Neuroprotective Potential of New Monoterpene-Adamatane Conjugates—A Pilot Study

Abstract

Neurodegenerative diseases, including Alzheimer’s disease, are marked by cholinergic dysfunction, oxidative stress, and reduced neurotrophic support, which drives the quest for multifunctional therapeutic agents. This pilot study presents four novel monoterpene–aminoadamantane conjugates (MACs 1–4) designed to combine the antioxidant and neuromodulatory characteristics of monoterpenes with the neuroprotective properties of aminoadamantane derivatives. Their physicochemical characteristics, blood–brain barrier permeability, and binding affinity to human acetylcholinesterase (AChE) were evaluated using molecular docking and in silico descriptor analysis. In vivo, the neuroprotective efficacy of the MACs was investigated in a scopolamine-induced dementia model in rats, employing behavioral tests. Biochemical assays conducted in the hippocampus and prefrontal cortex assessed AChE activity, antioxidant enzyme performance, lipid peroxidation levels, total glutathione content, and BDNF concentrations. The findings indicate that MAC1, MAC3, and MAC4 demonstrate favorable calculated blood–brain barrier permeability, strong predicted affinity for AChE, and significant in vivo alleviation of scopolamine-induced memory deficits, in conjunction with improvement of key markers of oxidative stress and cholinergic function. These results show that the structural hybridization of myrtenal with aminoadamantane frameworks produces promising multifunctional ligands that are relevant for Alzheimer’s-type neurodegeneration.

1. Introduction

Neurodegenerative diseases, including Alzheimer’s disease (AD), pose a significant global health challenge, marked by progressive neuronal degeneration, oxidative stress, synaptic dysfunction, and disrupted neurotransmission [1]. One of the primary pathological features of AD is characterized by the degeneration of cholinergic neurons and dysregulation of acetylcholine metabolism, in which AChE activity plays a contributory role, leading to cognitive impairments, learning difficulties, and memory loss [2]. Alzheimer’s disease encompasses additional significant pathological mechanisms, such as the buildup of beta-amyloid plaques, the presence of tau neurofibrillary tangles, mitochondrial dysfunction, and synaptic impairments. Current pharmacological treatments, such as donepezil, galantamine, and rivastigmine, provide symptomatic relief through AChE inhibition; however, they do not alter the course of the disease and are frequently linked to peripheral side effects [3]. Consequently, there is a growing interest in multifunctional compounds capable of targeting multiple pathological mechanisms, such as oxidative stress, neuroinflammation, and cholinergic dysfunction.

In this regard, monoterpenes and their derivatives have emerged as promising frameworks for the design of neuroprotective drugs. They demonstrate a wide range of biological activities, such as antioxidant [4], anti-inflammatory [5,6,7], analgesic [8,9,10], antitumor [11,12,13,14], and neuromodulatory [15,16,17] effects, rendering them suitable for multitarget strategies in the treatment of neurodegenerative diseases [18,19,20,21]. These compounds include the bicyclic monoterpenoid (-)-myrtenal (Figure 1), which has garnered increasing interest due to its ability to mitigate oxidative stress, restore mitochondrial function, and enhance cognitive performance in experimental dementia models [22].

In a recent study, we illustrated that two myrtenal–adamantane conjugates MAC-197 and MAC-198 containing 2- and 1-aminoadamantane fragments, respectively (Figure 1) significantly mitigated scopolamine (Scop)-induced cognitive deficits and oxidative damage in rats, indicating synergistic interactions between the terpenoid and aminoadamantane components [23]. MAC-197 and MAC-198 enhanced the memory of rats impaired by Scop, with results being more conclusive following repeated administration over a period of nine days. In terms of AChE activity, a notable reduction in the enzyme function, which was over-activated by Scop, was detected in the cerebral cortex, with MAC-198 demonstrating a more significant effect. In the hippocampus, the influence on this indicator maintained a consistent trend, although no statistically significant differences were found, except for MAC-198. The antioxidant properties of both compounds are closely related, yet not identical. For instance, only MAC-198 elevated the glutathione levels in the cerebral cortex of rats suffering from scopolamine-induced dementia, while both compounds restored catalase activity toward control levels.

These cumulative findings establish a robust basis for ongoing structural refinement of myrtenal derivatives. The integration of the neuroprotective monoterpenoid core with the aminoadamantane pharmacophore—recognized for its N-methyl-D-aspartate (NMDA) receptor modulation in some of its derivatives, and neurostabilizing attributes [24]—may yield innovative multifunctional molecules capable of concurrently modulating cholinergic and oxidative pathways. Additionally, computational methodologies such as in silico docking and pharmacokinetic modeling support the strategic design of these conjugates with desirable blood–brain barrier permeability and target specificity [25].

Building upon these insights, the current pilot study sought to develop a novel series of myrtenal–aminoadamantane conjugates (MACs 1–4, Figure 1) and to assess their neuroprotective potential through a combined in silico and in vivo methodology. From the structural point of view, MAC1 is an analog of MAC-198, but with memantine fragment instead of 1-aminoadamantane moiety. Memantine is used for the treatment of patients with moderate-to-severe AD [26,27] and it could potentially contribute to anti-AD-related activity. MAC2 is an amide analog of amine MAC-198. Synthesis of MAC3 and MAC4, analogs of MAC-197 and MAC1, respectively, includes the use of nopinal, the homolog of myrtenal, as a terpene precursor. This allowed us to obtain new conjugates with the linker extended by a methylene group. Thus, as compared with earlier studied compounds, we varied both the adamantane fragment and the linker. Molecular docking was conducted to evaluate binding affinity towards human AChE, followed by an analysis of pharmacokinetic descriptors and behavioral testing in a scopolamine-induced dementia model in rats. Concurrently, essential biochemical parameters—AChE activity, antioxidant enzyme profiles, lipid peroxidation levels, and brain-derived neurotrophic factor (BDNF) expression—were examined to clarify the mechanisms that facilitate cognitive recovery.

By merging synthetic and computational chemistry, behavioral pharmacology, and neurobiochemical evaluation, this study aims to identify multifunctional ligands that may hold therapeutic significance for AD’s-related dementia and other related neurodegenerative conditions.

2. Materials and Methods

2.1. Chemicals

(-)-Scopolamine hydrobromide ≥ 98% (TLC) (cat. No. S1875), 1-Adamantylamine, 97% (cat. No. 138576), and (1R)-(-)-Myrtenal, 98% (cat. No. 218243) were acquired from Sigma Aldrich (St. Louis, Missouri (MO), USA). The synthesis of MACs 1–4 (Figure 1) was conducted at the Department of Medicinal Chemistry at the Novosibirsk Institute of Organic Chemistry in Novosibirsk, Russia. All reagents and solvents were purchased from commercial sources and were used as received without further purification. Conversion of the reagents was determined using an Agilent 7820A gas chromatograph (GC, Agilent Technologies, Santa Clara, CA, USA) equipped with a flame ionization detector and an Agilent J&W HP-5 capillary column (30 m × 0.25 mm × 0.25 µm), the carrier gas was helium (flow rate of 2 mL min^–1, split ratio of 99:1), and the temperature ranged from 120 to 280 °C at a heating rate of 20 deg min^–1. Merck (Merck KGaA, Darmstadt, Germany) silica gel (63–200 µm) was used for column chromatography. The ¹H and ¹³C NMR spectra in CDCl₃ were recorded on a Bruker AV-400 spectrometer (400.13 and 100.61 MHz, respectively, Bruker, Billerica, MA, USA). The residual signals of the solvent were used as references (δH 7.24, δC 76.90 for CDCl₃). High-resolution mass spectra were recorded on a Thermo Scientific DFS instrument (Thermo Fisher Scientific Inc., Waltham, MA, USA) in full-scan mode over the m/z range of 15–500 by ionization with an electron impact of 70 eV, and direct introduction of samples. The specific rotation values [α]_D²⁵ were determined on a PolAAr 3005 polarimeter and expressed in (deg × mL)/(g × dm), while the concentration was expressed in g per 100 mL of solution. Cuvette length is 5 cm. The melting point was measured using a Mettler Toledo FP90 Thermal Analyzer with an FP81/HT Melting Point Apparatus (Mettler-Toledo GmbH, Im langacher, 8606, Greifensee, Switzerland). The atomic numbering in the compounds is provided for the assignment of signals in the NMR spectra and is different from the atomic numbering in the systematic name. All of the target free bases have a purity of no less than 95% (GC data). The analytical and spectroscopic studies were conducted at the Chemical Service Center for the collective use of the Siberian Branch of the Russian Academy of Sciences (SB RAS).

N-(((1R,5S)-6,6-Dimethylbicyclo[3.1.1]hept-2-en-2-yl)-methyl)adamantyl-1-carboxamide MAC2.

A solution of 1-adamantanecarboxylic acid chloride 4 (200 mg, 1 mmol) and triethylamine (170 µL, 1.2 mmol) in toluene (3 mL) were added dropwise to a solution of monoterpene amine 2 (160 mg, 1 mmol) in toluene (7 mL) at 0 °C. Then, the cooling was removed, and the reaction mixture was stirred for 10 min at room temperature, the solvent was distilled off, and the residue was dissolved in EtOAc (25 mL) and washed successively with aqueous solutions of 5% NaOH (20 mL), 5% HCl (20 mL), and a saturated NaCl solution (20 mL). The organic phase was dried over Na₂SO₄ and evaporated. Amide MAC2 was isolated by column chromatography with a yield of 131 mg, 42%.

White solid, m.p. 183.6–185.4 °C. ¹H-NMR (CDCl₃): 0.82 (s, 3H, 3H22), 1.12 (d, 1H, J_{20anti, 20syn} = 8.6 Hz, H_20anti), 1.26 (s, 3H, 3H21), 1.64–1.78 (m, 6H, 2H4, 2H6, 2H10), 1.84 (d, 6H, ³J = 3.0 Hz, 2H2, 2H8, 2H9), 1.99 (ddd, 1H, J_{19, 17} = J_{19, 20syn} = 5.6 Hz, J_{19, 15} = 1.5 Hz, H19), 1.99–2.03 (m, 3H, H3, H5, H7), 2.05–2.09 (m, 1H, H17), 2.11–2.21 (m, 1H, H16), 2.21–2.31 (m, 1H, H16′), 2.35 (ddd, 1H, J_{20syn, 20anti} = 8.6 Hz, J_{20syn, 17} = J_{20syn, 19} = 5.6 Hz, H_20syn), 3.63–3.73 (m, 1H, H-13), 3.73–3.84 (m, 1H, H13′), 5.32–5.35 (m, 1H, H-15), 5.45–5.52 (m, 1H, NH). ¹³C NMR (CDCl₃): 21.13 (q, C22), 26.00 (q, C21), 28.04 (d, C3, C5, C7), 31.02 (t, C16), 31.39 (t, C20), 34.42 (t, C4, C6, C10), 37.82 (s, C18), 39.26 (t, C2, C8, C9), 40.58 (s, C1), 40.65 (d, C17), 43.70 (t, C13), 43.90 (d, C19), 118.20 (d, C15), 144.82 (s, C14), 177.54 (s, C11). HRMS: m/z calcd for C₂₁H₃₁O₁N₁⁺ (M⁺) 313.2400, found 313.2396. [α]₅₈₉²⁵ = −19 (C = 0.8, MeOH).

General procedure for the synthesis of amines MACs 1, 3, 4.

To a solution of 2-adamantylamine or memantine (1 mmol) in MeOH (5 mL), a solution of corresponding aldehyde (2 mmol) in MeOH (3 mL) was added. The mixture was kept at room temperature for 12 h, then cooled in an ice bath and NaBH₄ (6 mmol) was added. After adding NaBH₄ to the reaction mixture, it was stirred for 4 h without cooling. The solvent was distilled off, and 8 mL of saturated NaCl solution was added to the residue in the flask. The product was extracted with Et₂O. The amines MACs 1, 3, and 4 were isolated from the reaction mixture through precipitation from an ether solution as hydrochloride salts, achieved by the addition of ethyl acetate that was saturated with hydrogen chloride. The resulting precipitate was subsequently filtered, washed with ether, and treated with a saturated sodium bicarbonate solution. The product was then extracted from the aqueous phase using ether. The organic phase was dried over Na₂SO₄, after which the drying agent was removed by filtration, and the solvent was distilled off using a rotary evaporator.

N-[((1R,5S)-6,6-Dimethylbicyclo[3.1.1]hept-2-en-2-yl)methyl]-5,7-dimethyladamantane-1-amine MAC1.

Yield 97 mg, 31%. Viscous oil. ¹H-NMR (CDCl₃): 0.76 (s, 3H, C22H3), 0.83 (s, 6H, C11H3, C12H3), 1.13 (d, 1H, ²J = 7.6 Hz, H21anti), 1.16–1.18 (m, 1H, H3), 1.26 (s, 3H, C23H3), 1.20–1.40 (m, 6H, 2H4, 2H6, 2H10), 1.64–1.80 (m, 5H, H2, 2H8, 2H9) 1.92–2.05 (m, 3H, H17, H18, H20), 2.12–2.29 (m, 3H, H2′, H17′), 2.39–2.45 (m, 1H, H21syn), 2.51 (brs, NH), 3.20–3.35 (m, 1H, H14), 3.45–3.60 (m, 1H, H14′) 5.88 (s, 1H, H16). ¹³C NMR (CDCl₃): 20.76 (q, C23), 23.33 (d, C3), 25.87 (q, C22), 29.67 (t, C10), 30.46 (t, C17), 31.34 (t, C21), 31.49 (s, C5, C7), 32.53 (q, C11, C12), 39.87 (s, C19), 41.76 (d, C18), 44.17 (t, C2), 44.32 (d, C20), 45.32 (t, C4), 48.64 (s, C1), 49.77 (t, C8, C9), 50.18 (t, C6), 59.61 (t, C14), 125.37 (d, C16), 139.53 (s, C15). HRMS: m/z calcd for C₂₂H₃₅N⁺ (M⁺) 313.2770, found 313.2764. [α]₅₈₉²⁵ = −16 (C = 0.6, MeOH).

N-[((1R,5S)-6,6-Dimethylbicyclo[3.1.1]hept-2-en-2-yl)ethyl]-5,7-dimethyladamantane-1-amine MAC3.

Yield 78 mg, 24%. Viscous oil. ¹H-NMR (CDCl₃): 0.72 (s, 9H, C11H3, C12H3, C23H3), 0.95–1.03 (m, 3H, 2H6, H22anti), 1.15 (s, 3H, C24H3), 1.06–1.19 (m, 10H, 2H2, 2H4, 2H8, 2H9, 2H10), 1.30–1.38 (m, 2H15), 1.88–1.92 (m, 1H, H19), 1.93–2.03 (m, 3H, H3, H21, NH), 2.05–2.15 (m, 2H, H18, H18′), 2.18–2.26 (m, 1H, H22syn), 2.42–2.53 (m, 2H14) 5.16 (s, 1H, H17). ¹³C NMR (CDCl₃): 21.08 (q, C24), 26.09 (q, C23), 30.17 (q, C11, C12), 30.30 (d, C3), 31.05 (s, C5, C7), 31.56 (t, C18), 32.07 (t, C22), 37.38 (t, C15), 37.73 (s, C20), 40.48 (d, C19), 40.80 (t, C2), 41.13 (t, C4), 42.80 (t, C10), 45.06 (t, C6), 48.80 (d, C21), 48.82 (s, C1), 50.74 (t, C14), 51.68 (t, C8, C9), 117.52 (d, C17), 146.01 (s, C16). HRMS: m/z calcd for C₂₃H₃₇N⁺ (M⁺) 327.2926, found 327.2924. [α]₅₈₉²⁴ = −18 (C = 0.9, MeOH).

N-[((1R,5S)-6,6-Dimethylbicyclo[3.1.1]hept-2-en-2-yl)ethyl]-adamantane-2-amine MAC4.

Yield 105 mg, 35%. Viscous oil. ¹H-NMR (CDCl₃): 0.74 (s, 3H, C21H3), 1.04 (d, 1H, ²J = 7.9 Hz, H20anti), 1.18 (s, 3H, C22H3), 1.38 (d, ²J = 12.6, 2H13), 1.57–1.64 (m, 4H, 2H5, 2H7), 1.65–1.69 (m, 1H, H8), 1.70–1.78 (m, 5H, H3, 2H9, 2H10), 1.81–1.89 (m, 2H, H4, H6), 1.90–1.97 (m, 1H, H2), 1.96–2.01 (m, 1H, H17), 2.02–2.10 (m, 2H, H19, NH), 2.06–2.20 (m, 2H, H16, H16′), 2.24–2.29 (m, 1H, H20syn), 2.44–2.57 (m, 2H12), 2.60 (brs, 1H, H1), 5.18 (s, 1H, H15). ¹³C NMR (CDCl₃): 20.90 (q, C22), 26.06 (q, C21), 27.63 (d, C3, C7, C9, C10), 31.54 (t, C20), 31.61 (t, C16), 32.17 (s, C5), 37.34 (t, C2, C8), 37.42 (s, C18), 37.75 (t, C13), 40.53 (d, C17), 45.24 (d, C19), 51.17 (t, C12), 67.71 (s, C1), 117.30 (d, C15), 146.22 (s, C14). HRMS: m/z calcd for C₂₁H₃₃N⁺ (M⁺) 299.2613, found 299.2615. [α]₅₈₉²⁴ = −22 (C = 0.7, MeOH).

General procedure for the synthesis of hydrochlorides of MACs 1, 3, 4.

Corresponding hydrochlorides were obtained by passing dry gaseous HCl through a solution of amines in dry diethyl ether in concentration of 100 mg/5 mL until precipitation ceased. The solution was stirred for 10 min; the precipitate was filtered, washed with dry diethyl ether (3 × 3 mL), and dried in a vacuum desiccator. Salts were formed in quantitative yield. White solids. Melting point data for the salts: 263.2–264.1 °C (MAC1); 246.7–247.8 °C (MAC3); 281.0–281.1 °C (MAC4).

2.2. Molecular Modeling and Screening

The compounds were docked against the crystal structure of AChE (PDB ID: 5HF9, Homo Sapiens, resolution 2.20 Å) [28], which were obtained from the Protein Data Bank (PDB) [29,30]. The GOLD (v2025.1, Cambridge Crystallographic Data Centre, Cambridge, UK) software suite was used to prepare the crystal structures for docking, i.e., the hydrogen atoms were added, and the co-crystallized ligands identified: 4-(aminocarbonyl)-1-[({2-[(E)-(hydroxyimino) methyl]pyridiunum-1-yl}methoxy) methyl]pydidinium (HI6). The docking center was defined as the position of HI6 with a 10 Å radius. None of the crystalline water molecules form hydrogen bonds with HI6 nor are structurally relevant in the binding pocket and were therefore deleted. Fifty docking runs were allowed for each ligand with default search efficiency (100%). The basic amino acids lysine and arginine were defined as protonated. Furthermore, aspartic and glutamic acids were assumed deprotonated. The GoldScore (GS) [31], ChemScore (CS) [32,33], Piecewise Linear Potential (ChemPLP) [34], and Astex Statistical Potential (ASP) [35] scoring functions were implemented to predict the binding modes and relative binding energies of the ligands using the GOLD v2025.1 software suite. The robustness of the docking protocol has been verified against the AChE structure used here [23]. The ligands were built in the HyperChem8.0.10 software (Hypercube, Inc., Gainesville, FL, USA, Copyright 1995–2011), and the geometries were optimized with the MM+ force field [36].

ChatGPT (v5) was used for the exploratory, qualitative evaluation of the charged state of the ligands at physiological pH. The sentence “Can I have the charges of the following compounds at a physiological pH?” was used for 25 ligands at a time. The outputs generated by ChatGPT were utilized exclusively for comparative and exploratory objectives and did not substitute or supersede the protonation states designated by recognized computational chemistry instruments.

The QikProp 6.2 [QikProp 6.2, 2021, 6.2] software package was used to calculate the molecular descriptors of the molecules. The reliability of QikProp for the calculated descriptors is established [37]. The Known Drug Indexes (KDI) were calculated from the molecular descriptors as described by Eurtivong and Reynisson [38]. For application in Excel, columns for each property were created, and the following equations were used to derive the KDI numbers for each descriptor:

KDI MW: =EXP(−((MW − 371.76)²)/(2 ∗ (112.76²)))

(1)

KDI Log P: =EXP(−((LogP − 2.82)²)/(2 ∗ (2.21²)))

(2)

KDI HD: =EXP(−((HD − 1.88)²)/(2 ∗ (1.7²)))

(3)

KDI HA: =EXP(−((HA − 5.72)²)/(2 ∗ (2.86²)))

(4)

KDI RB: =EXP(−((RB − 4.44)²)/(2 ∗ (3.55²)))

(5)

and

KDI PSA: =EXP(−((PSA − 79.4)²)/(2 ∗ (54.16²))).

(6)

These equations could simply be copied into Excel and the descriptor name (e.g., MW) substituted with the value in the relevant column. To derive KDI_2A, this equation was used:

KDI_2A = (KDI MW + KDI LogP + KDI HD + KDI HA + KDI RB + KDI PSA),

(7)

and for KDI_2B:

KDI_2B = (KDI MW × KDI LogP × KDI HD × KDI HA × KDI RB × KDI PSA)

(8)

The Gaussian 16 software suite [39] was used with unrestricted DFT. The B3LYP functional hybrid approach was employed [40,41,42] and a standard 6-31G(d,p) basis set [43,44] was used for geometry optimization and frequency analysis (keywords: opt freq). The zero-point vibrational energies (ZPEs) were scaled according to Wong (0.9804) [45]. In all cases, normal modes revealed no imaginary frequencies indicating that they represent minima on the potential energy surface. The subsequent energy calculations were then performed with the larger 6-311G(2df, p) basis set. Adiabatic ionization potentials (IPs) and adiabatic electron affinities (EAs) were calculated as described in Forseman and Frisch [46]. The energies and ZPEs are given in Table S7.

2.3. In Vivo Experiment

2.3.1. Experimental Animals

Male adult Wistar rats weighing between 180 and 220 g were utilized for the experiments and housed under standard laboratory conditions within plastic cages featuring a 12 h light/dark cycle, with access to drinking water and rodent food ad libitum, as well as optimal indoor temperature, humidity, and ventilation. The experimental protocols were conducted in accordance with the guidelines established by the Ethics Committee of the Bulgarian Food Safety Agency (Permit No. 432/25.04.2025) and in compliance with national laws and regulations (Regulation No. 20 of 1 November 2012 regarding the minimum standards for the protection and humane treatment of experimental animals, as well as the stipulations for facilities used for their care, breeding, and/or supply, effective from 1 January 2013, issued by the Ministry of Agriculture and Food, published in the State Gazette, No. 87 of 9 November 2012), while also conforming to the European Directive and the regulations set forth by the Ethics Committee of the Institute of Neurobiology of the Bulgarian Academy of Sciences.

2.3.2. Experimental Protocol

Laboratory rodents were separated in the following groups (n = 8): (1) Controls (Saline), (2) Scopolamine (Scop), (3) Scopolamine and Myrtenal (Scop + M), (4) Scopolamine and 1-Adamantylamine (Scop + AA), (5) Scopolamine and MAC1 (Scop + MAC1), (6) Scopolamine and MAC2 (Scop + MAC2), (7) Scopolamine and MAC3 (Scop + MAC3), and (8) Scopolamine and MAC4 (Scop + MAC4).

The neurodegeneration model was established through the intraperitoneal administration of scopolamine (2.0 mg/kg b.w.) over a period of 11 consecutive days, utilizing a solution of scopolamine hydrobromide dissolved in distilled water. Simultaneously, other substances were administered alongside scopolamine via separate intraperitoneal injections. The four novel MAC compounds were injected at a dosage of 1 mg/kg b.w. Myrt served as a reference compound at an effective dosage of 40 mg/kg b.w., as previously documented in our research [22]. Additionally, standard Adamantylamine (AA) was administered at a dose of 25 mg/kg b.w., as selected from the existing literature [24]. To formulate the solutions of Myrt and MACs, Tween 80 was utilized as a co-solvent. All solutions were prepared ex tempore and administered at the same time each day.

The experiment commenced with preliminary training in the behavioral assessments, as detailed below. Following a 24 h period after the final treatment, the relevant behavioral assessments were conducted—focusing on memory and learning, as well as spatial orientation. After the behavioral experiments, the rodents were euthanized via mild CO₂ inhalation in accordance with ethical standards for the treatment of experimental animals. Each rat’s brain was swiftly dissected, isolating the two brain regions linked to memory functions—the hippocampus and the prefrontal cortex. Subsequent to processing as outlined below (the Biochemical studies Section), the samples were prepared for the analysis of lipid peroxidation product levels; total glutathione content; the activity of the enzymes catalase, superoxide dismutase, and glutathione peroxidase; the activity of brain acetylcholinesterase; and the levels of BDNF.

2.3.3. Behavioral Tests

• The passive avoidance test was employed to assess both short-term and long-term memory capabilities [47,48]. In this experimental design, to prevent receiving an electric shock to their feet, the rodent was required to learn to remain in the brightly illuminated section of the apparatus and refrain from entering the favored dark section. Preliminary training (initial latency, IL) was conducted prior to the administration of the test substances. Each subject was positioned in the light compartment of the apparatus, with the barrier between the light and dark sections left open. Upon the rat’s entry into the dark area (with all four paws), the door was shut, and a mild electric shock (0.7 mA for 3 s) was delivered through the floor. The experimental phase consisted of an initial test conducted 1 h and 24 h after the first scopolamine application, and on day 12 after the start of scopolamine treatment; no further electric shocks were administered. As a measure of memory status, the latency time (up to 180 s), which indicated the duration taken by the animal to transit to the dark section of the apparatus, was used. A key indicator of memory performance is the variation in latency when compared to the initial learning phase (delta latency time = testing latency—initial learning latency).
• The Barnes maze test was employed to evaluate spatial memory [49]. The apparatus comprises a circular platform measuring 122 cm in diameter positioned 80 cm above the ground, featuring 20 peripheral holes, with a dark box located beneath one of these holes, which functions as an escape box. Upon placing the animal in the center of the open platform, a stimulus consisting of bright light (3000 lx) and loud noise (76 dB) compels the animal to seek shelter and conceal itself in the escape box. The assessment of spatial memory is conducted by comparing the duration it takes for the animals to locate and enter the hole containing the escape box (total latency) and the count of “incorrect” holes explored prior to reaching it (termed the “number of head dips”). The experiment is carried out in two phases: a “training” session and a “test” session, following a previously established protocol [50]. The “training” session started on the 4th day after the start of scopolamine application, while the “test session” took place on the 12th day. Training was conducted over a span of four consecutive days. On the first day, the animals were permitted to explore the platform freely for 5 min. During each of the subsequent 3 days, four trials (interspersed with 15 min intervals) were performed for each animal, with the animal initially placed at the center of the platform and allowed to search for the escape box. The duration taken to discover the hole with the escape box beneath and the number of “incorrect” holes explored prior to discovering the escape box were documented. On the second of the four training days, the experiment was executed without the presence of an aversive stimulus, while the following two days incorporated both light and noise. The test session adhered to the same protocol as the final training day, specifically involving 4 trials (separated by 15 min) for each animal, with the time taken to locate the hole containing the escape box and the number of “incorrect” holes explored recorded.
• The novel object recognition test [51,52].

This assessment is designed to evaluate alterations in non-spatial memory and serves as an indicator of visual (object) recognition memory. It focuses on the identification of both familiar and novel objects, leveraging the inherent curiosity of rodents and their inclination to investigate new and unfamiliar stimuli. The premise is that when a rat is presented with a familiar object alongside a new one, it will allocate less time to the familiar object. The experimental setup consists of a dark plexiglass chamber measuring 60 × 60 × 40 cm, allowing the rat to navigate freely without restrictions. This investigation was carried out over two consecutive days. On the initial day, the animal was carefully placed in the chamber and permitted to explore it freely for 5 min without any objects present. Subsequently, a second phase of 5 min occurred on the same day, during which the animal was reintroduced to the chamber containing two identical objects (A + A). The actual testing took place on the following day, when the animal was again placed in the same environment, but this time with two objects: one being the familiar object and the other a similar yet distinct object (A + B). The duration spent exploring both the old and new objects was recorded over a 5 min interval. A recognition index (RI), which serves as a measure of recognition memory, was computed using the formula:

RI = (N/(N + F)),

(9)

where N represents the time spent exploring the new object B, and F denotes the time spent exploring the familiar object A. RI values can range from 0 to 1, with an RI greater than 0.5 suggesting that the animals are capable of remembering and distinguishing between the objects; an RI less than 0.5 indicates potential impairments in recognition memory.

2.3.4. Biochemical Studies

• Tissue preparation: Following the extraction of the brains, they were rinsed with chilled 0.15 M KCl, and the hippocampus and cerebral cortex were separated. Each of the resulting preparations was homogenized individually in chilled 0.15 M KCl—10 mM potassium phosphate buffer (pH 7.4) and subjected to centrifugation for 10 min at 3000 rpm. A portion of the resulting post-nuclear supernatant fraction was utilized for subsequent spectrophotometric analysis of lipid peroxidation (LPO) product levels, AChE activity, and BDNF content. Another portion of the homogenate was utilized to evaluate the activity of antioxidant enzymes (CAT, SOD, and GPx). For this analysis, a supernatant was obtained by centrifugation at 12,000 rpm for 20 min. All procedures were conducted at temperatures ranging from 0 to +4 °C.
• Protein concentrations were assessed using the Lowry method [53], with the calculation curve based on bovine serum albumin as the standard. The method is based on the established capacity of copper ions in an alkaline environment to create complexes with aromatic amino acids, including tyrosine and tryptophan, which they bind to with significant affinity. Folin’s reagent is introduced to brain homogenate samples, with the objective of generating a blue-colored complex that exhibits an absorption peak at λ = 750 nm, the intensity of which correlates with the protein quantity present in the sample. To determine the protein concentration (in mg/mL), a calibration curve is developed utilizing bovine serum albumin.
• Brain-derived neurotrophic factor (BDNF) levels in the cortex and hippocampus were quantified utilizing an ELISA kit (cat. No. MBS703433-96, MyBioSource, Inc., San Diego, CA, USA). Following the provided protocol, 100 µL of BDNF standards (ranging from 0.156 to 10 ng/mL) and 100 µL of BDNF samples were dispensed into the wells of a microplate that had been precoated with BDNF. This was followed by a 90 min incubation in the dark at 37 °C. Afterward, the liquid from each well was discarded, and 100 µL of detection solution A was introduced; the plate was incubated once more for 45 min at 37 °C, and each well was washed three times with wash buffer. The same steps were repeated using detection solution B, followed by additional washing. The introduction of 3,3′,5,5′-tetramethylbenzidine (TMB) initiated a color reaction, which was halted 10 min later with stop solution, and the absorbance at 450 nm was immediately recorded using an MR-96A microplate reader.
• The activity of AChE in the cortex and hippocampus was assessed following the methodology established by Ellman et al. (1961) [54]. Brain supernatants were combined with a solution that included 1.0 mM acetyl thiocholine (AcSCh), 0.1 mM 5,5′-dithio-bis (2-nitrobenzoic acid) (DTNB), and 100 mM phosphate buffer (pH 8.0). The samples were then incubated for 5 min at 37 °C. The intensity of the yellow hue generated from the reaction between thiocholine and DTNB was quantified spectrophotometrically at a wavelength of λ = 412 nm. The findings were reported as AChE activity in µmol min/g protein.
• The method for the assessment of lipid peroxidation product levels is based on the examination of thiobarbituric acid reactive substances (TBARs), which are generated as a by-product of lipid peroxidation, utilizing TBA as a reagent [55]. The presence of malondialdehyde (MDA) in the sample is assessed, as it is one of several low-molecular-weight products that arise from the breakdown of certain primary and secondary products of lipid peroxidation, in addition to being produced from lipid hydroperoxides under the hydrolytic conditions of the reaction. In an acidic environment, at elevated temperatures, a colored complex is formed with an absorption maximum at λ = 532 nm. To achieve a solution with a protein concentration of 1 mg/mL, the isolated post-nuclear brain homogenate is diluted, and the samples are incubated for one hour at 37 °C. A mixture of acids (2.8% TCA + 5 N HCl + 2% TBA) in a ratio of 2:1:2 is introduced, followed by heating in a water bath for 15 min at boiling temperature. Subsequently, the samples are cooled, centrifuged for five minutes at 3000 rpm, and measured at λ = 532 nm. To determine the concentration of lipid peroxidation products resulting from the reaction with TBA, a calibration curve is established, using MDA as a standard. The findings are expressed as nmol MDA per mg of protein, employing a molar absorption coefficient of 1.56 × 10⁵ M⁻¹ cm⁻¹.
• The assessment of total glutathione (tGSH) content relies on the transformation of oxidized glutathione into its reduced form through its interaction with NADPH, a process that is facilitated by glutathione reductase (GR) [56]. When reduced glutathione interacts with dithio-nitro benzoate (DTNB), it yields a yellow compound that exhibits an absorption peak at λ = 412 nm. Oxidized glutathione serves as a reference standard for quantifying the glutathione concentration in the sample, expressed in ng/mg protein.
• The evaluation of Cu, Zn-superoxide dismutase (SOD) activity is predicated on the enzyme’s capacity to catalyze the dismutation of superoxide anion radicals into hydrogen peroxide [57]. The photochemical generation of superoxide anion radicals by riboflavin results in the reduction of nitroblue tetrazolium (NBT), leading to the formation of an insoluble formazan that imparts a blue hue to the reaction medium. The extent of NBT reduction inhibition in the presence of the enzyme preparation is quantified spectrophotometrically at λ = 560 nm, with the activity expressed in U/mg protein; one unit of SOD activity corresponds to the enzyme amount that achieves 50% inhibition of NBT reduction. To conduct the experiment, a mixture of the requisite reagents is prepared, consisting of 0.05 M potassium phosphate buffer (pH 7.8), 0.1 M methionine, 44 µM riboflavin, 1 mM NBT, and 30 mM KCN, and distributed into two sets of cuvettes. SOD is incrementally added to one group of cuvettes. The solution in these cuvettes is illuminated at λ = 560 nm for a duration of 6 min, while the control set, which lacks the enzyme preparation, is kept in the dark for the same period. The reduction of NBT in the experimental group is compared to that of the control, and the measurements are taken at λ = 560 nm.
• To assess catalase (CAT) activity, the enzyme’s ability to facilitate the breakdown of hydrogen peroxide into water and molecular oxygen is utilized [58]. A 10 mM solution of H₂O₂ in a 50 mM potassium phosphate buffer at pH 7.0 is prepared from the brain supernatant, to which the enzyme is subsequently added. The decrease in absorption is measured against a blank sample at λ = 240 nm, which correlates with the decomposition of H₂O₂ and serves as an indicator of catalase activity, quantified as ΔA240/min/mg protein.
• An assessment of glutathione peroxidase (GPx) activity [59] is conducted.

Glutathione peroxidase facilitates the reduction in organic peroxides, a process that involves reduced glutathione as a co-substrate. In the presence of GR and NADPH, oxidized glutathione is reduced. The oxidation of NADPH, which exhibits an absorption peak at λ = 340 nm, serves as a marker for GPx activity, evaluated by the reduction in absorbance at this wavelength. Brain supernatant samples were incubated in a reaction medium consisting of 0.05 M potassium phosphate buffer (pH 7.0); 1 mM EDTA; 1 mM NaN₃; 0.2 mM NADPH; 1 mM GSH; and 1 U/mL GR for 5 min at room temperature, with the reaction initiated by the addition of 1.5 mM t-butyl peroxide. The absorbance decrease was recorded at λ = 340 nm. Glutathione peroxidase activity was expressed as nmol NADPH oxidized per minute per mg protein, utilizing a molar extinction coefficient of 6.22 × 10³ M⁻¹ cm⁻¹.

2.4. Statistical Analysis

Results are presented as means ± the standard error of the mean (SEM). Statistical analyses of the data were conducted with GraphPad Prism 8.0 software (San Diego, CA, USA) using one-way analysis of variance (ANOVA), followed by Tukey’s multiple comparison post hoc test. For variables exhibiting deviations from normality, a nonparametric Kruskal–Wallis test was applied, followed by Dunn’s post hoc analysis, adjusted using the Benjamini–Hochberg method. Differences were deemed significant at p < 0.05.

3. Results

3.1. Synthesis of MACs

The synthesis of the starting (-)-myrtenylamine 2 and (-)-nopilaldehyde ((-)-nopinal) was carried out using known methods (Scheme 1) [60,61]. Amine 2 was obtained by reacting (-)-myrtenol with PBr₃ and then converting the resulting bromoderivative 1 via the Gabriel reaction. Oxidation of (-)-nopol was performed using 2-iodoxybenzoic acid (IBX).

Acid chloride 4 was obtained from 1-adamantanecarboxylic acid by refluxing with SOCl₂ (Scheme 2). Its subsequent reaction with amine 2 led to the formation of amide MAC2 in 42% yield.

Amines MAC1 and MAC3 were obtained by reacting memantine with (-)-myrtenal or nopynal. At a 1:1 ratio of amine to aldehyde, complete conversion of the starting amine was not achieved. To shift the equilibrium, the reaction was carried out with an excess of aldehyde. Reduction in the resulting imines with NaBH₄ led to the formation of the products MAC1 and MAC3, but their yields were low, 31 and 24%, respectively. Amine MAC4 was similarly obtained in 35% yield from (-)-nopinal and 2-adamantylamine. The moderate yields of amines MAC3 and MAC4 are due to the low stability of nopinal in reaction conditions.

Hydrochlorides of MAC1, MAC3, and MAC4 were obtained in quantitative yields by passing dry gaseous HCl through a solution of amines MAC1, MAC3, and MAC4 in dry diethyl ether. The spectral data of MACs 1–4 are given in Supplement.

Compound MAC2 as well as hydrochlorides of MAC1, MAC3, and MAC4 were then used in the biological experiments.

3.2. Physicochemical Properties Affecting Blood–Brain Barrier Diffusion

The calculated molecular descriptors MW (molecular weight), log P (water-octanol partition coefficient), HDs (hydrogen bond donors), HAs (hydrogen bond acceptors), PSA (polar surface area, Ų), and RBs (rotatable bonds) are given in Table S3. The values of the molecular descriptors lie within lead-like chemical space for HDs, HAs, and PSA, and in drug-like space RBs and MW. Log P is in Known Drug Space (KDS), except MAC2, which is drug-like (for the definition of lead-like, drug-like, and KDS regions, see ref. [62] and Table S4).

An analysis of ligands’ ability to cross the blood–brain barrier (log BB) showed that low HD, HA, and PSA values are favored in conjunction with high log P values [23]. The four ligands presented here all have these favorable attributes. It is known that compounds’ pKa values have profound impact on their membrane permeability [25]. As three of the ligands are secondary amines, formulated as HCl salts and are protonated at physiological pH; therefore, the collection of 206 organic compounds’ logBB values [23] were analyzed for their protonation state at pH—7.4. For this, ChatGPT (v5) was used; when compared with the physiological charge prediction module of Chemaxon for known drugs (n − 152) given in the DrugBank website [63,64,65], 84.2% of the drugs produced the same result. In addition, ChatGPT was also used to predict whether the compounds are zwitterions. Linear correlation graphs were built for each protonation state (−1, 0, +1, +2 and +3) and Pearson’s R² correlations are given in Table S5. For the +1 compounds, RB shows a negative trend with an increased number; MW is insignificant; HA, HD, and PSA are negatively correlated; and, finally Log P has a positive trend. According to these results MAC-1, 3, and 4 all have favorable values for blood–brain permeation. For the neutral compounds and MAC-2, RBs show a negative trend with an increased number; MW is insignificant; HA, HD, and PSA are negatively correlated; and finally, Log P has a positive trend. For the neutral compounds, the same correlations and trends are seen as for the +1 group apart from MW, which shows a weak negative trend. MAC-2 has higher HA and PSA values than the other MAC ligands and lower log P, which makes it less likely to cross the blood–brain barrier.

The Known Drug Indexes (KDIs) for the ligands were calculated to gauge the balance of the molecular descriptors (MW, log P, HD, HA, PSA, and RB). This method is based on the analysis of drugs in clinical use, i.e., the statistical distribution of each descriptor is fitted to a Gaussian function and normalized to 1, resulting in a weighted index. Both the summation of the indexes (KDI_2a) and multiplication (KDI_2b) methods were used [38] as shown for KDI_2a in Equation (10) and for KDI_2b in Equation (11); the numerical results are given in Table S2.

KDI_2a = I_MW + I_{log P} + I_HD + I_HA + I_RB + I_PSA

(10)

KDI_2b = I_MW × I_{log P} × I_HD × I_HA × I_RB × I_PSA

(11)

The KDI_2a values for the ligands range from 3.88 to 4.69 with a theoretical maximum of 6 and the average of 4.08 (±1.27) for known drugs. The KDI_2b range is from 0.04 to 0.20, with a theoretical maximum of 1 and with KDS average of 0.18 (±0.20). Oral bioavailability trends with both KDI_2a (R² − 0.158) and KDI_2b (R² − 0.084) [66]. The KDI values are relatively low as the ligands are optimized to cross the blood–brain barrier, e.g., with high log P for MAC-1, 3 and 4, which lowers their KDI values.

To check the redox stability of the ligands, the ionization potential (one-electron oxidation) and electron affinity (one-electron reduction) were derived for the new MACs conjugates and compared to the statistical distribution of known drugs [67]. The ionization potentials lie in the region of 6.9 to 7.2 eV, with 95% of drugs lying in the 6.0–9.0 eV range; the electron affinities are in the 1.0 to 1.4 eV, with drugs ranging from −1.5 to 2.0 eV (95% confidence interval) (see Table S6). The ligands all lie well within the confidence intervals so it can be concluded that they are redox-stable.

3.3. Effects of MACs on Scopolamine-Impaired Memory in Rats

3.3.1. Passive Avoidance Test

In the initial hour of treatment, Scop decreased the delta latency time compared to the control group, although the decrease failed to reach statistical significance (Figure 2a). The tested experimental compounds increased the parameter compared to the Scop group, although the values did not approach those of the control group. The two references, AA, produced a more pronounced increase in latency time than Myrt; however, only the change in latency time induced by MAC3 in comparison to the scopolamine group reached statistical significance (p < 0.05).

After 24 h, all substances induced changes in the indicator, nearing the values observed in the control group (Figure 2b). The latency time significantly increased for the compounds MAC2 (p < 0.05) and MAC3 (p < 0.01), which were 38% and 28% higher, respectively, than in the rats treated with the Sc + Myrt combination.

At the end of the experiment on the 12th day, scopolamine continued to demonstrate its detrimental impact on memory, with the indicator significantly lower than that of the controls (p < 0.01) (Figure 2c). The references and their newly synthesized derivatives also enhanced memory processes, with significant differences in the indicator noted only with the application of MAC3 (p < 0.05 compared to scopolamine).

Throughout the three time points, the assessment of short-term and long-term memory revealed the most favorable outcomes for MAC3.

3.3.2. Novel Object Recognition Test

The recognition index, derived from the test, exhibited a decrease of 22.35% due to scopolamine when compared to the control group (Figure 3). In the context of the administered toxic agent, the two referents demonstrated an enhancement in the indicator’s value—Myrt increased it by 27.7%, while AA increased it by 63.9% (p < 0.01). Furthermore, the novel compounds also contributed to the improvement of recognition memory, achieving index values comparable to those of healthy controls; among the four compounds, the most significant increases compared to the scopolamine group were observed for MAC2 (37.9%) and MAC4 (43.6%).

3.3.3. Barnes Maze Test

The assessment employed two metrics that define the condition of spatial memory and orientation in the experimental subjects, as outlined in the methodology section. The latency period needed to access the shelter pocket served as a measure of spatial orientation, whereas the count of errors made in locating the shelter (head dips) indicates the level of hesitation or disorientation.

As illustrated in the figure, the latency time reported does not exhibit notable variations among the different groups (Figure 4a). Conversely, the number of errors made before locating the refuge clearly reflects disorientation in rats treated with Scop, with an increase of 166.6% (p < 0.05) (Figure 4b). All substances administered concurrently with scopolamine lead to a reduction in the level of disorientation; specifically, a significant decrease of 67.5% is observed for Myrt (p < 0.05) and by 72.5% for MAC4 (p < 0.01). To obtain more comprehensive results, we computed an additional parameter: Delta head dips, defined as the difference between “number of head dips before reaching the escape box at the time of the experiment on the 12th day” and “the number of head dips before reaching the escape box on the last training day” for each particular group. This was aimed at accurately determining the protective effect of each substance against Scop (Figure 4c). It is evident that in healthy controls, the number of errors decreases during the experiment, indicating intact spatial orientation and memory. In contrast, in animals exposed to Scop, the number of errors significantly increased compared to the controls (p < 0.01); both referents yield very similar outcomes in maintaining the spatial orientation of the rats, as evidenced by a reduction in the number of errors relative to the training phase, with the indicator decreasing compared to Scop (p < 0.05). Among the new compounds, MAC4 demonstrates results comparable to the references (p < 0.05 vs. Sc); however, the most pronounced effect on spatial memory is observed with MAC3, where the indicator in this group approaches that of the controls, and the change is significantly less than that observed with Scop (p < 0.01).

3.4. Effects of MACs on Brain AChE Activity

3.4.1. Docking Study of MACs’ Affinity to AChE

Four myrtenal–adamantine derivatives were docked into the binding site of the AChE (PDB ID: 5HF9, resolution 2.20 Å) [28] enzyme. The robustness of the AChE docking scaffold has been previously established [23]. The scoring functions GoldScore (GS) [30], ChemScore (CS) [32,33], Piecewise Linear Potential (ChemPLP) [34], and Astex Statistical Potential (ASP) [35] in the GOLD (v2025.1) docking algorithm were used. The GOLD docking algorithm is reported to be an excellent molecular modeling tool [68,69].

The binding scores for the AChE binding pocket are given in Table S1. Comparison with our previous work reveals that the new ligands have higher scores than Myrt, MAC-197, and MAC-198 with values akin to those of the AD drugs Tacrine and Donepezil (see Table S2) [23].

The predicted binding poses of the ligands were investigated, and derivative MAC4 gave a clear dominant conformation predicted by the four scoring functions used as they all converged on one pose [70]; it is shown in Figure 5, with the adamantane moiety overlapping with the HI6 co-crystallized ligand and the Myrt inserted deep into the binding pocket.

Interestingly, this is the same conformation as was predicted for MAC-197 in our previous work, with the same H-bond predicted to the Tyr124 amino acid side chain and two lipophilic pockets hosting the myrtenal and adamantane moieties [23].

Although MAC-197 demonstrated greater efficacy compared to MAC-198 in that study, the current docking findings ought to be viewed as suggestive of advantageous theoretical binding rather than serving as a direct forecast of functional activity or in vivo efficacy. In the case of the newly synthesized MACs 1–3 conjugates, similar poses were sometimes anticipated; however, these frequently entailed reversed or displaced orientations of the myrtenal and adamantane moieties and did not show predicted hydrogen bonding, indicating potentially less stable or less advantageous theoretical interactions.

3.4.2. Evaluation of Brain AchE Activity In Vivo

In the cerebral cortex, Scop induced a notable activation of the enzyme that degrades acetylcholine, resulting in a reported increase of 72.3% (p < 0.01) when compared to control groups (Figure 6a). Conversely, as in our previous study with the other two compounds MAC-197 and MAC-198 [23], all other substances tested led to a decrease in the pathologically elevated enzyme activity; specifically, the reference compounds Myrt and AA lowered the activity by 16.00% and 21.37%, respectively, while the new conjugates achieved a significant reduction of 41.25% with MAC1 (p < 0.01 in comparison to Scop).

In the hippocampus, Scop similarly elevated the activity of acetylcholinesterase by 32.9% (Figure 6b); however, the new compounds brought the enzyme activity down to levels comparable to those of the controls, with the most pronounced suppression noted in MAC3, which reduced activity by 37.10% relative to Scop (p < 0.01).

The toxic effects of Scop, with the protective effects of the substances under investigation on brain AChE, are more prominently observed in the cortex, where the enzyme is expressed at lower levels (approximately 65 UI) compared to the hippocampus (around 115 UI). Surprisingly, in the cerebral cortex, MAC1 most effectively counteracted the effects of Scop, whereas in the hippocampus, MAC3 demonstrated the greatest antagonistic effect. The observed regional variations in AChE inhibition, where MAC1 demonstrates greater efficacy in the cortex and MAC3 in the hippocampus, may indicate differences in distribution, the penetration of the blood–brain barrier, metabolic processes, or local enzyme affinity. While the precise mechanism was not investigated in this study, it is reasonable to suggest that these factors could contribute to the tissue-specific effects that were noted.

It is important to note that although MAC4 demonstrated the most stable and convergent predicted binding pose during the docking analysis, the behavioral and AChE inhibition results revealed that MAC3 and MAC1 had more significant functional effects. This inconsistency underscores a well-known limitation of molecular docking methods, which provide estimates of theoretical binding modes and relative affinities but fail to account for critical factors influencing in vivo efficacy.

3.5. Effects of MACs on BDNF Levels

Scop diminished the levels of BDNF in both brain regions associated with memory functions—by 20.50% in the cortex and by 22.30% in the hippocampus (Figure 7). It is noteworthy that neither of the referents influences this measure, whereas the newly synthesized compounds mitigate the detrimental impact of Scop (without statistical significance), with the measure in some instances showing a borderline significant increase compared to the referents—MAC2 vs. AA (p = 0.0563), MAC4 vs. AA (p = 0.0316), and MAC4 vs. Myrt (p = 0.0563) in the cortex; MAC4 vs. Myrt (p = 0.0526) and MAC3 vs. Myrt (p = 0.0843) in the hippocampus. In the cerebral cortex, an elevated level of BDNF compared to the Sc group was observed in rats administered MAC2 (by 19.36%) and MAC4 (by 23.80%) (Figure 7a); similarly, in the hippocampus, this effect was again noted in MAC2 (by 20.76%) and MAC4 (by 34.10%), as well as in MAC3 (by 34.87%) (Figure 7b). Across both brain regions, the most favorable outcome was achieved with the application of MAC4.

3.6. Effects of MACs on Brain Oxidative Status

In the cerebral cortex, the levels of lipid peroxidation products were slightly influenced by Scop, resulting in an increase of 14.47% (Figure 8a); in contrast, the reference Myrt decreased this indicator by 25.84%, achieving a significance level of p < 0.01.

In the hippocampus, the changes in MDA content were more pronounced (Figure 8b). Scop elevated the indicator by 30.10% (p < 0.0001), while all other substances exhibited comparable statistically significant antioxidant effects. On the one hand, AA lowered MDA levels to a lesser degree by 17.20% (p < 0.001), and MAC2 demonstrated the weakest effect with a decrease of 10.90% (p < 0.05). Myrt, on the other hand, reduced the indicator by 22.90% (p < 0.0001), and the other three novel substances—MAC1 (decrease of 21%, p < 0.0001), MAC3 (decrease of 21.25%, p < 0.0001), and MAC4 (decrease of 22.60%, p < 0.0001)—yielded similar outcomes.

In both brain regions, Scop diminished the glutathione levels, thereby validating its detrimental impact through the induced oxidative stress—in the cortex by 32.00% (p < 0.01) (Figure 9a) and in the hippocampus by 30.80% (p < 0.05) (Figure 9b). Regarding recovery of the GSH levels decreased by Scop, only MAC2 demonstrated antioxidant effects, elevating the measure equally in both the cortex and the hippocampus—by 48.10% (p < 0.01) (Figure 9a) and by 48.40% (p < 0.01) (Figure 9b), respectively.

The detrimental impact of Scop, through the stimulation of SOD activity, is particularly evident in the cerebral cortex. Consequently, the antioxidant properties of both the referents and the newly synthesized compounds hold considerable significance (Figure 10a). Scop elevated the enzyme activity by 31.10% (p < 0.01) when compared to controls. The referents and conjugates significantly mitigated the pathologically elevated SOD activity induced by Scop, achieving a high level of significance at p < 0.0001—with Myrt reducing it by 51.90%, AA by 40.60%, MAC1 and 2 by approximately 33.00%, and MAC3 and 4—by 45.00%.

In the hippocampus, the results indicate a maintained trend, albeit with reduced intensity (Figure 10b). A statistically significant reduction in SOD activity relative to Scop was noted in treatments with MAC2 and MAC4, decreasing it by 42.50% and 41.88% (p < 0.05), respectively.

In accordance with the reduced levels of GSH, Scop also diminishes the activity of GPx. Similarly to its effect on SOD, statistically significant effects are noted concerning GPx in the cortex. In the cerebral cortex, Scop decreases the indicator by 47.29% (p < 0.01) (Figure 11a). Among the two references, the increase in enzyme activity compared to Scop achieved by AA—by 101.00%—is statistically significant (p < 0.001), whereas Myrt raised the indicator by 46.20% without statistical significance. MAC1 and 2 enhance the values compared to Scop by 79.20% (p < 0.05) and 79.60% (p < 0.01, respectively, while MAC3 increases it by 56.90%, and MAC4 by 73.54% (p < 0.05).

In the hippocampus, Scop decreased GPx activity by 34.90%, and the substances administered alongside it did not mitigate its detrimental effect (Figure 11b).

Interestingly, the activity of catalase was unaffected by Scop; however, the reference substances significantly reduced the indicator. In the cortex, both Myrt and AA decreased CAT activity (p < 0.05) by 28.80% and 24.10%, respectively (Figure 12a). A similar outcome was observed in the hippocampus, where Myrt diminished the indicator by 44.92% (p < 0.05), while AA reduced it by 31.74% (Figure 12b). The newly synthesized compounds did not exhibit significant effects on CAT activity in the two brain regions examined, and their impact was less pronounced than that of the reference substances.

4. Discussion

The current pilot study presents new evidence indicating that the MACs have significant neuroprotective effects in a rat model of scopolamine-induced dementia. Among the four conjugates evaluated, nopinal-derived MAC3 and MAC4 demonstrated the most substantial enhancements in memory and learning metrics across various behavioral paradigms, including the passive avoidance, novel object recognition, and Barnes maze assessments. In the passive avoidance test, MAC3 achieved a notable increase in latency time across all three time-points (1 h, 24 h, and 12 days). In the Barnes maze test, MAC4 exhibited significant differences in performance indicators, while in the novel object recognition test both MAC2 and MAC4 were prominent, albeit without reaching statistical significance. The differences noted in the impacts of MAC2 and MAC4 on various behavioral tasks may indicate a task-dependent sensitivity of different memory domains under conditions of scopolamine-induced cognitive impairment. Considering the exploratory nature of this research, these results should be regarded as suggestive rather than definitive evidence of selective effects on specific memory types.

Taken together, the behavioral findings indicate that individual MAC representatives may preferentially influence distinct forms of memory. Within the domain of cognitive functions, the passive avoidance test reflects emotionally conditioned associative memory, involving long-term retention of an aversive association between context and stimulus and relying heavily on limbic structures such as the amygdala. The Barnes maze evaluates hippocampus-dependent spatial long-term memory, which requires the formation and retention of reference spatial maps for navigation. The primary cognitive process assessed by the novel object recognition task is object recognition, commonly classified as a declarative-like (recognition) memory process in behavioral neuroscience. These distinctions offer a neurofunctional framework for understanding the varying behavioral profiles seen among the MACs.

The behavioral outcomes were further corroborated by biochemical data demonstrating modulation of brain AChE activity, BDNF expression, and antioxidant defense mechanisms, as evidenced by activities of SOD, and GPx, along with reduced lipid peroxidation, and elevated GSH content. In this context, an apparent dissociation emerged between in silico and in vivo findings related to AChE modulation. Docking studies assessing ligand affinity toward acetylcholinesterase revealed that MAC4 adopted a distinctly dominant binding conformation. However, in vivo analyses showed that MAC1 significantly decreased AChE activity in the cortex, whereas MAC3 exerted a comparable effect predominantly in the hippocampus.

With respect to the enhanced cognitive performance observed in the passive avoidance test, the reduction in cortical AChE activity induced by MAC1 may facilitate improved cholinergic neurotransmission within cortical circuits involved in attention, sensory processing, and contextual encoding. These mechanisms might indirectly enhance emotional learning that relies on the amygdala, leading to better behavioral outcomes. The well-established notion that cholinergic systems in the cortex and hippocampus serve distinct roles in learning and memory supports this region-specific interpretation. Accordingly, selective modulation of cortical cholinergic tone may significantly influence associative learning tasks even in the absence of marked hippocampal alterations. Conversely, the enhancement of passive avoidance performance observed with MAC3, together with reduced hippocampal AChE activity and minimal cortical effects, suggests that hippocampal cholinergic modulation can independently support synaptic plasticity and memory consolidation within hippocampal–amygdala networks.

Regarding BDNF expression, none of the tested compounds produced statistically significant changes. Nevertheless, MAC2 and MAC4 in the cortex, as well as MAC3 and MAC4 in the hippocampus, yielded values approaching those of the control group. It is well-recognized that BDNF regulation is highly complex and region-specific, and that a relatively high activation threshold governs BDNF transcription. Moderate changes in synaptic activity or cholinergic tone may thus be inadequate to trigger significant neurotrophin upregulation, which might account for the restricted BDNF response noted despite substantial cholinergic modulation.

The antioxidant properties of the novel compounds displayed heterogeneous and region-dependent manifestations. In the cortex, MDA levels were most substantially reduced by MAC4, whereas in the hippocampus all MACs demonstrated significant activity. Glutathione levels were most prominently elevated by MAC2 in both brain regions. SOD activity was influenced by all MACs in the cortex and by MAC2 and MAC4 in the hippocampus. GPx activity was modulated by MAC1, MAC2, and MAC4 in the cortex, while CAT activity remained unaffected by MACs. These findings indicate that the MACs exert region- and marker-specific antioxidant effects, with each compound exhibiting a distinct activity profile, reflecting the complex and independent regulation of antioxidant system components.

Antioxidant enzymes and redox markers function as an integrated yet non-synchronous network that responds to oxidative stress in a region-specific manner within the brain. Differences in metabolic demand, vulnerability to oxidative damage, and local regulatory signaling contribute to divergent antioxidant responses between the cortex and hippocampus. Accordingly, the distinct antioxidant signatures of the MAC compounds—such as selective cortical MDA reduction by MAC4, broad hippocampal activity across all MACs, preferential GSH elevation by MAC2, and variable modulation of SOD, GPx, and CAT—are consistent with the intricate regulatory architecture of endogenous antioxidant defense systems. Such differential responses are widely reported in preclinical models of oxidative stress and reflect distinct mechanisms of antioxidant action.

In summary, the present findings suggest that the newly synthesized MAC conjugates confer neuroprotection through synergistic modulation of cholinergic neurotransmission, partial engagement of neurotrophic pathways, and region-specific antioxidant mechanisms. These effects align with the dual pharmacological characteristics of their structural components and support the rationale for multitarget-directed ligand design in the development of disease-modifying strategies for Alzheimer’s disease.

4.1. Mechanistic Interpretation

The observed recovery in behavior and the normalization of biochemical parameters suggest that the conjugation of myrtenal with aminoadamantane moieties enhances both the inhibition of acetylcholinesterase and the maintenance of redox homeostasis. In silico docking studies demonstrated a strong binding affinity of the MAC ligands to the active site of human AChE (PDB ID: 5HF9), with hydrogen–bond interactions involving Tyr124, akin to those seen with well-known AChE inhibitors such as tacrine and donepezil. These results align with previous computational research that supports the GOLD docking algorithm as a dependable predictor of AChE–ligand binding interactions [68].

The neuroprotective effectiveness of the novel conjugates is comparable to the previously documented effects of myrtenal alone, which mitigated cognitive decline and oxidative damage in rats treated with scopolamine, but in significantly lower dose [22]. Notably, the present findings expand upon this evidence by demonstrating that the hybridization with aminoadamantane not only enhances AChE inhibitory potency but also improves blood–brain barrier permeability, as established through in silico analyses [25]. These findings align with our previous research, where myrtenal–adamantane derivatives enhanced behavioral and biochemical indicators of neurodegeneration [23].

From a mechanistic perspective, the improved antioxidant profile seen in MAC-treated animals supports the theory that oxidative stress is a one of the key factors in the development of Alzheimer’s disease [71,72]. By decreasing lipid peroxidation and restoring glutathione levels, MACs may safeguard neuronal membranes and mitochondria from oxidative damage, thus aiding in the maintenance of synaptic function. The simultaneous increase in BDNF in both hippocampal and cortical tissues further reinforces the idea that the conjugates enhance neurotrophic signaling, promoting neuronal survival and plasticity [73,74].

4.2. Comparative Analysis and Pharmacological Relevance

The synthesized conjugates exhibit a dual action—modulating cholinergic transmission while also influencing oxidative status—thus categorizing them within the novel class of multitarget-directed ligands, which represents a promising approach in the discovery of drugs for Alzheimer’s disease [75,76].

The physicochemical properties assessed for the MACs (MW, log P, PSA, HD, HA, RB) align with the characteristics of drug-like compounds, thereby reinforcing their potential for bioavailability and ability to penetrate the brain. The incorporation of adamantane—a well-regarded structure recognized for its role in stabilizing interactions within the central nervous system [24]—seems to enhance both pharmacokinetic and pharmacodynamic attributes, resulting in a class of compounds that demonstrate greater efficacy compared to their monoterpenoid precursors.

4.3. Limitations and Future Directions

While these findings are promising, the study possesses several limitations that are intrinsic to its pilot nature. Although the scopolamine model successfully mimics cholinergic dysfunction, it fails to encompass the multifactorial pathology associated with AD fully, especially regarding the roles of amyloid and tau pathology. Comprehensive pharmacokinetic, metabolic, and toxicity profiles of the MACs have yet to be defined.

Consequently, future research should build upon these findings by investigating dose–response relationships, assessing long-term efficacy, and validating molecular targets using additional models, such as amyloidogenic and transgenic paradigms. Structural optimization, informed by quantitative structure–activity relationship (QSAR) analyses and in silico predictions, could further enhance their pharmacological profiles.

Despite MAC4 showing the most advantageous docking profile, it did not achieve the highest efficacy in the behavioral and biochemical assessments. This inconsistency highlights a significant limitation of molecular docking, which fails to reliably predict in vivo effectiveness and must be approached with caution.

Moreover, the investigation of related hybrid scaffolds may uncover structural determinants that are essential for achieving a balance between AChE inhibition, antioxidant efficacy, and neurotrophic modulation.

5. Conclusions

This research illustrates that the newly developed myrtenal/nopinal–aminoadamantane/dimethylaminoadamantane conjugates function as potent multifunctional agents exhibiting both cholinergic and antioxidant properties. The noted improvements in behavior and biochemistry indicate that these compounds could signify a new class of neuroprotective candidates. These results build upon and expand our earlier investigations into myrtenal derivatives, affirming that the structural hybridization with aminoadamantane improves both efficacy, and potential for selective engagement of AChE. Additional preclinical assessments are necessary to confirm their therapeutic potential and safety profile, thereby facilitating the advancement of innovative multitarget medications for neurodegenerative diseases.