Deprotometalation-Iodolysis and Direct Iodination of 1-Arylated 7-Azaindoles: Reactivity Studies and Molecule Properties

Five protocols were first compared for the copper-catalyzed C-N bond formation between 7-azaindole and aryl/heteroaryl iodides/bromides. The 1-arylated 7-azaindoles thus obtained were subjected to deprotometalation-iodolysis sequences using lithium 2,2,6,6-tetramethylpiperidide as the base and the corresponding zinc diamide as an in situ trap. The reactivity of the substrate was discussed in light of the calculated atomic charges and the pKa values. The behavior of the 1-arylated 7-azaindoles in direct iodination was then studied, and the results explained by considering the HOMO orbital coefficients and the atomic charges. Finally, some of the iodides generated, generally original, were involved in the N-arylation of indole. While crystallographic data were collected for fifteen of the synthesized compounds, biological properties (antimicrobial, antifungal and antioxidant activity) were evaluated for others.


Introduction
Due to its similarity with indole and purine, 1H-pyrrolo[2,3-b] pyridine (7-azaindole) has aroused the interest of the chemical community [1][2][3][4][5][6][7][8], for example, for medicinal applications, as this motif can be found in molecules with of a broad spectrum of bioactivities. Mention may be made, for example, of compounds used to treat diseases involving the abnormal regulation of enzymes.
Among them, variolin B and certain meriolins are, respectively, good inhibitors of casein and cyclin-dependent kinases and therefore promising for the treatment of brain cancers (Figure 1, left) [9,10]. Also based on a 7-azaindole, GSK 1070916 is a potent inhibitor of Aurora kinases, which are involved in the regulation of mitosis and frequently overexpressed in cancer tumors (Figure 1, right) [11]. 7-Azaindole can also be present in the backbone of organic materials, for example exhibiting luminescence properties. Furthermore, due to the nucleophilicity of pyridine nitrogen, 7-azaindole derivatives can also act as ligands for catalysis [12].
While much work has been devoted to the functionalization of 7-azaindole, relatively little of it deals with the functionalization of the azaindole ring of 1-arylated derivatives. Dong and coworkers published studies in 2015-2016 in which tetracycles were formed from 1-aryl-7-azaindoles and diphenylacetylene by a rhodium(III)-catalyzed annulation reactions [13,14]. In 2016, tetracyclic heterocycles were also built by Ge, Li and coworkers from 1-arylated 7-azaindoles by using as a key step the rhodium(III)-catalyzed C-H oxidative olefination of the aryl group [15]. A year later, Mishra, Kim and coworkers converted a 1-arylated 7-azaindole to azaindoloquinoline by rhodium(III)-catalyzed C-H amination followed by intramolecular cyclization [16]. Metal-free functionalization of 1-arylated 7-azaindoles is also possible, as evidenced by the work of Xu, Dong and coworkers; in this case, another type of tetracyclic heterocycle was synthesized by TsOH-induced tandem [3 + 2] cyclization between 7-azaindoles and pyridotriazoles [17].
Our objective in the present study was to develop methods to introduce an iodine atom either at the 2-or at the 3-position of 1-aryl-7-azaindoles and to predict the outcome of these reactions [18,19] using pKa, atomic charges and HOMO orbital coefficients.
While much work has been devoted to the functionalization of 7-azaindole, relatively little of it deals with the functionalization of the azaindole ring of 1-arylated derivatives. Dong and coworkers published studies in 2015-2016 in which tetracycles were formed from 1-aryl-7-azaindoles and diphenylacetylene by a rhodium(III)-catalyzed annulation reactions [13,14]. In 2016, tetracyclic heterocycles were also built by Ge, Li and coworkers from 1-arylated 7-azaindoles by using as a key step the rhodium(III)-catalyzed C-H oxidative olefination of the aryl group [15]. A year later, Mishra, Kim and coworkers converted a 1-arylated 7-azaindole to azaindoloquinoline by rhodium(III)-catalyzed C-H amination followed by intramolecular cyclization [16]. Metal-free functionalization of 1-arylated 7-azaindoles is also possible, as evidenced by the work of Xu, Dong and coworkers; in this case, another type of tetracyclic heterocycle was synthesized by TsOH-induced tandem [3 + 2] cyclization between 7-azaindoles and pyridotriazoles [17].
1j, 35 1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered.
Concerning the double N-arylation reaction between 7-azaindole and 1,3-diiodobenzene, Method C also provided the expected product 2k with a higher yield (40%) than Method D (15%). However, despite long reaction times of 4-5 days, the product 2k resulting from a single N-arylation is still present (isolated in 20% yield in both cases) (Scheme 1, top). By contrast, when Method C was applied to the reaction between 7-azaindole and 1,4diiodobenzene, the expected product 2l was this time obtained with a high 80% yield (Scheme 1, bottom). The compounds 1e, 1i, 1j, 2k and 2l were unambiguously identified by X-ray diffraction ( Figure 2). Due to steric hindrance, their two or three rings have not been shown to be coplanar. While the torsion angles were at most 25° in the case of 1e (25.1°), 1i (2.3/10.9°), 1j (14.2°) and 2l (25.0° and 25.0°), a higher value was observed for 2k (22.7° and 35.8°). This could explain why lower yields were noticed for 2k compared to 2l. The X-ray diffraction data recorded for the compounds 1e and 1i deserve some additional comments. In the case of 1e, short intermolecular contacts were observed at the solid state between the pyridine nitrogen of azaindole and the hydrogen at C3 (2.644 Å), causing a linear chain, while a fluorine of the trifluoromethyl group is close to the hydrogens of two different azaindoles of another chain, one at C4 (2.651 Å) and the other at C6 (2.625 Å) ( Figure 3A). In the case of 1i, the molecules are arranged in pairs of two parallel azaindoles (separated by about 3.3 Å); these pairs are connected by short contacts between the pyridine nitrogen of the azaindole of one pair and the hydrogen at C3 of the azaindole of another pair (2.609 Å) ( Figure 3B). Overall, the XRD geometries are very close to those predicted by DFT calculations (see Supplementary Materials), including the most stable rotamer form. The X-ray diffraction data recorded for the compounds 1e and 1i deserve some additional comments. In the case of 1e, short intermolecular contacts were observed at the solid state between the pyridine nitrogen of azaindole and the hydrogen at C3 (2.644 Å), causing a linear chain, while a fluorine of the trifluoromethyl group is close to the hydrogens of two different azaindoles of another chain, one at C4 (2.651 Å) and the other at C6 (2.625 Å) ( Figure 3A). In the case of 1i, the molecules are arranged in pairs of two parallel azaindoles (separated by about 3.3 Å); these pairs are connected by short contacts between the pyridine nitrogen of the azaindole of one pair and the hydrogen at C3 of the azaindole of another pair (2.609 Å) ( Figure 3B)

Deprotometalation-Iodolysis of 1-Arylated 7-Azaindoles
Deprotolithiation-trapping sequences at the 2-position of 7-azaindoles have largely been developed after protection of the NH [3]. In 1997, Mérour and coworkers reported the first studies on the topic from the 1-phenylsulfonyl derivative; the key deprotolithiation step was carried out by using lithium diisopropylamide (2 equiv) in tetrahydrofuran (THF) at −25 °C for 0.5 h and was evidenced by subsequent trapping with various electrophiles [34]. However, probably because the sulfonamide function also activates the phenyl group [35], a second deprotonation at the phenyl ring was noticed with electrophiles such as chlorotrimethylsilane and chlorotrimethylstannane which are known for their greater compatibility with hindered lithium amides. In 2007, Kondo and coworkers identified mesityllithium as an alternative base to the lithium amide for similar substrates [36].
One way to deprotometalate the 2-position without protection/deprotection steps is to form a carbamate in situ. In 1999, Curtis and coworkers extended this approach, first developed in the indole series by Katritzky and coworkers [37], to 7-azaindole. They successively treated 7-azaindole with n-butyllithium and carbon dioxide before performing C2-deprotolithiation with tert-butyllithium in THF at −78 °C, subsequent electrophilic trapping (CO2) and acidic treatment; under these conditions, the carboxylic acid was obtained with a correct yield [38].
The 7-azaindoles 1-substituted by methyl, diethylaminomethyl and methoxymethyl groups can also be functionalized at their 2-position after deprotolithiation using tert-butyllithium in THF (addition of the base at −78 °C before warming to 0 °C). This was evidenced in 2008 by O'Shea, Tacke and coworkers who successfully employed 6-(dimethylamino)fulvene as an electrophile [39]. It is worth noting that 2-(trimethylsilyl)ethoxymethyl can also be employed as a protecting group to easily deprotonate the adjacent site [40,41]. As might be expected, when tri(isopropyl)silyl is present on 7-azaindole N1, the 5-membered ring is protected from deprotometalation [42].
In the absence of a substituent at N1, it is possible to reroute the reaction on the pyridine ring by benefiting from an anionic shielding in situ, thanks to an efficient directing group (such as N,N-diethylcarboxamide or N,N-diethylsulfonamide) at the 3-and possibly 4-position. This was demonstrated by Snieckus and coworkers in 2012 [43]. The same group more recently identified N,N-diisopropylcarboxamide as a directing group for the introduction of different substituents at C2 after deprotolithiation using lithium diisopropylamide (2 equiv) in THF at −78 °C. An elegant dance of directed metalation-groups

Deprotometalation-Iodolysis of 1-Arylated 7-Azaindoles
Deprotolithiation-trapping sequences at the 2-position of 7-azaindoles have largely been developed after protection of the NH [3]. In 1997, Mérour and coworkers reported the first studies on the topic from the 1-phenylsulfonyl derivative; the key deprotolithiation step was carried out by using lithium diisopropylamide (2 equiv) in tetrahydrofuran (THF) at −25 • C for 0.5 h and was evidenced by subsequent trapping with various electrophiles [34]. However, probably because the sulfonamide function also activates the phenyl group [35], a second deprotonation at the phenyl ring was noticed with electrophiles such as chlorotrimethylsilane and chlorotrimethylstannane which are known for their greater compatibility with hindered lithium amides. In 2007, Kondo and coworkers identified mesityllithium as an alternative base to the lithium amide for similar substrates [36].
One way to deprotometalate the 2-position without protection/deprotection steps is to form a carbamate in situ. In 1999, Curtis and coworkers extended this approach, first developed in the indole series by Katritzky and coworkers [37], to 7-azaindole. They successively treated 7-azaindole with n-butyllithium and carbon dioxide before performing C2-deprotolithiation with tert-butyllithium in THF at −78 • C, subsequent electrophilic trapping (CO 2 ) and acidic treatment; under these conditions, the carboxylic acid was obtained with a correct yield [38].
The 7-azaindoles 1-substituted by methyl, diethylaminomethyl and methoxymethyl groups can also be functionalized at their 2-position after deprotolithiation using tertbutyllithium in THF (addition of the base at −78 • C before warming to 0 • C). This was evidenced in 2008 by O'Shea, Tacke and coworkers who successfully employed 6-(dimethylamino)fulvene as an electrophile [39]. It is worth noting that 2-(trimethylsilyl)etho xymethyl can also be employed as a protecting group to easily deprotonate the adjacent site [40,41]. As might be expected, when tri(isopropyl)silyl is present on 7-azaindole N1, the 5-membered ring is protected from deprotometalation [42].
In the absence of a substituent at N1, it is possible to reroute the reaction on the pyridine ring by benefiting from an anionic shielding in situ, thanks to an efficient directing group (such as N,N-diethylcarboxamide or N,N-diethylsulfonamide) at the 3-and possibly 4-position. This was demonstrated by Snieckus and coworkers in 2012 [43]. The same group more recently identified N,N-diisopropylcarboxamide as a directing group for the introduction of different substituents at C2 after deprotolithiation using lithium diisopropylamide (2 equiv) in THF at −78 • C. An elegant dance of directed metalation-groups was also developed by the same authors to successively functionalize the 6 (directing group onto the N7 nitrogen) and 2 (directing group onto the N1 nitrogen) positions [44].
To our knowledge, the only 1-aryl-7-azaindole already involved in deprotolithiationtrapping is 1-(2-pyridyl)-7-azaindole. Wang and coworkers introduced a deuterium or a methyl group at its 2-position by reaction with lithium diisopropylamide (2.5 equiv) in diethyl ether at −78 • C for 40 min and subsequent trapping with heavy water or iodomethane, respectively [45]. Our goal in this part is to evaluate the scope of the deprotometalation of 1-aryl-7-azaindoles by using as substrates the compounds 1 described above. We used a common electrophile in order to easily compare the reactivities of our substrates, and we chose iodine since it is possible to take advantage of this heavy halogen to carry out postfunctionalizations such as Suzuki-Miyaura [3,46], Sonogashira [3] or even Heck [3] couplings.
Previous results from the group showed that 1-phenylindole could be efficiently deprotonated at C2 by using a base prepared from ZnCl 2 ·TMEDA (TMEDA = N,N,N ,Ntetramethylethylenediamine) and LiTMP (TMP = 2,2,6,6-tetramethylpiperidino) in a 1:3 ratio [47][48][49]. Indeed, after treatment of the substrate by this formed 1:1 LiTMP-Zn(TMP) 2 mixture of amides [47,50,51] in THF at room temperature (rt) and addition of iodine to intercept the heteroarylmetal obtained, 2-iodo-1-phenylindole was isolated with a yield of 92% [52]. Under these conditions, it is accepted that LiTMP deprotolithiates the heterocycle and that the heteroaryllithium is then trapped by a zinc species through transmetalation [47,50,51]. The calculated hydrogen atomic charges (H charges) and C-H pK a values [52] of 1-phenylindole ( Figure 4A) are in good agreement with the observed results, with both the most positive hydrogen and the most stabilized lithiated compound at C2. was also developed by the same authors to successively functionalize the 6 (directing group onto the N7 nitrogen) and 2 (directing group onto the N1 nitrogen) positions [44]. To our knowledge, the only 1-aryl-7-azaindole already involved in deprotolithiationtrapping is 1-(2-pyridyl)-7-azaindole. Wang and coworkers introduced a deuterium or a methyl group at its 2-position by reaction with lithium diisopropylamide (2.5 equiv) in diethyl ether at −78 °C for 40 min and subsequent trapping with heavy water or iodomethane, respectively [45]. Our goal in this part is to evaluate the scope of the deprotometalation of 1-aryl-7-azaindoles by using as substrates the compounds 1 described above. We used a common electrophile in order to easily compare the reactivities of our substrates, and we chose iodine since it is possible to take advantage of this heavy halogen to carry out postfunctionalizations such as Suzuki-Miyaura [3,46], Sonogashira [3] or even Heck [3] couplings.
Previous results from the group showed that 1-phenylindole could be efficiently deprotonated at C2 by using a base prepared from ZnCl2•TMEDA (TMEDA = N,N,N′,N′tetramethylethylenediamine) and LiTMP (TMP = 2,2,6,6-tetramethylpiperidino) in a 1:3 ratio [47][48][49]. Indeed, after treatment of the substrate by this formed 1:1 LiTMP-Zn(TMP)2 mixture of amides [47,50,51] in THF at room temperature (rt) and addition of iodine to intercept the heteroarylmetal obtained, 2-iodo-1-phenylindole was isolated with a yield of 92% [52]. Under these conditions, it is accepted that LiTMP deprotolithiates the heterocycle and that the heteroaryllithium is then trapped by a zinc species through transmetalation [47,50,51]. The calculated hydrogen atomic charges (H charges) and C-H pKa values [52]   To our knowledge, the acidic properties of 7-azaindoles have not yet been investigated. The most related experimental study is that of Fraser and coworkers [54], who determined the pKa values in THF of five-membered heteroaromatic C-H acids, including N-methylindole. In 2014-2015, we also computed the pKa values of N-arylated pyrroles and indoles [52] as well as 1-arylated benzotriazoles and indazoles [55] by using DFT.
In general (except for C5 and C6 which are reversed), the pKa distribution pattern of 1-aryl-7-azaindoles resembles that of 1-arylindoles [52]. In the case of 1-phenyl-7-azaindole (1a), the most positively charged hydrogens are located at phenyl C2′ (opposite to the pyridine nitrogen in bold) and azaindole C2 (next to the pyrrole nitrogen) while the most stabilized lithiated derivative is also at C2 ( Figure 4B). It is interesting to know which To our knowledge, the acidic properties of 7-azaindoles have not yet been investigated. The most related experimental study is that of Fraser and coworkers [54], who determined the pK a values in THF of five-membered heteroaromatic C-H acids, including N-methylindole. In 2014-2015, we also computed the pK a values of N-arylated pyrroles and indoles [52] as well as 1-arylated benzotriazoles and indazoles [55] by using DFT.
In general (except for C5 and C6 which are reversed), the pK a distribution pattern of 1-aryl-7-azaindoles resembles that of 1-arylindoles [52]. In the case of 1-phenyl-7azaindole (1a), the most positively charged hydrogens are located at phenyl C2 (opposite to the pyridine nitrogen in bold) and azaindole C2 (next to the pyrrole nitrogen) while the most stabilized lithiated derivative is also at C2 ( Figure 4B). It is interesting to know which of the lithiated intermediates is intercepted by Zn(TMP) 2 during such a LiTMPmediated deprotonation.
For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl 2 ·TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pK a value [53]) [56] ( Figure 4C). Table 2. Deprotometalation of the 1-arylated 7-azaindoles followed by iodolysis. of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). of the lithiated intermediates is intercepted by Zn(TMP)2 during such a LiTMP-mediated deprotonation. For this purpose, 1a was treated at rt for 2 h by the basic combination prepared in situ in THF from ZnCl2•TMEDA (1 equiv) and LiTMP (3 equiv), and iodine was added to quench the organometallic intermediate. Under these conditions, the 2-iodinated derivative 3a was isolated with a yield of 75% (Table 2, entry 1). This can result either from direct deprotolithiation at C2 or from deprotonation at another position (directed by the pyridine nitrogen) followed by isomerization, before transmetalation to a zinc species. This situation could parallel that of 2-chloropyridine for which the kinetic lithiated product is at C6 (due to the presence of a neighboring coordinating pyridine nitrogen) and the thermodynamic product at C3 (both highest H charge and lower pKa value [53]) [56] ( Figure  4C). 1 After purification (see experimental part). 2 Low yields were obtained, due to difficult separation. 3 The rest was 1f.
When a substituent such as OMe and especially CF3 is present at the 4-position of the phenyl, the hydrogens at C2′ still compete with those at C2 to be the most positively charged, but this time the pKa values are lowered at the remaining free sites of the phenyl ( Figure 5A,B). Consequently, other aryllithiums could be formed and intercepted by Zn(TMP)2 before isomerization to 2-(7-azaindolyl)lithium. It is therefore no surprise that inseparable mixtures coming from deprotonations at C2 and at the 4-methoxyphenyl were When a substituent such as OMe and especially CF 3 is present at the 4-position of the phenyl, the hydrogens at C2 still compete with those at C2 to be the most positively charged, but this time the pK a values are lowered at the remaining free sites of the phenyl ( Figure 5A,B). Consequently, other aryllithiums could be formed and intercepted by Zn(TMP) 2 before isomerization to 2-(7-azaindolyl)lithium. It is therefore no surprise that inseparable mixtures coming from deprotonations at C2 and at the 4-methoxyphenyl were obtained from 1b under these conditions (not shown). In the case of 1e, both the 2-iodinated derivative 3e and the 2,3 -diiodinated derivative 3e were obtained (in yields of 10 and 20%, respectively) from such a complex mixture (Table 2, entry 2); this could result from a second deprotonation at the phenyl of the 2-(7-azaindolyl)zinc species [57]. By contrast, from 1-(3,5-dimethylphenyl)-7-azaindole (1f), no competitive reaction on the phenyl ring took place, as expected from the calculated pK a values ( Figure 5C); the only 2-iodinated derivative 3f was isolated with a yield of 40% ( Table 2, entry 3). 1 After purification (see experimental part). 2 Low yields were obtained, due to difficult separation. 3 The rest was 1f.
When a substituent such as OMe and especially CF3 is present at the 4-position of the phenyl, the hydrogens at C2′ still compete with those at C2 to be the most positively charged, but this time the pKa values are lowered at the remaining free sites of the phenyl ( Figure 5A,B). Consequently, other aryllithiums could be formed and intercepted by Zn(TMP)2 before isomerization to 2-(7-azaindolyl)lithium. It is therefore no surprise that inseparable mixtures coming from deprotonations at C2 and at the 4-methoxyphenyl were obtained from 1b under these conditions (not shown). In the case of 1e, both the 2-iodinated derivative 3e and the 2,3′-diiodinated derivative 3e′ were obtained (in yields of 10 and 20%, respectively) from such a complex mixture (Table 2, entry 2); this could result from a second deprotonation at the phenyl of the 2-(7-azaindolyl)zinc species [57]. By contrast, from 1-(3,5-dimethylphenyl)-7-azaindole (1f), no competitive reaction on the phenyl ring took place, as expected from the calculated pKa values ( Figure 5C); the only 2-iodinated derivative 3f was isolated with a yield of 40% ( Table 2, entry 3). As indicated by the corresponding pKa values, 2-thienylmetals are easier to obtain due to their greater stability than, for example, phenylmetals ( Figure 6A). Therefore, when subjected to our lithium-zinc basic combination prior to iodolysis, 1g did not lead to the As indicated by the corresponding pK a values, 2-thienylmetals are easier to obtain due to their greater stability than, for example, phenylmetals ( Figure 6A). Therefore, when subjected to our lithium-zinc basic combination prior to iodolysis, 1g did not lead to the 2-iodinated product. Instead, 1-(5-iodo-2-thienyl)-7-azaindole (3g; isolated in 30% yield) and the diiodinated derivative 3g (40% yield) were formed (Table 2, entry 4). Again, due to similar pK a values ( Figure 6B) or a proximity effect [57], a mixture of the three products 3k, 3k and 3k" was obtained when 2k was processed in the same reaction conditions ( 2-iodinated product. Instead, 1-(5-iodo-2-thienyl)-7-azaindole (3g; isolated in 30% yield) and the diiodinated derivative 3g′ (40% yield) were formed ( Table 2, entry 4). Again, due to similar pKa values ( Figure 6B) or a proximity effect [57], a mixture of the three products 3k, 3k′ and 3k″ was obtained when 2k was processed in the same reaction conditions ( Table 2, entry 5). Thus, by calculating the pKa values of these 1-aryl-7-azaindoles, it is quite easy to predict whether the reaction leads to a main iodinated derivative or whether mixtures are expected.
Among the X-ray diffraction data collected to unambiguously assign a structure to the isolated products (Figure 7), a few elements are noticeable. Halogen bonds [58] be- Thus, by calculating the pK a values of these 1-aryl-7-azaindoles, it is quite easy to predict whether the reaction leads to a main iodinated derivative or whether mixtures are expected.
Among the X-ray diffraction data collected to unambiguously assign a structure to the isolated products (Figure 7), a few elements are noticeable. Halogen bonds [58] between pyridine nitrogen and iodine at C2 (2.932 Å) have been identified at the solid state in the case of 3a whereas they do not exist for the monoiodide 3a ( Figure 8A). Moreover, an intramolecular chalcogen bond [59] between pyridine nitrogen and thiophene sulfur (2.962 Å) exists for 3g ( Figure 8B). Finally, short contacts were observed in the case of 3g , between thiophene iodines (3.948 Å), as well as between iodine at C2 and thiophene iodine-bearing carbon (3.633 Å) ( Figure 8C). Thus, by calculating the pKa values of these 1-aryl-7-azaindoles, it is quite easy to predict whether the reaction leads to a main iodinated derivative or whether mixtures are expected.
Among the X-ray diffraction data collected to unambiguously assign a structure to the isolated products (Figure 7), a few elements are noticeable. Halogen bonds [58] between pyridine nitrogen and iodine at C2 (2.932 Å) have been identified at the solid state in the case of 3a′ whereas they do not exist for the monoiodide 3a ( Figure 8A). Moreover, an intramolecular chalcogen bond [59] between pyridine nitrogen and thiophene sulfur (2.962 Å) exists for 3g ( Figure 8B). Finally, short contacts were observed in the case of 3g′, between thiophene iodines (3.948 Å), as well as between iodine at C2 and thiophene iodine-bearing carbon (3.633 Å) ( Figure 8C).

Direct Iodination of 1-Arylated 7-Azaindoles
The incorporation of iodine atoms at the 3-position of 7-azaindole can be used to introduce different aryl or heteroaryl groups [60][61][62][63][64], as well as other functions [8,65]. Our objective in this part is to predict the outcome of this aromatic electrophilic substitution (SEAr) in the case of 1-arylated 7-azaindoles. Indeed, if 7-azaindole can be easily iodinated at its 3-position in DMF containing iodine after treatment by potassium hydroxide [43], the behavior of 1-aryl-7-azaindoles has only been the subject of very few studies.
Liu, Xu and coworkers showed in 2014, during the rhodium-catalyzed chlorination of 7-azaindoles, that 3-iodo-1-phenyl-7-azaindole could be obtained by reacting 1a with N-iodosuccinimide (NIS; 1 equiv) and potassium hydroxide (3 equiv) in acetonitrile at rt for 11 h [66]. Inspired by their protocol, we replaced NIS with iodine and isolated 1a with a yield of 65% (Table 3, entry 1). The present reaction takes place at the carbon site possessing the most negative atomic charge (C charge; Figure 9). Another way to rationalize the regioselectivity of SEAr reactions [67,68] is to use Fukui's concept (an aromatic compound reacts with an electrophile at its carbon having the highest orbital coefficient of HOMO in absolute value) [69] and thus to calculate the HOMO coefficients by applying Hückel's theory [70][71][72] (Figure 9). In the present case, both approaches converge toward a SEAr at C3.

Direct Iodination of 1-Arylated 7-Azaindoles
The incorporation of iodine atoms at the 3-position of 7-azaindole can be used to introduce different aryl or heteroaryl groups [60][61][62][63][64], as well as other functions [8,65]. Our objective in this part is to predict the outcome of this aromatic electrophilic substitution (S E Ar) in the case of 1-arylated 7-azaindoles. Indeed, if 7-azaindole can be easily iodinated at its 3-position in DMF containing iodine after treatment by potassium hydroxide [43], the behavior of 1-aryl-7-azaindoles has only been the subject of very few studies.
Liu, Xu and coworkers showed in 2014, during the rhodium-catalyzed chlorination of 7-azaindoles, that 3-iodo-1-phenyl-7-azaindole could be obtained by reacting 1a with N-iodosuccinimide (NIS; 1 equiv) and potassium hydroxide (3 equiv) in acetonitrile at rt for 11 h [66]. Inspired by their protocol, we replaced NIS with iodine and isolated 1a with a yield of 65% (Table 3, entry 1). The present reaction takes place at the carbon site possessing the most negative atomic charge (C charge; Figure 9). Another way to rationalize the regioselectivity of S E Ar reactions [67,68] is to use Fukui's concept (an aromatic compound reacts with an electrophile at its carbon having the highest orbital coefficient of HOMO in absolute value) [69] and thus to calculate the HOMO coefficients by applying Hückel's theory [70][71][72] (Figure 9). In the present case, both approaches converge toward a S E Ar at C3.        1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated  1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated 3g, 35

10
Molecules 2021, 26 1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated 1h 1.  Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated 4h, 36 4h, 45 11 12 Molecules 2021, 26 1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated  1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated 4i, 17 4i, 51 1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered.
Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated in yields ranging from 40% to 62% (Table 3, entries 2, 3, 4 and 7). It is interesting to note that the yield of the 3-iodinated product can be slightly improved by carrying out the reaction at 40 • C with an excess of iodine, as for example noticed in 1e (Table 3, entries 5 and 6).  1 After purification (see experimental part); when low yields were recorded, starting materials were in general recovered. Figure 9. Calculated C charges (in brackets) and HOMO coefficients obtained by using the HuLiS calculator [73] for 7-azaindole, 1-phenylindole and 1-phenyl-7-azaindole (1a).
Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated in yields ranging from 40% to 62% (Table 3, entries 2, 3, 4 and 7). It is interesting to note that the yield of the 3-iodinated product can be slightly improved by carrying out the reaction at 40 °C with an excess of iodine, as for example noticed in 1e (Table 3, entries 5 and 6).   . Calculated C charges (in brackets) and HOMO coefficients obtained by using the HuLiS calculator [73] for 7-azaindole, 1-phenylindole and 1-phenyl-7-azaindole (1a).
Next, we determined the charges and HOMO coefficients for the 1-arylated 7-azaindoles to attempt a prediction of the experimental results. For the 7-azaindoles bearing a substituted phenyl group 1b-1f, all the maximum HOMO coefficients and the most negative C charges were found at C3 (Figure 10). Experimentally, the 3-iodinated derivatives 4b, 4c, 4d and 4f were indeed the only products formed, as expected; they were isolated in yields ranging from 40% to 62% (Table 3, entries 2, 3, 4 and 7). It is interesting to note that the yield of the 3-iodinated product can be slightly improved by carrying out the reaction at 40 °C with an excess of iodine, as for example noticed in 1e (Table 3, entries 5 and 6).  Since thiophene is also a five-membered heteroaromatic prone to S E Ar, it was interesting to consider the iodination of 1-(2-thienyl)-7-azaindole (1g). Indeed, the calculations carried out as before showed a maximum HOMO coefficient and a most negative C charge at the 5-position of the thienyl ring ( Figure 10). Experimentally, we obtained the 5-iodinated derivative 3g (Table 3, entry 8), which is the compound already formed by deprotometalation-iodolysis ( Table 2, entry 4). With regard to 1-pyridyl-7-azaindoles 1h-1j, direct iodination at C3 is expected ( Figure 10). As assumed, by carrying out the reaction from the 3-and 2-pyridyl substrates, we observed the formation of the 3-iodinated derivatives 4h and 4i as the only reaction products (Table 3, entries 9-12). Therefore, it appears that HOMO coefficients and carbon atomic charges can be used to predict the regioselectivity of S E Ar iodination reactions.
Throughout this study, the regioselectivity was established by NMR and confirmed for the products 4c, 4e, 4h and 4i by X-ray diffraction ( Figure 11). For 4c, short intermolecular contacts were observed at the solid state between the nitrogen of the azaindole pyridine and the hydrogen at C2 (2.734 Å), at the origin of a linear chain, while these chains are linked by short chlorine-iodine contacts (3.569 Å) ( Figure 12A). The molecular networks of 4e, 4h and 4i all exhibit intermolecular halogen bonds [58] in which the iodine atoms are connected to the pyridine nitrogens. For 4e ( Figure 12B) and 4h ( Figure 12C), these weak interactions bind the heavy halogen and the nitrogen of the azaindole pyridine and thus create linear chains (with iodine-nitrogen distances alternating between 3.137 and 3.158 Å for the first and 3.497 Å for the second). In the case of 4i, the iodine is instead linked to the 2-pyridyl attached to the azaindole core, this time establishing a zig-zag chain (iodine-nitrogen distance of 3.283 Å) ( Figure 12D). networks of 4e, 4h and 4i all exhibit intermolecular halogen bonds [58] in which the iodine atoms are connected to the pyridine nitrogens. For 4e ( Figure 12B) and 4h (Figure 12C), these weak interactions bind the heavy halogen and the nitrogen of the azaindole pyridine and thus create linear chains (with iodine-nitrogen distances alternating between 3.137 and 3.158 Å for the first and 3.497 Å for the second). In the case of 4i, the iodine is instead linked to the 2-pyridyl attached to the azaindole core, this time establishing a zig-zag chain (iodine-nitrogen distance of 3.283 Å) ( Figure 12D). Figure 11. ORTEP diagrams (30% probability) of 4c, 4e, 4h and 4i. Figure 11. ORTEP diagrams (30% probability) of 4c, 4e, 4h and 4i. chains are linked by short chlorine-iodine contacts (3.569 Å) ( Figure 12A). The molecular networks of 4e, 4h and 4i all exhibit intermolecular halogen bonds [58] in which the iodine atoms are connected to the pyridine nitrogens. For 4e ( Figure 12B) and 4h (Figure 12C), these weak interactions bind the heavy halogen and the nitrogen of the azaindole pyridine and thus create linear chains (with iodine-nitrogen distances alternating between 3.137 and 3.158 Å for the first and 3.497 Å for the second). In the case of 4i, the iodine is instead linked to the 2-pyridyl attached to the azaindole core, this time establishing a zig-zag chain (iodine-nitrogen distance of 3.283 Å) ( Figure 12D).

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides.

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2).
Copper-catalyzed double N-arylation reactions are often difficult to perform due to competitive deiodination under the conditions required for second coupling [77]. However, despite a complex reaction mixture, the expected biscoupled product 6g was here isolated with a moderate yield of 20% (entry 3). Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides.

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides.

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides.

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides. (1) C (24 h) Figure 12. Short-contact network observed for 4c (A); halogen bond networks observed for 4e (B), 4h (C) and 4i (D).

N-Arylation of Indole by a Few of the Prepared Iodides
Because the N-arylation of indole with such iodinated 7-azaindoles can lead to original molecules with properties for potential applications [74][75][76], we selected the iodides 4e and 4i as well as the diiodide 3g′ for this purpose ( Table 4). As already generally observed in Table 1, Method C has proved to be superior to Method D; in fact, while the trifluoromethylated product 5e was only isolated with a yield of 30% (entry 1), that containing pyridine 5i was obtained with a yield greater than 50% (entry 2). Table 4. N-arylation of indole by a few of the prepared iodides. 1 After purification (see experimental part). 2 The rest was mainly recovered starting material. 3 A low yield was obtained, due to difficult separation.
Copper-catalyzed double N-arylation reactions are often difficult to perform due to competitive deiodination under the conditions required for second coupling [77]. However, despite a complex reaction mixture, the expected biscoupled product 6g′ was here isolated with a moderate yield of 20% (entry 3).

Biological Evaluation
As 7-azaindole is present in bioactive molecules, some of the compounds synthesized have been screened for their biological properties [77][78][79]. The compounds 1a-c, 1e, 1f,  1h-j, 2k, 2l, 3a, 3g, 3g′ and 5i were evaluated for their antimicrobial activity against 1 After purification (see experimental part). 2 The rest was mainly recovered starting material. 3 A low yield was obtained, due to difficult separation.
Copper-catalyzed double N-arylation reactions are often difficult to perform due to competitive deiodination under the conditions required for second coupling [77]. However, despite a complex reaction mixture, the expected biscoupled product 6g′ was here isolated with a moderate yield of 20% (entry 3).
In the case of 5i, crystals suitable for X-ray diffraction ( Figure 13A) made it possible to detect several short intermolecular contacts at the solid state. First, 2-pyridyl nitrogens are close to azaindole H6 (2.638 Å), the origin of linear chains. The 2-pyridyl ring is almost coplanar with the azaindole backbone (6.37 • ), with its nitrogen positioned outward, while the indole ring makes a 50.6 • twist angle, presumably to reduce the steric hindrance between the two bicyclic cores. These linear chains are in parallel plans successively separated by 3.368 Å (azaindole C4-pyridine C2"; see Table 4, entry 2 for numbering) and 3.399 Å (azaindole C5-pyridine C2"). Finally, short distances can be found between pyridine H6" and both indole C7 (2.795 Å) and Ca (2.878 Å), and between pyridine H3" and indole C2 (2.866 Å) ( Figure 13B). 1 After purification (see experimental part). 2 The rest was mainly recovered starting material. 3 A low yield was obtained, due to difficult separation.
Copper-catalyzed double N-arylation reactions are often difficult to perform due to competitive deiodination under the conditions required for second coupling [77]. However, despite a complex reaction mixture, the expected biscoupled product 6g′ was here isolated with a moderate yield of 20% (entry 3).

Biological Evaluation
As 7-azaindole is present in bioactive molecules, some of the compounds synthesized have been screened for their biological properties [77][78][79]. The compounds 1a-c, 1e, 1f,  1h-j, 2k, 2l, 3a, 3g, 3g and 5i were evaluated for their antimicrobial activity against bacteria and for their antifungal activity (Table 5). No clear effect on microbial growth of strains of E. coli, P. aeruginosa and S. aureus was detected . For 1i and 1j, an effect on the growth of E. faecium was noticed. However, the most significant growth inhibitions were found for L. monocytogenes (compound 1j) and C. dubliniensis (compounds 1a, 1b, 1h and 1i).
500 µg 2 0 0 0 0 0 0 -Reference compound 28 5 28 5 18 6 24 6 30 7 10 8 - 1 The diameters of zones of inhibition are given in mm. 2 10 µL/well. 3  The antioxidant properties of a few selected compounds were finally evaluated. As shown in Table 6, the compounds 1b, 1c, 1f, 1i, 1j and 5i have about 50% activity. The hemolytic activity of 1b, 1f, 1i and 1j, which were found active against Candida dubliniensis or Listeria monocytogenes, was also evaluated and found to be less than 15% at a concentration of 5 µg/µL. This low toxicity on human red blood cells, for compounds exhibiting an antimicrobial effect, demonstrates a specific antibacterial or antifungal effect and thus highlights a possible therapeutic interest. Concerning 1a, 1h and 3a, all active against Candida dubliniensis, the hemolytic activity tested at a concentration of 10 µg/µL was found to be approximately 22%, 65% and 32%, respectively (data not shown).

General Information
Column chromatography separations were achieved on silica gel (40-63 µm). Melting points were measured on a Kofler apparatus. InfraRed (IR) spectra were taken on an ATR Spectrum 100 spectrometer (Perkin-Elmer, Waltham, MA, USA) and the main absorption wavenumbers are given in cm −1 . 1 H and 13 C nuclear magnetic resonance (NMR) spectra were recorded on an Avance III spectrometer (291 K) at 300 and 75 MHz, respectively (Bruker, Billevica, MA, USA). 1 H chemical shifts (δ) are given in ppm relative to the solvent residual peak, and 13 C chemical shifts are relative to the central peak of the solvent signal [80]. ZnCl 2 ·TMEDA was prepared as reported previously [81].

Computational Details
The DFT calculations were performed by using GAUSSIAN 09 package [88]. The B3LYP hybrid functional was employed. All optimized geometries were obtained by using the 6-31G(d) basis set without any symmetry constraints. The vibrational frequencies were calculated at the same level of theory in order to characterize stationary points and calculate zero-point vibrational energies (ZPVE) and thermal corrections. The total energy of species was found by using the 6-311 + G(d,p) basis set. Further, the gas-phase Gibbs energies (G 0 298 ) were calculated by using Equation (1), as follows: The gas-phase acidity ∆G acid was defined as the Gibbs energy of deprotonation of the corresponding substrate R-H (R-H(g) → R − (g) + H + (g)): The solvent effect was simulated within the polarized continuum model (PCM) with the default parameters for THF [89]. The PCM energies were calculated at the B3LYP/6-311 + G(d,p) level by using geometries optimized for isolated structures.
The following homodesmic reaction was composed for the pK a values calculation: R-H(s) + Het − (s) → R − (s) + Het-H(s) where Het-H is N-methylindole. The latter was chosen as reference compound due to its structural similarity and since its pK a (THF) = 38.1 reported by Fraser et al. [54] was expected to be close to the substrates under consideration. Consequently, the Gibbs energy of the homodesmic reaction (∆G r,s ) and the pK a value are related by the following equation: The atomic charges were calculated by using Mulliken population analysis. MO coefficients were generated by using the HuLiS calculator [73]. This was adapted from a reported protocol [27]. A mixture of 7-azaindole (0.18 g, 1.5 mmol), aryl iodide (1.0 mmol) or diiodide (0.50 mmol), Cu (13 mg, 0.20 mmol), Cs 2 CO 3 (0.65 g, 2.0 mmol) in acetonitrile (1 mL) was heated at reflux under Ar (the reaction time is given in the product description). The reaction mixture was cooled to rt before addition of EtOAc (20 mL) and filtration. The solvent was removed under reduced pressure, and the crude was purified by chromatography over silica gel (the eluent is given in the product description).

General Procedure 2 Using Copper(I) Iodide without Ligand
This was adapted from a reported protocol [28]. A mixture of 7-azaindole (0.12 g, 1.0 mmol), aryl iodide (1.1 mmol), CuI (19 mg, 0.10 mmol), K 2 CO 3 (0.41 g, 3.0 mmol) and LiCl (42 mg, 1.0 mmol) in DMF (1 mL) was heated at 120 • C for 24 h under Ar. The reaction mixture was cooled to rt before addition of an aqueous saturated solution of NH 4 Cl (20 mL). Extraction with EtOAc (3 × 20 mL), drying over MgSO 4 , removal of the solvent under reduced pressure, and purification of the crude over silica gel (the eluent is given in the product description) gave the product.

General Procedure 3 Using Copper(I) Iodide with Ligand
This was adapted from reported protocols [29]. A mixture of 7-azaindole (0.12 g, 1.0 mmol), aryl iodide (1.2 mmol) or diiodide (0.60 mmol), CuI (9.5 mg, 50 µmol), K 3 PO 4 (0.42 g, 2.0 mmol) and DMEDA (11 µL, 0.10 mmol) in DMF (1 mL) was degassed and heated at reflux under Ar (the reaction time is given in the product description). The reaction mixture was cooled to rt. The residue was taken with EtOAc (20 mL) and filtrated over celite. Removal of the solvent under reduced pressure and purification of the crude over silica gel (the eluent is given in the product description) gave the product.

General Procedure 4 Using Copper(I) Oxide
This was adapted from a reported protocol [30]. A mixture of 7-azaindole (0.24 g, 2.0 mmol), aryl iodide (1.0 mmol) or diiodide (0.50 mmol), Cu 2 O (14 mg, 0.10 mmol) and Cs 2 CO 3 (0.65 g, 2.0 mmol) in DMSO (1 mL) was heated at 110 • C under Ar (the reaction time is given in the product description). The reaction mixture was cooled to rt. The residue was taken with EtOAc (20 mL) and filtrated over celite. Removal of the solvent under reduced pressure and purification of the crude over silica gel (the eluent is given in the product description) gave the product.

General Procedure 5 Using Copper(I) Iodide and Microwaves
This was adapted from a reported protocol [31]. A mixture of 7-azaindole (0.12 g, 1.0 mmol), aryl iodide (1.1 mmol), CuI (19 mg, 0.10 mmol) and Cs 2 CO 3 (0.33 g, 1.0 mmol) in DMF (1 mL) was heated in a Whirpool M571 domestic microwave oven at 350 W (the reaction time is given in the product description). The reaction mixture was cooled to rt before addition of an aqueous saturated solution of NH 4 Cl (20 mL). Extraction with EtOAc (3 × 20 mL), drying over MgSO 4 , removal of the solvent under reduced pressure, and purification of the crude over silica gel (the eluent is given in the product description) gave the product.