Total Synthesis of Lycogarubin C and Lycogalic Acid

Abstract

Two complementary concise total syntheses of lycogarubin C (1) and lycogalic acid (2, aka chromopyrrolic acid) are detailed utilizing a 1,2,4,5-tetrazine → 1,2-diazine → pyrrole Diels–Alder strategy and enlisting acetylenic dienophiles.

Lycogarubin C (1) and lycogalic acid (2) were first identified as natural products in 1994, having been isolated independently by Steglich¹ and Akazawa² from Lycogala epidendrum, a slime mold (Figure 1). More recently, lycogalic acid, also referred to as chromopyrrolic acid (CPA),³ has been identified as a common intermediate in the biosynthesis of the indolo[2,3-a]carbazole alkaloids including rebeccamycin (6) and staurosporine (7) that exhibit broad spectrum activity as inhibitors of protein kinases as well as Topoisomerase I.⁴ As the efforts to elucidate the details of this biosynthetic pathway have progressed, the oxidation of chromopyrrolic acid (2) to 4 via 3 has attracted considerable interest since it involves an unusual oxidative aryl–aryl coupling reaction.^3,5 Moreover, in exploration of the individual enzyme-catalyzed steps in the pathway, 5 was isolated as an aerobic product following the oxidative coupling of 2 effected by RebP/StaP.⁶ As an off pathway intermediate that does not lead to formation of 4, it is likely that 5 and related compounds may well constitute the newest members of this class of natural products. As a result, we initiated efforts on the synthesis of 1 and 2 that in turn may serve as synthetic as well biosynthetic precursors to these potential newest members of this class of natural products.

Figure 1.

Natural products.

Complementary to reports of the synthesis of 1 or 2,^1,7–9 we anticipated that 1 and 2 would be readily accessible through use of a 1,2,4,5-tetrazine → 1,2-diazine → pyrrole Diels–Alder strategy that appears ideally suited for their preparation.¹⁰ Thus, the inverse electron demand Diels–Alder reaction of a 1,2-bis(indol-3-yl)acetylene (8) with dimethyl 1,2,4,5-tetrazine-3,6-dicarboxylate (9)¹¹ followed by a reductive ring contraction reaction of the resulting 1,2-diazine¹² to a dimethyl pyrrole-2,5-dicarboxylate could directly provide 1 or a protected penultimate precursor (Figure 2). Moreover, the potential use of the mono methyl esters derived from such dimethyl pyrrole-2,5-dicarboxylates to directly access products like 5 via a unique oxidative decarboxylation reaction¹³ provided the additional incentive for us to pursue the synthesis of 1 and 2. The recent disclosure of Fu and Gribble⁹ reporting that this direct strategy was not successful and their development of a clever alternative, using an olefinic versus acetylenic dienophile, provided the incentive for us to disclose our related, but more successful observations utilizing acetylenic dienophiles.

Figure 2.

Initial synthetic strategy.

The initial route explored entailed implementing the Diels–Alder reaction of the 1,2-bis-[(N-methoxycarbonyl)-indol-3-yl]acetylene (8) with 1,2,4,5-tetrazine 9, Scheme 1. The preparation of the indole substituted acetylene 8 began with iodination of indole followed by immediate methyl carbamate protection of the sensitive indole providing 10. Stepwise Sonogashira coupling of 10 first with trimethylsilylacetylene (82%), TMS deprotection of 11 (Bu₄NF, THF, 80%), and subsequent coupling of the resulting acetylene 12 again with 10 provided 8 (85%).

Scheme 1.

Two syntheses of 1 and 2.

The Diels–Alder reaction of acetylene 8 with 9 provided 13 (65%) in a reaction that proved sluggish requiring 15 d in refluxing toluene (110 °C) with repetitive additions of the 1,2,4,5-tetrazine 9 every 3 d as it slowly decomposes at this temperature. Use of higher reaction temperatures simply accelerated the decomposition of the 1,2,4,5-tetrazine 9 and did not lead to improvements in the rate or conversions to 13. Notably and although this result merits the examination of alternative approaches to the preparation of the 1,2-diazine 13, it was not as unsuccessful as reported by Fu and Gribble.⁹ In fact, such 1,2- diarylacetylenes exhibit a reactivity that is dependent on the electronic character of the aryl groups. For example, although alkoxyphenyl substituents convey sufficient reactivity to such alkynes making their use synthetically attractive,12d,e,g the unsubstituted diphenylacetylene itself reacts with 9 only slowly. We found that 8 exhibits a reactivity that is slightly lower than that of diphenylacetylene, and that it not as reactive as a number of more productive acetylenic dienophiles.

The acetylene adopted for an alternative approach to 13 was 1,2-bis(tributylstannyl)acetylene (14).¹⁴ The reaction of 14 with dimethyl 1,2,4,5-tetrazine-3,6-dicarboxylate (9) proceeded smoothly in dioxane under mild thermal conditions (45 °C, 24 h) and provided the Diels–Alder product 15 in exceptional conversions (97%). Subsequent Stille coupling of 10 with the resulting 1,2-diazine 15 proceeded effectively and twice providing the same key 4,5-bis(indol-3-yl)-1,2-diazine 13 in good yield (70%). In the optimization of this reaction, (Ph₃P)₂PdCl₂ proved more effective than (Ph₃P)₄Pd, the addition of CuI or CuCl₂ improved the initially modest conversions, and the additional inclusion of LiCl further improved the reaction eliminating a side reaction of proto deiodination.

Treatment of 13 with Zn/HOAc (30 equiv Zn, HOAc–CH₂Cl₂ 1:1, 25 °C, 12 h) cleanly effected the reductive ring contraction reaction providing pyrrole 16 (68%) and completing the 1,2,4,5-tetrazine → 1,2-diazine → pyrrole conversions originally envisioned. Selective removal of the indole N-methoxylcarbonyl groups under mild conditions (2 equiv of LiOH, MeOH/THF/H₂O 2:2:1, 24 °C, 12 h) provided lycogarubin C (1) in good to excellent conversion (65–89%), whereas exhaustive hydrolysis of 16 (7 equiv of KOH, THF/H₂O 1:1, 45 °C, 24 h) or hydrolysis of 1 (3.5 equiv of KOH, THF/H₂O 1:1, 45 °C, 16 h) afforded lycogalic acid (2) in superb conversion (95%).

Thus, two complementary syntheses of 1 and 2 based on a 1,2,4,5-tetrazine → 1,2-diazine → pyrrole Diels–Alder strategy using acetylenic dienophiles are disclosed that extend our use of heterocyclic azadiene Diels–Alder reactions^12,15 to a key biosynthetic precursor to the indolo[2,3-a]carbazole alkaloids.