Synthesis 2015; 47(11): 1547-1556
DOI: 10.1055/s-0034-1378698
feature
© Georg Thieme Verlag Stuttgart · New York

Total Syntheses of (–)-Alstolucines A, B, and F, (–)-Echitamidine, and (–)-N-Demethylalstogucine

Christiana N. Teijaro
Department of Chemistry, Temple University, 1901 N. 13th St., Philadelphia, PA 19122, USA   Email: randrade@temple.edu
,
Senzhi Zhao
Department of Chemistry, Temple University, 1901 N. 13th St., Philadelphia, PA 19122, USA   Email: randrade@temple.edu
,
Praveen Kokkonda
Department of Chemistry, Temple University, 1901 N. 13th St., Philadelphia, PA 19122, USA   Email: randrade@temple.edu
,
Rodrigo B. Andrade*
Department of Chemistry, Temple University, 1901 N. 13th St., Philadelphia, PA 19122, USA   Email: randrade@temple.edu
› Author Affiliations
Further Information

Publication History

Received: 04 February 2015

Accepted after revision: 09 March 2015

Publication Date:
09 April 2015 (online)

 


Abstract

The first enantioselective total syntheses of (–)-alstolucinces A, B, and F, (–)-echitamidine, and (–)-N-demethylalstogucine are reported. This article details the development of our first- and second-generation approaches toward the ABCE tetracyclic core of the strychnos alkaloids and the application thereof to the aforementioned targets. Key steps involve our sequential one-pot biscyclization method that constructs the C and E rings of the tetracyclic core and Rawal’s application of the intramolecular Heck reaction to secure the pentacyclic framework common amongst all targets.


#

Biographical Sketches

Zoom Image

Christiana Teijaro obtained her B.A. in chemistry and psychology in 2012 from St. Mary’s College, St. Mary’s City, MD. She is a National Science Foundation Graduate Research Fellow currently pursuing her Ph.D. in organic chemistry under the supervision of Prof. Andrade at Temple University in Philadelphia, PA. Her research is centered on the total synthesis of indole alkaloids and understanding their function in biological systems. Her interests consist of methodology development applied to total synthesis, medicinal chemistry, and chemical biology.

Zoom Image

Senzhi Zhao obtained his Bachelor of Engineering degree in Applied Chemistry from Xuzhou Normal University, Xuzhou, China. He earned his M.S. degree in organic chemistry from Nanjing University, Nanjing, China in 2007. He is currently pursuing his Ph.D. in organic chemistry under the direction of Prof. Andrade. His research interests include development of new synthetic methodology, total synthesis of natural products and medicinal chemistry.

Zoom Image

Praveen Kokkonda obtained his B.S. in chemistry from Kakatiya University in 2003, Warangal, India, and M.S. in organic chemistry from Osmania University in 2006, Hyderabad, India. After completion of his M.S. degree, he worked as a Senior Chemist in GVK Bio sciences Pvt. Ltd, Hyderabad, India for three years. Praveen is currently a graduate student pursuing his Ph.D. under the supervision of Prof. Andrade. His research is focused on the total synthesis of indole alkaloids.

Zoom Image

Rodrigo Andrade obtained a B.A. degree in biophysics in 1996 from the Johns Hopkins University in Baltimore, MD. In 2001, he obtained his Ph.D. in organic chemistry from the Massachusetts Institute of Technology in Cambridge, MA under the supervision of Prof. Peter Seeberger. He was an NIH Postdoctoral Fellow in the laboratory of Prof. Stephen Martin at the University of Texas at Austin from 2003–2006. In August 2006, he began his independent research at Temple University in the Department of Chemistry and was promoted to Associate Professor in 2011. His research areas include chemical synthesis, chemical biology, and medicinal and bioorganic chemistry.

Indole alkaloids represent a large class of biologically active natural products. The structural diversity of these small molecules has long been a source of inspiration for the development of novel synthetic methodology and a driver of progress in total synthesis.[1]

In 2010, Kam and co-workers isolated 25 strychnos alkaloids from Alstonia spatulata of which five (i.e., alstolucines A–E) were novel.[2] The known alkaloid N-demethyl­alstogucine (5),[3] along with alstolucines A (6), B (3), and F (4) were found to reverse multidrug resistance in vincristine-resistant KB cells (Figure [1]). The structural complexity of these targets, coupled with their novelty and biological significance, motivated us to prepare these in asymmetric fashion, which we recently communicated.[4]

Zoom Image
Figure 1 ABCE core 1, known and novel pentacyclic strychnos alkaloids

In 1994, Kuehne, reported the racemic syntheses of echitamidine (24) by the stereoselective hydride reduction of alstolucines B (3). Moreover, 3 was equilibrated to alstolucine F (4) in 2:1 ratio via treatment with sodium methoxide in methanol.[5]

Inspection of the targets reveals a pentacyclic framework with a pendant group at C20 that serves to differentiate each congener. Moreover, these features are reminiscent of the ABCE tetracycle 1 and the classic alkaloid (–)-aku­ammicine (2). The building blocks and inspiration for the synthesis of (–)-alstolucines A (6), B (3), F (4), (–)-echitamidine (24), and (–)-N-demethylalstogucine (5) stemmed from our synthetic knowledge of 1 and 2.

Akuammicine (2) is a common synthetic target among monoterpenoid indole alkaloids.[6] The isolation of aku­ammicine from Picralima klaineana (i.e., akumma) seeds was first described in 1932 by Henry.[7]

Akumma seeds are widely employed in traditional medicine throughout Africa, particularly for the treatment of malaria. More recently, akuammicine (2) was shown to possess additional biological activity relating to glucose uptake,[8] μ- and κ-opioid receptors,[9] as well as cytotoxicity in vincristine-resistant KB cells.[2]

The synthesis of akuammicine (2) and other alkaloids, such as strychnine, have led to a wealth of chemistries. Our approach to the tetracyclic core of the alkaloids 1 and aku­ammicine (2) was inspired by the work of Heathcock[10] and Rawal.[11] In 1991, Rawal employed an elegant strategy to form the D ring of the pentacyclic core of the alkaloids via an intramolecular Heck reaction. In 2000, Heathcock reported the synthesis of (±)-aspidospermidine in which a 2-haloacetamide gramine derivative was employed to close the C ring. This inspiration was used for the formation of the C ring of the ABCE tetracycle 1 and the biscyclization in our approach to akuammicine (2).

Based on this inspiration a first generation asymmetric approach to akuammicine (Scheme [1]) was devised. The most efficient manner to akuammicine is to employ Rawal’s approach using the intramolecular Heck reaction as an end-game approach.

Zoom Image
Scheme 1 Retrosynthetic analyses for the synthesis of (–)-akuammicine (2)

To gain entry to intermediate 7, we envisioned using a biscyclization approach employing Heathcock’s tactic in closing ring C followed by a novel intramolecular aza-Morita­[12] or aza-Baylis–Hillman (IABH)[13] to close ring E to give the tetracyclic core; however, this would leave ring C at the amide oxidation state requiring subsequent deoxygenation. Arriving at intermediate 8 could be accomplished by N-alkylation of the primary amine 9 followed by amidation using the appropriate side chains. Chiral amine 9 could be accessed using Yus allylation chemistry[14] after removal of Ellman’s N-tert-butyl sulfinamide[15] chiral auxiliary and deprotection of the indole nitrogen.

Zoom Image
Scheme 2 Reagents and conditions: (a) (R)-N-tert-butanesulfinamide, In(0), Ti(OEt)4 then allyl bromide, THF, 87%, dr 10:1; (b) 4 M HCl, dioxane then Mg(0), MeOH, 75% over 2 steps; (c) (Z)-2-iodobut-2-enyl bromide, Cs2CO3; (d) BrCH2COCl, Et3N, 83% over 2 steps; (e) methyl acrylate, Hoveyda–Grubbs 2nd generation catalyst (10 mol%), CH2Cl2, 80%; (f) AgOTf, DTEMP then DBU, toluene, r.t., 60%; (g) Lawesson’s reagent, 87%; (h) Et3OBF4 then NaBH4, MeOH, 92% over 2 steps; (i) Pd(OAc)2, Ph3P, Et3N, 87%.

The synthesis of akuammicine (2) was accomplished starting from known compound 10 (Scheme [2]),[1] [15] which was easily synthesized from commercially available indole-3-carbaldehyde in one step.[16] (R)-N-tert-Butanesulfinamide and 10 were condensed in the presence of titanium(IV) ethoxide and indium(0). Upon the formation of the chiral imine, allyl bromide was added to effect the Barbier formation. Stereoselective addition of the resulting allyl–indium species into the imine afforded compound 11 in 87% yield with dr 10:1.[14] Treatment of 11 with hydrochloric acid removed the chiral auxiliary to give the free, primary amine. Further treatment with magnesium(0) effectively removed the tosyl group from the indole nitrogen to afford gramine 9 in 75% over two steps.

Amine 9 was alkylated with (Z)-2-iodobut-2-enyl bromide[17] and cesium carbonate. The resulting secondary amine was reacted with bromoacetyl chloride and triethylamine to yield intermediate 8 in 83% over two steps. Cross metathesis of 8 with methyl acrylate in the presence of Hoveyda–Grubbs 2nd generation catalyst[18] furnished compound 12 in 80% yield, setting the stage for the biscyclization. The first attempts at the spirocyclization of ring C using Heathcock’s original conditions, treatment of 12 with only silver(I) triflate, did not lead to product formation but instead decomposition due to triflic acid generated under the reaction conditions. To address this issue, a wide range of bases were screened [e.g., pyridine, Et3N, i-Pr2NEt, 2,6-lutidine, 2,6-di-tert-butyl-4-methylpyridine (DTBMP), NaH, and t-BuOK] with the best result being the use of 2,6-di-tert-butyl-4-methylpyridine which gave the product in 95% yield and with dr 13:1. Next, we focused on the formation of the E ring; typical conditions to effect the intramolecular aza-Morita[12] or aza-Baylis–Hillman (IABH)[13] were employed (e.g., Bu3P, DMAP, DABCO, Et3N, i-Pr2NEt) in various solvents with no product formation. It was only upon treatment of the tricycle with DBU that intermediate 7 was formed in 90% yield. Due to the fact that DBU was the only reagent capable of effecting this transformation, coupled with the fact that it is the most basic, led us to hypothesize an alternative mechanism, namely an intramolecular vinylogous Mannich reaction followed by an isomerization. In 2012, Kwon reported a similar intramolecular Morita reaction using trimethylphosphine. The use of DBU in their system, despite being more hindered, led to an unproductive, base-mediated elimination as opposed to cyclization.[19] Altogether, these data support the Mannich mechanism with DBU.

We were able to effect a sequential one-pot variant of the reaction (i.e., 12 to 7) in 60% overall yield. Treatment of 7 with Lawesson’s reagent afforded thioamide 13 in 87% yield. Chemoselective alkylation of sulfur with triethyloxonium tetrafluoroborate and subsequent sodium borohydride reduction effectively removed the sulfur to yield 14 in 92% over two steps.[20] Of mention is the fact the use of trimethyloxonium tetrafluoroborate resulted in partial N-methylation of the indoline nitrogen. Finally, the intramolecular Heck reaction of enoate 14 using a modification of Rawal’s conditions, namely palladium(II) acetate, triphenylphosphine, and triethylamine as solvent, afforded aku­ammicine (2) in 87%.

Zoom Image
Scheme 3 Retrosynthetic analyses for the synthesis of (–)-akuammicine (2)

The need for the reduction of lactam 7 to pyrrolidine 14 afforded an opportunity to avoid this redox step and streamline our approach. To this end, we devised a second-generation asymmetric synthesis of ABCE tetracycle 1.[21] Inspiration for this new route (Scheme [3]) came from Ellman’s asymmetric synthesis of (–)-aurantioclavine.[22] Specifically, in a key step it was found that 3-(2-hydroxyethyl)indoles quickly undergo cyclization to form spiro[cyclopropyl]indolenines using Mitsunobu conditions.[21] We hypothesized that Mitsunobu activation of N-Boc-N-(2-hydroxyethyl)-substituted gramine 15 would access the desired spiroindolenine with the correct oxidation state on the C ring (Scheme [3]).

Following the first-generation approach, the late stage intramolecular Heck reaction would be applied after alkylating ABCE tetracycle 1. We envisioned arriving at 1 via a revised biscyclization of intermediate 15 where first ring C would form under Mitsunobu conditions followed by an intramolecular Mannich reaction to construct the E ring. Intermediate 15 could be readily made from 16 by N-Boc protection then cross metathesis. Intermediate 16 would in turn be accessed from 9, which we had already synthesized in the first-generation approach using the Yus allylation strategy.[14]

Starting from gramine 9, various alkylation reactions were attempted using substrates such as 2-bromoethanol and various other ethanol derivatives with little success. Recourse to reductive amination ultimately proved successful (Scheme [4]).[1] [4] Ethyl glyoxalate and 9 were condensed in the presence of 4-Ǻ molecular sieves to form the imine, which was then reduced using lithium aluminum hydride to give amino alcohol 16. Finally, N-Boc protection of 16 using di-tert-butyl carbonate and Hünig’s base afforded 17 in 57% yield over two steps.

Zoom Image
Scheme 4 Reagents and conditions: (a) 50% ethyl glyoxalate in toluene then LiAlH4; (b) Boc2O, i-Pr2NEt, CH2Cl2, 57% over 2 steps; (c) methyl acrylate, Hoveyda–Grubbs 2nd generation catalyst (10 mol%), CH2Cl2, 80%; (d) Ph3P, DEAD then DBU, 80 °C, 12 h, 56%; (e) TFA, CH2Cl2, quant.; (f) (Z)-2-iodobut-2-enyl bromide, K2CO3, MeCN, 71%; (g) Pd(OAc)4, Ph3P, Et3N, 87%.

Alcohol 17 was then subjected to cross metathesis with methyl acrylate using Hoveyda–Grubbs 2nd generation catalyst[18] to provide 15 in 80% yield, setting the stage for the biscyclization. Under Mitsunobu conditions,[22] 15 cyclized forming ring C and the spirocenter with full stereocontrol. Upon further treatment with DBU in the same pot, the E ring cyclized to afford Boc-protected ABCE tetracycle 18 in 56% yield. Removal of the N-Boc group using trifluoroacetic acid furnished ABCE tetracycle 1. Alkylation of 1 with (Z)-2-iodobut-2-enyl bromide generated compound 14 in 71% yield. Using Rawal’s endgame strategy of the intramolecular Heck cyclization,[11] akuammicine (2) was obtained in 87% yield.

Zoom Image
Scheme 5 Retrosynthetic approach to alstolucine B

With routes to the ABCE tetracycle 1 and akuammicine (2) in hand, we turned our focus to the alstolucines. Retrosynthetic analysis of alstolucine B (3) suggested it could be derived from 19 via oxidation (Scheme [5]). Indoline 19 in turn would be prepared via an intramolecular Michael reaction of Mannich base 20, which would be derived from tetracycle 1. Finally, access to epimeric alstolucine F (4) from 3 would be possible using Kuehne’s protocol for equilibration.[4]

The synthesis of 3 began with the site-selective intermolecular aza-Michael reaction[23] between 1 and methyl vinyl ketone (MVK) to afford Mannich base 20 in 58% yield (Scheme [6]). Cyclization of the D ring was best accomplished by treating 20 with sodium hexamethyldisilazanide at –10 °C for two hours, which furnished pentacycle 19 in 25% yield. Finally, Swern oxidation of indoline 19 delivered alstolucine B (3) in 20% yield.

Zoom Image
Scheme 6 Reagents and conditions: (a) MVK (1.0 equiv), MeOH, –78 °C, 3 h, 58%; (b) 1 M NaHMDS in THF (0.95 equiv), THF, –10 °C, 2 h, 25%; (c) oxalyl chloride (2.4 equiv), DMSO (4.2 equiv), Et3N (5.0 equiv), –60 °C to r.t., 2 h, 20%.

Numerous conditions were screened to optimize the intramolecular Michael addition of 20 [24] (Table [1]). Upon determining the effects of substrate concentration, addition rate, addition sequence, solvents, equivalents of base, and counterions, the best conditions found were the addition of 1 M sodium hexamethyldisilazanide in tetrahydrofuran (0.95 equiv) to the substrate (0.01 M in THF) over 10 minutes at –2 °C to 0 °C. Upon stirring for two hours at this temperature, compounds 19a and 19b were formed as a mixture of two inseparable diastereomers in 25% yield with a ratio of 1.4:1 to 1.7:1. Furthermore, if a protecting group such as the benzyloxycarbonyl (Z) group was used on the indoline nitrogen prior to the Michael addition, the resulting diastereomers at C16 were separable. The diastereomers differed in the stereochemical configuration at C16 as determined through NOE based on the known stereocenters (Figure [2]). Ultimately this is of no consequence as the stereocenter is destroyed following oxidation.

Table 1 Intramolecular Michael Addition Conditions

Entry

Base

Solvent

Temp (°C)

Time (h)

Yield (%)

1

DBU

THF

r.t.

48

a

2

NaOMe

MeOH

0 to r.t.

 5

a

3

Triton B (t-Bu4NOH)

DME

reflux

16

a

4

BEMPb

THF

r.t.

24

a

5

pyrrolidine

MeCN

r.t.

48

a

6

NaH

THF

 70

 5

20

7

NaHMDS

THF

–78 to r.t.

15

a

8

NaHMDS

THF

–10 to 0

 2

5–25

a Not observed.

b 2-(tert-Butylimino)-2-(diethylamino)-1,3-dimethylperhydro-1,3,2-diazaphosphorine.

In an attempt to optimize the D ring cyclization, we studied the effects of protecting the indole nitrogen. It was determined that addition of either a methyl, ethyl, or benzyl carbamate on the indole nitrogen (not shown) did slightly increase the yield of the intramolecular Michael addition to 29%, 39%, and 60% respectively. At this stage, we turned our attention to indoline 19 oxidation (Table [2]).

Zoom Image
Figure 2 NOE data for compound 19b

Table 2 Indoline Oxidation

Entry

Oxidant

Solvent

Temp (°C)

Time (h)

Yield (%)

1

Pb(OAc)4

CH2Cl2

–10

 1

a

2

DDQ

dioxane

r.t., reflux

18

a

3

PdCl2, Et3N

MeOH

r.t.

96

traceb

4

DMS, NCS, Et3N

CH2Cl2

–78

 4

traceb

5

(COCl)2, DMSO, Et3N

CH2Cl2

–78 to r.t.

 1.5

traceb

6

(COCl)2, DMSO, Et3N

CH2Cl2

–60 to 0

 2

20

a Not observed.

b Observed by LC-MS.

Previous work on the oxidation on indolines with similar core ABCE structures had shown modest to good yields when using lead tetraacetate or 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) as the oxidants.[25] Upon trying these conditions on substrate 19, varying the equivalents of oxidant, temperature, and reaction time, no product formation was observed. At this time we turned to oxidation conditions shown to oxidize simple indoline systems employing palladium(II) chloride, Corey–Kim, and Swern conditions, all showing trace amounts of product formation.[17b] [26] Upon optimizing the Swern oxidation conditions, particularly temperature, alstolucine B (3) was prepared in 20% yield.

The low yields plaguing our synthetic route to the alstolucines justified a newer, more efficient approach. As we had an efficient route to structurally similar akuammicine (2) in place, we reasoned that manipulations of the ethylidene in 2 would be the most straightforward route to accessing the alstolucines. A concise route would certainly feature anti-Markovnikov hydroboration followed by a subsequent oxidation; however, Levy and co-workers had shown that akuammicine (2) under standard hydroboration conditions underwent Markovnikov addition.[27] Numerous attempts to alter the hydroboration conditions using bulkier boranes, such as thexylborane,[28] were unsuccessful with no product formation. In addition, one might envision using ozonolysis on the pendent alkene followed by a Wittig or Horner–Wadsworth–Emmons reaction[29] to install the pendant ketone; however, these attempts were ineffective with only the ozonolysis followed by the Horner–Wadsworth–Emmons forming a trace amount of alstolucine B (3) or F (4). With this in mind, we envisioned a more conservative, second-generation route (Scheme [7]) consisting of dihydroxylation, oxidation of the resulting secondary alcohol followed by selective deoxygenation of the tertiary alcohol to afford alstolucine B (3).

Zoom Image
Scheme 7 Second-generation retrosynthesis of alstolucine B

To realize our new approach (Scheme [8]),[4] akuammicine (2) was subjected to Upjohn dihydroxylation conditions to afford 19,20-dihydroxyakuammicine (22) in 86% yield.[30] The resulting secondary alcohol was oxidized using Corey–Kim conditions to give 19-hydroxyalstolucine B (21) in 66% yield.[31] Employing elegant work by Molander on the acyloin reduction,[32] we set out to deoxygenate the α-hydroxy group using samarium diiodide chemistry. After initial attempts to reduce the free hydroxy were unsuccessful (i.e., the more step-efficient option), recourse was made to the α-acetoxy variant by acylating the acyloin using acetic anhydride, triethylamine, and catalytic 4-(dimethylamino)pyridine to access 19-acetoxyalstolucine B (23) in 95% yield. Subjecting 23 to samarium diiodide furnished a readily separable mixture of alstolucines B (3) and F (4) in 1.6:1 ratio with an overall yield of 71%.

Zoom Image
Scheme 8 Reagents and conditions: (a) NMO (5 equiv), OsO4 (10 mol%), t-BuOH–THF–MeOH (3:2:1), 18–36 h, 86%; (b) NCS (1.5 equiv), DMS (1.65 equiv), Et3N (2.0 equiv), CH2Cl2, 0 °C, 3 h, 66%; (c) Ac2O (1.1 equiv), Et3N (1.2 equiv), DMAP (0.1 equiv), 0 °C to r.t., 18 h, 95%; (d) 1.0 M SmI2 in THF (5.0 equiv), MeOH–THF (1:2), –78 °C to r.t., 71%; (e) NaBH4, MeOH, 85%; (f) NaBH4, CeCl3·7H2O, MeOH, 83%; (g) EtCO2Cl, Et3N, CH2Cl2, 73%.

Subsequent reduction of alstolucine B (3) with sodium borohydride resulted in the formation of (–)-echitamidine (24) as a single diastereomer in 85% yield. The Luche reduction of alstolucine F (4) led to (–)-N-demethylalstogucine (5) in 83% as a single diastereomer.[33] Upon reaction of 5 with ethyl chloroformate in the presence of triethylamine, alstolucine A (6) was formed in 73% yield.

In summary, we have achieved the first enantioselective total syntheses of (–)-alstolucines A (6), (–)-B (3), and (–)-F (4), (–)-echitamidine (24), and (–)-N-demethylalstogucine (5) using our second-generation biscyclization method for the synthesis of ABCE tetracycle 1 and a dihydroxylation/acyloin reduction sequence to modify the ethylidene side chain of akuammicine (2). The newly synthesized alkaloids, 36, were shown to resensitize vincristine-resistant KB cells to vincristine[2] and further investigation into the biological activity of 36 is currently being pursued. Those results will be reported in due course.

All reactions containing air or water sensitive reagents were performed in flame-dried or oven-dried glassware under an argon or N2 atmosphere. CH2Cl2 and THF were passed through two columns of neutral alumina and toluene was passed through one column of neutral alumina and one column of Q5 reactant. Prior to use, methyl acrylate was distilled, Et3N was distilled from CaH2 and 4-Å molecular sieves were activated by flame-drying under vacuum. For cross-metathesis­ and SmI2 reactions, all solvents were deaerated by bubbling argon through for at least 1 min/mL. (Z)-2-Iodobut-2-enyl bromide was prepared according to the procedure of Cook.[17b] All other reagents and solvents for workup procedures were purchased from commercial sources and used without further purification. TLC was performed on Analtech 60F254 silica gel plates. Detection was performed using UV light, KMnO4 stain, PMA stain, and subsequent heating. Flash column chromatography was performed according to the procedure of Still[34] using ICN Silitech 32-63 D 60Ǻ silica gel with the indicated solvents. 1H and 13C NMR spectra were recorded at the indicated field strength in CDCl3 at r.t. and internally referenced to residual solvent signals.


#

Methyl 3-(3-Oxobutyl)-2,3,3a,4,6,7-hexahydro-1H-pyrrolo[2,3-d]carbazole-6-carboxylate (20)

To a stirred solution of 1 (670 mg, 2.48 mmol) in anhydrous MeOH (20 mL) at –78 °C was added a solution of methyl vinyl ketone (0.209 mL, 2.48 mmol) in anhydrous MeOH (20 mL) over 15 min. The mixture was stirred at –78 °C for a further 3 h. The reaction was quenched with H2O (10 mL) at –78 °C. The resulting mixture was concentrated in vacuo to remove MeOH, followed by extraction with CH2Cl2 (2 × 10 mL). The combined organic phases were dried (Na2SO4) and concentrated under vacuum. The residue was purified by flash column chromatography (MeOH–CH2Cl2, 1:100 to 3:100) to give 20 (490 mg, 58%) as a pale yellow gum.

IR (neat): 3392, 2952, 2362, 1704, 1250, 1098, 743 cm–1.

1H NMR (500 MHz, CDCl3): δ = 7.05–7.01 (m, 3 H), 6.69 (td, J = 7.5, 1.0 Hz, 1 H), 6.56 (dd, J = 8.0, 1.0 Hz, 1 H), 4.53 (s, 1 H), 4.26 (s, 1 H), 3.76 (s, 3 H), 3.18–3.04 (m, 2 H), 2.95 (dd, J = 4.5, 2.9 Hz, 1 H), 2.66–2.51 (m, 4 H), 2.44–2.37 (m, 1 H), 2.29–2.22 (m, 1 H), 2.21–2.14 (m, 1 H), 2.13 (s, 3 H), 1.98–1.90 (m, 1 H).

13C NMR (126 MHz, CDCl3): δ = 208.23, 167.28, 150.06, 139.25, 132.55, 129.98, 128.06, 122.79, 118.46, 109.05, 63.64, 61.27, 53.29, 51.68, 50.69, 48.32, 42.55, 37.68, 30.08, 24.89.

HRMS (FAB): m/z [M + H]+ calcd for C20H25N2O3: 341.1865; found: 341.1860.


#

(±)-2,16-Dihydroalstolucine B (19)

To a stirred solution of 20 (120 mg, 0.352 mmol) in THF (30 mL) at –10 °C was added 1 M NaHMDS in THF (0.33 mL, 0.33 mmol) in THF (2 mL) over 10 min. The resulting mixture was stirred at –10 °C for a further 2 h. The reaction was quenched with sat. aq NH4Cl (10 mL) and diluted with H2O (5 mL). The resulting mixture was extracted with CH2Cl2 (2 × 10 mL). The combined organic phases were dried (Na2SO4) and concentrated under vacuum. The residue was purified by flash column chromatography (MeOH–CH2Cl2, 5:100) to give an inseparable mixture of two diastereomers (30 mg, 25%) as a pale yellow solid.

IR (neat): 3381, 2949, 1723, 1706, 1482, 1170, 742 cm–1.

1H NMR (500 MHz, CDCl3): δ = 7.07 (td, J = 7.6, 1.2 Hz, 1 H), 6.99 (d, J = 7.4 Hz, 1 H), 6.75 (td, J = 7.4, 0.9 Hz, 1 H), 6.64 (d, J = 7.6 Hz, 1 H), 4.37 (s, 1 H), 3.75 (s, 1 H), 3.73–3.72 (m, J = 8.4 Hz, 4 H), 3.35–3.30 (m, 1 H), 3.10–2.91 (m, 3 H), 2.79 (s, 1 H), 2.73 (dd, J = 13.7, 5.0 Hz, 1 H), 2.34 (dd, J = 10.1, 3.3 Hz, 1 H), 2.22–2.18 (dd, J = 12.8, 6.0 Hz, 1 H), 2.16–2.08 (m, 4 H), 1.69 (ddd, J = 12.8, 11.3, 8.0 Hz, 1 H), 1.41 (dt, J = 14.4, 2.3 Hz, 1 H).

13C NMR (126 MHz, CDCl3): δ = 208.30, 174.53, 149.34, 129.76, 128.35, 122.17, 118.96, 109.68, 62.14, 57.69, 53.52, 52.05, 51.35, 51.27, 47.55, 45.42, 38.88, 28.91, 26.74, 23.46.

HRMS (FAB): m/z [M + H]+ calcd for C20H25N2O3: 341.1865; found: 341.1862.


#

(±)-Alstolucine B (3)

A solution of DMSO (23.8 mg, 0.304 mmol) in CH2Cl2 (0.5 mL) was added under argon to a solution of oxalyl chloride (14.6 mg, 0.115 mmol) in CH2Cl2 (2 mL) at –78 °C. The resulting mixture was stirred at –78 °C for 20 min, followed by addition of a solution of 19 (26 mg, 0.1 mmol) in CH2Cl2 (0.5 mL). The stirring was continued at –78 °C for a further 20 min, and then a solution of Et3N (38 mg, 0.4 mmol) in CH2Cl2­ (0.5 mL) was added. The mixture was allowed to warm to r.t. over 2 h. Then the reaction was quenched with water (3 mL). The organic layer was separated and the aqueous layer was extracted with CH2Cl2 (2 × 5 mL). The combined organic phases were dried (Na2SO4) and concentrated under vacuum. The residue was purified by flash column chromatography (MeOH–CH2Cl2, 1:100 to 5:100) to give an intermediate that was dissolved in CH2Cl2 (5 mL) and subsequently washed with 1 M aq HCl (2 mL) and 10% aq NaOH (2 mL) to give of 3 (5 mg, 20%) as a white foam. 1H and 13C NMR spectra were identical with reported literature values.[2] [4]


#
#

Acknowledgment

We acknowledge the previous efforts of Dr. Gopal Sirasani and Mr. Surendrachary Munagala on this project. We thank Dr. Richard Pederson (Materia, Inc.) for catalyst support. This research was supported by the National Science Foundation (CHE-1111558) and the Drug Discovery Initiative (DDI) Grant from the Moulder Center for Drug Discovery, Temple University (R.B.A.). This material is based upon work supported by the National Science Foundation Graduate Research Fellowship Program under Grant No. (DEG-12262) awarded to C.N.T. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do no necessarily reflect the views of the National Science Foundation.

Supporting Information



Zoom Image
Zoom Image
Zoom Image
Zoom Image
Zoom Image
Figure 1 ABCE core 1, known and novel pentacyclic strychnos alkaloids
Zoom Image
Scheme 1 Retrosynthetic analyses for the synthesis of (–)-akuammicine (2)
Zoom Image
Scheme 2 Reagents and conditions: (a) (R)-N-tert-butanesulfinamide, In(0), Ti(OEt)4 then allyl bromide, THF, 87%, dr 10:1; (b) 4 M HCl, dioxane then Mg(0), MeOH, 75% over 2 steps; (c) (Z)-2-iodobut-2-enyl bromide, Cs2CO3; (d) BrCH2COCl, Et3N, 83% over 2 steps; (e) methyl acrylate, Hoveyda–Grubbs 2nd generation catalyst (10 mol%), CH2Cl2, 80%; (f) AgOTf, DTEMP then DBU, toluene, r.t., 60%; (g) Lawesson’s reagent, 87%; (h) Et3OBF4 then NaBH4, MeOH, 92% over 2 steps; (i) Pd(OAc)2, Ph3P, Et3N, 87%.
Zoom Image
Scheme 3 Retrosynthetic analyses for the synthesis of (–)-akuammicine (2)
Zoom Image
Scheme 4 Reagents and conditions: (a) 50% ethyl glyoxalate in toluene then LiAlH4; (b) Boc2O, i-Pr2NEt, CH2Cl2, 57% over 2 steps; (c) methyl acrylate, Hoveyda–Grubbs 2nd generation catalyst (10 mol%), CH2Cl2, 80%; (d) Ph3P, DEAD then DBU, 80 °C, 12 h, 56%; (e) TFA, CH2Cl2, quant.; (f) (Z)-2-iodobut-2-enyl bromide, K2CO3, MeCN, 71%; (g) Pd(OAc)4, Ph3P, Et3N, 87%.
Zoom Image
Scheme 5 Retrosynthetic approach to alstolucine B
Zoom Image
Scheme 6 Reagents and conditions: (a) MVK (1.0 equiv), MeOH, –78 °C, 3 h, 58%; (b) 1 M NaHMDS in THF (0.95 equiv), THF, –10 °C, 2 h, 25%; (c) oxalyl chloride (2.4 equiv), DMSO (4.2 equiv), Et3N (5.0 equiv), –60 °C to r.t., 2 h, 20%.
Zoom Image
Figure 2 NOE data for compound 19b
Zoom Image
Scheme 7 Second-generation retrosynthesis of alstolucine B
Zoom Image
Scheme 8 Reagents and conditions: (a) NMO (5 equiv), OsO4 (10 mol%), t-BuOH–THF–MeOH (3:2:1), 18–36 h, 86%; (b) NCS (1.5 equiv), DMS (1.65 equiv), Et3N (2.0 equiv), CH2Cl2, 0 °C, 3 h, 66%; (c) Ac2O (1.1 equiv), Et3N (1.2 equiv), DMAP (0.1 equiv), 0 °C to r.t., 18 h, 95%; (d) 1.0 M SmI2 in THF (5.0 equiv), MeOH–THF (1:2), –78 °C to r.t., 71%; (e) NaBH4, MeOH, 85%; (f) NaBH4, CeCl3·7H2O, MeOH, 83%; (g) EtCO2Cl, Et3N, CH2Cl2, 73%.