Synthesis of 8-Fluoro-3,4-dihydroisoquinoline and Its Transformation to 1,8-Disubstituted Tetrahydroisoquinolines

A simple procedure for the synthesis of 8-fluoro-3,4-dihydroisoquinoline is described below, based on a directed ortho-lithiation reaction. This key intermediate was then applied in various transformations. Fluorine–amine exchange afforded the corresponding 8-amino-3,4-dihydroisoquinolines, suitable starting compounds for the synthesis of 1-substituted 8-amino-tetrahydroisoquinolines. On the other hand, reduction and alkylation reactions of 8-fluoro-3,4-dihydroisoquinoline led to novel 1,2,3,4-tetrahydroisoquinoline derivatives that can be used as building blocks in the synthesis of potential central nervous system drug candidates.


Introduction
Isoquinolines and their partly saturated congeners (i.e., dihydro-and tetrahydroisoquinolines) constitute an important class of natural and synthetic compounds exhibiting biological activity. N-Acylated 1,8-disubstituted 1,2,3,4-tetrahydroisoquinolines 1 [1] and 2 [2] (Figure 1) proved to be potent calcium channel blockers for the treatment of chronic pain. Nomifensine (3), a norepinephrine-dopamine reuptake inhibitor [3], was launched as an antidepressant, without sedative effects.  The observed biological activity and our interest in the elaboration of a simple synthesis of isoquinoline derivatives monosubstituted on the aromatic ring at position 8 prompted us to develop an efficient synthesis of 8-fluoro-3,4-dihydroisoquinoline key intermediate and its further transformation to 1,2,3,4-tetrahydroisoquinolines bearing various cyclic amino substituents at position 8 and diverse lipophilic substituents at position 1 (4, Figure 1).
In general, the syntheses of tetrahydroisoquinolines exhibiting one single substituent on the benzene ring at position 8 require individual solutions. Compounds 1 (see Figure 1) were synthesized starting from N-acylated arylethylamines 5 (Scheme 1) [1]. The 4-bromo substituent in dihydroisoquinoline intermediate 6 serves as a protecting group, preventing formation of the regioisomeric isoquinoline, which would be the preferred product in the course of Bischler-Napieralski cyclization. Reduction of the C=N double bond of intermediate 6 with sodium borohydride leads to the corresponding tetrahydroisoquinoline 7. The bromine protecting group could then be removed by catalytic hydrogenation to give compound 8, which was transformed in two steps to products 1. The observed biological activity and our interest in the elaboration of a simple synthesis of isoquinoline derivatives monosubstituted on the aromatic ring at position 8 prompted us to develop an efficient synthesis of 8-fluoro-3,4-dihydroisoquinoline key intermediate and its further transformation to 1,2,3,4-tetrahydroisoquinolines bearing various cyclic amino substituents at position 8 and diverse lipophilic substituents at position 1 (4, Figure 1).
In general, the syntheses of tetrahydroisoquinolines exhibiting one single substituent on the benzene ring at position 8 require individual solutions. Compounds 1 (see Figure 1) were synthesized starting from N-acylated arylethylamines 5 (Scheme 1) [1]. The 4-bromo substituent in dihydroisoquinoline intermediate 6 serves as a protecting group, preventing formation of the regioisomeric isoquinoline, which would be the preferred product in the course of Bischler-Napieralski cyclization. Reduction of the C=N double bond of intermediate 6 with sodium borohydride leads to the corresponding tetrahydroisoquinoline 7. The bromine protecting group could then be removed by catalytic hydrogenation to give compound 8, which was transformed in two steps to products 1. The disadvantage of the synthetic route shown in Scheme 1 is that in order to introduce a new substituent into position 1 of the isoquinoline, the whole synthetic route has to be repeated with the new substituent in place of the cyclohexyl group.
A simple and widely applicable method has been published for the synthesis of 1-substituted tetrahydroisoquinolines 9 by treatment of 3,4-dihydroisoquinoline (10) with Grignard or organolithium reagents (Scheme 2) [4]. Organolithium reagents react much faster, even under milder conditions, than the corresponding Grignard reagents. The disadvantage of the synthetic route shown in Scheme 1 is that in order to introduce a new substituent into position 1 of the isoquinoline, the whole synthetic route has to be repeated with the new substituent in place of the cyclohexyl group.

Results and Discussion
After surveying the literature, we concluded that an efficient synthetic route to target compounds 4 should involve introduction of the lipophilic substituents into position 1 by treating 8-(cyclic amino)-3,4-dihydroisoquinoline 22 with the appropriate lithium reagent (Scheme 5). 8-Fluoro-3,4-dihydroisoquinoline (23) was selected as the precursor of compound 22, expecting that the C=N double bond activates the fluorine atom towards nucleophilic substitution by cyclic amines. To the best of our knowledge, compound 23 is not described in the literature.

Results and Discussion
After surveying the literature, we concluded that an efficient synthetic route to target compounds 4 should involve introduction of the lipophilic substituents into position 1 by treating 8-(cyclic amino)-3,4-dihydroisoquinoline 22 with the appropriate lithium reagent (Scheme 5). 8-Fluoro-3,4-dihydroisoquinoline (23) was selected as the precursor of compound 22, expecting that the C=N double bond activates the fluorine atom towards nucleophilic substitution by cyclic amines. To the best of our knowledge, compound 23 is not described in the literature. Schlosser and coworkers reported a short and efficient synthesis of 8-methoxy-3,4dihydroisoquinoline. Lithiation of N-pivaloyl-3-methoxyphenylethylamine (24) with butyllithium in diethyl ether at 25 °C for 2 h, followed by treatment with dimethylformamide gave aldehyde 25 (Scheme 6). Acid-catalyzed cyclization of the latter was accompanied by the loss of the pivaloyl moiety resulting in 8-methoxy-3,4-dihydroisoquinoline hydrochloride (26 • HCl) in 79% overall yield [21]. The ortho-directing ability of fluorine in lithiation reactions of aromatic compounds is well documented in the literature [22][23][24][25][26][27]. Based on this, we succeeded in extending the aforesaid method for the synthesis of 8-fluoro-3,4-dihydroisoqunoline (23) by a significant modification of the reaction conditions in the metalation step. Acylation of 2-(3-fluorophenyl)ethylamine (27) with pivaloyl chloride led to pivaloylamide 28 (Scheme 7). The lithiation was performed at −78 °C in order to prevent aryne formation by LiF elimination. Due to the poor solubility of compound 28 in diethyl

Results and Discussion
After surveying the literature, we concluded that an efficient synthetic route to target compounds 4 should involve introduction of the lipophilic substituents into position 1 by treating 8-(cyclic amino)-3,4-dihydroisoquinoline 22 with the appropriate lithium reagent (Scheme 5). 8-Fluoro-3,4-dihydroisoquinoline (23) was selected as the precursor of compound 22, expecting that the C=N double bond activates the fluorine atom towards nucleophilic substitution by cyclic amines. To the best of our knowledge, compound 23 is not described in the literature. Schlosser and coworkers reported a short and efficient synthesis of 8-methoxy-3,4dihydroisoquinoline. Lithiation of N-pivaloyl-3-methoxyphenylethylamine (24) with butyllithium in diethyl ether at 25 °C for 2 h, followed by treatment with dimethylformamide gave aldehyde 25 (Scheme 6). Acid-catalyzed cyclization of the latter was accompanied by the loss of the pivaloyl moiety resulting in 8-methoxy-3,4-dihydroisoquinoline hydrochloride (26 • HCl) in 79% overall yield [21]. The ortho-directing ability of fluorine in lithiation reactions of aromatic compounds is well documented in the literature [22][23][24][25][26][27]. Based on this, we succeeded in extending the aforesaid method for the synthesis of 8-fluoro-3,4-dihydroisoqunoline (23) by a significant modification of the reaction conditions in the metalation step. Acylation of 2-(3-fluorophenyl)ethylamine (27) with pivaloyl chloride led to pivaloylamide 28 (Scheme 7). The lithiation was performed at −78 °C in order to prevent aryne formation by LiF elimination. Due to the poor solubility of compound 28 in diethyl The ortho-directing ability of fluorine in lithiation reactions of aromatic compounds is well documented in the literature [22][23][24][25][26][27]. Based on this, we succeeded in extending the aforesaid method for the synthesis of 8-fluoro-3,4-dihydroisoqunoline (23) by a significant modification of the reaction conditions in the metalation step. Acylation of 2-(3-fluorophenyl)ethylamine (27) with pivaloyl chloride led to pivaloylamide 28 (Scheme 7). The lithiation was performed at −78 • C in order to prevent aryne formation by LiF elimination. Due to the poor solubility of compound 28 in diethyl ether at this low temperature, THF was used as the solvent. Subsequent treatment with DMF afforded formyl derivative 29, demonstrating that the lithiation occurred at the common ortho site of the substituents. Cyclization of aldehyde 29 in acidic medium occurred with simultaneous loss of the pivaloyl moiety to give 8-fluoro-3,4-dihydroisoquinoline (23), which was prepared as the hydrochloride hydrate (23 · HCl · H 2 O). Next, 8-fluoro-3,4-dihydroisoquinoline hydrochloride hydrate (23 • HCl • H2O) was heated with morpholine, pyrrolidine and piperidine at 80 °C for several hours in a sealed tube (Scheme 8). In the case of morpholine and pyrrolidine, the required products 22a and 22b were isolated in moderate yields (51% and 49%). Unexpectedly, piperidine derivative 22c was obtained in significantly lower yield (17%). Higher reaction temperatures and longer reaction times did not improve the yields: tarring was observed and HPLC-MS measurements indicated formation of the dehydrogenated Next, 8-fluoro-3,4-dihydroisoquinoline hydrochloride hydrate (23 · HCl · H 2 O) was heated with morpholine, pyrrolidine and piperidine at 80 • C for several hours in a sealed tube (Scheme 8). In the case of morpholine and pyrrolidine, the required products 22a and 22b were isolated in moderate yields (51% and 49%). Unexpectedly, piperidine derivative 22c was obtained in significantly lower yield (17%). Higher reaction temperatures and longer reaction times did not improve the yields: tarring was observed and HPLC-MS measurements indicated formation of the dehydrogenated byproduct (the corresponding 8-aminoisoquinoline). Analogous reactions starting from base 23 were significantly slower indicating that protonation of the N-2 atom increased the electrophilicity of the C-8 atom, i.e., it promotes the nucleophilic attack at this position. Next, 8-fluoro-3,4-dihydroisoquinoline hydrochloride hydrate (23 • HCl • H2O) was heated with morpholine, pyrrolidine and piperidine at 80 °C for several hours in a sealed tube (Scheme 8). In the case of morpholine and pyrrolidine, the required products 22a and 22b were isolated in moderate yields (51% and 49%). Unexpectedly, piperidine derivative 22c was obtained in significantly lower yield (17%). Higher reaction temperatures and longer reaction times did not improve the yields: tarring was observed and HPLC-MS measurements indicated formation of the dehydrogenated byproduct (the corresponding 8-aminoisoquinoline). Analogous reactions starting from base 23 were significantly slower indicating that protonation of the N-2 atom increased the electrophilicity of the C-8 atom, i.e., it promotes the nucleophilic attack at this position.
Finally we realized that some simple 8-fluoroisoquinoline derivatives easily available from 8fluoro-3,4-dihydroisoquinoline (23) are not described in the literature. Therefore we treated Next, 8-fluoro-3,4-dihydroisoquinoline hydrochloride hydrate (23 • HCl • H2O) was heated with morpholine, pyrrolidine and piperidine at 80 °C for several hours in a sealed tube (Scheme 8). In the case of morpholine and pyrrolidine, the required products 22a and 22b were isolated in moderate yields (51% and 49%). Unexpectedly, piperidine derivative 22c was obtained in significantly lower yield (17%). Higher reaction temperatures and longer reaction times did not improve the yields: tarring was observed and HPLC-MS measurements indicated formation of the dehydrogenated byproduct (the corresponding 8-aminoisoquinoline). Analogous reactions starting from base 23 were significantly slower indicating that protonation of the N-2 atom increased the electrophilicity of the C-8 atom, i.e., it promotes the nucleophilic attack at this position.
Finally we realized that some simple 8-fluoroisoquinoline derivatives easily available from 8-fluoro-3,4-dihydroisoquinoline (23) are not described in the literature. Therefore we treated compound 23 with sodium borohydride to obtain tetrahydroisoquinoline 30 (Scheme 10). Methylation of compound 23 with methyl iodide gave isoquinolinium derivative 31, which was reduced with sodium borohydride to 8-fluoro-2-methyl-1,2,3,4-tetrahydroisoquinoline 32. To the best of our knowledge, isoquinoline derivatives 30-32 are new. Although compound 30 has been mentioned in the literature [28], its preparation and characterization were not described.

General
All melting points were determined on a Büchi B-540 (Flawil, Switzerland) capillary melting

Experimental Section
General All melting points were determined on a Büchi B-540 (Flawil, Switzerland) capillary melting point apparatus and are uncorrected. IR spectra were obtained on a Bruker ALPHA FT-IR spectrometer (Billerica, MA, USA) in KBr pellets of as a film. 1 H-NMR, 13 C-NMR and 19 F-NMR spectra were recorded at 295 K on a Bruker Avance III HD 600 (Billerica, MA, USA) (600, 150 and 564.7 MHz for 1 H-, 13 C-and 19 F-NMR spectra, respectively) or at ambient temperature on a Bruker Avance III 400 (Billerica, MA, USA) (400 and 100 MHz for 1 H and 13 C-NMR spectra, respectively) spectrometer. CDCl 3 or CD 3 OD was used as the solvent, tetramethylsilane (TMS) for 1 H, 13 C-NMR or trichlorofluoromethane (CFCl 3 ) for 19 F-NMR as the internal standard. Chemical shifts (δ) and coupling constants (J) are given in ppm and in Hz, respectively. Mass spectra were recorded on a Bruker O-TOF MAXIS Impact mass spectrometer (Billerica, MA, USA) coupled with a Dionex Ultimate 3000 RS HPLC (Sunnyvale, CA, USA) system with a diode array detector. The reactions were followed by analytical thin-layer chromatography on silica gel 60 F 254 (Darmstadt, Germany) and HPLC-MS on a Shimadzu LC-20 HPLC equipment (Kyoto, Japan). Purifications by flash chromatography were carried out using Merck 107736 silica gel 60 H (Darmstadt, Germany) using a hexane-ethyl acetate or dichloromethane-methanol solvent system. All reagents were purchased from commercial sources.   (23). To a vigorously stirred mixture of 23 · HCl · H 2 O (5.89 g, 28.9 mmol) in dichloromethane (100 mL) and water (50 mL) aqueous sodium carbonate (10%, 20 mL) was added. The layers were separated and the aqueous layer was extracted with dichloromethane (3 × 25 mL). The combined organic layer was extracted with water (2 × 50 mL) and brine (50 mL) and dried over MgSO 4 . The solvent was evaporated to afford the title compound 8-(Pyrrolidin-1-yl)-3,4-dihydroisoquinoline (22b). A mixture of 23 · HCl · H 2 O (3.50 g, 17.2 mmol) and pyrrolidine (4.23 mL, 3.66 g, 51.6 mmol) was stirred for 8 h at 80 • C in a sealed tube. After the reaction mixture was cooled, dichloromethane (60 mL) was added, and the resulting mixture was extracted with water (3 × 20 mL). The combined organic layer was dried over MgSO 4 , and evaporated.   8-Fluoro-2-methyl-1,2,3,4-tetrahydroisoquinoline (32). Sodium borohydride (80 mg, 2.09 mmol) was added to a solution of 31 (508 mg, 1.75 mmol) in methanol (14 mL), and the reaction mixture was cooled with an ice/water bath. After stirring for 1 h at room temperature, water (6 mL) was added, and the resulting mixture was extracted with dichloromethane (3 × 8 mL). The combined organic layer was dried over MgSO 4 , and evaporated. The residue was triturated in hexane and filtered. The filtrate was evaporated to afford the title compound (251 mg, 87%) as a colorless oil. IR (film): ν = 2924, 1468 cm −1 .
Starting from this, 1-alkyl(phenyl)-8-(cyclic amino)-1,2,3,4-tetrahydroisoquinolines were prepared by a fluorine-amine exchange reaction followed by the addition of alkyl(phenyl)lithium reagents to the C=N double bond. The synthetic route, based on simple model compounds, provides the basis for the preparation of a compound library containing more complex analogues as potential central nervous system drug candidates. 8-Fluoro-3,4-dihydroisoquinoline is also an advantageous precursor of some new 1,2,3,4-tetrahydroisoquinoline building blocks, optionally substituted at the nitrogen atom. Funding: This research received no external funding.

Conflicts of Interest:
The authors declare no conflict of interest.