Rapid Construction of a Chloromethyl-Substituted Duocarmycin-like Prodrug

The construction of duocarmycin-like compounds is often associated with lengthy synthetic routes. Presented herein is the development of a short and convenient synthesis of a type of duocarmycin prodrug. The 1,2,3,6-tetrahydropyrrolo[3,2-e]indole-containing core is here constructed from commercially available Boc-5-bromoindole in four steps and 23% overall yield, utilizing a Buchwald–Hartwig amination followed by a sodium hydride-induced regioselective bromination. In addition, protocols for selective mono- and di-halogenations of positions 3 and 4 were also developed, which could be useful for further exploration of this scaffold.

The existing synthetic pathways are elaborative and/or give the wrong substitution pattern ( Figure 2). Furthermore, in our early attempts to use Boc-5-nitroindole 9 as starting material, we faced several problems, such as over-reduction when reducing the nitro group (i.e., the generation of indoline), the generation of complex mixtures when performing the halogenation reaction on the aniline, and problems with controlling the mono-Boc protection of the aniline.
In our approach, we envisioned that the desired di-Boc-protected 5-aminoindole intermediate 12 ( Figure 3) could be synthesized from commercially available Boc-5-bromoindole 11 via a Buchwald-Hartwig amination with tBu-carbamate followed by a regioselective bromination. This strategy would considerably shorten the route and also overcome the problems related to the nitro reduction and mono-Boc protection of the aniline nitrogen; vide supra. The existing synthetic pathways are elaborative and/or give the wrong substitution pattern ( Figure 2). Furthermore, in our early attempts to use Boc-5-nitroindole 9 as starting material, we faced several problems, such as over-reduction when reducing the nitro group (i.e., the generation of indoline), the generation of complex mixtures when performing the halogenation reaction on the aniline, and problems with controlling the mono-Boc protection of the aniline. In our approach, we envisioned that the desired di-Boc-protected 5-aminoindole intermediate 12 (Figure 3) could be synthesized from commercially available Boc-5-bromoindole 11 via a Buchwald-Hartwig amination with tBu-carbamate followed by a regioselective bromination. This strategy would considerably shorten the route and also over-  In our approach, we envisioned that the desired di-Boc-protected 5-aminoindole intermediate 12 ( Figure 3) could be synthesized from commercially available Boc-5-bromoindole 11 via a Buchwald-Hartwig amination with tBu-carbamate followed by a regioselective bromination. This strategy would considerably shorten the route and also overcome the problems related to the nitro reduction and mono-Boc protection of the aniline nitrogen; vide supra.

Results and Discussion
The Pd(OAc) 2 /XPhos-catalyzed Buchwald-Hartwig amination of Boc-5-bromoindole (11) with tBu-carbamate performed well, and compound 13 could be isolated in 78% yield (Scheme 1). Performing the subsequent halogenation under acidic conditions (i.e., NXS/TsOH) on the Boc-protected aniline gave the wrong regioisomer, although with complete selectivity, and the 3-bromo (14) and 3-iodo (15) products could be isolated in 74% and 71% yields, respectively, using the two different halogen sources. We envisioned that the deprotonation of the Boc-protected aniline with NaH prior to the halogenation might render the aromatic ring sufficiently electron-rich to direct the halogenation to the right position (see Supporting Information). Gratifyingly, that strategy gave the desired 4-bromo analog 12 in 65% yield with complete regioselectivity. All attempts to introduce iodine in this position failed, even when using a more electrophilic I + source (i.e., N-Iodosaccharin [23]), other solvents, or elevated temperatures.

Results and Discussion
The Pd(OAc)2/XPhos-catalyzed Buchwald-Hartwig amination of Boc-5-bromoindole (11) with tBu-carbamate performed well, and compound 13 could be isolated in 78% yield (Scheme 1). Performing the subsequent halogenation under acidic conditions (i.e., NXS/TsOH) on the Boc-protected aniline gave the wrong regioisomer, although with complete selectivity, and the 3-bromo (14) and 3-iodo (15) products could be isolated in 74% and 71% yields, respectively, using the two different halogen sources. We envisioned that the deprotonation of the Boc-protected aniline with NaH prior to the halogenation might render the aromatic ring sufficiently electron-rich to direct the halogenation to the right position (see supporting information). Gratifyingly, that strategy gave the desired 4bromo analog 12 in 65% yield with complete regioselectivity. All attempts to introduce iodine in this position failed, even when using a more electrophilic I + source (i.e., N-Iodosaccharin [23]), other solvents, or elevated temperatures.

Materials and Methods
General Methods: All solvents and reagents were used as received from commercial suppliers. N-Bromosuccinimide (NBS) was recrystallized from hot water and dried under vacuum for 24 h and then stored under cold and dark conditions. Sodium hydride was used as 60% dispersion in mineral oil. Column chromatography was employed on normal-phase silica gel (230-400 mesh, 60 Å; the eluents are given in brackets). 1 H-and 13 C-NMR spectra were recorded on a 400 MHz spectrometer at 298 K and calibrated using the residual peak of the solvent as an internal standard CDCl3 (CHCl3 H 7.26 ppm, CDCl3 C 77.16 ppm). HRMS was performed using a microTOF instrument with electrospray ionization (ESI), and sodium formate was used as a calibration chemical. Optical rotations were measured on a polarimeter at 589 nm (D line of sodium) and 20 °C. Chiral chromatography was performed on supercritical fluid chromatography equipment, using mixtures of MeOH and supercritical CO2 as eluents.

Materials and Methods
General Methods: All solvents and reagents were used as received from commercial suppliers. N-Bromosuccinimide (NBS) was recrystallized from hot water and dried under vacuum for 24 h and then stored under cold and dark conditions. Sodium hydride was used as 60% dispersion in mineral oil. Column chromatography was employed on normal-phase silica gel (230-400 mesh, 60 Å; the eluents are given in brackets). 1 H-and 13 C-NMR spectra were recorded on a 400 MHz spectrometer at 298 K and calibrated using the residual peak of the solvent as an internal standard [CDCl 3 (CHCl 3 δ H 7.26 ppm, CDCl 3 δ C 77.16 ppm)]. HRMS was performed using a microTOF instrument with electrospray ionization (ESI), and sodium formate was used as a calibration chemical. Optical rotations were measured on a polarimeter at 589 nm (D line of sodium) and 20 • C. Chiral chromatography was performed on supercritical fluid chromatography equipment, using mixtures of MeOH and supercritical CO 2 as eluents.