Efficient Electrochemical N-Alkylation of N-Boc-Protected 4-Aminopyridines: Towards New Biologically Active Compounds

The use of electrogenerated acetonitrile anion allows the alkylation of N-Boc-4-aminopyridine in very high yields, under mild conditions and without by-products. The high reactivity of this base is due to its large tetraethylammonium counterion, which leaves the acetonitrile anion “naked.” The deprotection of the obtained compounds led to high yields in N-alkylated 4-aminopyridines. Nonsymmetrically dialkylated 4-aminopyridines were obtained by subsequent reaction of monoalkylated ones with t-BuOK and alkyl halides, while symmetrically dialkylated 4-aminopyridines were obtained by direct reaction of 4-aminopyridine with an excess of t-BuOK and alkyl halides. Some mono- and dialkyl-4-aminopyridines were selected to evaluate antifungal and antiprotozoal activity; the dialkylated 4-aminopyridines 3ac, 3ae and 3ff showed antifungal towards Cryptococcus neoformans; whereas 3cc, 3ee and 3ff showed antiprotozoal activity towards Leishmania infantum and Plasmodium falciparum.


Introduction
N-Alkylated 4-aminopyridine is a common moiety in biologically active molecules. It is present, in fact, in compounds with different activities such as inhibitors of p38 MAP kinase [1], inhibitors of HIV-EP1 cellular transcription factor [2], inhibitors of coagulation Factor Xa [3], and -chemokine receptor CCR5 antagonists in anti-HIV therapy [4]; in particular we have focused our work on the development of new CYP51 inhibitors, active both on fungal strains [5] and Trypanosoma Cruzi [6]. Many literature data evidenced that the pyridine group can efficaciously replace the hemeiron chelating azole moiety present in classical azole CYP51 inhibitors and, therefore, the alkylation of 4-aminopyridine (4AP) represents an important goal in organic synthesis to develop novel classes of antifungal and antiparasitic drugs [7,8].
Due to the wide presence of these products, the alkylation of 4-aminopyridine (4AP) is therefore an important goal in organic synthesis. Different approaches to obtain N-alkylated 4-aminopyridines have been reported in the literature. Some examples are the efficient condensation of 4AP with alcohols catalyzed by benzaldehyde [9] or copper [10,11] or magnetite [12], the reaction of 4AP with an acyl chloride, and the following reduction of the amide with LiAlH 4 [13].
The most straightforward method, however, is the direct alkylation of 4AP with alkyl halides, although it suffers from some drawbacks. The two different nitrogen atoms compete in the alkylation reaction and usually the more nucleophilic pyridine nitrogen atom reacts faster, leading to the corresponding pyridinium salt (Scheme 1) [14,15].
In these case, the use of a very strong base is therefore necessary: n-BuLi was successfully used by Singh [16], obtaining N-methyl-and N-ethyl-4-aminopyridines in 74-80% yields. A viable alternative is the enhancement of the nucleophilicity of the amine nitrogen atom (versus the pyridine one), allowing the use of weaker bases. An example is the activation of 2-aminopyridine as formyl or Boc derivative at the amine nitrogen atom [17], with subsequent deprotonation using sodium hydride, alkylation, and deprotection with trifluoroacetic acid. The deprotonation of N-Boc-2aminopyridine with NaH needs a careful control of the temperature (0-5 ∘ C) and is carried out in anhydrous DMF, with a vigorous stirring required to keep the suspension fluid.
In this context, we envisaged the possibility to alkylate N-Boc-4-aminopyridine (N-Boc-4AP) using milder reaction conditions, that is, using electrogenerated tetraethylammonium cyanomethanide (Et4N +− CH 2 CN) [18]. This base, the acetonitrile anion, can be easily obtained by cathodic galvanostatic reduction of a solution of acetonitrile containing tetraethylammonium hexafluorophosphate as supporting electrolyte (Scheme 2), without by-products (the reagent is the electron), and it was successfully used by us in a good variety of reactions [19], such as the selective N-alkylation of bifunctional compounds [20], the Gewald reaction [21], the synthesis of -lactams [22], and the synthesis of carbamates [23]. The actual mechanism for the formation of acetonitrile anion is not known, but a hypothesis based on the direct reduction of the tetraethylammonium cation has been reported (Scheme 2) [24].
The high reactivity of this base is ascribable to the large tetraethylammonium counterion, which renders the acetonitrile anion extremely reactive. Moreover, its reaction as a base gives no by-products, as the protonation restores the molecule of solvent.

Results and Discussion
The reaction of electrogenerated acetonitrile anion with 4AP, followed by an alkyl halide, leads to poor yields in desired compound with the pyridinium salt being the major product. This prevents the direct use of − CH 2 CN with 4AP. On the other hand, if the amine nitrogen is activated as Boc derivative (N-Boc-4AP), the deprotonation/alkylation reaction using acetonitrile anion leads to products 1 in very high yields (Scheme 3 and Table 1, entries 1-6). The classic deprotection with trifluoroacetic acid allows obtaining the desired products 2 (Scheme 3 and Table 1, entries 1-6).
The data in Table 1 highlight that the reaction of deprotonation of N-Boc-4AP using electrogenerated acetonitrile anion, alkylation with both alkyl and benzyl halides, and deprotection with trifluoroacetic acid is very efficient, with overall yields of 78-86%. However, when the alkylating agent is a bromoacetophenone, the yields in alkylated product are lower and in most cases the deprotection reaction leads to the dealkylation of the starting material (Table 1, entries 7-10).
As many biologically active compounds contain the dialkylated 4-aminopyridine moiety, we tried to carry out a second alkylation on products 2a-j using acetonitrile anion but, as expected, the high nucleophilicity of the pyridine nitrogen led to the synthesis of the corresponding pyridinium salt.
We thus carried out this second alkylation using strong bases, the most efficient being t-BuOK in DMSO (Scheme 4), although the yields in dialkylated 4AP were not very high. The results of this reaction are reported in Table 2.
In order to obtain symmetrically dialkylated 4AP, 4AP was subjected to deprotonation with t-BuOK in DMSO, adding an excess of alkylating agent. The reaction led to a mixture of mono-and dialkylated 4-aminopyridines, in moderate to acceptable yields. The results are reported in Table 3.

Biological Activity
A selection of synthesized compounds was in vitro tested to evaluate antifungal activity against different strains of C. albicans, C. parapsilosis, and Cryptococcus neoformans; data are reported in Table 4. As can be evidenced the nonsymmetrical dialkylated 4APs 3ac and 3ae showed a moderate antifungal activity towards C. albicans and C. parapsilosis with MIC 50 values of 32 g/mL and showed an interesting activity against Cryptococcus neoformans, with MIC 50 values of 0.4 and 4 g/mL, respectively. Otherwise, the symmetrical dialkylated 4APs 3cc, 3ee and the Boc-protected monoalkylated 4APs 1b, 1e, 1f showed poor antifungal activity with MIC 50 and MIC 100 ≥ 64 g/mL.
Furthermore, the symmetrical dialkylated 4APs 3cc, 3ee, and 3ff were in vitro tested to evaluate the activity against Trypanosoma cruzi, Trypanosoma brucei, Leishmania infantum, and Plasmodium falciparum; the results are summarized in Table 5.
As can be evidenced, all tested compounds showed a moderate activity versus P. falciparum and an interesting activity towards L. infantum with IC 50 values lower than the reference drug miltefosine; otherwise, they resulted scarcely active against T. cruzi and T. brucei. Moreover, these compounds also showed low toxic activity versus growing MRC-5 cells.

Conclusion
In conclusion, we demonstrated the usefulness of electrogenerated acetonitrile anion in the alkylation of N-Boc 4aminopyridines, both from the point of view of the high yields and of the cleanliness of the reaction (no by-products). The deprotection of N-Boc 4-aminopyridines allowed obtaining monoalkylated 4-aminopyridine in very high yields. The following alkylation, by means of t-BuOK and alkyl halides, led to nonsymmetrically dialkylated 4-aminopyridine, while symmetrically dialkylated products were obtained directly from 4-aminopyridine by reaction with an excess of t-BuOK and alkyl halide.   At the end of the electrolysis, 1 mmol of alkylating agent was added. After 2 h at rt, usual workup afforded the products. Deprotection was carried out as described in the experimental part. All the yields are in isolated products. b When compounds 2g-i were subjected to deprotection with trifluoroacetic acid, a large amount of 4AP was obtained.
Overall reaction: Possible mechanism: Furthermore, it can also be concluded that the monoalkylation of the 4AP leads to inactive products and otherwise interesting activity against fungi and some protozoa can be obtained by dual, symmetrical, or nonsymmetrical dialkyla-

General.
Acetonitrile was distilled twice from P 2 O 5 and CaH 2 . Commercially available reagents were used without further purification. The Boc protection of 4-aminopyridine was carried out following the literature [25].  All the experiments were carried out in a divided glass cell separated through a porous glass plug filled up with a layer of gel (i.e., methyl cellulose 0.5% volume dissolved in DMF-Et 4 NPF 6 1.0 mol dm −3 ); Pt spirals (apparent areas 0.8 cm 2 ) were used both as cathode and anode. MeCN-Et 4 NPF 6 0.1 mol dm −3 was used as solvent-supporting electrolyte system (catholyte: 20 cm 3 ; anolyte: 5 cm 3 ). 1 mmol of N-Boc-4-aminopyridine was present in the catholyte. After 145 C were passed, the current was switched off and 1 mmol of alkylating agent was added to the catholyte. The solution was kept under stirring at room temperature for 2 hours; then the solvent was evaporated under reduced pressure and the residue was purified by flash column chromatography, using a mixture of ethyl acetate/light petroleum ether 2/8 in volume, obtaining the pure products.

4-[N-(tert-Butoxycarbonyl)amino]pyridine N-
Flash column chromatography was carried out using Merck 60 kieselgel (230-400 mesh) under pressure. GC-MS measurements were carried out on SE 54 capillary column using a Fisons 8000 gas chromatograph coupled with a Fisons MD 800 quadrupole mass selective detector. 1 H and 13 C NMR spectra were recorded at room temperature using a Bruker AC 200 spectrometer using CDCl 3 as internal standard.

Deprotection of Compounds 1a-j.
To a solution of 1 (1 mmol) in CH 2 Cl 2 (5 cm 3 ), kept at 0 ∘ C, 1 cm 3 of CF 3 COOH was added. This mixture was allowed to stir for 3 h at 0 ∘ C. The solution was then mixed with aqueous sodium carbonate till pH 8 and extracted with ethyl acetate. The solvent

Alkylation of Compounds 2a,c,e.
To a solution of 2 (1 mmol) in anhydrous DMSO (2 cm 3 ), kept at rt under N 2 , 1.5 mmol of t-BuOK was added. This mixture was allowed to stir for 20 min at rt; then 1 mmol of alkyl halide was added and the solution was kept under stirring at rt for 4 h. The solution was then mixed with water and extracted with dichloromethane. The solvent was removed under reduced pressure and the mixture was purified by flash chromatography, yielding pure compound 3.

Dialkylation of 4-Aminopyridine.
To a solution of 4AP (1 mmol) in anhydrous DMSO (2 cm 3 ), kept at rt under N 2 , 2 mmol of t-BuOK was added. This mixture was allowed to stir for 20 min at rt; then 2 mmol of alkyl halide was added and the solution was kept under stirring at rt for 4 h. The solution was then mixed with water and extracted with dichloromethane. The solvent was removed under reduced pressure and the mixture was purified by flash chromatography, yielding pure compound 3.  [26,27].

N,N-Di
Antifungal Susceptibility Assays. In vitro antifungal susceptibility was evaluated using the CLSI broth microdilution methods [26,27]. Fluconazole and Amphotericin B were used as reference drugs. The final concentration ranged from 0.125 to 64 g/mL. The compounds were dissolved previously in DMSO at concentrations 100 times higher than the highest desired test concentration and successively diluted in test medium in accordance with the procedures of the CLSI [28]. Microdilution trays containing 100 L of serial twofold dilutions of compounds in RPMI 1640 medium (Sigma-Aldrich, St. Louis, MO, USA) were inoculated with an organism suspension adjusted to attain a final inoculum concentration of 1.0 × 10 3 -1.5 × 10 3 cells/mL for yeasts and 0.4 × 10 4 -5 × 10 4 CFU/mL for dermatophytes. The panels were incubated at 35 ∘ C and observed for the presence of growth at 48 h (Candida spp.) and 72 h (C. neoformans and dermatophytes).
The minimal inhibitory concentration (MIC) was, for yeasts, the lowest concentration that showed ≥ 50% growth inhibition compared with the growth control and, for dermatophytes, the lowest concentration that showed ≥ 80% growth inhibition compared with the growth control. The MIC 100 was the lowest drug concentration that prevented 100% of growth with respect to the untreated control. According to CSI protocols, the fluconazole MIC 50 and the amphotericin B MIC 100 were calculated (22,23). The results were expressed as the geometric mean (G M) of the MIC values.
All assays were performed in triplicate. Compounds were tested at 5 concentrations (64, 16, 4, 1, and 0.25 g/mL) to establish a full dose titration and determine the IC 50 (inhibitory concentration 50%). The final in-test concentration of DMSO did not exceed 0.5%, which is known not to interfere with the different assays [29].