Synthesis of new lophine–carbohydrate hybrids as cholinesterase inhibitors: cytotoxicity evaluation and molecular modeling

A series of selective butyrylcholinesterase inhibitors were obtained. The absence of in vitro cytotoxicity and good ADME-Tox profile make these compounds new promising prototypes for the treatment of Alzheimer's disease.

Abstract

In this study, we synthesized nine novel hybrids derived from d-xylose, d-ribose, and d-galactose sugars connected by a methylene chain with lophine. The compounds were synthesized by a four-component reaction to afford the substituted imidazole moiety, followed by the displacement reaction between sugar derivatives with an appropriate N-alkylamino-lophine. All the compounds were found to be the potent and selective inhibitors of BuChE activity in mouse serum, with compound 9a (a d-galactose derivative) being the most potent inhibitor (IC₅₀ = 0.17 μM). According to the molecular modeling results, all the compounds indicated that the lophine moiety existed at the bottom of the BuChE cavity and formed a T-stacking interaction with Trp231, a residue accessible exclusively in the BuChE cavity. Noteworthily, only one compound exhibited activity against AChE (8b; IC₅₀ = 2.75 μM). Moreover, the in silico ADME predictions indicated that all the hybrids formulated in this study were drug-likely, orally available, and able to reach the CNS. Further, in vitro studies demonstrated that the two most potent compounds against BuChE (8b and 9a) had no cytotoxic effects in the Vero (kidney), HepG2 (hepatic), and C6 (astroglial) cell lines.

Introduction

Alzheimer's disease (AD) is the most common form of neurodegenerative dementia characterized by a decline in cognitive function and memory. Due to the progressive aging of the population, AD is becoming a healthcare burden of epidemic proportions.1 Although the pathogenesis of AD is still poorly understood, many hypotheses have been proposed over the last three decades. Among them, the cholinergic hypothesis is the earliest approved, which recognizes that increasing the brain levels of acetylcholine (ACh) to enhance cholinergic neurotransmission is an efficacious approach for AD treatment.2,3 Since 1998, over 100 drugs have been tested in clinical trials and only 4 have been approved for clinical use. All the approved drugs in clinical practice may provide modest symptomatic benefits, but do not treat the underlying process.4–6

AD is manifested as the selective loss of cholinergic neurons, reduced levels of the neurotransmitter ACh in the brain, and overactivity of enzymes acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE). These pathological findings support the development of the so-called first-generation drug treatments for AD. Until now, four cholinesterase inhibitors (ChEIs) have been approved to treat AD, namely, tacrine, galantamine, rivastigmine, and donepezil. All of them can block ACh decomposition via AChE inhibition and increase ACh concentration in the synaptic cleft. These mainstay drugs for AD symptomatic treatment yield beneficial effects on memory retention, at least in the short term.4

In a normal brain, AChE is the main enzyme responsible for the rapid hydrolysis of ACh and terminates neurotransmission at cholinergic synapses.7–9 Nevertheless, in the advanced stages of AD, AChE levels in the brain can be found to decline, while BuChE levels are found to increase up to 120% of the normal levels, indicating that BuChE plays a critical role in ACh hydrolysis during the later stages of AD. In this sense, the development of selective BuChE inhibitors has become imperative.10

Tacrine (Fig. 1a), the first approved ChEI, was removed from the US market after reports of hepatic toxicity. Over the years, a substantial number of tacrine derivatives and its analogues have been synthesized in an effort to limit the side-effects concomitant with increasing efficiency.4,11–14

Fig. 1. a) Tacrine, bis(7)-tacrine, and lophine (imidazole ring is colored in red); b) lophine–carbohydrate hybrids formulated in this study.

All the current kinetic models for both AChE and BuChE propose the existence of at least two substrate-binding sites: catalytic anionic site (CAS) near the bottom of the active-site gorge and peripheral anionic site (PAS) near its entrance.15,16 In 1996, Pang et al. designed the first hybrid molecule for the potential treatment of AD based on computational studies. They envisaged that methylene chain spacers joining two units of tacrine moiety allowed a double interaction of the compound with the enzyme to simultaneously target the CAS and PAS of AChE. Based on these data, Pang and co-workers performed the synthesis of alkylene-linked bis-tacrine compounds and the heptylene-linked bis(7)-tacrine dimer (Fig. 1a) was found to be almost 1500 times more potent against AChE than tacrine.17,18 Since then, several dual-binding hybrids have been developed; however, most of them have been obtained from a combination of well-known drugs.

Therefore, AChE inhibitors with novel structural diversity are urgently needed.4,19

The imidazole moiety is found in the main structure of certain biologically active compounds, playing important roles in biochemical processes with pharmacological properties.20,21 The imidazole derivative, namely, 2,4,5-triphenyl-1H-imidazole (lophine, Fig. 1a), can be used as a fluorescent-labeling reagent.22 Recently, hybrids containing lophine and pyrimidine nuclei connected by a methylene chain exhibited photophysical features that were successfully used to explore their interaction with the bovine serum albumin (BSA) protein and exhibited a significant suppression mechanism.23 In the recent years, several tri- and tetra-substituted imidazoles have been explored by their neuro-related biological activities such as anticonvulsant, antidepressant, and ChEIs.24 Lophine-based hybrids in connection with tacrine by alkyl linkers showed considerable ChEI inhibitory activity at the nanomolar 50% inhibiting concentration (IC₅₀) scale.25

Carbohydrates are versatile building blocks that allow the substitution at various positions of the sugar ring as well as further structural modifications. Compounds containing carbohydrate scaffolds have been studied as promising candidates for AD therapy.26–29 Furthermore, carbohydrates play an essential role in molecular recognition and binding affinity via interactions with proteins, particularly CH/π packing with aromatic amino acids and polar interactions. Therefore, based on a dual-binding site strategy where the designed molecules simultaneously bind to amino acid residues present in both CAS and PAS, we proposed that cyclic carbohydrate derivatives and lophine cooperatively act to provide enzyme-binding affinity.

Taking into consideration the potential of both lophine and carbohydrate derivatives in order to formulate novel ChEIs and as a part of our investigation on novel bioactive hybrids,25,26,30 in the present study, we describe the synthesis of a series of novel lophine hybrids with natural-based d-xylose, d-ribose, and d-galactose, aiming to achieve high AChE and BuChE inhibitory activities (Fig. 1b).

Results and discussion

Synthesis

The lophine moiety was obtained by the one-pot four-component condensation of tert-butyl-(n-aminoalkyl)carbamate (1a–e), benzil, benzaldehyde, and ammonium acetate (NH₄OAc) using InCl₃ as the catalyst based on a protocol earlier described by our group (Scheme 1).25 The reaction afforded compounds 2a–e in good yields, i.e., 79–96%. Subsequently, the carbamate-protecting group was removed by acid hydrolysis in methanol, yielding n-(lophin-1-yl)alkylamines 3a–e.

Scheme 1. One-pot four-component reaction to obtain the lophine nucleus.

With the 3a–e amines, we turned our attention to their connection with the carbohydrate derivatives through a nucleophilic substitution reaction using appropriate sugar-based tosylates.31–34 The reaction was performed in isopropanol in a sealed flask; under these conditions, tosylates 4, 5, and 6 were converted into lophine-based hybrids derived from d-xylose, d-ribose, and d-galactose, respectively (Scheme 2).

Scheme 2. Synthesis of hybrids containing lophine and carbohydrate derivatives.

For these synthesized hybrids, a similar pattern of chemical shifts can be observed using proton NMR spectroscopy. The corresponding peaks of the lophine nucleus were observed from 7.0 to 8.0 ppm; the protons of the sugar moiety were assigned from 2.0 to 5.0 ppm and the peaks matching to the linker's methylenes exhibited chemical shifts below 1.6 ppm. The main variations were distinguished for anomeric protons: 5.92 ppm (7c, d-xylose), 4.95 (8b, d-ribose), and 5.53 (9a, d-galactose), as shown in Fig. S51.† These values are consistent as expected for these carbohydrate derivatives.31–34

As a design element already reported in our earlier work, we have selected the functionalization of sugar at the primary position, protecting the anomeric position.26 In this way, the performed structural changes did not involve the control of stereochemistry. Further functionalizations can be performed, if needed, to refine the structure and improve the activity of compounds.

AChE and BuChE inhibition evaluation

The inhibitory activities against AChE and BuChE of the new lophine–carbohydrate derivatives are listed in Table 1 and are expressed as IC₅₀ values. Tacrine was synthesized according to the literature protocol and taken as the control compound.35 In these set of experiments, only one hybrid (8b), a d-ribose derivative connected to lophine by a heptylene chain, demonstrated AChE inhibitory activity with an IC₅₀ value of 2.75 μM. On the other hand, all the hybrids under investigation showed selective BuChE inhibition, 9a (a d-galactose derivative) being the most potent inhibitor (IC₅₀ = 0.17 μM), bearing a hexamethylene chain between the lophine and d-galactose-derivative moieties. This indicated that this linker allows an optimal interaction between the CAS and PAS of BuChE.

Table 1. Inhibitory activities toward AChE and BuChE and the IC₅₀ ratios of the compounds under investigation.

Entry	Cmpd	n	IC50 (μM) [confidence interval]		IC50 ratio BuChE/AChE
			AChE	BuChE
1	7a	4	n.a.	0.30 [0.18–0.48]	—
2	7b	5	n.a.	0.28 [0.17–0.44]	—
3	7c	6	n.a.	0.71 [0.22–2.33]	—
4	7d	7	n.a.	0.40 [0.28–0.57]	—
5	7e	8	n.a.	1.30 [0.94–1.80]	—
6	8a	6	n.a.	0.40 [0.31–0.50]	—
7	8b	7	2.75 [2.37–3.19]	0.50 [0.37–0.68]	0.18
8	9a	6	n.a.	0.17 [0.10–0.32]	—
9	9b	7	n.a.	0.62 [0.42–0.91]	—
10	Tacrine	—	0.03 [0.01–0.10]	0.06 [0.04–0.09]	1.8
11	Lophine		n.a.	n.a.

Molecular modeling

The ensemble docking strategy adopted in this work was successfully validated with all the reference ligands docked, yielding root-mean-square deviation (RMSD) ≤2 Å in the corresponding protein conformation with the lowest glide score (Table 2). The co-crystallized ligand of the ; 1ZGC complex was docked with a glide score of –12.989 kcal mol^–1 and RMSD of 1.211 Å in the ; 1Q84 conformation corresponding to a docking score slightly better than that obtained within its native conformation (score = –12.328 kcal mol^–1 and RMSD = 1.451 Å).

Table 2. Ensemble docking results of the reference ligands for the four AChE conformations and BuChE.

Reference ligand	Conformation a	Glide score b	RMSD c
1ZGC	1Q84 (AChE)	–12.989 (–12.328 d )	1.211 (1.451 d )
1Q84	1Q84 (AChE)	–14.797	1.271
2CKM	2CKM (AChE)	–12.026	1.077
4EY7	4EY7 (AChE)	–12.242	0.986
5K5E	5K5E (BuChE)	–8.765	1.262

^aConformation with the lowest glide score in the ensemble docking.

^bGiven in kcal mol^–1.

^cGiven in Å.

^dGlide score and RMSD against the native conformation.

According to the docking results, the hybrid compounds interact with AChE and BuChE with inverted orientations: at the AChE-binding site, the carbohydrate moiety is packed between Trp84 and Phe330 and the lophine moiety is located in the PAS; in the case of BuChE, the lophine moiety interacts at the base of the cavity and the carbohydrate moiety is located at the entrance of the gorge.

Despite the docking scores obtained for the compounds of around –9 kcal mol^–1 against AChE, all of them, except 8b, were experimentally inactive. One possible reason for this lies in the high desolvation cost of the carbohydrate moieties becoming buried while not yielding hydrogen bonds with any residues inside the binding cavity, which was found in the docking experiments; therefore, this negatively impacted the activity of these compounds.

Interestingly, only 8b exhibited activity against AChE (IC₅₀ = 2.75 μM). Compound 8a (n = 6) also contains a ribose moiety as the carbohydrate group, only differing from 8b (n = 7) with respect to the linker size. Both these compounds were predicted to interact more favorably within the ; 2CKM conformation of AChE with similar scores (–8.674 kcal mol^–1 and –8.836 kcal mol^–1 for 8a and 8b, respectively). According to the glide binding modes (Fig. 2), these compounds have the ribose moiety packed between Trp84 and Phe330, whereas the lophine moiety has one phenyl group packed between Tyr70 and Trp279 from PAS. The main difference between them is that inactive 8a has the remaining phenyl groups of the lophine moiety totally exposed to the solvent, whereas in the 8b binding mode, there is a “bend” in the linker, allowing the interaction of a second phenyl ring with many aromatic residues on a hydrophobic pocket in the middle of the binding cavity, mainly formed by Try121, Phe290, Phe331, and Phe330.

Fig. 2. Docking results of 8a (A and C) and 8b (B and D) against the AChE conformation ; 2CKM. Two-dimensional diagrams were generated with the 2D ligand interaction diagram tool from Maestro.

Remarkably, all the compounds that are BuChE-selective inhibitors form a T-stacking interaction with Trp231—a residue accessible exclusively in the BuChE cavity. With regard to AChE, this residue is buried and inaccessible to inhibitors due to its vicinity to aromatic residues, e.g., Phe295 and Phe297. The BuChE inhibitor co-crystallized in the ; 5K5E conformation also makes this specific interaction. The most potent BuChE inhibitor, namely, 9a (IC₅₀ = 0.17 μM) (Fig. 3), contains a galactose moiety as the carbohydrate and a linker size (n = 6) sufficient to accommodate the entire compound interacting with BuChE: the lophine moiety is located at the bottom of the cavity interacting with Trp231 and Trp82 through T-stacking interactions (∼3.0 Å). The positively charged amino group is hydrogen-bonded with the Thr284 main chain (2.93 Å) and the galactose group makes Van der Waals contacts with certain hydrophobic side chains at the entrance of the binding site (e.g., Pro285 and Ala277). Compound 9b (IC₅₀ = 0.62 μM) was predicted to interact similarly to that of 9a with BuChE and differs only on the linker size (n = 7). We believe that the differences in the binding affinities might be due to an entropic penalty on the linker size without additional favorable interactions.

Fig. 3. (A) Docking results of 9a (cyan) and 9b (salmon) against the BuChE conformation ; 5K5E. (B) The two-dimensional diagram of 9a was generated with the 2D ligand interaction diagram tool from Maestro.

Therefore, on the basis of the docking results, we believe that the lophine moiety can become a promising fragment in the design of potent and selective BuChE inhibitors. The specific interaction with Trp231, together with the adequate fitting of the hydrophobic lophine moiety at the bottom of the BuChE-binding cavity, explains the selectivity of our compounds against BuChE over AChE. Furthermore, the carbohydrate moiety does not have a strong influence on the binding affinities, since it is mainly exposed to the solvent in BuChE; in certain cases, it performs polar interactions with residues already accessible to the solvent at the entrance of the binding site.

ADME-Tox predictions

The pharmacokinetics profile predicted for the lophine–carbohydrate derivatives synthesized in this work indicate that they are drug-like, orally available, and can reach the CNS (ESI,† Table S4). However, such compounds were predicted as poorly soluble in water according to the QPlogPo/w, QPlogS, and CIQPlogS descriptors. All these compounds, including bis(7)-tacrine and donepezil, were predicted to block hERG K⁺ channels. However, we aimed to perform the structural optimization of the most promising BuChE inhibitors guided by the docking results to improve the pharmacokinetics and toxicological profiles.

Evaluation of cytotoxicity in in vitro models

An important step in the development of new compounds is the determination of their toxicological potential. Over the years, several cell lines have been developed and have been widely used as experimental models in drug toxicity assays.36

A primary goal of in vitro systems is to provide a rapid and reliable tool to screen for potential toxicity chemicals since these assays must have sensitivity, simplicity, agility, cost-effectiveness, and versatility. The use of diverse in vitro models may provide additional information if the compounds present differential effects, or have different potencies, in distinct cell types. In addition, the relative sensitivity of each in vitro model toward the toxic effects of compounds may offer additional information about the possible targets and mechanisms involved in compound-induced toxicity.37,38

>Lophine–carbohydrate derivatives that showed the best inhibitory activities (7b, 8b, and 9a) and tacrine have had their cytotoxicity effects evaluated in cellular models of VERO (kidney), HepG2 (hepatic), and C6 (astroglial) cell lines in order to provide information on nephrotoxicity, hepatotoxicity, and neurotoxicity of the selected compounds, respectively.

The results showed that compounds 8b and 9a had no cytotoxic effects in all the tested cellular models, suggesting a degree of safety (see Table 4, Experimental; ESI,† Fig. S64–S66). Further, it was possible to recognize a large difference between the necessary concentrations for inhibitory activities against AChE (IC₅₀ = 2.75 μM to 8b) and BuChE (IC₅₀ = 0.50 μM to 8b and IC₅₀ = 0.17 μM to 9a), as well as the concentration at which all the cell lines were exposed to (up to 250 μM) in the cytotoxic assay: no cytotoxic effect was observed. For the 8b derivative, concentrations that were 90 and 500 times higher than those active for AChE and BuChE were used, respectively; further, for the 9b derivative, concentrations that were 1400 times higher than the active one for BuChE were used without any reduction in cell viability. These data reinforce the wide range of safety of these compounds.

Table 4. Cellular viabilities after treatment with the selected compounds for 24 h using MTT assay (IC₅₀) in Vero, HepG2, and C6 cell lines.

IC50 ± SEM (μM)
Compound	Vero	HepG2	C6
8b	>250.0	>250.0	>250.0
9a	>250.0	>250.0	>250.0
7b	115.4 ± 4.2	116.4 ± 0.1	>250.0
Tacrine	168.9 ± 14.0	114.9 + 17.9	450.7 ± 4.2

Concerning the 7b derivative and tacrine, a dose-dependent profile of cytotoxicity can be observed for both these compounds by using the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium (MTT) assay after 24 h of treatment (Fig. S64–S66†). These two compounds presented a similar cytotoxicity profile, demonstrating more pronounced nephro (Vero) and hepatotoxic (HepG2) effects, as well as low neurotoxicity (C6) (Table 4). The Vero and HepG2 cell lines were more sensitive to the effects of the 7b derivative and tacrine than to compounds 8b and 9a: a 50% reduction in cellular viability was observed between 115.4 and 168.9 μM (Table 3).

Table 3. Absorbed fraction of standard drugs used as controls and target compounds by the PAMPA-TGI technique.

Compounds	Pe literature (10–6 cm s–1)	Pe experimental (10–6 cm s–1)	Fa experimental (%)	Classification	Log P
Aciclovir	0	0.60	20.40	Low	—
Atenolol	0	0.16	5.90	Low	—
Ceftriaxone	0	0.38	13.36	Low	—
Coumarin	22.1	26.19	99.99	High	—
Diclofenac	12.5	12.60	99.16	High	—
Hydrocortisone	3.4	5.57	87.94	High	—
Norfloxacin	0.9	0.46	15.91	Low	—
Ranitidine	0.5	0.62	20.87	Low	—
Sulfasalazine	0.1	0.71	23.51	Low	—
Verapamil	7.4	5.43	87.30	High	—
8b	—	2.28	57.99	Medium	8.03
9a	—	0.54	18.70	Low	9.86

Parallel artificial membrane permeability assay (PAMPA)

PAMPA is a non-cell-based and low-cost alternative to cellular models: it exhibits high tolerance to a wider pH range, higher dimethyl sulfoxide (DMSO) content, and amenability to high throughput as compared to cell-based methods39 since absorption is one of the main physicochemical properties that determine oral bioavailability.

Kansy et al.40 developed this method as a rapid assay using a 96-well-plate-based in vitro system for the evaluation of passive transcellular permeability, where the molecules traverse the cell membrane in the intestinal wall to access passive diffusion-driven blood circulation guided by the concentration gradient. Lipophilic compounds diffuse via the cell as they have high permeability through the lipid membrane.39,41 Therefore, compound dissolution in the intestinal fluid and permeation through the intestine are factors that determine drug absorption.41

The results presented in Table 3 show the fraction absorbed for each compound evaluated by the PAMPA technique, where we can observe that compound 8b (57.99%) has medium permeability and 9a has low (18.70%) permeability. Certain parameters may interfere with the permeability of the compounds, such as solubility, log P, log D, polar surface area, hydrogen-bond donor groups, and hydrogen-bond acceptor groups.60 It is worth noting that compound 9a exhibited a low permeability value despite having a very high theoretical log P value of 9.86, as calculated using the ACD/Percepta software. One of the reasons for this phenomenon might be the low water solubility that would not allow the compound to be available for membrane absorption, which justifies its low permeability.

Experimental

Chemistry

¹H NMR and ¹³C NMR spectra were recorded in a CDCl₃ solution by using a Varian VNMRS 300 MHz spectrometer. The assignment of chemical shifts is based on standard NMR experiments (¹H; ¹³C; ¹H, ¹H-COSY; ¹H; and ¹³C-HMQC). Chemical shifts (δ) are expressed in parts per million from the peak of tetramethylsilane (δ = 0.00 ppm) as the internal standard in ¹H NMR or from the solvent peak of CDCl₃ (δ = 77.23 ppm) in ¹³C NMR. NMR multiplicities are expressed as s (singlet), d (doublet), dd (doublet of doublets), ddd (doublet of doublet of doublets), t (triplet), and m (multiplet); br (broad) and coupling constants (J) are expressed in Hz. The IR spectra were recorded on a Varian 640-IR spectrometer using KBr pellets. Specific rotations were measured on a JASCO P-2000 polarimeter with 1.0 mL cell at a temperature of 20 °C. The high-resolution mass spectrometry with electrospray ionization (HRMS-ESI) data on the positive mode were collected by using a UHPLC-QTOF/MS Bruker Impact II. The samples were infused from a 100 mL Hamilton syringe at a flow rate from 5 to 10 mL min^–1 depending on the sample. The instrument settings were as follows. Capillary voltage: 3000 V; cone voltage: 33 V; extraction cone voltage: 2.5 V; and desolvation gas temperature: 100 °C. Nitrogen was used as the desolvation gas. Methanol (HPLC-grade Tedia) was used as the solvent for the analyzed samples and filtered prior to injection. Purification by column chromatography was carried out on 60 Å silica gel (70–230 mesh) and analytical thin-layer chromatography (TLC) was conducted on aluminum plates with 0.2 mm of silica gel 60F-254 (MACHEREY-NAGEL). Solvents were obtained from Tedia and Nuclear, and the reagents were purchased from Sigma-Aldrich, Acros Organics, and TCI. The compounds 1a–e (ref. 42) and tosylates 3, 4, and 5 (ref. 31–34) were prepared according to earlier reported procedures.

General procedure for the preparation of tert-butyl-(n-(2,4,5-triphenyl-1H-imidazol-1-yl)alkyl)carbamates (2a–e)

A mixture of tert-butyl-(n-aminoalkyl)carbamate 1a–e (2.0 mmol), benzil (2.0 mmol), benzaldehyde (2.0 mmol), ammonium acetate (2.0 mmol), and InCl₃ (0.30 mmol) in absolute ethanol (2 mL), placed in a round-bottom flask fitted with a condenser and drying tube, was stirred at 78 °C. After 12 h, one more equivalent of benzil, benzaldehyde, and ammonium acetate was added and the mixture was maintained at 78 °C for an additional 12 h. The solvent was removed under reduced pressure, and the crude product was purified by column chromatography (eluting with hexane–ethyl acetate–triethylamine at 90 : 9 : 1, 80 : 19 : 1, 70 : 29 : 1, 60 : 39 : 1, and 50 : 49 : 1) to yield the desired product.

tert-Butyl (4-(2,4,5-triphenyl-1H-imidazol-1-yl)butyl)carbamate (2a)

Amine 1a was treated with benzil, benzaldehyde, and ammonium acetate according to a general procedure to yield desired product 2a as a white solid (79% yield); m.p. 102–103 °C; IR (KBr) ν_max/cm^–1: 3352, 3222, 3050, 2968, 2928, 2860, 1691, 1527, 1171, 774, 692; ¹H NMR (300 MHz, CDCl₃) δ 7.70 (dd, J = 8.0, 1.4 Hz, 2H), 7.57–7.41 (m, 10H), 7.26–7.11 (m, 3H), 4.21 (br, 1H), 3.93 (t, J = 6.0 Hz, 2H), 2.88–2.70 (m, 2H), 1.69 (s, 2H), 1.42 (s, 9H), 1.19–1.08 (m, 2H); ¹³C NMR (75 MHz, CDCl₃) δ 155.7, 147.7, 137.8, 134.4, 131.4, 131.3, 131.0, 129.5, 129.2, 129.0, 128.8, 128.7, 128.6, 128.5, 128.1, 127.9, 126.8, 126.3, 125.5, 44.2, 39.5, 28.4, 27.7, 26.6; HRMS-ESI: calcd for [M – H]⁺ 468.2646, found 468.2659.

tert-Butyl (5-(2,4,5-triphenyl-1H-imidazol-1-yl)pentyl)carbamate (2b)

Amine 1b was treated with benzil, benzaldehyde, and ammonium acetate according to a general procedure to afford desired product 2b as a white solid (80% yield); m.p. 59–60 °C; IR (KBr) ν_max/cm^–1: 3344, 3050, 2976, 2852, 1697, 1499, 1163, 768, 700; ¹H NMR (300 MHz, DMSO-d₆) δ 7.72 (d, J = 6.6 Hz, 2H), 7.59–7.37 (m, 10H), 7.24–7.04 (m, 3H), 6.62 (s, 1H), 3.87 (t, J = 7.3 Hz, 2H), 2.74–2.57 (m, 2H), 1.33 (s, 9H), 1.29–1.19 (m, 2H), 1.09–0.93 (m, 2H), 0.86 (d, J = 7.0 Hz, 2H); ¹³C NMR (75 MHz, DMSO-d₆) δ 155.9, 147.1, 137.0, 135.2, 131.7, 131.5, 131.3, 130.2, 129.6, 129.3, 129.2, 129.1, 128.9, 128.6, 128.5, 127.5, 126.5, 125.6, 77.8, 44.7, 29.8, 29.0, 28.7, 23.4; HRMS-ESI: calcd for [M – H]⁺ 482.2802, found 482.2805.

tert-Butyl (6-(2,4,5-triphenyl-1H-imidazol-1-yl)hexyl)carbamate (2c)

Amine 1c was treated with benzil, benzaldehyde, and ammonium acetate according to a general procedure to yield desired product 2c as a white oil (94% yield); IR (KBr) ν_max/cm^–1: 3344, 3058, 2968, 2928, 2846, 1703, 1499, 1355, 1246, 1172, 774, 692; ¹H NMR (300 MHz, DMSO-d₆) δ 7.49 (d, J = 6.6 Hz, 2H), 7.37–7.13 (m, 10H), 7.01–6.85 (m, 3H), 6.44 (t, J = 5.4 Hz, 1H), 3.65 (t, J = 7.3 Hz, 2H), 2.62–2.44 (m, 2H), 1.13 (s, 9H), 1.02 (s, 2H), 0.86 (s, 2H), 0.63 (s, 4H); ¹³C NMR (75 MHz, DMSO-d₆) δ 156.0, 147.2, 137.0, 135.2, 131.7, 131.6, 131.3, 130.2, 129.6, 129.3, 129.2, 129.1, 128.9, 128.6, 128.5, 128.2, 127.5, 127.0, 126.5, 125.7, 77.7, 44.6, 30.0, 29.5, 28.7, 25.7, 25.6; HRMS-ESI: calcd for [M – H]⁺ 496.2959, found 496.2965.

tert-Butyl (7-(2,4,5-triphenyl-1H-imidazol-1-yl)heptyl)carbamate (2d)

Amine 1d was treated with benzil, benzaldehyde, and ammonium acetate according to a general procedure to afford desired product 2d as a white oil (96% yield); IR (KBr) ν_max/cm^–1: 3344, 3064, 2928, 2852, 1697, 1171, 692, 570; ¹H NMR (300 MHz, DMSO-d₆) δ 7.72 (d, J = 6.9 Hz, 2H), 7.60–7.41 (m, 10H), 7.26–7.04 (m, 3H), 6.71 (t, J = 5.4 Hz, 1H), 3.88 (t, J = 7.4 Hz, 2H), 2.88–2.67 (m, 2H), 1.37 (s, 9H), 1.29–1.07 (m, 4H), 1.03–0.81 (m, 6H); ¹³C NMR (75 MHz, DMSO-d₆) δ 156.0, 147.2, 137.0, 135.2, 131.8, 131.6, 131.3, 130.2, 129.6, 129.3, 129.2, 129.1 (2C), 128.7, 128.5, 126.5, 125.7, 77.7, 44.5, 29.8, 29.7, 28.7, 28.0, 26.2, 25.9; HRMS-ESI: calcd for [M – H]⁺ 510.3115, found 510.3117.

tert-Butyl (8-(2,4,5-triphenyl-1H-imidazol-1-yl)octyl)carbamate (2e)

Amine 1e was treated with benzil, benzaldehyde, and ammonium acetate according to a general procedure to afford desired product 2e as a white oil (95% yield); IR (KBr) ν_max/cm^–1: 3344, 3050, 2921, 2860, 1697, 1505, 1465, 1164, 760, 686; ¹H NMR (300 MHz, DMSO-d₆) δ 7.71 (d, J = 7.6 Hz, 2H), 7.61–7.36 (m, 10H), 7.28–7.06 (m, 3H), 6.74 (t, J = 5.4 Hz, 1H), 3.87 (t, J = 7.3 Hz, 2H), 2.90–2.73 (m, 2H), 1.36 (s, 9H), 1.27–1.16 (m, 4H), 1.05–0.75 (m, 8H); ¹³C NMR (75 MHz, DMSO-d₆) δ 156.0, 147.2, 137.0, 135.2, 131.8, 131.6, 131.3, 130.2, 129.6, 129.3, 129.2, 129.1, 128.5, 126.6, 125.7, 77.8, 44.6, 29.9, 29.8, 28.8, 28.6, 28.3, 26.6, 25.8; HRMS-ESI: calcd for [M – H]⁺ 524.3272, found 524.3273.

General procedure for the preparation of n-(2,4,5-triphenyl-1H-imidazol-1-yl)alkyl-1-amines (3a–e)

Here, 10 mL of an aqueous solution of HCl 10% was added to a solution of carbamate 2a–e (1.0 g) in methanol (100 mL) and the mixture was stirred in reflux for 24 h. Thereafter, the solvent was removed under pressure and the remaining oil was basified with Na₂CO₃ until pH 11. The product was extracted with dichloromethane, and the organic layer was dried with Na₂SO₄ and concentrated under vacuum. The crude oil was purified by a chromatography column (eluting with chloroform–methanol–ammonium hydroxide at 93 : 6.5 : 0.5) to afford the desired amine.

4-(2,4,5-Triphenyl-1H-imidazol-1-yl)butan-1-amine (3a)

Intermediate 2a was reacted according to a general procedure to afford desired product 3a as a white solid (59% yield); m.p. 119–120 °C; IR (KBr) ν_max/cm^–1: 3044, 2920, 2846, 1595, 1438, 768, 692; ¹H NMR (300 MHz, CDCl₃) δ 7.69 (dd, J = 8.0, 1.6 Hz, 2H), 7.56–7.38 (m, 10H), 7.23–7.08 (m, 3H), 3.91 (t, J = 7.5 Hz, 2H), 2.34 (t, J = 7.0 Hz, 2H), 1.44–1.28 (m, 2H), 1.23 (s, 2H), 1.14–0.99 (m, 2H); ¹³C NMR (75 MHz, CDCl₃) δ 147.6, 137.8, 134.5, 131.5, 131.5, 131.0, 129.5, 129.2, 129.1, 128.8, 128.7, 128.6, 128.0, 126.8, 126.2, 44.5, 41.1, 30.1, 27.7; HRMS-ESI: calcd for [M – H]⁺ 468.2646, found 468.2659.

5-(2,4,5-Triphenyl-1H-imidazol-1-yl)pentan-1-amine (3b)

Intermediate 2b was reacted according to a general procedure to afford desired product 3b as a colorless oil (75% yield); IR (KBr) ν_max/cm^–1: 3433, 3058, 2922, 2851, 1595, 1503, 1437, 768, 692; ¹H NMR (300 MHz, CDCl₃) δ 7.67 (dd, J = 8.0, 1.6 Hz, 2H), 7.56–7.38 (m, 10H), 7.22–7.06 (m, 3H), 3.88 (t, J = 7.5 Hz, 2H), 2.39 (t, J = 6.0 Hz, H), 1.48 (br, 2H), 1.32 (dt, J = 15.0, 7.5 Hz, 2H), 1.14–0.86 (m, 4H); ¹³C NMR (75 MHz, CDCl₃) δ 147.7, 137.8, 134.6, 131.6, 131.5, 131.0, 129.6, 129.2, 129.1, 128.9, 128.7, 128.6, 128.4, 128.1, 126.9, 126.3, 125.7, 44.6, 41.7, 32.6, 30.2, 23.5; HRMS-ESI: calcd for [M – H]⁺ 382.2278, found 382.2278.

6-(2,4,5-Triphenyl-1H-imidazol-1-yl)hexan-1-amine (3c)

Intermediate 1c was reacted according to a general procedure to yield desired product 2c as a colorless oil (63% yield); IR (KBr) ν_max/cm^–1: 3058, 2968, 2920, 2846, 1697, 1260, 1163, 734, 500; ¹H NMR (300 MHz, CDCl₃) δ 7.68 (dd, J = 7.9, 1.4 Hz, 2H), 7.58–7.37 (m, 10H), 7.22–7.07 (m, 3H), 3.86 (t, J = 7.5 Hz, 2H), 2.48 (t, J = 7.0 Hz, 2H), 1.55 (s, 2H), 1.38–1.26 (m, 2H), 1.24–1.08 (m, 2H), 1.01–0.86 (m, 4H); ¹³C NMR (75 MHz, CDCl₃) δ 147.5, 137.5, 134.4, 131.4, 131.4, 130.8, 129.4, 129.0, 128.9, 128.7, 128.5, 127.9, 126.6, 126.0, 44.4, 41.7, 33.0, 30.1, 25.8, 25.7; HRMS-ESI: calcd for [M – H]⁺ 396.2434, found 396.2431.

7-(2,4,5-Triphenyl-1H-imidazol-1-yl)heptan-1-amine (3d)

Intermediate 2d was reacted according to a general procedure to afford desired product 3d as a colorless oil (84% yield); IR (KBr) ν_max/cm^–1: 3427, 3064, 2928, 2852, 1595, 1444, 774, 692; ¹H NMR (300 MHz, CDCl₃) δ 7.67 (dd, J = 8.0, 1.4 Hz, 2H), 7.58–7.35 (m, 10H), 7.21–7.04 (m, 3H), 3.85 (t, J = 7.5 Hz, 2H), 2.48 (t, J = 7.1 Hz, 2H), 2.30 (s, 2H), 1.41–1.14 (m, 4H), 0.90–0.85 (m, 6H); ¹³C NMR (75 MHz, CDCl₃) δ 147.4, 137.4, 134.4, 131.2, 130.8, 129.5, 128.9, 128.8, 128.6, 128.4, 127.8, 126.6, 126.0, 44.4, 41.5, 32.8, 30.0, 28.1, 26.1, 25.8; HRMS-ESI: calcd for [M – H]⁺ 410.2591, found 410.2591.

8-(2,4,5-Triphenyl-1H-imidazol-1-yl)octan-1-amine (3e)

Intermediate 2e was reacted according to a general procedure to afford desired product 3e as a colorless oil (87% yield); IR (KBr) ν_max/cm^–1: 3434, 3064, 2920, 2852, 1561, 768, 692; ¹H NMR (300 MHz, CDCl₃) δ 7.69 (dd, J = 8.0, 1.6 Hz, 2H), 7.57–7.38 (m, 10H), 7.23–7.07 (m, 3H), 3.87 (t, J = 7.5 Hz, 2H), 2.60 (t, J = 7.1 Hz, 2H), 1.63 (s, 2H), 1.40–1.27 (m, 4H), 1.19–0.86 (m, 8H); ¹³C NMR (75 MHz, CDCl₃) δ 147.5, 137.5, 134.5, 131.4, 131.4, 130.9, 129.5, 129.0, 128.9, 128.7, 128.5, 127.9, 126.7, 126.1, 44.6, 41.9, 33.3, 30.1, 28.8, 28.4, 26.5, 25.9; HRMS-ESI: calcd for [M – H]⁺ 424.2747, found 424.2746.

General procedure for the preparation of lophine–carbohydrate hybrids

A mixture containing 0.3 mmol of tosylate from d-xylose (4), d-ribose (5), or d-galactose (6) and 0.6 mmol of n-(2,4,5-triphenyl-1H-imidazol-1-yl)alkyl-1-amines (3a–e) was dissolved in isopropyl alcohol (1.0 mL) and was maintained under stirring at 83 °C for 72 h in a sealed flask. Thereafter, the reaction was quenched by saturated NaHCO₃ (20 mL), washed with brine (20 mL) and water (20 mL), and the product was extracted with dichloromethane. The organic layer was dried by Na₂SO₄, the solvent was removed under a vacuum, and the crude product was purified by column chromatography, eluting with hexane:ethyl acetate of 1:1 until removal of the remaining tosylate, followed by eluting with chloroform:methanol of 98:2 to afford the desired product.

N¹-(5-Deoxy-1,2-O-isopropylidene-α-d-xylofuranose)-5-(N⁴-(2,4,5-triphenyl-1H-imidazol-1-yl)butan-1-amine) (7a)

Intermediate 3a and tosylate 4 were reacted according to a general procedure to afford desired product 7a as a yellow solid (45% yield); m.p. 67–68 °C; [α]20D = +19.0° (c 0.237, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3059, 2983, 2937, 2853, 1434, 1372, 1213, 1161, 1069, 1001, 781, 697; ¹H-NMR (CDCl₃, 300 MHz) δ 7.70 (dd, J = 7.9, 1.3 Hz, 2H), 7.56–7.41 (m, 10H), 7.24–7.11 (m, 3H), 5.91 (d, J = 3.7 Hz, 1H), 4.46 (d, J = 3.6 Hz, 1H), 4.22 (d, J = 2.8 Hz, 1H), 4.14 (s, 1H), 3.93 (t, J = 7.4 Hz, 3H), 3.21 (dd, J = 13.0, 3.5 Hz, 1H), 2.76 (d, J = 12.9 Hz, 1H), 2.39–2.11 (m, 2H), 1.49 (s, 3H), 1.39–1.27 (m, 5H), 1.16–1.02 (m, 2H); ¹³C NMR (75 MHz, CDCl₃) δ 147.8, 137.9, 134.6, 131.5, 131.1, 129.6, 129.3, 129.0, 128.8, 128.1, 127.0, 126.4, 111.5, 105.1, 86.0, 78.1, 76.9, 48.6, 48.4, 44.4, 28.1, 26.9, 26.3; HRMS-ESI: calcd for [M – H]⁺ 540.2857; found 540.2851.

N¹-(5-Deoxy-1,2-O-isopropylidene-α-d-xylofuranose)-5-(N⁵-(2,4,5-triphenyl-1H-imidazol-1-yl)pentan-1-amine) (7b)

Intermediate 3b and tosylate 4 were reacted according to a general procedure to yield desired product 7b as a yellow solid (67% yield); m.p.: 53–54 °C; [α]20D = –107.6° (c 0.196, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3053, 2982, 1419, 1260, 1068, 890, 734; ¹H-NMR (CDCl₃, 300 MHz) δ 7.72–7.63 (m, 2H), 7.57–7.36 (m, 10H), 7.24–7.09 (m, 3H), 5.91 (d, J = 3.6 Hz, 1H), 4.45 (d, J = 3.7 Hz, 1H), 4.23 (d, J = 2.8 Hz, 1H), 4.15 (s, 1H), 4.06 (s, 1H), 3.88 (t, J = 7.5 Hz, 2H), 3.27 (dd, J = 12.9, 3.4 Hz, 1H), 2.81 (d, J = 11.9 Hz, 1H), 2.48–2.11 (m, 2H), 1.47 (s, 3H), 1.31 (s, 5H), 1.08 (s, 2H), 0.96 (d, J = 6.4 Hz, 2H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.8, 138.0, 134.7, 131.7, 131.6, 131.2, 129.7, 129.4, 129.3, 129.1, 128.9, 128.2, 127.0, 126.4, 111.6, 105.2, 86.2, 78.4, 77.0, 49.2, 48.6, 44.6, 30.2, 28.7, 27.0, 26.3, 23.8; HRMS-ESI: calcd for [M – H]⁺ 554.3013, found 554.3012.

N¹-(5-Deoxy-1,2-O-isopropylidene-α-d-xylofuranose)-5-(N⁶-(2,4,5-triphenyl-1H-imidazol-1-yl)hexan-1-amine) (7c)

Intermediate 3c and tosylate 4 were reacted according to a general procedure to afford desired product 7c as a yellow solid (72% yield); m.p.: 55–56 °C; [α]20D = –7.1° (c 0.210, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3434, 2923, 2838, 1649, 1555, 1064, 998, 686; ¹H-NMR (CDCl₃, 300 MHz) δ 7.67 (d, J = 6.5 Hz, 2H), 7.57–7.35 (m, 10H), 7.14 (m, 3H), 5.92 (d, J = 3.5 Hz, 1H), 4.45 (d, J = 3.5 Hz, 1H), 4.23 (d, J = 2.6 Hz, 1H), 4.16 (s, 1H), 4.00 (s, 1H), 3.88 (t, J = 7.5 Hz, 2H), 3.28 (dd, J = 12.9, 3.3 Hz, 1H), 2.85 (d, J = 12.8 Hz, 1H), 2.81–2.65 (m, 1H), 2.45–2.32 (m, 2H), 1.47 (s, 3H), 1.30 (s, 5H), 1.15 (s, 2H), 0.93 (s, 4H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.7, 137.8, 134.6, 131.6, 131.5, 131.1, 129.7, 129.3, 129.1, 128.9, 128.7, 128.1, 126.9, 126.3, 111.4, 105.1, 86.1, 78.2, 77.0, 49.4, 48.5, 44.6, 30.1, 29.1, 26.9, 26.2, 26.1, 25.9; HRMS-ESI: calcd for [M – H]⁺ 568.3170, found 568.3164.

N¹-(5-Deoxy-1,2-O-isopropylidene-α-d-xylofuranose)-5-(N⁷-(2,4,5-triphenyl-1H-imidazol-1-yl)heptan-1-amine) (7d)

Intermediate 3d and tosylate 4 were reacted according to a general procedure to afford desired product 7d as a yellow solid (61% yield); m.p.: 54–55 °C; [α]20D = –46.7° (c 0.120, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3393, 3050, 2928, 2843, 1609, 1450, 1364, 1072, 1010, 981; ¹H-NMR (CDCl₃, 300 MHz) δ 7.78–7.64 (m, 2H), 7.64–7.38 (m, 10H), 7.25–7.08 (m, 3H), 5.94 (d, J = 3.7 Hz, 1H), 4.48 (d, J = 3.6 Hz, 1H), 4.27 (d, J = 2.6 Hz, 1H), 4.18 (s, 1H), 3.88 (t, 2H), 3.48 (s, 1H), 3.36 (dd, J = 12.8, 3.4 Hz, 1H), 2.91 (d, J = 12.8 Hz, 1H), 2.65–2.33 (m, 1H), 1.48 (s, 3H), 1.40–1.20 (m, 7H), 1.12–0.88 (m, 8H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.8, 137.8, 134.7, 131.6, 131.6, 131.1, 129.7, 129.3, 129.2, 129.0, 128.8, 128.2, 127.0, 126.4, 111.6, 105.2, 86.1, 78.3, 77.0, 49.6, 48.6, 44.7, 30.4, 29.4, 28.5, 27.0, 26.8, 26.3, 26.1; HRMS-ESI: calcd for [M – H]⁺ 582.3326, found 582.3329.

N¹-(5-Deoxy-1,2-O-isopropylidene-α-d-xylofuranose)-5-(N⁸-(2,4,5-triphenyl-1H-imidazol-1-yl)octan-1-amine) (7e)

Intermediate 3e and tosylate 4 were reacted according to a general procedure to afford desired product 7e as a yellow oil (63% yield); [α]20D = +38.7° (c 0.281, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3420, 3055, 2928, 2852, 1471, 1068, 1007, 768, 690; ¹H-NMR (CDCl₃, 300 MHz) δ 7.68 (dd, J = 8.0, 1.6 Hz, 2H), 7.55–7.37 (m, 10H), 7.23–7.07 (m, 3H), 5.94 (d, J = 3.7 Hz, 1H), 4.48 (d, J = 3.7 Hz, 1H), 4.27 (d, J = 2.8 Hz, 1H), 4.23–4.10 (m, 2H), 3.90–3.81 (m, 2H), 3.36 (dd, J = 13.0, 3.5 Hz, 1H), 2.93 (d, J = 11.8 Hz, 1H), 2.65–2.38 (m, 2H), 1.54–0.79 (m, 15H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.6, 137.7, 134.5, 131.5, 131.5, 131.0, 129.6, 129.2, 129.0, 128.8, 128.6, 128.0, 126.8, 126.2, 111.4, 105.0, 86.0, 78.1, 76.9, 49.5, 48.4, 44.6, 30.2, 29.7, 29.3, 28.9, 28.4, 26.8, 26.1, 26.0; HRMS-ESI: calcd for [M – H]⁺ 596.3483, found 596.3481.

N¹-(1-O-Methyl-5-deoxy-2,3-O-isopropylidene-β-d-ribofuranoside)-5-(N⁶-(2,4,5-triphenyl-1H-imidazol-1-yl)hexan-1-amine) (8a)

Intermediate 3c and tosylate 5 were reacted according to a general procedure to afford desired product 8a as a yellow oil (45% yield); [α]20D = –88.4° (c 0.301, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3417, 2928, 2843, 1561, 1108, 693; ¹H-NMR (CDCl₃, 300 MHz) δ 7.69 (dd, J = 7.9, 1.6 Hz, 2H), 7.56–7.39 (m, 10H), 7.24–7.10 (m, 3H), 4.95 (s, 1H), 4.58 (dd, J = 11.8, 6.1 Hz, 2H), 4.27 (t, J = 6.9 Hz, 1H), 3.95–3.82 (m, 2H), 3.31 (s, 3H), 2.66 (dd, J = 10.9, 9.0 Hz, 2H), 2.50–2.38 (m, 2H), 1.62 (br, 3H), 1.48 (s, 3H), 1.40–1.18 (m, 5H), 1.05–0.90 (m, 4H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.8, 137.9, 134.7, 131.7, 131.7, 131.2, 129.7, 129.3, 129.2, 129.0, 128.8, 128.2, 127.0, 126.3, 112.4, 109.8, 86.3, 85.5, 82.9, 55.2, 53.1, 49.6, 44.8, 30.5, 29.8, 26.6, 26.5, 26.2, 25.1; HRMS-ESI: calcd for [M – H]⁺ 582.3326, found 582.3327.

N¹-(1-O-Methyl-5-deoxy-2,3-O-isopropylidene-β-d-ribofuranoside)-5-(N⁷-(2,4,5-triphenyl-1H-imidazol-1-yl)heptan-1-amine) (8b)

Intermediate 3d and tosylate 5 were reacted according to a general procedure to afford desired product 8b as a yellow oil (69% yield); [α]20D = –10.3° (c 0.340, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3415, 2934, 2858, 1593, 1102, 697; ¹H-NMR (CDCl₃, 300 MHz) δ 7.69 (dd, J = 8.0, 1.5 Hz, 2H), 7.56–7.38 (m, 10H), 7.23–7.07 (m, 3H), 4.95 (s, 1H), 4.65–4.53 (m, 2H), 4.28 (t, J = 6.9 Hz, 1H), 3.88 (t, 2H), 3.32 (s, 3H), 2.66 (d, J = 7.4 Hz, 2H), 2.56–2.44 (m, 2H), 1.55 (s, 3H), 1.48 (s, 3H), 1.31 (m, 5H), 1.01 (m, 6H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.7, 137.8, 134.7, 131.7, 131.6, 131.1, 129.7, 129.3, 129.1, 128.9, 128.7, 128.7, 128.1, 126.9, 126.3, 112.4, 109.8, 86.3, 85.5, 82.8, 55.2, 53.1, 49.8, 44.7, 30.4, 29.9, 28.6, 26.9, 26.6, 26.2, 25.1; HRMS-ESI: calcd for [M – H]⁺ 596.3483, found 596.3481.

N¹-(6-Deoxy-1,2:3,4-di-O-isopropylidene-α-d-galactopyranose)-6-(N⁶-(2,4,5-triphenyl-1H-imidazol-1-yl)hexan-1-amine) (9a)

Intermediate 3c and tosylate 6 were reacted according to a general procedure to afford desired product 9a as a yellow oil (64% yield); [α]20D = –6.33° (c 0.300, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3427, 2980, 2923, 2855, 1377, 1202, 1066, 687; ¹H-NMR (CDCl₃, 300 MHz) δ 7.68 (d, J = 6.6 Hz, 2H), 7.47 (m, 10H), 7.23–7.06 (m, 3H), 5.53 (d, J = 5.0 Hz, 1H), 4.58 (dd, J = 7.9, 1.9 Hz, 1H), 4.30 (dd, J = 4.9, 2.1 Hz, 1H), 4.16 (d, J = 7.9 Hz, 1H), 3.95–3.80 (m, 3H), 2.90–2.76 (m, 1H), 2.69 (dd, J = 12.5, 3.9 Hz, 1H), 2.56–2.30 (m, 2H), 1.93 (br, 1H), 1.51 (s, 3H), 1.44 (s, 3H), 1.42–1.15 (m, 10H), 0.96 (m, 4H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.7, 137.7, 134.7, 131.6, 131.6, 131.1, 129.7, 129.3, 129.1, 128.9, 128.7, 128.7, 128.1, 126.9, 126.3, 109.2, 108.5, 96.4, 72.1, 70.9, 70.6, 66.8, 49.8, 49.7, 44.7, 30.3, 29.6, 26.4, 26.2, 26.2, 26.1, 25.0, 24.4; HRMS-ESI: calcd for [M – H]⁺ 638.3588, found 638.3585.

N¹-(6-Deoxy-1,2:3,4-di-O-isopropylidene-α-d-galactopyranose)-6-(N⁷-(2,4,5-triphenyl-1H-imidazol-1-yl)heptan-1-amine) (9b)

Intermediate 3d and tosylate 6 were reacted according to a general procedure to yield desired product 9b as a yellow oil (70% yield); [α]20D = +17.0° (c 0.206, CH₂Cl₂); IR (KBr) ν_max/cm^–1: 3429, 2917, 2843, 1646, 1060, 694; ¹H-NMR (CDCl₃, 300 MHz) δ 7.73–7.65 (m, 2H), 7.59–7.38 (m, 10H), 7.24–7.07 (m, 3H), 5.54 (d, J = 5.0 Hz, 1H), 4.59 (dd, J = 7.9, 2.1 Hz, 1H), 4.31 (dd, J = 5.0, 2.2 Hz, 1H), 4.18 (d, J = 7.9 Hz, 1H), 4.00–3.79 (m, 3H), 2.95–2.67 (m, 2H), 2.64–2.38 (m, 2H), 1.97 (br, 1H), 1.63–1.24 (m, 18H), 1.18–0.87 (d, J = 30.3 Hz, 4H); ¹³C-NMR (CDCl₃, 75 MHz) δ 147.8, 137.8, 134.8, 131.7, 131.6, 131.2, 129.8, 129.3, 129.2, 129.0, 128.8, 128.2, 127.0, 126.3, 109.3, 108.7, 96.5, 72.2, 71.0, 70.7, 66.8, 49.9, 44.8, 30.4, 29.9, 28.7, 27.0, 26.3, 26.2, 25.1, 24.5; HRMS-ESI: calcd for [M – H]⁺ 652.3745, found 652.3737.

AChE and BuChE inhibition assays

Swiss Webster mice provided by the Central Animal Laboratory of the Oswaldo Cruz Foundation/RJ (Fiocruz) were used in this study. Animals were held in a light/dark cycle of 12/12 h and fed ad libitum. All the experiments were performed in accordance with the guidelines of the Ethics Committee for Animal Use of Fiocruz (CEUA; license number L044/2015 with additional license number LA-010/2016). After euthanasia, the brain and blood were collected and used for the preparation of biological fractions used in enzymatic assays. A cerebral homogenate was prepared in distilled water and then centrifuged at 20 000 × g for 60 min at 4 °C. The supernatant was discarded and the pellet was resuspended in 2 mL of 1% Triton X-100. Subsequently, this fraction was again centrifuged at 14 000 rpm (20 800g) for 90 min. The pellet was discarded and the supernatants were used for enzymatic assays. After allowing the whole blood to clot at room temperature, the sample was centrifuged at 2000 × g for 10 min in a refrigerated centrifuge (4 °C). The resulting supernatant, called the serum fraction, was maintained in polypropylene tubes until the enzymatic assays. The activities of the AChE enzyme in the brain and BuChE in the serum of mice were determined using the modified Ellman method.43 Both brain extract and blood serum were added to their respective reaction media containing sodium phosphate buffer (0.1 M, pH 7.5) in a 96-well plate, as well as the synthesized compounds to be analyzed. Control groups did not have this addition of compounds. 5,5′-Dithio-bis-(2-nitrobenzoic acid) (DTNB) at a concentration of 0.32 mM was added and the reaction medium was preincubated at 25 °C for 10 min. The enzymatic assays were initiated by the addition of substrates, acetylthiocholine iodide, or butyrylthiocholine iodide at a final concentration of 1.5 mM, and further incubation was carried out for 10 min. After incubation, the optical density was measured at a wavelength of 412 nm and blank wells (without enzymatic activity) had their values discounted. The compounds were first solubilized in 100% DMSO. Thereafter, they were diluted again in Milli-Q® water and tested against inhibition on the enzymatic activities at the final concentrations ranging from 0.01 to 10 000 nM. Data analysis was performed using the GraphPad Prism software, version 6.0 (GraphPad Software, Inc., San Diego, CA).

Ensemble docking

In this work, the docking experiments were performed with the molecular docking program Glide under the standard precision (SP) mode.44 All the nonpolar atoms of the protein were softened using a scaling factor of 0.80 of the van der Waals radii of the atoms with the partial charge between –0.25 and 0.25. The soft docking strategy is commonly used to include a small degree of protein flexibility through the suavization of the scoring function for pose prediction.45,46 Further, we redocked the reference ligands (i.e., the co-crystallized compounds) into their respective AChE and BuChE conformations to validate the docking protocol adopted herein. We selected four representative conformations of AChE for the ensemble docking strategy to consider the significant conformational changes mainly observed on the PAS.47,48 This approach consists of docking the compounds into each representative conformation of the receptor, aiming to implicitly consider large-scale protein movements.49,50 The protein conformations selected in this work were ; 1ZGC (Torpedo californica),51; 2CKM (Torpedo californica),52; 1Q84 (Mus musculus),53 and ; 4EY7 (Homo sapiens).54 All the inhibitors from the four representative conformations of AChE act as dual inhibitors interacting with both CAS and PAS. Four structural water molecules were explicitly considered after the alignment of the protein–ligand complexes and visual inspection of structural water (Table 5). For BuChE, we selected the conformation complexed with the largest inhibitor, namely, 6QS (PDB ID ; 5K5E), and maintained three structural waters (Table 5). The binding mode with the lowest glide score among the four AChE representative structures was selected for each compound.

Table 5. Four structural waters considered in the docking experiments (Wat1, Wat2, Wat3, and Wat4).

PDB ID	Wat1	Wat2	Wat3	Wat4
1ZGC a	1468	1489	1531	1481
1Q84 a	1708	1715	1735	1755
2CKM a	2062	2061	2035	2054
4EY7 a	729	722	731	737
5K5E b	727	740	731	—

^aAChE.

^bBuChE.

Preparation of structures

The three-dimensional structures of the compounds were prepared using ChemBio3D Ultra 14 Suite (PerkinElmer, Waltham, MA, USA, 2014), and isomers, protonation states, and tautomers of the ligands were determined using LigPrep/Epik from Maestro at pH 7.0 ± 0.4 (small-molecule drug discovery suite 2019-1, Schrödinger, LLC, New York, NY, 2019).55,56 The AChE and BuChE structures were prepared with the protein preparation wizard tool from Maestro.57 The protonation states of the amino acid residues were determined using PROPKA at pH 7. Interestingly, Glu199 located near the catalytic triad was predicted to be neutral due to a high pK_a value (∼10) for all the AChE and BuChE structures. Recently, Wan et al. proposed that the protonated form of Glu199 can interact with conserved water and stabilize the catalytic triad in the molecular simulations of the BuChE–tacrine complex.58 The optimization of the hydrogen bond network between the protein and reference ligand was performed to adjust the orientation of the hydrogen atoms, followed by energy minimization with fixed heavy atoms.

ADME-Tox predictions

The ADME-Tox properties of all the compounds and the reference ligands (donepezil and bis(7)-tacrine) were determined by using the QikProp tool in the normal mode (Schrödinger Release 2019-1: QikProp, Schrödinger, LLC, New York, NY, 2019). QikProp accurately predicts diverse pharmaceutically relevant properties such as log P, log S, overall CNS activity, Caco-2 and MDCK cell permeabilities, and log IC₅₀ for hERG K⁺ channel blockage based on the three-dimensional structure of the compounds, providing reference values for comparing their properties to those of 95% of known drugs.

Evaluation of cytotoxicity in in vitro models

Cell culture and treatment

The Vero (African green monkey kidney cells), HepG2 (liver hepatocellular carcinoma), and C6 (astroglial) cell lines were obtained from the Rio de Janeiro Cell Bank (BCRJ). All the cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum (FBS) (Vero cells and HepG2) or 5% FBS (C6 cells), 0.5% amphotericin B (250 μg mL^–1), 0.5% penicillin (100 IU mL^–1) and streptomycin (10 mg mL^–1), at 37 °C, in a humid atmosphere containing 5% CO₂. The cells were seeded in 96-well plates (3 × 10⁴ cell per well) and incubated for 24 h. For cytotoxic studies, the selected compounds (3.1–250.0 μM) and tacrine (standard drug; 12.5 to 500 μM) were dissolved in a culture medium containing 0.5% DMSO, and the cells were exposed to them for 24 h. After the treatment, the cells were subjected to the cell viability assay (MTT assay). The cells treated with DMSO (final concentration: 0.5%) were used as the control. The experiments were performed in triplicate for each concentration and tested at least in three independent experiments.

Cell viability assays

The cell viabilities of the different cell lines were assessed by the MTT assay.59–61 Briefly, after 24 h of treatment with the selected compounds, the medium was withdrawn and 100 μL of 0.5 mg mL^–1 MTT medium solution was added to each well. The plates were incubated for 1 h at 37 °C and 5% CO₂; subsequently, MTT was removed and the colored product was solubilized in 100 μL DMSO. The optical density of each well was measured at 570 and 630 nm using the SpectraMax 190 M2 microplate spectrophotometer system (Molecular Devices). The results were expressed in IC₅₀ as compared to the control (DMSO vehicle at 0.5%).

PAMPA assay

Initially, a 10 mM solution of each compound (test and control) in DMSO was prepared. Then, in a 5 mL glass vial, 250 μL of the freshly prepared solution was homogenized with 4750 μL PBS (pH 6.6) at 10 mM. The solution was then filtered (PVDF filter: 0.45 μM) and reserved. Subsequently, 180 μL of PBS (pH 7.4) : DMSO (95 : 5) solution was added to the wells of the recipient plate and 5 μL of soybean l-α-phosphatidylcholine lipid solution (20 mg lipid per mL in dodecane) in the wells of the donor plate. After 5 min, 180 μL of the reserved solution containing each compound was added in triplicate to the donor plate.

The donor plate was then carefully placed over the recipient forming a sandwich system and it was stirred at 50 rpm for 8 h at room temperature (±25 °C) in a closed container containing 10 mL PBS (pH 7.4). After this period, the donor plate was removed and the contents present on the recipient plate were transferred to a UV reading plate, which were read (SpectraMax 5®, Molecular Devices) at the preestablished wavelengths for each compound. The blank was prepared in the presence of 180 μl PBS (pH 7.4) : DMSO (95 : 5) solution.62,63

The optical density values of the receptor plate contents at each wavelength were compared with the initial values, acquired by reading the initially reserved solutions containing each compound. It is noteworthy that the experiments were performed in triplicate and two different analyses (n = 2) were performed in the presence of the controls, namely, acyclovir, atenolol, ceftriaxone, coumarin, diclofenac, hydrocortisone, norfloxacin, ranitidine, sulfasalazine, and verapamil, which correspond to the reference with the absorbed fraction values described; this yielded an experimentally reliability standard.64

For the data analysis, the optical density values obtained at the reading at each selected wavelength for each of the compounds were analyzed in comparison with the control values, which were used to elaborate the line equation and determine the permeability coefficient (Pe) by using a spreadsheet prepared earlier in the Excel® program. The absorbed fraction (Fa%) of each compound under investigation was determined and classified according to the percentage of absorption: low absorption (0–29%); medium absorption (30–69%); and high absorption (70–100%).65

Conclusions

In summary, we synthesized a novel series of lophine–carbohydrate hybrids and evaluated their AChE and BuChE inhibitory activities. In addition, we investigated the binding mode of the compounds to ChEI using molecular modeling and tested them against a panel of human cell lines to measure their cytotoxic effects. All the compounds inhibited BuChE; 9a (a d-galactose derivative) was the most potent inhibitor (IC₅₀ = 0.17 μM). Only one compound (8b) exhibited activity against AChE (IC₅₀ = 2.75 μM). The BuChE-selectivity character of lophine-based carbohydrate derivatives observed in this work is probably due to the T-stacking interaction with Trp231, a residue that is buried and inaccessible in AChE. Further, it may have contributed to the high desolvation cost of buried sugar moieties without requiring hydrogen bonds with any residues inside the binding cavity of AChE as soon as the hydrophobic lophine moiety does not fit into the bottom of the gorge. In addition, the most potent compounds, namely, 9a and 8b, had no cytotoxic effects in all the cellular models under investigation, without any reduction in the cell viability up to 500 times higher concentrations than those for active AChE and BuChE.

Considering the several limitations of the current therapies toward neurodegenerative disorders, new, effective, and safe inhibitors are required, particularly those that can act through multiple receptor sites, but do not induce undesirable effects. In this sense, a simple synthesis approach, BuChE selectivity, and high potent inhibitory activity of new compounds formulated in this study, in addition to the absence of in vitro cytotoxicity and good ADME-Tox profile, ensure that the lophine–carbohydrate derivatives proposed in this study represent new promising prototypes in the development of novel therapies toward neurodegenerative disorders.

Ethical statement

All animal procedures were performed in accordance with the Guidelines for Care and Use of Laboratory Animals of Oswaldo Cruz Foundation (FIOCRUZ) and approved by the Animal Ethics Committee of FIOCRUZ (CEUA, license number L044/2015 with the additive license LA-010/2016).

Conflicts of interest

There is no conflict of interest to declare.