Multigram Scale Synthesis of A21, A New Antibiotic Equally Eective and Less Toxic than Amphotericin B Jose ́ David Flores-Romero, Josue ́ Rodríguez-Lozada, Manuel Ló pez-Ortiz, Ricardo Magañ a, Iva ́ n Ortega-Blake, § Ignacio Regla,* , and Mario Ferna ́ ndez-Zertuche* , Instituto de Investigació n en Ciencias Ba ́ sicas y Aplicadas, Centro de Investigaciones Químicas, Universidad Autó noma del Estado de Morelos, Av. Universidad 1001, Chamilpa, 62209 Cuernavaca, Morelos, Mexico Facultad de Estudios Superiores, Universidad Nacional Autó noma de Me ́ xico, Batalla del 5 de Mayo y Fuerte de Loreto, 09230 Iztapalapa, Mé xico DF, Mé xico § Instituto de Ciencias Físicas, Universidad Nacional Autó noma de Mé xico, Av. Universidad 1001, Chamilpa, 62209 Cuernavaca, Morelos, Mexico * S Supporting Information ABSTRACT: A21 (2) is a new polyene macrolide Amphotericin B amide antibiotic derived from amphoter- icin B AmB (1), which has been tested extensively on preclinical trials showing the same antimycotic eective- ness and increased margin of safety over AmB (1). We present the multigram scale synthesis, isolation, purity assessment by HPLC, and key aspects of its character- ization by NMR studies of A21 (2). INTRODUCTION Amphotericin B, AmB (1)(Figure 1), has been the most eective agent against systemic fungal infections (SFI) for over 60 years, 1 and recently its use has been extended to treat parasitic infections like leishmaniases. 2 This antifungal agent belongs to the macrocyclic drug family known as polyene macrolides, which also includes amphotericin A, natamycin, and nystatine; their chemical structures comprise a macrocyclic lactone ring, a polyene chain, multiple hydroxyl groups, and an amino glycoside moiety. 3 The use of polyene macrolides in medical practice, and of 1 in particular, has been widely established. However, its clinical use has serious limitations due to the adverse eects that occur after his administration, mainly damage to the kidney (nephrotoxicity), 4 damage to the liver (hepatotoxicity), 5 and anemia, 6 by breaking the red blood cells. When these problems arise on patients under treatment with 1, it is necessary to suspend treatment. Nevertheless, the lack of an eective alternative treatment for SFI has settled the clinical importance of 1 as unchanged. 7 During the past decade, a large amount of research has been devoted to the understanding of the mechanism of action. It is widely accepted that 1 associates with membrane sterols to form barrel-stave ionic pores, which increases membrane permeability and modies its potentials, leading to K + ions leaking and cell death as a nal step. 8 In recent years, this mechanism of action has been questioned by the evidence of high biological activity of synthetically modied 1 molecules suggesting that ionic pore formation could be a secondary mechanism and that 1 works as a sterol sponge, extracting ergosterol from the fungal membrane and leading to cell death. 9 Understanding the mechanism of action of 1 as an antibiotic opens the possibility to carry out chemical modications on its structure in order to reduce its toxicity, maintaining at the same time its maximum activity as an antibiotic. Numerous attempts have been reported in the literature to modiy the structure of 1 in order to reduce its toxicity. These include the modication of the polyene chain, 10 derivatization of the amino group on the mycosamine ring system, 11 preparation of amides at the carboxylic function, 12 removal of the exocyclic carboxylic function, 13 synthesis of uorinated derivatives, 14 and covalent dimers with carbonyl-amine linkage. 15 More recently, elegant methodologies have been developed to remove the OH group at C-3 16 or at C-35. 17 Our research group has made numerous attempts to modify the chemical structure of 1. As a result, we have found that, through the preparation of several amide type derivatives, it is possible to develop a new antibiotic with these characteristics. Specically, we found that an amide derivative prepared by the coupling of the (L)-histidine methyl ester to 1 leads to a derivative we call A21 (2)(Figure 2). This new derivative maintains the same eectiveness as 1 as an antifungal agent and a superior prole of toxicity. For example, the lethal dose 50 (LD 50 ) of 1 in Balb-C mice by intravenous administration for 48 h is 4.13 mg/kg, whereas for 2 it is 49.82 mg/kg, showing an increased margin of safety. Received: June 13, 2016 Published: July 19, 2016 Figure 1. Structure of AmB (1). Communication pubs.acs.org/OPRD © 2016 American Chemical Society 1529 DOI: 10.1021/acs.oprd.6b00211 Org. Process Res. Dev. 2016, 20, 1529-1532