poses presented RMSD values lower than 1.5 Å (0.893 Å for CP-99,994, 1.242 Å for Aprepitant, and 1.075 Å for Netupitant) when compared to the experimentally cocrystallized binding pose (see Figure S1 ). These results indicate that the used molecular docking protocol using AutoDock Vina is satisfactory for inferring the correct binding modes and the interactions of such ligands with NKR1.

Molecular structures of the ligands were optimized in the ground state at the DFT level with the B3LYP 47 and the 6-31G (d,p) basis set 48 implemented in the Gaussian 09 Rev.D.O1 package programs. 49 Molecular docking calculations were performed by AutoDock Vina 50 and AutodockTools software. 51 The structure of NKR1 was retrieved from the Protein Data Bank (PDB ID: 6HLO), and all water molecules and cocrystallized ligand were removed from crystallographic structures to prepare the docking receptor. The best docking poses and interactions involved in the binding mode were visualized with Discovery Studio Visualizer (Accelrys Software Inc.). 52 

Hz, 1H), 1.52 (d, J = 6.5 Hz, 3H), 1.41 (s, 3H), 1.35 (s, 3H) ppm; 13 C NMR (125 MHz

HRMS: calcd for C 25 H 25 O 5 F 7 Na

CDCl 3 ): δ 7.90 (bs, 2H), 7.82 (bs, 1H), 7.22−7.17 (m, 2H)

After stirring overnight, the solvent was evaporated under reduced pressure. The residue obtained was purified by flash chromatography (EtOAc/hexane, 1:1) to obtain 3.36 g of 23α (6.55 mmol, quantitative yield) as a white solid

Hz, 1H), 3.71 (dd, J = 3.5, 8.1 Hz, 1H), 3.59−3.56 (m, 1H), 3.47−3.43 (m, 2H), 2.84−2.80 (m, 1H), 1.53 (d, J = 6.5 Hz, 3H) ppm

To a solution of 1.47 g of 22β (2.75 mmol) in methanol (60.00 mL) at room temperature, a catalytic amount of CSA was added. After stirring overnight, the solvent was evaporated under reduced pressure. The residue obtained was purified by flash chromatography (EtOAc) to obtain 1

91−93°C; 1 H NMR (500 MHz, CDCl 3 ): δ 7.89 (bs, 2H), 7.83 (bs, 1H), 7.18−7.15 (m, 2H), 6.99− 6.95 (m, 2H)

CDCl 3 ): δ 162.7 (d, J CF = 246.5 Hz), 145.9, 133.4 (d, J CF = 3.3 Hz), 132.1 (q, J CF = 33.3 Hz), 129.8 (d, J CF = 8.2 Hz), 126.8 (q, J CF = 3.0 Hz)

HNO97 (human tongue cancer cells), A64-CLS (human submaxillary gland adenoma cells), AN3Ca (human endometrial adenocarcinoma cells), Sk-OV-3 (human ovarian cancer cells), KATO III (gastric cancer), Sk-Br-3 (HER2-positive breast cancer), T24 (bladder cancer), and MeWo (Melanoma; BRAF WT) were purchased from Cell Lines Service (CLS). MCF7 (human breast adenocarcinoma cells) and MCF 10 (human mammary epithelial cells) cell lines were a gift from

OV-3, HepG2, VC8B2, and VC8 were maintained in Dulbecco's modified Eagle's medium (DMEM) highglucose medium. PC-3 and T24 were grown in DMEM-F12. UACC-62 was maintained in RPMI 1640. Except for MCF 10, all media were supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 μg/mL streptomycin. The MCF 10 cell line was maintained in a 1:1 mixture of Dulbecco's modified Eagle's medium and Ham's F12 medium

Met(O2)11]-SP (1 μM) was used and the incubation time was extended to 30 min. The specific ligand binding to the receptors is defined as the difference between the total binding and the nonspecific binding determined in the presence of an excess of unlabeled ligand. All data were averaged from five independent experiments, and the results are expressed as a percent of control specific binding [(measured specific binding/control specific binding) × 100] and as a percent inhibition of control specific binding {100 −

Tachykinins and their receptors: contributions to physiological control and the mechanisms of disease

A comprehensive map of molecular drug targets

Tachykinins and their receptors: contributions to physiological control and the mechanisms of disease

Tachykinin NK1 receptor antagonists act centrally to inhibit emesis induced by the chemotherapeutic agent cisplatin in ferrets

Primary afferent tachykinins are required to experience moderate to intense pain

Neuropeptide receptors as potential drug targets in the treatment of inflammatory conditions

Distinct mechanism for antidepressant activity by blockade of central substance P receptors

The substance P/NK-1 receptor system: NK-1 receptor antagonists as anti-cancer drugs

The NK-1 receptor is expressed in human melanoma and is involved in the antitumor action of the NK-1 receptor antagonist aprepitant on melanoma cell lines

Substance P enhances forskol in stimulated cyclic AMP production in human UC11MG astrocytoma cells

Neurokinin-1 receptor expression and its potential effects on tumor growth in human pancreatic cancer

Increased expression of preprotachykinin-I and neurokinin receptors cells: Implications for bone marrow metastasis

Aprepitant as a combinant with dexamethasone reduces the inflammation via neurokinin 1 receptor antagonism in severe to critical Covid-19 patients and potentiates respiratory recovery: a novel therapeutic approach

A potent nonpeptide antagonist of the substance P (NK1) receptor

Antagonists at the neurokinin receptors -recent patent literature

Structural optimization affording 2-(R)-(1-(R)-3,5-bis(trifluoromethyl)-phenylethoxy)-3-(S)-(4-fluoro)phenyl-4-(3-oxo-1,2,4-triazol-5-yl)-methylmorpholine, a potent, orally active, long-acting morpholine acetal human NK-1 receptor antagonist

Fernández, I. Design, synthesis and biological studies of a library of NK1-receptor ligands based on a 5-arylthiosubstituted 2-amino-4,6-diaryl-3-cyano-4H-pyran core: switch from antagonist to agonist effect by chemical modification

Antagonistas de los receptores NK1 derivados de hidratos de carbono, metodo de obtencioń y uso medico. WIPO Patent WO2016189179A12016. (25) (a)

Identification of a series of 3-(benzyloxy)-1-azabicyclo[2.2.2] octane human NK1 antagonists

Derivation of a three-dimensional pharmacophore model of substance P antagonists bound to the neurokinin-1 receptor

Biological and conformational evaluation of bifunctional compounds for opioid receptor agonists and neurokinin1 receptor antagonists possessing two penicillamines

Phosphinite thioglygosides as useful ligands for palladium catalyzed asymmetric substitution: synthesis of both enantiomers using natural sugars as catalyst precursors

Selective excitation 1D-NMR experiments for the assignement of the absolute configuration of secondary alcohols

Recent advances in pharmacotherapy of chemotherapy-induced nausea and vomiting

Antitumoral action of the neurokinin-1 receptor antagonist L-733 060 on human melanoma cell lines

NK1R Antagonist decreases inflammation and metastasis of breast carcinoma cells metastasized to liver but not to brain; phenotype-dependent therapeutic and toxic consequences

Case report: sodium dichloroacetate (DCA) inhibition of the "Warburg effect" in a human cancer patient: complete response in non-Hodgkin's lymphoma after disease progression with rituximab-CHOP

The two faces of reactive oxygen species in cancer

Reactive oxygen species are second messengers

of neurokinin signaling in peripheral sensory neurons

Inhibition of DNA synthesis is a potent mechanism by which cytostatic drugs induce homologous recombination in mammalian cells

Biological and pharmacological aspects of the NK1-receptor

In vitro and in vivo pharmacological characterization of the novel NK1 receptor selective antagonist netupitant

The NK1 receptor antagonist NKP608 inhibits proliferation of human colorectal cancer cells via Wnt signaling pathway

Functional characterization of neurokinin-1 receptors on human U373MG astrocytoma cells

The Log P parameter as a molecular descriptor in the computer-aided drug design−an overview

DeepBindRG: a deep learning based method for estimating effective protein-ligand affinity

Halogen bonding in ligand−receptor systems in the framework of classical force fields

Halogen interactions in protein−ligand complexes: implications of halogen bonding for rational drug design

c) López-Lázaro, M. Two preclinical tests to evaluate anticancer activity and to help validate drug candidates for clinical trials

Synergistic antipancreatic tumor effect by simultaneously targeting hypoxic cancer cells with HSP90 inhibitor and glycolysis inhibitor

Selfconsistent molecular orbital methods XII. Further extensions of gaussiantype basis sets for use in molecular orbital studies of organic molecules

AutoDock vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading

AutoDock4 and AutoDockTools4: automated docking with sSelective receptor flexibility

Discovery studio visualizer, version 4.0; software for viewing, sharing, and analyzing protein and modeling data

Author Contributions # R.R. and P.L. contributed equally to this work.

The authors declare no competing financial interest. ■ ABBREVIATIONS NK1R, neurokinin 1 receptor; SP, substance P; CarbNK1-RAnt, carbohydrate-based NK1R antagonists; GalNK1RAnt, galactose-derived NK1R antagonist; AraNK1RAnt, arabinosederived NK1R antagonist