Baran’s Partial Synthesis of Ouabain Aglycone from Cortisone Acetate

Na⁺,K⁺-ATPase, often referred to as the sodium-potassium pump, is an integral membrane protein that catalyzes the enzymatic ATP-dependent transport of three sodium ions out of the cell and two potassium ions in, generating a Na⁺ gradient across the cell. The resulting gradient drives a number of vital processes including the transport of glucose and other nutrients across the cell membrane. Apart from its transport function, Na⁺,K⁺-ATPase can also act as a functional receptor for the initiation of a variety of signaling events that can alter cellular functions and cell growth in a cell-specific manner. Some of these Na⁺,K⁺-ATPase-mediated signaling pathways can be activated in certain mammals by circulating endogenous cardiotonic steroids, which are inhibitory ligands of the sodium-potassium pump. Altered plasma levels of endogenous cardiotonic (cardenolide) steroids have been associated with the development of pathological conditions such as chronic renal failure, hypertension and congestive heart failure. Moreover, certain cardenolide steroids such as ouabain and digoxin are widely prescribed therapeutic treatments for heart failure and other cardiac conditions. The practical chemical synthesis of structurally complex cardiotonic steroids and analogues thereof remains a significant obstacle to the task of improving the narrow therapeutic index of this important class of cardiac drugs.

Recently, Phil Baran’s group at the Scripps Research Institute disclosed a truly groundbreaking synthesis of the polyhydroxylated steroid ouabagenin (the aglycone of ouabain, structure depicted above) starting from a readily available steroidal precursor. The route was demonstrated on a moderate laboratory scale: 500 mg of a late-stage intermediate was obtained from a single batch run. Baran refers to his synthetic approach to ouabagenin as ‘quasibiomimetic,’ utilizing the concept of ‘redox relay,’ broadly defined as the transfer of redox information from one carbon position to another site within a given molecular framework. In this case, the 11-oxo functionality resident in the starting material cortisone is sequentially relayed, first, to the angular C19 position and then subsequently to C1 and C5 on the A-ring through the intermediacy of structures 3 and 4. Intermediate 3 is the product of Norrish type II photochemical functionalization the C19 methyl group by the proximal C11 carbonyl to furnish the cyclobutanol system (of 3) after stereocontrolled recombination of a transiently generated 1,4-biradical. A related photochemically-mediated C-C bond formation (shown below) to selectively functionalize the C18 angular methyl group of the ouabain derivative 5 was previously described in 1999 by Corey and Stoltz.

As a temporary aside, it must be noted tangentially that the ‘endogenous ouabain’ structure elucidation story is a complicated and fascinating one. For the interested reader, a leading reference can be found here. In brief, during the isolation, structural analysis and physiological characterization of ‘endogenous ouabain,’ Nakanishi and co-workers uncovered a facile ouabain-borate coordination phenomenon (depicted below) that occurs readily on exposure to borosilicate glassware. The polyhydroxylated steroid framework was shown to serve as a polydentate ligand capable of complexation to soluble inorganic borate derived from either borosilicate glass or the chemical reagent sodium tetraborate. Baran and co-workers insightfully exploited ouabain’s unique capacity for borate complexation in their A-ring protecting group strategy, highlighted below.

Returning to the Baran synthetic route to ouabagenin, stereoselective Super-Hydride® reduction of the less hindered C3 ketone of the advanced dione 1 was accompanied by formation of a cyclic ethyl boronic ester (8) which was elegantly utilized as a robust protecting group for the A-ring cis-1,3-diol. Next, reduction of the internal C11 carbonyl with lithium/ammonia under thermodynamically favorable conditions provided the desired 11-OH a-stereoisomer 9 in good yield (70%). The advanced intermediates 8 and 9 are reminiscent of the ouabain-borate complexes (see above) that were previously discovered and characterized by Nakanishi and co-workers. The bridging ethyl boronic ester functionality resident in 8 and 9 constitutes an unconventional yet effective A-ring diol blockade that was shown to be sufficiently stable throughout the entirety of the endgame reaction sequence leading to the target structure. A final redox relay from C17 (of 9) to C14 with inversion of the C/D ring junction configuration was now necessitated in order to obtain a 14b-hydroxy-androstane derivative suitable for completion of the synthesis of ouabagenin.

     Two other research groups have previously disclosed potentially applicable methodologies for the transformation of readily available sapogenin or androstane systems into C/D cis-fused cardenolide-type steroids. Baran’s laboratory chose to develop a new solution to this longstanding synthetic problem. Their innovative method for elaboration of the eastern cardenolide substructural framework must be assessed as superior to that of Heathcock and Wiesner (both of the previous technologies are discussed in detail here) with regard to step economy and modularity. Wiesner’s method, previously regarded as the state of the art in semi-synthetic approaches to cardiotonic steroids, requires about 8 steps to obtain a C/D cis-fused 14b-hydroxy-cardenolide derivative from a synthetic intermediate derived from testosterone. Baran’s methodology requires only three synthetic operations to accomplish the same objective: (1) Saegusa oxidation for D-ring (D^14,15) dehydrogenation; (2) a novel olefin isomerization (conversion of 10 into 11) and (3) Mukaiyama olefin hydration to secure the advanced intermediate 12. Moreover, the corresponding vinyl iodide (13) obtained from 12 by implementation of Barton’s protocol constitutes a valuable medicinal chemistry branch point for the installation of a variety of heterocyclic appendages to the C17 position in the course of forthcoming structure-activity relationship (SAR) studies. Wiesner’s analogous ‘furan methodology,’ developed in the early 1980’s, is far less flexible with regard to synthetic accessibility of C17 ouabain analogues. The advanced vinyl iodide 13 can be converted to ouabagenin in just four additional steps. The overall route proceeds in 20 steps from andrenosterone (2) in 0.56% yield. It is a testament to the practicality and potential medicinal relevance of Baran’s synthetic cardenolide technology that this work, a steroid semi-synthesis (not exactly a new concept…), was published in the journal Science in the year 2013.