Evaluation of the Local Anesthetic Activity, Acute Toxicity, and Structure–Toxicity Relationship in Series of Synthesized 1-Aryltetrahydroisoquinoline Alkaloid Derivatives In Vivo and In Silico

Isoquinoline alkaloids constitute one of the most common classes of alkaloids that have shown a pronounced role in curing various diseases. Finding ways to reduce the toxicity of these molecules and to increase their therapeutic margin is an urgent matter. Here, a one-step method for the synthesis of a series of 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines was performed in 85–98% yield by the Pictet–Spengler reaction. This was accomplished using the reaction between 3,4-dimethoxyphenylethylamine and substituted benzaldehydes boiling in trifluoroacetic acid. Furthermore, 1-(3′-amino-, 4′-aminophenyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines were obtained in 94% and 97% yield by reduction in 1-(3′-nitro-, 4′-nitrophenyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines with SnCl2 × 2H2O. The structures of the substances obtained were confirmed by infrared (IR) and nuclear magnetic resonance (1H and 13C NMR) spectra. ADMET/TOPKAT in silico study concluded that the synthesized compounds exhibited acceptable pharmacodynamic and pharmacokinetic properties without carcinogenic or mutagenic potential but with variable hepatotoxicity. The acute toxicity and structure–toxicity relationship (STR) in the series of 20 derivatives of 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines (3a–r, 4a, b) was studied via determination of acute toxicity and resorptive action in white mice employing intragastric step-by-step administration. The first compound, 1-phenyl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline hydrochloride (3a), showed the highest toxicity with LD50 of 280 mg/kg in contrast to 1-(3′-bromo -4′-hydroxyphenyl)-6,7-methylenedioxy-1,2,3,4-tetrahydroisoquinoline hydrochloride (3e) which proved to be the safest of the compounds studied. Its toxicity was 13.75 times lower than that of the parent compound 3a. All compounds investigated showed high local anesthetic activity on rabbit eyes in the concentrations studied. Only 3r, 3n, and 4a caused eye irritation and redness. All investigated derivatives (except 4b) in 1% concentration were more active than lidocaine, providing longer duration of complete anesthesia. Therefore, based on the obtained results of in silico tests, local anesthesia, and acute toxicity, a conclusion can be drawn that the experimental compounds need further extensive future investigations and possible modifications so that they can act as promising drug candidates.


Introduction
Isoquinoline alkaloids are one of the common classes of alkaloids. They are biosynthesized from tyrosine and phenylalanine and basically include isoquinoline or tetrahydroisoquinoline ring as the main structural part in their skeleton [1]. However, the isoquinoline alkaloids do not constitute a homogeneous group of structurally related compounds due to the various levels of oxygenation, distribution, and intramolecular rearrangements along with the additional existence of rings attached to the base system. They could be divided into eight subgroups: aporphine, phthalideisoquinoline, benzylisoquinoline, benzo[c]phenanthridine, protopine, emetine, protoberberine, and morphinan alkaloids [2]. Meanwhile, isoquinoline alkaloids have recently received great attention as they constitute the genetic precursors of a plethora of naturally occurring compounds that have shown a pronounced role in the treating of various diseases [3]. The 1,2,3,4-tetrahydroisoquinoline derivatives have a wide range of biological activities, including antidepressant [4], immunomodulatory [5], anti-Alzheimer and anti-Parkinson [6], inhibiting NO production [7], anti-HIV [8], as well as antifungal [9] activity.
Aporphine alkaloids such as boldine, magnoflorine, glaucine, and oxoglaucine have demonstrated potent anti-inflammatory activity in various in vitro models, including antiinflammatory activity by inhibiting cyclooxygenase (COX) enzymes, inhibiting prostaglandin synthesis, and downregulating the expression of proinflammatory cytokines (TNF-, IL-6) [10]. Furthermore, recent studies have suggested its neuroprotective effects in ischemic brain damage in mice, and anti-inflammatory activity in colitis models by downregulation of proinflammatory cytokines production [11]. In addition, tetrandrine possesses calciumantagonistic and hypotensive effects [12]. Neferine improved cell viability and mitochondrial function and reduced cell apoptosis and the production of reactive oxygen species in LPS-treated H9c2 cells. In addition, neferine significantly upregulated Bcl-2 expression and suppressed cleaved caspase 3 activity in LPS-induced mouse heart tissue and H9c2 cells [13]. Selective antagonists of the orexin 1 (OX1) receptor have been proposed in the range of 6-and 7-substituted tetrahydroisoquinolines [14]. Noscapine is a phthalideisoquinoline alkaloid isolated from the opium poppy, Papaver somniferum. It has long been used as an antitussive agent, but has more recently been found to possess microtubule-modulating properties and anticancer activity [15]. The vasodilator mechanism of protopine is related to decreasing effects on [Ca 2+ ] i and increasing effects on cAMP and cGMP, as well as its influence on the PKC, has been proposed [16]. The 1-aryl-6,7-dihydroxy tetrahydroisoquinoline shows significant affinity towards the D 2 receptor, with a K i value of 31 nM. This significant affinity can be attributed to the presence of a thiomethyl group, and it is the most active 1-aryl-6,7-dihydroxy tetrahydroisoquinoline derivative reported to date [17]. Several 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline derivatives displayed potent anticonvulsant effects in different animal models of epilepsy [18]. In addition, 1,2,3,4-tetrahydroisoquinoline (THIQ) and aryl-substituted derivatives of THIQ are potent inhibitors of the enzyme that catalyzes the formation of epinephrine-phenylethanolamine N-methyltransferase [19].
Several THIQ based natural products have been previously explored for their antitumor properties, making it a vital scaffold for anticancer drug design. Various medicinal chemistry strategies employed for the design and development of THIQ analogs as in-hibitors or modulators of relevant anticancer targets, the common strategies employed for the synthesis of the core scaffold are also used [21].
Tetrahydroisoquinolines, which belong to a group of cyclized condensation adducts of biogenic amines with aldehydes, are referred to as mammalian alkaloids. They include salsolinol and tetrahydropapaveroline (THP) that are derived from dopamine through condensation with acetaldehyde and dopaldehyde (3,4-dihydroxyphenylacetaldehyde), respectively. THP is a putative dopaminergic neurotoxin that is implicated in the pathology of Parkinson's disease [22]. In [23], the authors exclude possibility that salsolinol under physiological conditions could be an endogenous factor involved in the neurogenerative processes; conversely, it can exert a protective action on nerve cells in the brain.
Continuing the search for biologically active substances in the series of 1-aryltetrahydroisoquinolines, the process was optimized, and six more new derivatives were obtained in addition to the already known ones.
The aim of this study is the search for promising local anesthetic substances with low toxicity. Finding ways to reduce the toxicity of these molecules and increase their therapeutic margin is an urgent matter. In an earlier study conducted by the authors, a number of 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline derivatives were synthesized in 35-86% yield by Pictet-Spengler reaction from 3,4-dimethoxyphenylethylamine (1) and substituted benzaldehydes in two steps via imine production by boiling in benzene with apparatus Dean-Stark additive [31]. Pictet-Spengler reaction is a chemical reaction mode in which the condensation of β-arylethylamine with ketone or aldehyde occurs and consequently ring closure is achieved. Continuing our study, a more efficient onestep method was adopted to obtain 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines (3a-r) in 85-98% yield by boiling in trifluoroacetic acid. Furthermore, the reduction of 1-(nitrophenyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines was also performed to obtain compounds 4a, b.
In addition, the 'structure-toxicity' relationship in the series of 20 derivatives of 1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline (3a-r, 4a, b) was also performed in vivo by determination of acute toxicity and resorptive action in white mice by stepwise intragastric administration. This was carried out with the aim of determining the value of introducing multiple functional groups onto the original compound (3a) in reducing its toxicity. Moreover, the synthesized series of 1-aryltetrahydroisoquinoline derivatives underwent absorption, distribution, metabolism, excretion, and toxicity (ADMET) prediction and toxicity prediction using computer assisted technology (TOPKAT) by Discovery Studio 2016 (Accelrys Inc., San Diego, CA, USA) in order to determine the pharmacokinetic, pharmacodynamic, and toxicity characteristics.
In addition, methods have been used to reduce the nitro group, such as NaBH 4 /Cu(NO 3 ) 2 ×3H 2 O, SnCl 2 ×2H 2 O [32], or NH 2 -NH 2 by microwave irradiation [33]. 1-(3 -Nitro-, 4nitrophenyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinolines were reduced with SnCl 2 ×2H 2 O and 1-(3 -amino-, 4 -aminophenyl)-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline with yields of in 94% and 97% respectively were obtained as shown in Scheme 2. The spectral characteristics of compounds 3a-r, 4a, b confirm the formation of substituted 1-aryltetrahydroisoquinolines. This includes the presence of the active hydrogen band in the IR spectra and in 1 H NMR spectra of aromatic protons singlets which are H-5 at δ H 6.23-6.91 and H-8 at δ H 6.09-6.80, as well as the singlet of the methine proton H-1 at δ H 4.78-5.92. It should be noted that protons H-8 resonate in a stronger field than H-5, and the chemical shift of these protons depends slightly on the nature of the substituents in the C ring (only the presence of the nitro group causes a weak field shift of H-5 by~0.5 ppm). Whereas substituents of the second type, as well as ortho-substituents, significantly affect the chemical shift of the H-1 proton.

In Silico ADMET/TOPKAT Prediction
ADMET/TOPKAT evaluation was performed in silico using Discovery Study 2016 (Accelrys Inc., San Diego, CA, USA) for the synthesized series of 1-aryltetrahydroisoquinoline derivatives. The ADMET evaluation results shown in Table 1 showed good human intestinal absorption level for all tested compounds, as thus they were within the 95% absorption ellipses as revealed by the ADMET graph ( Figure 1). Furthermore, all 1-aryltetrahydroisoquinoline derivatives tested showed good to optimal levels of solubility. In addition, all 1-aryltetrahydroisoquinoline derivatives tested exhibited medium to high blood-brain barrier (BBB) penetration and therefore allocated within the 95% of confidence ellipse of the BBB ( Figure 1). Meanwhile, all compounds revealed more than 90% plasma protein binding (PPB) pattern. Furthermore, they showed no inhibition of cytochrome P450 2D6 (CYP2D6) reflecting its inability to trigger uncontrolled drug-drug interactions but, unfortunately, showed hepatotoxicity with varying degrees (Table 1).

In Silico ADMET/TOPKAT Prediction
ADMET/TOPKAT evaluation was performed in silico using Discovery Stud celrys Inc., San Diego, CA, USA) for the synthesized series of 1-aryltetrahydroi derivatives. The ADMET evaluation results shown in Table 1 showed good huma absorption level for all tested compounds, as thus they were within the 95% abso ses as revealed by the ADMET graph ( Figure 1). Furthermore, all 1-aryltetrahydroi derivatives tested showed good to optimal levels of solubility. In addition, all 1droisoquinoline derivatives tested exhibited medium to high blood-brain barrier tration and therefore allocated within the 95% of confidence ellipse of the BBB Meanwhile, all compounds revealed more than 90% plasma protein binding (PP Furthermore, they showed no inhibition of cytochrome P450 2D6 (CYP2D6) reflec bility to trigger uncontrolled drug-drug interactions but, unfortunately, showed city with varying degrees (Table 1). 0, 1, 2, and 3 indicates good, moderate, low, and very low absorption, respectively; 0, 1, 2, 3, 4, and 5 indicates extremely low, very low but possible, low, good, optimal, and too soluble, respectively; 0, 1, 2, 3, and 4 denote very high, high, medium, low, and undefined, penetration via BBB respectively. PBB-plasma protein binding, FALSE means less than 90%, TRUE means more than 90; NI-non-inhibitor; NT-non-toxic. and 3p that triggered severe eye irritation. Meanwhile, the tested compounds showed ra chronic LOAEL (lowest observed adverse effect level) values ranging between 0.004 and 0.026 g/kg body weight. From ADMET/TOPKAT in silico evaluation, it was obvious tha the examined synthesized compounds exhibited acceptable pharmacodynamic and phar macokinetic properties but unfortunately caused hepatotoxicity with no carcinogenic o mutagenic potential. Figure 1. ADMET plot for a series of synthesized 1-aryltetrahydroisoquinoline derivatives display ing 95% and 99% confidence limit ellipses corresponding to the blood-brain barrier (BBB) and th human intestinal absorption (HIA) models.  ADMET plot for a series of synthesized 1-aryltetrahydroisoquinoline derivatives displaying 95% and 99% confidence limit ellipses corresponding to the blood-brain barrier (BBB) and the human intestinal absorption (HIA) models.
For TOPKAT prediction displayed in Table 2, all examined synthesized series of 1aryltetrahydroisoquinoline derivatives revealed no mutagenicity with respect to the in silico Ames Mutagenicity evaluation. They also revealed no carcinogenicity in both rat male and female FDA. In addition, most of the compounds caused no irritation to the skin with the exception of 3i, 3m, and 3n that showed mild skin irritancy. However, most of the compounds showed mild ocular irritation with the exception of 3c, 3d, 3e, 3f, 3l, 3m, and 3p that triggered severe eye irritation. Meanwhile, the tested compounds showed rat chronic LOAEL (lowest observed adverse effect level) values ranging between 0.004 and 0.026 g/kg body weight. From ADMET/TOPKAT in silico evaluation, it was obvious that the examined synthesized compounds exhibited acceptable pharmacodynamic and pharmacokinetic properties but unfortunately caused hepatotoxicity with no carcinogenic or mutagenic potential.

Biological Evaluation
Acute toxicity studies have shown a similar pattern of resorptive action where the introduction of subtoxic and toxic doses resulted in depression of general state, suppression of motor activity, short-term respiration rate, head tremor, convulsions, and death of some animals in which the LD 50 of the compounds tested was shown in Table 3. Table 3. Acute toxicity of the studied compounds in mice after intragastric administration (N = 10). The first compound,1-phenyl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline hydrochloride (3a), showed the maximum toxicity with LD 50 of 280 mg/kg in contrast to 1-(3 -bromo-4 -hydroxyphenyl)-6,7-methylenedioxy-1,2,3,4-tetrahydroisoquinoline hydrochloride (3e) which was found to be the safest of the studied compounds. Its toxicity was 13.75 times lower than that of compound 3a. However, introduction of the metylenedioxy group into the aryl radical at the C-3 and C-4 atoms of 3i did not lead to a change in acute toxicity.

Compound
Conversely 3k, the additional introduction of a chlorine atom into the C-6 3k compound resulted in a 1.86-fold decrease of acute toxicity, and the introduction of a bromine atom (3j) at this position resulted in a decrease than 2.5 times the acute toxicity. Meanwhile, the presence of an oxy or methoxy group at the C-2 position (3b and 3p), a hydroxyl group at C-2, and a bromine atom at C-5 (3f) did not result in a significant decrease of toxicity. Simultaneous administration of bromine at C-2, methoxyl at C-4, and hydroxyl at C-5 (3h) reduced acute toxicity about 2-fold compared with 3a, while compound 3n with a nitro group at C-3 had showed an LD50 value of 460.5 ± 598.5 mg/kg. Introduction of methoxy groups into the phenyl radical at C-4 and C-5 positions with simultaneous bromination at the C-2 position (compound 3g) resulted in no significant change in acute toxicity compared to 3a. However, the introduction of methoxy groups into the phenyl radical at C-4 and C-5 (compound 3d), bromination of the parent compound at C-3 (compound 3o), as well as chlorination or nitration at C-4 (compounds 3m, 3r) reduced the acute toxicity (3a) by about 2.5-2.6 times, and the amino group in C-4 by 3.75 times (4a). Substitution of the nitro group with a dimethylamino radical reduced toxicity nearly 4.5-fold (compound 3q). The presence of a hydroxy-at C-3 and a methoxy group at C-4 resulted in a 3.5-fold decrease of LD 50 (compound 3c). Simultaneous introduction of bromine atom at C-5 , hydroxyl at C-4, and methoxy group at C-3 into the initial molecule (3a) reduced the toxicity by 5.6-fold (compound 3l), and the nitro group in C-4 -by nearly 7-fold (compound 3m).
The local anesthetic activity of the studied compounds was evaluated on rabbit corneal model simulating surface anesthesia. During the experiments, no local irritant effects on rabbit eyes were observed in the concentrations studied, with the exception of compounds 3r, 3n, and 4a which caused reversible irritation and redness of the eyes. As shown in Table 4, all studied derivatives (except 4b) at a concentration 1% were more active than lidocaine, providing a longer duration of complete anesthesia. The compounds 3f, 3h, 3l, 3n, 3p, and 3q showed a duration of complete anesthesia for 1.6, 1.7, 1.4, 2.6, 3.0, and 2.2 times longer than lidocaine. The compounds 3c-f, 3l, 3o, and 3p revealed a longer total duration of anesthesia than lidocaine. Table 4. Local anesthetic activity of the studied compounds in rabbit corneal model in compared to lidocaine.

General Methods
Solvents and reagents were purchased from Sigma-Aldrich (Sigma-Aldrich GmbH, Steinheim, Germany). Starting materials or some benzaldehydes containing bromine and chlorine were synthesized corresponding to the benzaldehyde in the laboratory. The purity of the reagents was 95-98%. Mass spectra were recorded using an Agilent 1200 Infinity high performance liquid chromatograph (Agilent Technologies, Santa Clara, CA USA) and a 6420 Triple Quad LC-MS mass detector (Agilent Technologies, Santa Clara, CA USA USA) (electrospray ionization +ESI TIC Scan). IR spectra were recorded using the FTIR System 2000 spectrometer (PerkinElmer, Inc., Waltham, MA, USA,) in KBr tablets. NMR spectra were recorded on a JNM-ECX400 and JNM-ECZ600R spectrometer (Jeol, Akishima, Japan) at operating frequencies of 400 and 600 MHz; for 1 H assessment the solvents were CDCl 3 , CCl 4 , CD 3 COOD, and DMSO-d 6 . Meanwhile, TMS (0 ppm) was used as an internal standard in NMR spectra. In the 13
The spectra of all tested compounds were represented in Supplementary Materials Figures S1-S57.

In Silico ADMET/TOPKAT Evaluation
The synthesized series of 1-aryltetrahydroisoquinoline derivatives was subjected to ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) prediction as well as Toxicity Prediction by Komputer Assisted Technology (TOPKAT) using Discovery Studio 2016 (Accelrys Inc., San Diego, CA, USA) with the aim of determining the pharmacokinetic, pharmacodynamic, and toxicity characteristics. Water solubility, plasma protein binding (PPB) prediction, human intestinal absorption (HIA), cytochrome P450 2D6 inhibition (CYP2D6), blood-brain barrier (BBB) penetration, and level of hepatotoxicity were chosen as ADMET parameters. Meanwhile, rat chronic LOAEL (lowest observed adverse effect level), Ames mutagenicity, together with skin and eye irritation were chosen as TOPKAT descriptors [34,35].

Biological Evaluation
Local anesthetic activity was evaluated using the rabbit eye cornea terminal anesthesia test on 2.0-2.5 kg male chinchilla rabbits after introduction of total 0.4 mL (0.2 mL twice with an interval of 30 s) of 1.0 % solutions of the studied substances. Lidocaine was used as an etalon drug. Each substance was tested on three rabbits.
The acute toxicity and resorptive action were determined in white mice weighing 20-24 g by gradual oral administration at doses of 1-5000 mg/kg.
The animals were kept under standard nursery conditions with a natural 12-h lightdark cycle, at an air temperature of 20 ± 2 • C. The animals had unlimited access to food and water. The experiments were performed in accordance with the International Convention for the Protection of Vertebrate Animals used for Experimental and Scientific Purposes (Strasbourg, 1986) [36]. Each substance was tested on 10 mice and LD 50 was determined graphically by the well-known Litchfield-Wilcoxon method [37,38].

Conclusions
From the current study, it was concluded that the synthesized compounds showed acceptable pharmacodynamic and pharmacokinetic properties without carcinogenic or mutagenic potential but with variable hepatotoxicity as revealed from ADMET/ TOPKAT in silico study. Furthermore, in vivo studies showed that the introduction of substituents in the C-3 and C-4 positions of the phenyl radical of the initial 3a structure led to a decrease of the acute toxicity of the obtained derivatives. Substitution of hydrogen at the C-2 position for various groups does not significantly decrease the acute toxicity. The degree of acute toxicity depends on the mutual arrangement and nature of the substituents in the phenyl radical. Among the compounds studied, the least toxic compounds are those having substituents at C-3 and C-4 atoms, or at C-3 hydroxyl, bromo or methoxy, or at C-4 hydroxyl, nitro group, amino group, or methoxy. The exception is the methoxylation of both positions simultaneously (3i). All studied compounds showed high local anesthetic activity on rabbit eyes in the investigated concentrations. Only substances 3r, 3n, and 4a caused eye irritation and redness. All studied derivatives (except 4b) at a concentration of 1% were more active than lidocaine, providing a longer duration of complete anesthesia. Thus, based on the results obtained from the in silico, local anesthetic, and acute toxicity studies, it was clearly obvious that the investigated compounds (3c, 3d, 3e, 3l) need further in-depth future investigations so that they may act as promising drug candidates.