Synthesis of Pyridoxine-Derived Dimethylpyridinols Fused with Aminooxazole, Aminoimidazole, and Aminopyrrole

Described in this paper are studies on the preparation of three classes of dimethylpyridinols derived from pyridoxine fused with aminooxazole, aminoimidazole, and aminopyrrole. The key feature of this synthetic strategy is the manipulation of hydroxymethyl moiety of C(5)-position of the pyridoxine starting material along with the installation of an amino group at C(6)-position. Efficient and practical synthesis for the oxazole- and imidazole-fused targets was accomplished, while the instability of the pyrrole-fused one was observed.


Introduction
Phenotypic screening campaigns have long been crucial to the identification of biologically active small molecules in conjunction with target-based approaches [1,2]. Because of their biological relevancy and favorable physicochemical properties, novel scaffolds from nature are frequently used to inspire the preparation of compound collections for phenotypic screening [3][4][5][6]. Pyridoxine (1) is a form of vitamin B 6 that scavenges reactive oxygen species and regulates diverse cellular metabolisms, including amino acid biosynthesis and fatty acid biosynthesis ( Figure 1) [7][8][9][10][11]. Inspired by pyridoxin (1), our group has pursued derivative syntheses along with biological investigations to explore the pharmacological potential of the 6-aminopyridin-3-ol scaffold (2). The class of aminopyridinols (3) was found to have anti-angiogenic activity, which is linked to cancer and age-related diseases, as well as anti-inflammatory activity against ulcerative colitis. [12,13]. Notably, remarkable anticancer activity was observed in novel structures where the 6-aminopyridin-3-ol core was hybridized with biologically active agents such as α-tocopherol and sunitinib. An α-tocopherol-hybridized compound 10 suppressed lung tumor growth by dual inhibition of NADPH oxidase 2 and receptor tyrosine kinases. A sunitinib-hybridized compound 11 was also discovered as an apoptosis-inducing anticancer agent by controlling the p53 level, with a safer cytotoxicity profile than sunitinib [14]. Therefore, these results showcase that the 6-amino-pyridin-3-ol scaffolds indeed have the potential to provide favorable biological activities and demand further studies to expand the scope of the scaffolds.

Results and Discussion
A general outline of the synthetic strategy for the preparation of dimethylpyridinols fused with aminooxazole (12), aminoimidazole (13), and aminopyrrole (14) is shown in Scheme 2. This began with the preparation of known primary alcohol 15, which was used as a common intermediate for the syntheses of three final products, according to the threestep sequence starting from pyridoxine (1) established by us [10,31]. Scheme 1. Pyridoxine-derived five-membered heterocycle-fused pyridinols.

Results and Discussion
A general outline of the synthetic strategy for the preparation of dimethylpyridinols fused with aminooxazole (12), aminoimidazole (13), and aminopyrrole (14) is shown in Scheme 2. This began with the preparation of known primary alcohol 15, which was used as a common intermediate for the syntheses of three final products, according to the three-step sequence starting from pyridoxine (1) established by us [10,31]. Scheme 1. Pyridoxine-derived five-membered heterocycle-fused pyridinols.

Results and Discussion
A general outline of the synthetic strategy for the preparation of dimethylpyridinols fused with aminooxazole (12), aminoimidazole (13), and aminopyrrole (14) is shown in Scheme 2. This began with the preparation of known primary alcohol 15, which was used as a common intermediate for the syntheses of three final products, according to the threestep sequence starting from pyridoxine (1) established by us [10,31].

Scheme 2.
A strategy for the divergent synthesis of dimethylpyridinols fused with aminooxazole (12), aminoimidazole (13), and aminopyrrole (14)  First, for the synthesis of the aminooxazole-containing analogue 12, the hydroxymethyl moiety of 15 was converted to amine 17 in three steps [31]. The amino group of 17 was replaced by hydroxy group to afford 18 as we previously reported [31]. Then, an amine group was successfully introduced at the C(2)-position of 18 in the presence of free alcohol at the C(3)-position, without additional protection/deprotection steps [31]. The resulting 2-amino-3-hydroxy pyridinol 19 was nicely constructed to test the formation of the aminooxazole moiety. Gratefully, the formation of the aminooxazole ring was accomplished by using cyanogen bromide in almost quantitative yield (compound 23) [34], which was followed by debenzylation to afford the first target compound 12 (Scheme 3). First, for the synthesis of the aminooxazole-containing analogue 12, the hydroxymethyl moiety of 15 was converted to amine 17 in three steps [31]. The amino group of 17 was replaced by hydroxy group to afford 18 as we previously reported [31]. Then, an amine group was successfully introduced at the C(2)-position of 18 in the presence of free alcohol at the C(3)-position, without additional protection/deprotection steps [31]. The resulting 2-amino-3-hydroxy pyridinol 19 was nicely constructed to test the formation of the aminooxazole moiety. Gratefully, the formation of the aminooxazole ring was accomplished by using cyanogen bromide in almost quantitative yield (compound 23) [34], which was followed by debenzylation to afford the first target compound 12 (Scheme 3).
Having observed the successful aminooxazole formation, we applied a similar approach to achieve analogue synthesis, which is the installation of corresponding functional groups followed by cyclization as outlined in Scheme 2. Since aminopyridinol in- Having observed the successful aminooxazole formation, we applied a similar approach to achieve analogue synthesis, which is the installation of corresponding functional groups followed by cyclization as outlined in Scheme 2. Since aminopyridinol intermediates (15~17) with diverse functionalities were generated during the preparation of aminooxazolopyridinol 12, we envisaged that these intermediates could serve as great branching points for the synthesis of aminoimidazole 13 and aminopyrrole 14.
The synthesis of aminoimidazolopyridinol 13 initiated with the preparation of the diamine compound 20 (Scheme 4). We first attempted to introduce an amino group at 2-position of the pyridine ring in the primary amide compound 16 via phthalimidation and subsequent Hofmann rearrangement (the left side of Scheme 4). After m-CPBA oxidation of 16 to pyridine N-oxide 24, the nucleophilic addition of phthalimide to the O-p-toluenesulfonylated pyridinium intermediate followed by the elimination of p-toluenesulfonic acid afforded 25. Removal of the N-phthalyl group in 25 was smoothly proceeded by the treatment of hydrazine to give free amine 26 [31]. The trial for Hofmann rearrangement of the primary amide compound 26 to obtain the diamine compound 20 was performed in a mixed solvent with alkaline water and tetrahydrofuran, employing hypochlorite for activation of the primary amide. However, contrary to our expectations, the desired diamine compound 20 was not produced under the reaction conditions. A cyclic urea compound 27 was instead obtained in a moderate yield (40%).  Since we experienced such a fact in this trial, we expected that using water as a solvent would increase the probability of transformation of isocyanate intermediate 32 to carbamic acid 33, which then spontaneously decomposed to the desired diamine 20 with the liberation of carbon dioxide. However, based on the results we observed, the intramolecular cyclization outcompeted the hydroxide (or water) addition to the isocyanate 32. Several attempts to cleave the urea of compound 27 to obtain to form a five-membered cyclic urea 27. Since we experienced such a fact in this trial, we expected that using water as a solvent would increase the probability of transformation of isocyanate intermediate 32 to carbamic acid 33, which then spontaneously decomposed to the desired diamine 20 with the liberation of carbon dioxide. However, based on the results we observed, the intramolecular cyclization outcompeted the hydroxide (or water) addition to the isocyanate 32. Several attempts to cleave the urea of compound 27 to obtain the diamine 20 under acidic conditions with high temperatures only gave a debenzylated compound quantitatively instead of the desired product. To avoid such a troublesome intramolecular cyclization issue, we decided to change the synthetic sequence. The conversion of the primary amide group at C(3)-position to a free amino group was performed prior to the installation of another free amino group at C(2)-position as depicted on the right side of Scheme 4. Hofmann rearrangement of 16 followed by N-oxide formation (28) and phthalimide substitution proceeded smoothly to afford 29, which was then treated with hydrazine to produce the diamine compound 20. The aminoimidazole backbone of compound 30 was successfully constructed by the treatment of 20 with cyanogen bromide. Finally, the aminoimidazole-fused pyridinol 13 was obtained by debenzylation.
Next, we investigated the synthesis of aminopyrrole-containing analog 14 (Scheme 6). The initial attempt includes the substitution of primary alcohol of the common intermediate 15 with nitrile functionality (21) [31] and phthalimidation of N-oxide compound 34. Reactions proceeded smoothly, generating compound 35, and phthalimide cleavage using hydrazine successfully furnished the requisite substrate 36 for aminopyrrole formation. Having installed the 2-amino group together with the nitrile group in 36, trials were made for the cyclization to form a 2-amino-7-azaindole framework. According to a datan Scheme 5. Plausible mechanism for the formation of cyclic urea compound 27.
To avoid such a troublesome intramolecular cyclization issue, we decided to change the synthetic sequence. The conversion of the primary amide group at C(3)-position to a free amino group was performed prior to the installation of another free amino group at C(2)-position as depicted on the right side of Scheme 4. Hofmann rearrangement of 16 followed by N-oxide formation (28) and phthalimide substitution proceeded smoothly to afford 29, which was then treated with hydrazine to produce the diamine compound 20. The aminoimidazole backbone of compound 30 was successfully constructed by the treatment of 20 with cyanogen bromide. Finally, the aminoimidazole-fused pyridinol 13 was obtained by debenzylation.
Next, we investigated the synthesis of aminopyrrole-containing analog 14 (Scheme 6). The initial attempt includes the substitution of primary alcohol of the common intermediate 15 with nitrile functionality (21) [31] and phthalimidation of N-oxide compound 34. Reactions proceeded smoothly, generating compound 35, and phthalimide cleavage using hydrazine successfully furnished the requisite substrate 36 for aminopyrrole formation. To avoid such a troublesome intramolecular cyclization issue, we decided to change the synthetic sequence. The conversion of the primary amide group at C(3)-position to a free amino group was performed prior to the installation of another free amino group at C(2)-position as depicted on the right side of Scheme 4. Hofmann rearrangement of 16 followed by N-oxide formation (28) and phthalimide substitution proceeded smoothly to afford 29, which was then treated with hydrazine to produce the diamine compound 20. The aminoimidazole backbone of compound 30 was successfully constructed by the treatment of 20 with cyanogen bromide. Finally, the aminoimidazole-fused pyridinol 13 was obtained by debenzylation.
Next, we investigated the synthesis of aminopyrrole-containing analog 14 (Scheme 6). The initial attempt includes the substitution of primary alcohol of the common intermediate 15 with nitrile functionality (21) [31] and phthalimidation of N-oxide compound 34. Reactions proceeded smoothly, generating compound 35, and phthalimide cleavage using hydrazine successfully furnished the requisite substrate 36 for aminopyrrole formation. Having installed the 2-amino group together with the nitrile group in 36, trials were made for the cyclization to form a 2-amino-7-azaindole framework. According to a database search result using the SciFinder n on the transformation of reactants 38 to products 39, no example has been known in the case of an unsubstituted compound at C(#)-position Having installed the 2-amino group together with the nitrile group in 36, trials were made for the cyclization to form a 2-amino-7-azaindole framework. According to a database search result using the SciFinder n on the transformation of reactants 38 to products 39, no example has been known in the case of an unsubstituted compound at C(#)-position (Scheme 7). Even if the search range was expanded to cases where there was a substituent other than hydrogen at the C(#)-position, only a couple of examples were found including the synthesis of compound 41 [35,36]. Formation of 2-aminoindoles, such as compound 43, from C(#)-unsubstituted-2-(2-aminopyridin-3-yl)acetonitrile compounds, such as 42, was also found to be rare [37,38].
Molecules 2022, 26, x FOR PEER REVIEW 43, from C(#)-unsubstituted-2-(2-aminopyridin-3-yl)acetonitrile compounds, suc was also found to be rare [37,38]. The first trial was done under basic conditions using sodium methoxide, wh been successfully applied in the preparation of 2-aminoindole ring formation (Sch [37]. Contrary to the literature in which the 2-aminoindole 43 was obtained in 70 from 42 (Scheme 7), the reaction of compound 36 under the same reaction conditio messy, and the desired product 37 was not obtained. It can be presumed that the ence in nucleophilicity between the amine attached to benzene in compound 42 amine attached to pyridine in compound 36 might be one of the main reasons for cess or failure of the 2-aminopyrrole cyclization reaction under similar reaction tions. Interestingly, the formation of a very small amount of a by-product 44 was ob instead. We speculate that this side reaction may have proceeded as follows (Sch The first trial was done under basic conditions using sodium methoxide, which has been successfully applied in the preparation of 2-aminoindole ring formation (Scheme 8) [37]. Contrary to the literature in which the 2-aminoindole 43 was obtained in 70% yield from 42 (Scheme 7), the reaction of compound 36 under the same reaction conditions was messy, and the desired product 37 was not obtained. It can be presumed that the difference in nucleophilicity between the amine attached to benzene in compound 42 and the amine attached to pyridine in compound 36 might be one of the main reasons for the success or failure of the 2-aminopyrrole cyclization reaction under similar reaction conditions. Interestingly, the formation of a very small amount of a by-product 44 was observed instead. We speculate that this side reaction may have proceeded as follows (Scheme 9). Formation of α-ketonitrile intermediate 47 might have been facilitated by the deprotonation with methoxide at the site adjacent to the nitrile group in 36, which is also a benzylic position. After the successive aerobic oxidation, the cyano group of 47 might be replaced by methoxide anion affording compound 44.
To avoid such a deprotonation/oxidation problem, we exposed compound 36 to hot acetic acid. However, the desired 2-amino-7-azaindole compound 37 was not obtained under the acidic conditions either. Instead, its N-acetyl compound 44 was obtained, albeit in small quantities, and the formation of a lactam 46 was also observed. Use of an additive like zinc acetate to activate the cyclization improved the formation of compound 44 to some extent. Unlike the other two targets, aminooxazolopyridinol and aminoimidazolopyridinol, which were readily prepared, an aminopyrrolopyridine-containing compound like 37 might be presumed to have quite poor chemical stability. The free amino compound 37 was not obtained under both basic and acidic reaction conditions, and even the isolated pure N-acetylated compound 44 was observed to gradually decompose in solution over time. Naturally, all efforts made for N-deacetylation were not fruitful. messy, and the desired product 37 was not obtained. It can be presumed that the difference in nucleophilicity between the amine attached to benzene in compound 42 and the amine attached to pyridine in compound 36 might be one of the main reasons for the success or failure of the 2-aminopyrrole cyclization reaction under similar reaction conditions. Interestingly, the formation of a very small amount of a by-product 44 was observed instead. We speculate that this side reaction may have proceeded as follows (Scheme 9) Formation of α-ketonitrile intermediate 47 might have been facilitated by the deprotonation with methoxide at the site adjacent to the nitrile group in 36, which is also a benzylic position. After the successive aerobic oxidation, the cyano group of 47 might be replaced by methoxide anion affording compound 44. Scheme 8. Attempts on the cyclization towards 2-aminopyrrole. To avoid such a deprotonation/oxidation problem, we exposed compound 36 to hot acetic acid. However, the desired 2-amino-7-azaindole compound 37 was not obtained under the acidic conditions either. Instead, its N-acetyl compound 44 was obtained, albeit in small quantities, and the formation of a lactam 46 was also observed. Use of an additive like zinc acetate to activate the cyclization improved the formation of compound 44 to some extent. Unlike the other two targets, aminooxazolopyridinol and aminoimidazolopyridinol, which were readily prepared, an aminopyrrolopyridine-containing compound like 37 might be presumed to have quite poor chemical stability. The free amino compound 37 was not obtained under both basic and acidic reaction conditions, and even the isolated pure N-acetylated compound 44 was observed to gradually decompose in solution over time. Naturally, all efforts made for N-deacetylation were not fruitful.
Based on our observations, a possible mechanism for the generation of 45 and 46 was proposed in Scheme 10. The reaction is initiated by intramolecular cyclization of 36 to form a 1,3-dihydro-2H-pyrrole-2-imine 37′, which undergoes tautomerization to yield a 2aminopyrrole 37. Under the reflux condition, the electron-rich amino group in 37 further reacts with acetic acid to generate N-acetyl adduct 45. Another pathway involves the nucleophilic addition of acetic acid to the electrophilic imine-carbon in 37′ and the subse- Based on our observations, a possible mechanism for the generation of 45 and 46 was proposed in Scheme 10. The reaction is initiated by intramolecular cyclization of 36 to form a 1,3-dihydro-2H-pyrrole-2-imine 37 , which undergoes tautomerization to yield a 2-aminopyrrole 37. Under the reflux condition, the electron-rich amino group in 37 further reacts with acetic acid to generate N-acetyl adduct 45. Another pathway involves the nucleophilic addition of acetic acid to the electrophilic imine-carbon in 37 and the subsequent elimination of ammonia to afford 48. Acidolysis of the resulting acetate 48 followed by tautomerization provides a lactam 46. Since high temperature and long reaction time are necessary to make the intramolecular cyclization of 36 happen, the reversible reactions illustrated in Scheme 10 might occur and generate the mixture of 45 and 46. Unlike our system, compound 40 contains an additional ester group at the C(#)-position (Scheme 7) [35]. This carbonyl functional group may accelerate cyclization/tautomerization events and make the resulting amine of 41 less nucleophilic, preventing the formation of an N-acetyl adduct of 41.
by tautomerization provides a lactam 46. Since high temperature and long reaction time are necessary to make the intramolecular cyclization of 36 happen, the reversible reactions illustrated in Scheme 10 might occur and generate the mixture of 45 and 46. Unlike our system, compound 40 contains an additional ester group at the C(#)-position (Scheme 7) [35]. This carbonyl functional group may accelerate cyclization/tautomerization events and make the resulting amine of 41 less nucleophilic, preventing the formation of an Nacetyl adduct of 41.

General
Unless otherwise specified, all the materials obtained from Merck (Kenilworth, NJ, USA), TCI (Tokyo, Japan), and Alfa Aesar (Ward Hill, MA, USA) were used without further purification. Inert gas conditions were applied to the sensitive reactions towards air and moisture. The progress of the reactions was traced by TLC analysis using silica gel 60 F 254 plates (Merck, Kenilworth, NJ, USA). Visualization of the TLC spots was done using a UV lamp (254 nm, Spectroline Corp., Westbury, NY, USA) and staining solutions (Anisaldehyde solution and KMnO 4 solution) prepared by us using commercially available reagents. Products were purified by flash column chromatography using silica gel 60 (70-230 mesh, Merck, Kenilworth, NJ, USA) or by using the Biotage 'Isolera One' (Biotage, Uppsala, Sweden) system with indicated solvents. Melting points were determined using a Büchi melting point B-540 apparatus (Büchi Labortechnik, Flawil, Switzerland) and were unchanged. 1 H and 13 C NMR spectra were obtained using a Bruker-250 spectrometer ( 1 H and 13 C frequencies were 250 and 63 MHz, respectively, Bruker Corp., Billerica, MA, USA), and a Bruker Avance Neo 400 spectrometer ( 1 H and 13 C frequencies were 400 and 100 MHz, respectively, Bruker Corp., Billerica, MA, USA). 1 H and 13 C NMR spectra of all synthesized molecules are available in the Supplementary Materials ( Figures S1-S26). Chemical shifts (δ) were expressed in ppm calibrated to residual solvent signals and the coupling constant (J) in hertz. HR-ESIMS was performed using a Thermo Scientific Q Exactive hybrid Quadrupole-Orbitrap mass spectrometer (Waltham, MA, USA) coupled to a Thermo Scientific Vanquish UHPLC system (Waltham, MA, USA) at the Core Research Support Center for Natural Products and Medical Materials (CRCNM).

Methyl 2-amino-5-(benzyloxy)-4,6-dimethylnicotinate (44)
To a solution of compound 36 (30 mg, 0.11 mmol) in MeOH (3 mL), we added 25% NaOMe in MeOH (0.25 mL, 1.1 mmol). The mixture was stirred at 50 • C for 1 h under argon atmosphere and cooled to room temperature. The mixture was poured into water and extracted with EtOAc. The organic solution was dried over MgSO 4 , filtered, and concentrated. The residue was purified by silica gel column chromatography (Hexanes:EtOAc = 1:1) to give 44 (3 mg, 10%) as a white solid. R f 0. 30   To a solution of compound 36 (13 mg, 0.048 mmol) in glacial acetic acid, we added zinc acetate (45 mg, 0.24 mmol). The mixture was refluxed under argon atmosphere for 12 h. The mixture was concentrated, and the residue was diluted with H 2 O. The aqueous layer was extracted with CH 2 Cl 2 , and the combined organic solution was dried over MgSO 4 , filtered, and concentrated. The residue was purified by silica gel preparative thin layer chromatography to give 45 (7 mg, 45%) as a brown solid. R f 0.29 (Hexanes:EtOAc:MeOH = 10:10:1); 1

Conclusions
In this study, the aminooxazole-, aminoimidazole-, and aminopyrrole-containing pyridinols were divergently synthesized starting from pyridoxine. The synthetic strategy relied on the conversion of hydroxymethyl functionality in the common intermediate 15 into phenol, aniline, and cyanomethyl structures, respectively. The installation of nitrogencontaining functional group at α-position in pyridine was achieved via either diazotization or phthalimidation. These synthetic routes enable the rapid construction of the three different aminobicyclic pyridinol derivatives in an efficient and divergent way. With the two chemical handles including the 5 -hydroxy group and arylamino group, these backbones may serve as a valuable starting point for the generation of diverse sets of compound libraries through appendage diversification.