Logo Beilstein Institut

Regioselective alkylation of a versatile indazole: Electrophile scope and mechanistic insights from density functional theory calculations

  1. ‡,1 ,
  2. ‡,1 ,
  3. 1 ORCID Logo ,
  4. 1,§ ORCID Logo ,
  5. 2 and
  6. 1,¶ ORCID Logo
1Department of Discovery Chemistry, BioCryst Pharmaceuticals Inc., Discovery Center of Excellence, 2100 Riverchase Center Building 200, Suite 200 Birmingham, AL, 35244, USA
2Department of Computational Chemistry and Structural Biology, BioCryst Pharmaceuticals Inc., Discovery Center of Excellence, 2100 Riverchase Center Building 200, Suite 200 Birmingham, AL, 35244, USA
  1. Corresponding author email
  2. ‡ Equal contributors
§ Phone: 205-403-4675
¶ Phone: 205-444-4628
Associate Editor: P. Schreiner
Beilstein J. Org. Chem. 2024, 20, 1940–1954. https://doi.org/10.3762/bjoc.20.170
Received 15 Mar 2024, Accepted 26 Jul 2024, Published 09 Aug 2024
Full Research Paper
cc by logo
PDF
Album
Supp Info
Cite

Abstract

Herein, we report a pair of regioselective N1- and N2-alkylations of a versatile indazole, methyl 5-bromo-1H-indazole-3-carboxylate (6) and the use of density functional theory (DFT) to evaluate their mechanisms. Over thirty N1- and N2-alkylated products were isolated in over 90% yield regardless of the conditions. DFT calculations suggest a chelation mechanism produces the N1-substituted products when cesium is present and other non-covalent interactions (NCIs) drive the N2-product formation. Methyl 1H-indazole-7-carboxylate (18) and 1H-indazole-3-carbonitrile (21) were also subjected to the reaction conditions and their mechanisms were evaluated. The N1- and N2-partial charges and Fukui indices were calculated for compounds 6, 18, and 21 via natural bond orbital (NBO) analyses which further support the suggested reaction pathways.

Keywords: DFT; indazole; indazole substitution; mechanism; N1-substituted indazole; N2-substituted indazole; regioselectivity

Introduction

Indazoles constitute an important class of heterocycles with interesting biological and medicinal properties. Indazole, also called benzpyrazole, is a heterocyclic organic compound commonly found as a structural motif in natural products, pharmaceuticals, agrochemicals, and bioactive compounds [1-6]. Indazole-containing compounds possess a wide range of pharmacological activities, such as anti-inflammatory, anti-arrhythmic, antitumor, antifungal, antibacterial, and anti-HIV activities [7-13]. For example, two N1-substituted bioactive indazoles are found in Figure 1, danicopan (1), a complement factor D inhibitor for the treatment of paroxysmal nocturnal hemoglobinuria, and CPI-637 (2), an inhibitor of both cyclic-AMP response element binding protein (CBP) and adenoviral E1A binding protein [14-16]. The N2-substituted indazole analogs pazopanib (3), an FDA-approved tyrosine kinase inhibitor used for the treatment of renal cell carcinoma, and Takeda’s MCHR1 antagonist 4 further exemplify indazole’s biological importance. Generally, direct alkylation of 1H-indazoles leads to a mixture of N1- and N2-substituted products [17-20]. Procedures that selectively produce either N1- or N2-substituted indazoles would provide greater synthetic utility for this valuable heterocycle. These examples suggest a common intermediate such as methyl 5-bromo-1H-indazole-3-carboxylate (6) could be used to generate such compounds. An alkylation strategy that uses the vast array of commercially available alcohols as precursors of alkylating reagents presents an opportunity to find new ways to diversify indazole reactivity by simple modifications of reaction conditions. Mechanistic understanding would allow for further diversification in related systems.

[1860-5397-20-170-1]

Figure 1: Indazole-containing bioactive molecules.

The indazole ring presents annular tautomerism regarding the position of the NH hydrogen atom: 1H-indazole (5a, benzenoid 1H-indazole tautomer) and 2H-indazole (5b, quinonoid 2H-indazole tautomer) (Figure 2) [21-23]. Since 1H-indazole is thermodynamically more stable than 2H-indazole, 5a is the predominant tautomer [24-26]. Conventionally, indazoles are employed as nucleophiles in chemical transformations, and a mixture of both N1- and N2-alkylated products is formed, depending on the reaction conditions, with little selectivity in regards to substituent effects [27-33].

[1860-5397-20-170-2]

Figure 2: Tautomerism of indazole.

Considering the importance of indazoles as a widely used pharmacophore in medicinal chemistry and the challenges in obtaining either N1- or N2-alkylated indazoles as the dominant regioisomer [30,34-39], we were interested in exploring the regioselectivity with methyl 5-bromo-1H-indazole-3-carboxylate (6, Figure 1) as a multifunctional model compound for N1/N2 discrimination studies. The existing approaches for generating N-substituted indazoles of compound 6 often lead to a mixture of N1- and N2-alkylated indazoles with either low selectivity or moderate yields depending on the conditions used. For example, Takahashi et al. obtained N1- and N2-substituted indazole analogs in 44% and 40% yields, respectively, by treating compound 6 with methyl iodide and potassium carbonate in dimethylformamide (DMF) at room temperature for 17 h [40]. Other works have shown poor selectivity when 6 and other isomers similar to 6 were reacted with isopropyl iodide and potassium carbonate, isopropyl bromide and cesium carbonate, and bromocyclohexane with potassium carbonate, and only afforded yields not higher than 52% in various solvents [41-43].

Recently, Alam and Keeting [37] explored the regioselectivity in the alkylation of variously substituted indazoles similar to 6. They observed high N1-selectivity using NaH in THF with pentyl bromide and electron-deficient indazoles, postulating a coordination of the indazole N2-atom and an electron-rich oxygen atom in a C-3 substituent with the Na+ cation from NaH. Under anhydrous conditions the yields ranged from 44% at room temperature to 89% when warmed to 50 °C. No information was provided to justify any N2-selectivity or the lack thereof. Should an N2–Cs+–O ion pair exist, this could reasonably account for all the reported results presented herein (vide infra). Additionally, using Cs2CO3 in dioxane provided no products at room temperature (see Table 1) presumably due to the low solubility of Cs2CO3 in dioxane. They also provided a single example of a Mitsunobu reaction utilizing n-pentanol, dibutyl azodicarboxylate (DBAD), and PPh3.

Table 1: Representative reactions from reference [37].

Electrophile Solvent Reagents Temp
(°C) 		Time
(h) 		Major product isolated yield N1:N2
n-C5H11Br 1,4-dioxane Cs2CO3 rt 16 0% n.a.
n-C5H11Br THF NaH 0 → 50 24 89% >99:1
n-C5H11OH THF DBAD, PPh3 0 → rt 2 58% 1:2.9

Therefore, there is still a great need to develop an operationally simple and mild method to selectively generate N1- or N2-substituted indazole analogs when the substituents appear to favor one over the other. Ideally, it would be greatly beneficial if the desired high regioselectivity on N1 or N2 could be achieved when commercially available chemicals, such as alcohols, react with 6 under different reaction conditions. In this paper, we report a concise and efficient approach to prepare N1- and N2-substituted analogs with high selectivity and excellent yields (>84%) from the same substrates: alcohols and 5-bromo-1H-indazole-3-carboxylate (6).

Results and Discussion

We commenced our studies by investigating yields of N1- and N2-substituted products of conventional indazole alkylation reactions using our model substrate methyl 5-bromo-1H-indazole-3-carboxylate (6) and confirmed structures of the corresponding N1-substituted and N2-substituted products. In Scheme 1 compound 6 was treated with isopropyl iodide (7) in DMF in the presence of sodium hydride to provide products 8 and 9 in 38% and 46% yields, respectively. The structures of both compounds were unambiguously assigned using X-ray crystallography and 1H and nuclear Overhauser effect (NOE) NMR spectroscopy (see Supporting Information File 1).

[1860-5397-20-170-i1]

Scheme 1: NMR, NOE, and yield data of compounds 8 and 9.

From the yield of products 8 and 9, we decided to explore new reaction conditions to improve the yields of the N1-substituted indazole analogs. As shown in Scheme 2, compound 12 was prepared by treating ethanol (10) with tosyl chloride (11) in the presence of 4-dimethylaminopyridine (DMAP) and triethylamine. Used as a model system, the subsequent reaction of sulfonate 12 with compound 6 under varied conditions afforded products P1 and P2. We investigated effects of reagent stoichiometry, bases, reaction time, and temperature on the yields of product P1, as summarized in Table S1 in Supporting Information File 1. Stoichiometric manipulation of 12 to 6 in DMF at 90 °C provided the N1-substituted product P1 in 52–60% yields. The yields of P1 formation were largely unaffected in DMF with temperatures ranging from room temperature to 110 °C. Varying the equivalents of Cs2CO3 showed little effect (averaging 52% yield), however, using NaH and lowering the temperature to rt lowered the P1 yield, averaging 32%. Finally, the effect of other solvents at 90 °C was investigated and the results are summarized in Table 2. Entry 1 shows the best conditions for the above reaction in DMF. The use of chlorobenzene slightly improved the yield to 66%. The reaction in dioxane at 90 °C, entry 6, had a 96% yield. This was a surprising outcome as Keeting observed no reaction in dioxane at rt, suggesting the concentration of Cs2CO3 is significantly increased at this temperature.

[1860-5397-20-170-i2]

Scheme 2: Synthesis of compounds P1 and P2.

Table 2: Effect of solvent on indazole N1 yield.a

Entry Solvent P1 (isolated yield, %)
1 DMF 60
2 DMSO 54
3 NMP 42
4 chlorobenzene 66
5 toluene 56
6 dioxane 96

aDMF: dimethylformamide; DMSO: dimethyl sulfoxide; NMP: N-methyl-2-pyrrolidone. Reaction conditions: 1.5 equiv 12, 1.0 equiv 6, 2.0 equiv Cs2CO3, 90 °C, 2 h.

With the promising yield results of P1, we next explored the scope of this transformation using a variety of alcohols (13a–q, Table 3) and report their regioselectivity as determined by crude LC–MS. Sulfonates 14a–q were prepared as described above or purchased (see Supporting Information File 1). The subsequent reactions with compound 6 afforded the N1-substituted indazole analogs 15a–q with excellent yields (>90%), except for 15m, which failed to form after multiple attempts likely due to an instability of the electrophile 14m under optimized conditions (conditions A: 1.5 equiv tosylate, 1.0 equiv 6, 2.0 equiv Cs2CO3, 90 °C, 2 h). Compounds containing linear or branched alkyl substitutions (15a–g), varied sizes of cycloalkane or saturated heterocycles (15h–p), including S- and R-tetrahydrofuran substitutions (15j, 15k) were isolated in excellent yields (>90%). Azetane 14m was unreactive towards alkylation in the presence of Cs2CO3.

Table 3: Scope of transformation and regioselectivity.

[Graphic 1]
R Conditions A, major product Isolated yield
15a–q (%) 		N1/N2
(LC–MS) 		Conditions B, major product Isolated yield
16a–q (%) 		N2/N1
(LC–MS)
CH3 15a 90 7.5 16a 92 4.0
[Graphic 2] 15b 96 12.5 16b 93 2.0
[Graphic 3] 15c 90 7.5 16c 91 8.6
[Graphic 4] 15d 94 13.1 16d 92 3.9
[Graphic 5] 15e 95 9.5 16e 97 8.2
[Graphic 6] 15f 96 13.8 16f 93 5.1
[Graphic 7] 15g 96 12.6 16g 95 3.4
[Graphic 8] 15h 94 7.7 16h 90 9.6
[Graphic 9] 15i 96 13.4 16i 95 12.1
[Graphic 10] 15j 94 14.8 16j 91 8.0
[Graphic 11] 15k 96 12.4 16k 97 3.9
[Graphic 12] 15l 91 13.9 16l 90 4.0
[Graphic 13] 15m n.r. – 16m 93 –
[Graphic 14] 15n 94 >99 16n 96 8.7
[Graphic 15] 15o 94 10.2 16o 98 5.6
[Graphic 16] 15p 95 15.8 16p 95 4.9
[Graphic 17] 15q 96 20.2 16q 91 2.2

To explore the possibility of N2-selectivity, we hypothesized that the phosphine intermediate of a Mitsunobu reaction could provide chelation control, directing alkylation to the indazole N2-atom while using identical alcohols as described above. Thus, we subjected 6 to simple and mild Mitsunobu conditions for the preparation of N2-substituted indazole analogs 16a–q. By directly reacting compound 6 with alcohols 13a–q (2 equiv), diethyl azodicarboxylate (DEAD, 2 equiv), and triphenylphosphine (TPP, 2 equiv) in THF at 50 °C (conditions B), the corresponding N2-substituted products were isolated in excellent yields (>90%) and high regiocontrol.

Crude product ratios as determined by LC–MS (averaging integrations at 254 nm and 260 nm) had an average error (standard deviation) of 3.2% (2.76%) and 13.7% (7.54%) for conditions A and B, respectively. The N1-isomer overlapped with OPPh3 contributing to the increased error and standard deviation. Compound 6 was completely consumed and not detected (see Supporting Information File 1).

Mechanistic considerations

Alam and Keeting proposed a deprotonated intermediate that utilized the indazole N2 and C=O from an ester substituent at C-3 as a bidentate ligand to the Na+ cation from NaH. The tight ion pair would direct alkylation under conditions A to N1. As this and other postulations exist, we explored the possible mechanisms of each reaction conditions computationally. All calculations were performed in implicit THF at the reaction condition temperature using Gaussian 16: SMD(THF)-PBE0/def2-TZVP // SMD(THF)-PBE0/def2-SVP, def2-TZVP(Cs) at 50 °C (w/MeOPPh3+) or 90 °C (w/Cs+), utilizing Goodvibes to calculate thermochemistry. The energy of the N1- and N2-tautomers of 6 differ by 3.1 kcal/mol at 50 °C, favoring the N1-tautomer, implying a 7:1 distribution of isomers in solution. Concerning conditions A, compound 6 + Cs2CO3 was found to favor the deprotonated indazole with a free Cs+ ion by 8.6 kcal/mol (see Figure 3), however, when all ions are discrete (2 × Cs+ and CO32−) the reaction becomes endergonic by 6.9 kcal/mol, presumably due to entropic penalties. Three of four computed resonance forms were all found to be of approximately equal energy. Only the E-enolate form 6 (-N1H-E) was slightly higher in energy by 0.06 kcal/mol likely due to electrostatic destabilization of the oxyanion with N2, however, this difference is negligible. These data suggest that deprotonation occurs prior to alkylation and that deprotonation of either indazole tautomer leads to anions of identical or highly similar energy. Furthermore, as seen in Figure 4, a total, five coordinated complexes were found to be at least 4.5 kcal/mol more stable than the uncoordinated anion when calculated as isolated structures. When the calculation is performed as a reaction of E and Z-enolates with Cs+ ion, two coordinated complexes 6(N-H)NNCs-E and-6(N-H)NOCs-Z are exergonically formed by 9.7 and 10.9 kcal/mol, respectively.

[1860-5397-20-170-3]

Figure 3: DFT-calculated deprotonation of 6 with Cs2CO3 in implicit THF with the temperature of the calculation set to 90 °C to simulate the dioxane conditions (top) and energy differences of four enolate resonance structures of 6 calculated as discrete structures. The hybrid is identified as 6(N-H) (bottom).

[1860-5397-20-170-4]

Figure 4: DFT-calculated Cs+-coordinated complexes with different enolate forms of 6(N-H) calculated as isolated compounds (top) and calculated intermediates of the reactions of 6(-N1H-Z) and 6(-N1H-E) with Cs+ (bottom).

We then searched for transition state (TS) structures that would produce both the N1- and N2-products from CH3OTs as a model system. When the Boltzmann average of the cesium-coordinated intermediates is calculated, a 3:1 ratio of 6(N-H)NOCs-Z:6(N-H)NNCs-E is found. This average is 10.6 kcal/mol more stable than 6(N-H), and was subsequently set to 0 kcal/mol leading to the energy diagram in Figure 5. Two TSs leading to each product were found, all four of which utilized a coordinating Cs+ cation. The N1-s-cis and N1-s-trans TSs were the lowest in energy (27.5 kcal/mol and 29.1 kcal/mol, respectively), leading to two conformations of the N1-product with highly similar energy (averaging −16.8 kcal/mol). The N2-s-cis and N2-s-trans TSs leading to the N2-product were higher in energy and led to the higher energy N2 products. The critical difference between N1-s-cis and N2-s-cis is the presence of the N2–Cs+–O non-covalent interaction (NCI) in N1-s-cis, which accounts for the 2.1 kcal/mol difference in energy. Calculations showed that the sulfonate oxygens also chelate the cesium ion in both TSs. Thus, nitrogen NCIs with cesium, or lack thereof, seem to drive N1-product formation, which is both kinetically and thermodynamically favorable under conditions A.

[1860-5397-20-170-5]

Figure 5: DFT-calculated reaction coordinate diagram for the reaction of 6 under conditions A. Concerning conditions B, we began our calculations under the assumption that MeOPPh3+ was already present in THF at 50 °C. The deprotonation by the dimethyl azodicarboxylate (DMAD) anion (DMAD−) to form 6(N-H) and 17 was found favorable by 14.7 kcal/mol (Figure 6).

[1860-5397-20-170-6]

Figure 6: DFT-calculated energy for the deprotonation of 6 by the DMAD anion.

Postulating that O or N-dative interactions with phosphorus were responsible for the high N2-selectivity in an analogous fashion to conditions A, we searched for intermediates and TSs that included this possibility. No O–P or N2–P-coordinated intermediates were found. The coordinated intermediate N1-P was found considerably endergonic with a ΔG of +8.0 kcal/mol compared to 6(N-H) (see Figure 7). A synchronous TS (N1-P-TS) leading to the N2-product was found starting from N1-P; however, the reaction barrier was 52.1 kcal/mol and thus a highly unlikely pathway. We again searched for TSs that led to both the N1- and N2-products but lacked any dative preorganization. However, under these reaction conditions, we found that the TS leading to the N2-product, N2-s-cis, was lower in energy than its N1-analog, N1-s-trans, by 1.1 kcal/mol (see Figure 8). This energy difference appears to be driven by stabilizing non-covalent interactions. Specifically, the carbonyl O in N2-s-cis shows NCIs with one of the benzene rings of PPh3 as well as a hydrogen bond-like NCI with a H-atom of the electrophilic methyl. Thus, the partitioning between transition states favor the N2-pathway over the N1-pathway by a product ratio of 4.5:1, which supports a pathway producing the observed experimental N2-product ratios with greater than 80% yield. The N1-product was again found to be lower in energy by 4.4 kcal/mol than the N2-product.

[1860-5397-20-170-7]

Figure 7: DFT-calculations concerning a coordinated Mitsunobu reaction pathway.

[1860-5397-20-170-8]

Figure 8: Reaction coordinate diagram of 6(N-H) reacting under conditions B. All calculated energies in kcal/mol. Ball-and-stick transition-state structures are provided for the lowest energy N1- and N2-transition states with favorable NCIs shown as red dashed lines.

To further explore whether the reaction mechanisms followed the chelation pathway proposed in Figure 5, we hypothesized that 18 (Figure 9) would provide a model for exploring the mechanism further. If chelation between an electron-rich oxygen atom from a substituent and a Lewis acid (such as Cs+ or P+) were taking place, we would expect regioselectivity for this substrate to be reversed (19-OCs, 20-OP), such that conditions A would produce the N2-product and conditions B would produce the N1-product. This was found to be the case as can be observed in Figure 9. The N2-product 19 was isolated in 93% yield under conditions A, and the N1-product 20 was isolated in >99% yield under conditions B albeit with low conversion. Both conditions provided >98:2 regioselectivity for their respective major products as determined by LC–MS.

[1860-5397-20-170-9]

Figure 9: Reaction of 18 under conditions A and B (top), and proposed chelation/coordination pathways to account for regioselectivity (bottom); black two-headed curved arrows indicate the observed NOEs of the major product; adetermined by LC–MS; bbased on recovered starting material.

We again turned to DFT calculations to explore the mechanisms of these reactions and found very similar results to the parent system (Figure 10). Under conditions A, deprotonation and cesium coordination were heavily favored by 11.1 kcal/mol (18(N-H)). Transition states 18-N2-Cs and 18-N1-Cs were found leading to N1- and N2-products, respectively. NCIs between the cesium cation with the ester and sulfonate oxygens, and critically N1, position the electrophilic methyl group 2.1 Å from N2, lowering the TS energy by ΔΔG‡ = 2.6 kcal/mol in 18-N1-Cs. Interestingly, the difference in product energies is quite small, only favoring the N1-product, 18-N1, by 0.6 kcal/mol. Concerning conditions B, the difference between the neutral indazole and the deprotonated indazole was only −0.2 kcal/mol (Figure 11). Again, no preorganized intermediates were found. The NCIs were consistent with the parent system. The hydrogen bond between the H on the electrophilic methyl group and an ester oxygen was found in the transition state leading to the N1 product, was conserved (18-N1-OMe, Figure 11), in a total of 4 NCIs. The only relevant NCIs in 18-N2-OMe were between an aryl hydrogen, N1 and an ester oxygen, the result of which is ΔΔG‡ = 2.0 kcal/mol favoring the N1 transition state 18-N1-OMe among the NCIs found.

[1860-5397-20-170-10]

Figure 10: DFT-calculated reaction coordinate diagram for the reaction of 18 under conditions A.

[1860-5397-20-170-11]

Figure 11: DFT-calculated reaction coordinate diagram for the reaction of 18 under conditions B. Ball-and-stick transition-state structures are provided for the lowest energy N1 and N2 transition states with favorable NCIs shown as red dashed lines.

To further probe whether the dominant discriminating factor was chelation or other NCIs, compound 21 was also subjected to the same reaction conditions (Scheme 3). As this cyano compound is not capable of forming an N2–Cs+–NCN ion pair or dative bond, we were curious to observe product ratios. Compound 21 produced the N1-product 22 as the major regioisomer regardless of which conditions were employed. Though separation of products was challenging, resulting in a lower isolated yield under conditions B, the regioselectivity was greater than 99:1 under both conditions.

[1860-5397-20-170-i3]

Scheme 3: Reaction of 21 under conditions A and B; amultiple purifications; bdetermined by LC–MS.

DFT calculations again showed the deprotonated analog of 21 to be heavily favored in complex with Cs+ by 20.6 kcal/mol under conditions A, and 13.6 kcal/mol under conditions B. Transition states were found for the reaction of 21 under both conditions (see Supporting Information File 1). However, in this case, neither chelation nor other NCIs resulted in an energy difference between the lowest energy N1- and N2-transition states greater than 1 kcal/mol (Figure 12). Thus, we employed natural bond orbital (NBO) analysis to estimate the partial charge of N1 and N2 in both the neutral and deprotonated states using Gaussian 16: NBO7, SMD(THF)-PBE0/def2-SVP. Additionally, Parr and Yang developed a now widely used index of nucleophilicity from robust electron population methods such as NBO analyses, referred to as the Fukui index (f-) [44]. The larger the Fukui index, the greater the nucleophilicity, and is thus inversely proportional to the partial charge. Our calculations showed that N1 was more electronegative and had a larger Fukui index in both neutral and deprotonated states, not only in 21, but in 18 and 6 also (Table 4). These data suggest that in the absence of an electron-withdrawing group responsible for either cation chelation or favorable NCI stabilization, nucleophilicity would dictate regioselectivity outcomes. This also implies that the favorable NCIs and chelation are stronger driving forces towards transition-state energy partitioning than nucleophilicity alone.

[1860-5397-20-170-12]

Figure 12: DFT-calculated transition-state structures and energies of 21 under conditions A (top) and conditions B (bottom).

Table 4: DFT-calculated electronegativities and Fukui indices.

Compound Partial charge Fukui index (f-)
N1 N2 N1 N2
6 −0.33525 −0.24443 0.61445 0.52779
6(N-H) −0.38846 −0.29832 0.65078 0.56750
18 −0.36252 −0.30508 0.15932 0.03122
18(N-H) −0.42226 −0.36313 0.24944 0.00431
21 −0.33075 −0.23976 0.06065 0.16123
21(N-H) −0.37927 −0.30008 0.20158 0.01133

These results suggest chelation is a highly plausible driving force for regioselectivity in the alkylation of methyl indazole-3- or -7-carboxylates. When the ester substituent is placed at the 3- or 7-position, the chelation of Cs+ or NCIs with ROPPh3+ and the associated nitrogens will drive regioselectivity to or away from that nitrogen, leading to excellent selectivity. These data support the claim made by Alam and Keeting that a tight ion pair drives N1-selectivity when electron-withdrawing groups that can coordinate the cation are present at the 3-position. When 3-cyanoindazole is employed and no bidentate coordination is possible with N2, the nucleophilicity of N1 drives the regioselectivity. Additionally, these data show the importance of NCIs in understanding mechanisms where regioselectivity outcomes are unexpected. Lastly, it should be noted that these reactions are likely irreversible due to the ≈50–60 kcal/mol barriers of the reverse reactions and near-absent nucleophilic character of TsO− and triphenylphosphine oxide, precluding any thermodynamic versus kinetic arguments for regioselectivity.

Conclusion

We have established highly regioselective N1- and N2-alkylations of methyl 5-bromo-1H-indazole-3-carboxylate from diverse commercially available alcohols with excellent yields (>84%). The unique quality of this work is the production of working mechanisms that support the observed regioisomeric ratios. This work presents the first comprehensive DFT mechanistic study on these systems which differentiate formation of either N1- or N2-substituted indazoles in excellent yields from the same carbon sources through reagent control.

Experimental

General methods

All materials were obtained from commercial suppliers and used without further purification unless otherwise noted. Anhydrous solvents were obtained from Sigma-Aldrich and used directly. Reactions involving air- or moisture-sensitive reagents were performed under a nitrogen or argon atmosphere. Silica gel chromatography was performed using prepacked silica gel columns (RediSep® Rf, Teledyne ISCO). An aluminum block atop a hotplate with a thermocouple was used to heat reactions to the specified temperatures. NMR spectra were acquired on Bruker 300 MHz spectrometers equipped with 5 mm BBFO probes. HRMS data were acquired using an Agilent 6530 LC/Q-TOF using a Dual AJS/ESI ion source, and the isotope 79 was used for HRMS analysis for any bromine-containing compounds.

Synthesis

All procedures and spectra can be found in Supporting Information File 1.

General procedure for the N1-alkylation using alkyltosylates

Preparation of methyl 5-bromo-1-methy-1H-indazole-3-carboxylate (15a)

To a solution of methyl 5-bromo-1H-indazole-3-carboxylate (300 mg, 1.176 mmol) in dioxane (10 mL) at room temperature was added cesium carbonate (766 mg, 2.352 mmol) followed by the necessary tosylate (1.5 equiv). The resulting mixture was stirred for 2 hours at 90 °C. The mixture was poured into EtOAc (500 mL) and washed with water (100 mL) and brine. The organic layer was dried and concentrated, the obtained residue was purified by chromatography (silica [24 g], eluting with EtOAc in hexane from 0–70%) to give methyl 5-bromo-1-alkyl-1H-indazole-3-carboxylate, in 90–98%. For 15a, 284.5 mg, 90%, as a white solid. Mp 141.3 °C; 1H NMR (300 MHz, DMSO-d6) δ 8.19 (dd, J = 1.9, 0.7 Hz, 1H), 7.82 (dd, J = 9.0, 0.7 Hz, 1H), 7.65 (dd, J = 8.9, 1.9 Hz, 1H), 4.17 (s, 3H), 3.92 (s, 3H); 13C{1H} NMR (75 MHz, DMSO-d6) δ 161.8, 139.4, 132.7, 129.4, 124.1, 123.0, 116.0, 113.0, 51.8, 36.6; IR (KBr disk): 1722, 1466, 1433, 1395, 1354, 1289, 1200, 1183, 1153 cm−1; HRESIMS (m/z): [M + H]+ calcd for C10H10BrN2O2+, 268.9921; found, 268.9902.

General procedure for the N2-alkylation using Mitsunobu conditions

Preparation of methyl 5-bromo-2-methyl-2H-indazole-3-carboxylate (16a)

To a solution of methyl 5-bromo-1H-indazole-3-carboxylate (1.384 g, 5.43 mmol) in THF (15 mL) was added triphenylphosphine (2.85 g, 10.85 mmol) and methanol (0.5 mL, 12.4 mmol) at 0 °C, followed by adding DEAD (1.718 mL, 10.85 mmol). The resulting mixture was stirred for 10 min at 0 °C, warmed to 50 °C, and stirred for 2 h. After TLC showed completion, the solvent was removed, and the residue was purified by chromatography (silica [24 g], eluting with ethyl acetate in hexane from 0–60%) to give methyl 5-bromo-2-alkyl-2H-indazole-3-carboxylate in 90–97% yield. For 16a, 291 mg, 92%, as a light pink solid. Mp 110.7 °C; 1H NMR (300 MHz, DMSO-d6) δ 8.14 (d, J = 2.0 Hz, 1H), 7.77 (dd, J = 9.1, 0.7 Hz, 1H), 7.49 (dd, J = 9.0, 1.9 Hz, 1H), 4.42 (s, 3H), 3.98 (s, 3H); 13C{1H} NMR (75 MHz, DMSO-d6) δ 159.4, 144.7, 129.3, 123.6, 123.2, 122.8, 120.0, 118.0, 64.2, 52.2, 41.4, 14.4, 13.9; IR (KBr disk): 1708, 1459, 1442, 1392, 1326, 1252, 1196 cm−1; HRESIMS (m/z): [M + H]+ calcd for C10H10BrN2O2+, 268.9921; found, 268.9918.

Supporting Information

Supporting Information File 1: Characterization of all compounds (1H NMR, 13C NMR, LC–MS, IR), and crystallographic methods and data for products P1 and P2.
Format: PDF Size: 38.4 MB Download
Supporting Information File 2: DFT methods, relative energy comparisons, TS imaginary frequencies, and XYZ coordinates.
Format: PDF Size: 1.2 MB Download
Supporting Information File 3: GoodVibes outputs.
Format: XLSX Size: 47.7 KB Download

Acknowledgements

The synthesis efforts discussed in this paper were critically enabled by the support of a diverse set of talented teams, functional leaders, and highly motivated scientists, without whom this work would not have been possible. We would like to thank Dr. Liliana Gallegos for help with quantum mechanics calculations, Drs. Minwan Wu and Trung Nguyen of BioCryst Pharmaceuticals, Inc., and Dr. Eric V. Anslyn, Norman Hackerman Professor of Chemistry, The University of Texas at Austin for helpful discussions, Dr. Ken Belmore at the University of Alabama, Tuscaloosa for NMR assistance, and Dr. Nattamai Bhuvanesh at Texas A&M University for X-ray diffraction analysis.

Funding

All studies were funded by BioCryst Pharmaceuticals Inc.

Conflict of Interest

The authors declare the following competing financial interest(s): All authors are employees/former employees of BioCryst Pharmaceuticals Inc. and may hold stock in the same.

Author Contributions

Pengcheng Lu: conceptualization; data curation; investigation; methodology; writing – original draft; writing – review & editing. Luis Juarez: investigation. Paul A. Wiget: conceptualization; data curation; formal analysis; investigation; methodology; project administration; supervision; validation; visualization; writing – original draft; writing – review & editing. Weihe Zhang: conceptualization; data curation; investigation; methodology; project administration; supervision; writing – original draft; writing – review & editing. Krishnan Raman: investigation; validation. Pravin L. Kotian: supervision; writing – review & editing.

Data Availability Statement

All data that supports the findings of this study is available in the published article and/or the supporting information to this article.

References

  1. Dong, J.; Zhang, Q.; Wang, Z.; Huang, G.; Li, S. ChemMedChem 2018, 13, 1490–1507. doi:10.1002/cmdc.201800253
    Return to citation in text: [1]
  2. Gaikwad, D. D.; Chapolikar, A. D.; Devkate, C. G.; Warad, K. D.; Tayade, A. P.; Pawar, R. P.; Domb, A. J. Eur. J. Med. Chem. 2015, 90, 707–731. doi:10.1016/j.ejmech.2014.11.029
    Return to citation in text: [1]
  3. Wan, Y.; He, S.; Li, W.; Tang, Z. Anti-Cancer Agents Med. Chem. 2018, 18, 1228–1234. doi:10.2174/1871520618666180510113822
    Return to citation in text: [1]
  4. Cerecetto, H.; Gerpe, A.; Gonzalez, M.; Aran, V. J.; de Ocariz, C. O. Mini-Rev. Med. Chem. 2005, 5, 869–878. doi:10.2174/138955705774329564
    Return to citation in text: [1]
  5. Gao, M.; Xu, B. Chem. Rec. 2016, 16, 1701–1714. doi:10.1002/tcr.201600020
    Return to citation in text: [1]
  6. Brown, N., Ed. Bioisosteres in Medicinal Chemistry; Methods and Principles in Medicinal Chemistry; Wiley-VCH: Weinheim, Germany, 2012. doi:10.1002/9783527654307
    Return to citation in text: [1]
  7. Hu, H.; Wang, X.; Chan, G. K. Y.; Chang, J. H.; Do, S.; Drummond, J.; Ebens, A.; Lee, W.; Ly, J.; Lyssikatos, J. P.; Murray, J.; Moffat, J. G.; Chao, Q.; Tsui, V.; Wallweber, H.; Kolesnikov, A. Bioorg. Med. Chem. Lett. 2015, 25, 5258–5264. doi:10.1016/j.bmcl.2015.09.052
    Return to citation in text: [1]
  8. Wang, X.; Kolesnikov, A.; Tay, S.; Chan, G.; Chao, Q.; Do, S.; Drummond, J.; Ebens, A. J.; Liu, N.; Ly, J.; Harstad, E.; Hu, H.; Moffat, J.; Munugalavadla, V.; Murray, J.; Slaga, D.; Tsui, V.; Volgraf, M.; Wallweber, H.; Chang, J. H. J. Med. Chem. 2017, 60, 4458–4473. doi:10.1021/acs.jmedchem.7b00418
    Return to citation in text: [1]
  9. Lai, A.; Kahraman, M.; Govek, S.; Nagasawa, J.; Bonnefous, C.; Julien, J.; Douglas, K.; Sensintaffar, J.; Lu, N.; Lee, K.-j.; Aparicio, A.; Kaufman, J.; Qian, J.; Shao, G.; Prudente, R.; Moon, M. J.; Joseph, J. D.; Darimont, B.; Brigham, D.; Grillot, K.; Heyman, R.; Rix, P. J.; Hager, J. H.; Smith, N. D. J. Med. Chem. 2015, 58, 4888–4904. doi:10.1021/acs.jmedchem.5b00054
    Return to citation in text: [1]
  10. Liu, Y.; Lang, Y.; Patel, N. K.; Ng, G.; Laufer, R.; Li, S.-W.; Edwards, L.; Forrest, B.; Sampson, P. B.; Feher, M.; Ban, F.; Awrey, D. E.; Beletskaya, I.; Mao, G.; Hodgson, R.; Plotnikova, O.; Qiu, W.; Chirgadze, N. Y.; Mason, J. M.; Wei, X.; Lin, D. C.-C.; Che, Y.; Kiarash, R.; Madeira, B.; Fletcher, G. C.; Mak, T. W.; Bray, M. R.; Pauls, H. W. J. Med. Chem. 2015, 58, 3366–3392. doi:10.1021/jm501740a
    Return to citation in text: [1]
  11. Zhu, W.; Chen, H.; Wang, Y.; Wang, J.; Peng, X.; Chen, X.; Gao, Y.; Li, C.; He, Y.; Ai, J.; Geng, M.; Zheng, M.; Liu, H. J. Med. Chem. 2017, 60, 6018–6035. doi:10.1021/acs.jmedchem.7b00076
    Return to citation in text: [1]
  12. Czodrowski, P.; Mallinger, A.; Wienke, D.; Esdar, C.; Pöschke, O.; Busch, M.; Rohdich, F.; Eccles, S. A.; Ortiz-Ruiz, M.-J.; Schneider, R.; Raynaud, F. I.; Clarke, P. A.; Musil, D.; Schwarz, D.; Dale, T.; Urbahns, K.; Blagg, J.; Schiemann, K. J. Med. Chem. 2016, 59, 9337–9349. doi:10.1021/acs.jmedchem.6b00597
    Return to citation in text: [1]
  13. Bermudez, J.; Fake, C. S.; Joiner, G. F.; Joiner, K. A.; King, F. D.; Miner, W. D.; Sanger, G. J. J. Med. Chem. 1990, 33, 1924–1929. doi:10.1021/jm00169a016
    Return to citation in text: [1]
  14. Al-Bogami, A. S. Res. Chem. Intermed. 2016, 42, 5457–5477. doi:10.1007/s11164-015-2379-5
    Return to citation in text: [1]
  15. Taylor, A. M.; Côté, A.; Hewitt, M. C.; Pastor, R.; Leblanc, Y.; Nasveschuk, C. G.; Romero, F. A.; Crawford, T. D.; Cantone, N.; Jayaram, H.; Setser, J.; Murray, J.; Beresini, M. H.; de Leon Boenig, G.; Chen, Z.; Conery, A. R.; Cummings, R. T.; Dakin, L. A.; Flynn, E. M.; Huang, O. W.; Kaufman, S.; Keller, P. J.; Kiefer, J. R.; Lai, T.; Li, Y.; Liao, J.; Liu, W.; Lu, H.; Pardo, E.; Tsui, V.; Wang, J.; Wang, Y.; Xu, Z.; Yan, F.; Yu, D.; Zawadzke, L.; Zhu, X.; Zhu, X.; Sims, R. J., III; Cochran, A. G.; Bellon, S.; Audia, J. E.; Magnuson, S.; Albrecht, B. K. ACS Med. Chem. Lett. 2016, 7, 531–536. doi:10.1021/acsmedchemlett.6b00075
    Return to citation in text: [1]
  16. Risitano, A. M.; Kulasekararaj, A. G.; Lee, J. W.; Maciejewski, J. P.; Notaro, R.; Brodsky, R.; Huang, M.; Geffner, M.; Browett, P. Haematologica 2021, 106, 3188–3197. doi:10.3324/haematol.2020.261826
    Return to citation in text: [1]
  17. Scott, L. J. Drugs 2017, 77, 1029–1034. doi:10.1007/s40265-017-0752-y
    Return to citation in text: [1]
  18. Baddam, S. R.; Uday Kumar, N.; Panasa Reddy, A.; Bandichhor, R. Tetrahedron Lett. 2013, 54, 1661–1663. doi:10.1016/j.tetlet.2013.01.030
    Return to citation in text: [1]
  19. Igawa, H.; Takahashi, M.; Shirasaki, M.; Kakegawa, K.; Kina, A.; Ikoma, M.; Aida, J.; Yasuma, T.; Okuda, S.; Kawata, Y.; Noguchi, T.; Yamamoto, S.; Fujioka, Y.; Kundu, M.; Khamrai, U.; Nakayama, M.; Nagisa, Y.; Kasai, S.; Maekawa, T. Bioorg. Med. Chem. 2016, 24, 2486–2503. doi:10.1016/j.bmc.2016.04.011
    Return to citation in text: [1]
  20. Igawa, H.; Takahashi, M.; Ikoma, M.; Kaku, H.; Kakegawa, K.; Kina, A.; Aida, J.; Okuda, S.; Kawata, Y.; Noguchi, T.; Hotta, N.; Yamamoto, S.; Nakayama, M.; Nagisa, Y.; Kasai, S.; Maekawa, T. Bioorg. Med. Chem. 2016, 24, 2504–2518. doi:10.1016/j.bmc.2016.04.013
    Return to citation in text: [1]
  21. Stadlbauer, W. Indazole (Benzopyrazole). Houben-Weyl, Methoden der Organischen Chemie; Thieme: Stuttgart, Germany, 1994; Vol. E8b, pp 764–864. doi:10.1055/b-0035-110489
    Return to citation in text: [1]
  22. Stadlbauer, W. Sci. Synth. 2002, 227–324. doi:10.1055/sos-sd-012-00277
    Return to citation in text: [1]
  23. Escande, A.; Lapasset, J.; Faure, R.; Vincent, E.-J.; Elguero, J. Tetrahedron 1974, 30, 2903–2909. doi:10.1016/s0040-4020(01)97464-2
    Return to citation in text: [1]
  24. Faure, R.; Vincent, E.-J.; Elguero, J. Heterocycles 1983, 20, 1713–1716. doi:10.3987/r-1983-09-1713
    Return to citation in text: [1]
  25. Catalán, J.; de Paz, J. L. G.; Elguero, J. J. Chem. Soc., Perkin Trans. 2 1996, 57–60. doi:10.1039/p29960000057
    Return to citation in text: [1]
  26. Foces‐Foces, C.; Hager, O.; Jagerovic, N.; Jimeno, M. L.; Elguero, J. Chem. – Eur. J. 1997, 3, 121–126. doi:10.1002/chem.19970030119
    Return to citation in text: [1]
  27. Luo, G.; Chen, L.; Dubowchik, G. J. Org. Chem. 2006, 71, 5392–5395. doi:10.1021/jo060607j
    Return to citation in text: [1]
  28. Haydl, A. M.; Xu, K.; Breit, B. Angew. Chem., Int. Ed. 2015, 54, 7149–7153. doi:10.1002/anie.201501758
    Return to citation in text: [1]
  29. Lin, M.-H.; Liu, H.-J.; Lin, W.-C.; Kuo, C.-K.; Chuang, T.-H. Org. Biomol. Chem. 2015, 13, 11376–11381. doi:10.1039/c5ob01747e
    Return to citation in text: [1]
  30. Hunt, K. W.; Moreno, D. A.; Suiter, N.; Clark, C. T.; Kim, G. Org. Lett. 2009, 11, 5054–5057. doi:10.1021/ol902050m
    Return to citation in text: [1] [2]
  31. Slade, D. J.; Pelz, N. F.; Bodnar, W.; Lampe, J. W.; Watson, P. S. J. Org. Chem. 2009, 74, 6331–6334. doi:10.1021/jo9006656
    Return to citation in text: [1]
  32. Schmidt, A.; Beutler, A.; Snovydovych, B. Eur. J. Org. Chem. 2008, 4073–4095. doi:10.1002/ejoc.200800227
    Return to citation in text: [1]
  33. Zhang, S.-G.; Liang, C.-G.; Zhang, W.-H. Molecules 2018, 23, 2783. doi:10.3390/molecules23112783
    Return to citation in text: [1]
  34. Mei, Y.; Yang, B. Indian J. Heterocycl. Chem. 2017, 27, 275–280.
    Return to citation in text: [1]
  35. Clemens, J.; Bell, E. L.; Londregan, A. T. Synthesis 2022, 54, 3215–3226. doi:10.1055/s-0040-1719917
    Return to citation in text: [1]
  36. He, H.; Yan, J.; Jin, J.; Yan, Z.; Yan, Q.; Wang, W.; Jiang, H.; Wang, H.; Chen, F. Chem. Commun. 2022, 58, 6429–6432. doi:10.1039/d2cc01404a
    Return to citation in text: [1]
  37. Alam, R. M.; Keating, J. J. Beilstein J. Org. Chem. 2021, 17, 1939–1951. doi:10.3762/bjoc.17.127
    Return to citation in text: [1] [2] [3]
  38. Cheung, M.; Boloor, A.; Stafford, J. A. J. Org. Chem. 2003, 68, 4093–4095. doi:10.1021/jo0265434
    Return to citation in text: [1]
  39. Genung, N. E.; Wei, L.; Aspnes, G. E. Org. Lett. 2014, 16, 3114–3117. doi:10.1021/ol5012423
    Return to citation in text: [1]
  40. Takahashi, T.; Umino, A.; Iijima, D.; Takamatsu, H. Novel 2-Aminopyridine and 2-Aminopyrimidine Derivatives and Medicinal Use Thereof. WO Pat. Appl. WO2015163435A1, Oct 29, 2015.
    Return to citation in text: [1]
  41. Xu, X. Heterocyclic fxr modulators. WO Pat. Appl. WO2018075207A1, April 16, 2018.
    Return to citation in text: [1]
  42. Yang, X.; Li, F.; Konze, K. D.; Meslamani, J.; Ma, A.; Brown, P. J.; Zhou, M.-M.; Arrowsmith, C. H.; Kaniskan, H. Ü.; Vedadi, M.; Jin, J. J. Med. Chem. 2016, 59, 7617–7633. doi:10.1021/acs.jmedchem.6b00855
    Return to citation in text: [1]
  43. Yu, L.; Wei, Y. Indazole-formamide-pyridone derivative and preparation method and use thereof. Chin. Patent CN107573327B, Jan 12, 2018.
    Return to citation in text: [1]
  44. Parr, R. G.; Yang, W. J. Am. Chem. Soc. 1984, 106, 4049–4050. doi:10.1021/ja00326a036
    Return to citation in text: [1]

© 2024 Lu et al.; licensee Beilstein-Institut.
This is an open access article licensed under the terms of the Beilstein-Institut Open Access License Agreement (https://www.beilstein-journals.org/bjoc/terms), which is identical to the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0). The reuse of material under this license requires that the author(s), source and license are credited. Third-party material in this article could be subject to other licenses (typically indicated in the credit line), and in this case, users are required to obtain permission from the license holder to reuse the material.

Other Beilstein-Institut Open Science Activities
Journal Logo
Institut Logo
Preprint Logo
Symposia Logo