DOI: 10.1002/chem.201405980

Communication

& Total Synthesis

Synthesis of ()-Spongiolactone Enabling Discovery of a More Potent Derivative Natalie L. Harvey,[a] Joanna Krysiak,[b] Supakarn Chamni,[a, c] Sung Wook Cho,[a, d] Stephan A. Sieber,[b] and Daniel Romo*[a] cells (K562, IC50 12  1 mm) and potential immunosuppressive activity through inhibition of human peripheral blood mononuclear cells (PBMC, IC50 30  10 mm).[2] Despite the presence of the resident b-lactone, with potential as a protein-reactive, acylating pharmacophore, neither synthetic nor mechanism-ofaction studies have been reported for the spongiolactones.

Abstract: An eleven-step synthesis of ()-spongiolactone from 1,3-cyclohexanedione is reported that relies on a diastereoselective, nucleophile-catalyzed aldol lactonization (NCAL) process with an advanced ketoacid intermediate that installed the anticipated b-lactone pharmacophore of the natural product. In addition, a stereoselective cyclohexenyl zinc addition to a substituted cyclohexanone simultaneously installed two fully substituted vicinal stereocenters. The reported synthesis enabled preliminary structure–activity studies that revealed a regio- and stereoisomeric derivative of spongiolactone with greater antiproliferative activity towards a leukemia (K562) cell line. Furthermore, unusual antiproliferative selectivity of these spongiolactone derivatives toward the K562 cell line was observed with no inhibition of the breast, liver, and lung cancer cell lines tested.

Discovered in 1974, the spongiane class of marine natural products, comprised of a general tetracyclic skeleton (1), provides an interesting array of structural diversity, for example, gracilin A (2).[1] However, while this family of natural products continues to expand,[2] much remains to be gleaned by increasing the accessibility of these compounds. Presenting a highly rearranged spongiane skeleton and a unique fused blactone, spongiolactone (3) was first isolated in 1986 by Sica and co-workers from the sponge Spongionella gracilis (Figure 1).[3] In 2009, a related congener was isolated by Jaspars, C3’-norspongiolactone (4), differing only by a methyl group in the side-chain ester, and found to possess antiproliferative activity toward human chronic mylogenous leukemia

Figure 1. General spongiane skeleton (1) and family members: gracilin A (2), spongiolactone (3), and C3’-norspongiolactone (4).

In line with our group’s interest in the synthesis of b-lactone-containing natural products and subsequent mechanism of action studies,[4] we targeted spongiolactone. In particular, we envisioned an opportunity to employ our previously described nucleophile-catalyzed, aldol lactonization (NCAL) process[5] on a late-stage, highly functionalized aldehyde acid. Furthermore, given the presence of the b-lactone embedded within a compact and stereochemically rich tricyclic core, we recognized the potential of this natural product as an activitybased probe.[6] Herein, we report the first total synthesis of spongiolactone which features, in addition to its brevity, a complex cyclohexenyl zinc addition to a cyclohexanone to install a stereogenic, quaternary carbon center and preliminary structure–activity relationship (SAR) data for this natural product and derivatives. Our retrosynthesis of spongiolactone (3) commenced with a late-stage NCAL process with racemic aldehyde acid 5 (Scheme 1). The required aldehyde acid ()-5 would be secured through standard functional group manipulations following a challenging SE’ cyclohexenyl zinc addition to a-substituted ketone ()-7 with the stereochemical outcome predicated on a chair-like, closed transition state arrangement[7] guided

[a] N. L. Harvey, Dr. S. Chamni, Dr. S. W. Cho, Prof. Dr. D. Romo Department of Chemistry, Texas A&M University P.O. Box 30012, College Station, TX 77842 (USA) E-mail: [email protected] Homepage: http://www.chem.tamu.edu/rgroup/romo/index.html [b] Dr. J. Krysiak, Prof. Dr. S. A. Sieber Department of Chemistry, TU Mnchen Lichtenbergstraße 4 D-85748 Garching (Germany) [c] Dr. S. Chamni Present address: Department of Pharmacognosy and Pharmaceutical Botany, Chulalongkorn University, Bangkok (Thailand) [d] Dr. S. W. Cho Present address: Dow Chemical Co., Hwaseong, Gyeonggi-do (Korea) Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/chem.201405980. Chem. Eur. J. 2014, 20, 1 – 5

These are not the final page numbers! ÞÞ

1

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Communication by the a-tert-butyl propanoate, C8-substituent (spongiolactone numbering). The trisubstituted, all anti-cyclohexanone ()-7 would be accessed through a Michael addition expected to occur anti to the existing g-benzyloxymethylene C13-substituent and subsequent protonation/equilibration at C8. a-Alkylation of enol ether 9, derived from 1,3-cyclohexanedione, with benzyloxymethyl chloride followed by application of a Stork– Danheiser enone synthesis would deliver cyclohexenone ()-8. The synthesis commenced with a scalable two-stage, onepot sequence involving conversion of 1,3-cyclohexanedione (10) to b-keto enol ether 9 through an initial Michael addition

with tert-butyl acrylate followed by O-alkylation in 90 % overall yield (Scheme 2). Application of a kinetic alkylation with benzyl chloromethyl ether exclusively afforded a-benzyloxymethylene adduct ()-11 through careful control of reaction temperature to ensure kinetic deprotonation. A Stork–Danheiser cyclohexenone synthesis[8] involving Luche reduction[9] and acidic workup directly provided cyclohexenone ()-8. Addition of isobutenyl cuprate to this enone in the presence of Me3SiCl provided a single diastereomer (> 19:1 by 500 MHz 1H NMR spectroscopy) of an intermediate silylenol ether[10] which was typically directly desilylated through an acidic workup to yield the anti,anti-trisubstituted cyclohexanone ()-7 as an inseparable 7:1 mixture of diastereomers, epimeric at the C8-stereocenter. With the desired trisubstituted cyclohexanone ()-7 in hand, we next studied the challenging SE’ addition of cyclohexenyl zinc reagent 12 c that would install two adjacent quaternary carbons including one all-carbon stereocenter. Commercially available cyclohexenol 12 a was converted to the labile allyl chloride 12 b through standard conditions and this intermediate was utilized directly for cyclohexenyl zinc formation as attempts to purify led to decomposition. Various methods were studied for allyl zinc formation,[7] however many of these procedures failed to generate active zinc reagent as evidenced by titration with an iodine solution as described previously.[7a] Neither attempted activation of Zn0 powder with HCl[12] nor flame drying both Zn0 and LiCl under vacuum led to serviceable concentrations of allyl zinc reagent (< 0.01 m). The addition of LiCl was previously shown by Knochel to facilitate organozinc formation.[13] Addition of Me3SiCl has also commonly been employed to assist with organozinc formation,[14] however the exclusion of this additive ultimately enabled preparation of consistent concentrations (0.2–0.5 m in THF) of the targeted zinc reagent ()-12 c. Numerous attempts to add cyclohexenyl zinc

Scheme 1. Retrosynthetic analysis of spongiolactone employing a late-stage NCAL process with aldehyde acid ()-5 and a complex cyclohexenyl zinc addition to cyclohexanone ()-7.

Scheme 2. Synthesis of spongiolactone ()-3 from 1,3-cyclohexanedione. Inset: the ORTEP representation of the single crystal X-ray structure of ()-13; thermal ellipsoids shown at 50 % probability. The isovalerate group (R) is removed for clarity.[11] NCS = N-chlorosuccinimide; HMPA = hexamethylphosphoramide.

&

&

Chem. Eur. J. 2014, 20, 1 – 5

www.chemeurj.org

2

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Communication ()-12 c to cyclohexanone ()-7 under a variety of reaction conditions were unsuccessful with elevated temperatures leading to reaction with both the ketone and ester of ()-7. Following extensive optimization studies, we eventually found that in situ generation of the allyl zinc ()-12 c in the presence of ketone ()-7 led to the greatest chemoselectivity and conversion. Under these conditions, using a large excess of zinc reagent, the desired tertiary alcohol ()-6 bearing two vicinal fully substituted centers was obtained as an inseparable 4:1 mixture of diastereomers in 35 % yield. While yields are modest, ketone ()-7 (56 %) could be easily recovered and recycled. Some degree of facial selectivity is anticipated for the major diastereomer of ()-7, as shown for one enantiomer of both ketone and allyl zinc reagent, proceeding through cyclic transition state A, as previously proposed for related additions to cyclohexanones.[7] This transition state arrangement allowed us to propose the stereochemistry of new stereocenters C9 and C10 as shown in alcohol ()-6 which was verified by X-ray crystallography following subsequent transformations leading to a crystalline intermediate (vide infra). Hydrogenation with Pearlman’s catalyst enabled chemoselective reduction of the cyclohexene with concomitant hydrogenolysis of the benzyl ether of ()-6 delivering a primary alcohol, which was directly esterified with isovaleryl chloride to provide ester ()-13. The latter intermediate was crystalline and enabled verification of the relative stereochemistry by single crystal X-ray analysis (see inset, Scheme 2). Conditions to selectively cleave the tert-butyl ester over the isovaleric ester were identified employing TiCl4,[15] and subsequent ozonolysis afforded aldehyde acid ()-5. With aldehyde acid ()-5 in hand, we subjected this complex substrate to the NCAL process. In situ activation of the carboxylic acid with modified Mukaiyama’s reagent, 2-bromoN-n-propyl-pyridinium-1,1,1-trifluoromethanesulfonate[5b] and triethylamine, serving as both the nucleophile (Lewis base) and Brønsted base, provided the desired tricyclic b-lactone ()-14 along with its separable diastereomer ()-15 in a 3:1 ratio (39 %), respectively. The final step involved dehydration of tertiary alcohol ()-14 employing thionyl chloride to afford spongiolactone ()-3 along with the alkene regioisomer ()-16 in an approximately 2.7:1 ratio favoring the more substituted alkene found in the natural product (Scheme 2). Characterization data for synthetic spongiolactone correlated well with data previously reported for both spongiolactone[3] and norspongiolactone.[2, 16] We next studied the antiproliferative activity of synthetic spongiolactone and congeners, possessing the presumed blactone pharmacophore, toward various cancer cell lines. While no activity was observed against HepG2 (liver hepatocellular carcinoma), A549 (lung adenocarcinoma), or MCF-7 (breast adenocarcinoma) cells, the K562 (human chronic myelogenous leukemia) cell line was sensitive to optically active spongiolactone ((+)-3)[17] as previously observed for 3’-norspongiolactone (Table 1).[2] Significant differences in IC50 values were observed between (+)-spongiolactone ((+)-3, 129  10 mm) and its alkene regioisomer (+)-16 (> 500 mm), however, note the regioisomer presents a topologically distinct trans-fused [4.3.0] bicyChem. Eur. J. 2014, 20, 1 – 5

www.chemeurj.org

These are not the final page numbers! ÞÞ

clic system. Most noteworthy is the increased potency of the regioisomeric, bis-epimeric (C6, C15) analogue ()-17 (29  6 mm, entry 3) relative to the optically active natural product. Simpler tricyclic b-lactones ()-18 and diastereomeric ()-19, which retain the expected pharmacophore,[18] were inactive (entries 4 and 5). On the other hand, reintroduction of the isovalerate side chain in derivative ()-20 restored bioactivity (entry 6), despite being devoid of the cyclohexyl substituent and also mindful that bioactive analogues 17 and 20 are racemic.

Table 1. Cytotoxicity of (+)-spongiolactone and derivatives against the K562 (human chronic myelogenous leukemia) cell line.

Entry

b-Lactone[a]

IC50  SEM [mm][b]

1 2 3 4 5 6

spongiolactone ((+)-3) regio spongiolactone ((+)-16) regio, bis-epi spongiolactone (()-17) benzyl ether ()-18 diastereomeric benzyl ether ()-19 isovaleric ester ()-20

129  10 > 500 29  4 > 500 > 500 297  49

[a] For details of the synthesis of spongiolactone derivatives 16–20 see the Supporting Information. [b] All IC50 values were determined by a MTT metabolic activity assay performed in triplicate. MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide. SEM = standard error of the mean.

In summary, we completed the first total synthesis of spongiolactone (3), in eleven steps from 1,3-cyclohexanedione, that includes application of the NCAL process with an advanced, complex aldehyde acid providing the tricyclic b-lactone core of spongiolactone. Furthermore, in situ generation of a cyclohexenyl zinc reagent enabled addition to a sterically hindered cyclohexanone with concomitant installation of two vicinal, fully substituted stereocenters, including an all-carbon quaternary center. Preliminary SAR investigations of (+)-spongiolactone and congeners led to identification of a regioisomeric, bis-epimeric analogue ()-17 that has greater potency than the natural product and also a simplified, bioactive tricyclic, b-lactone core devoid of the cyclohexyl substituent. The selective antiproliferative activity of spongiolactone and congeners toward K562 cells compared to liver, breast, and lung carcinoma cell lines (> 500 mm) is noteworthy. The described modular and concise synthetic strategy will facilitate further SAR studies and biological target discovery through activity-based protein profiling, with spongiolactone-derived cellular probes. These studies will be reported in due course.

3

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

&

&

Communication Acknowledgements This work was supported by NSF (CHE-1112397), the Welch Foundation (A-1280) and ERC starting (SAS) grants. We thank Professors Sica and Jaspars for NMR spectra of spongiolactone and norspongiolactone, respectively, and Ms. Sandra Fiorentini for technical assistance.

[6]

[7]

[8] [9] [10] [11]

Keywords: allyl zinc reagent · beta-lactone · cytotoxicity · spongiane diterpenoid · total synthesis [1] a) R. A. Keyzers, P. T. Northcote, O. A. Zubkov, Eur. J. Org. Chem. 2004, 419 – 425; b) R. A. Keyzers, P. T. Northcote, M. T. Davies-Coleman, Nat. Prod. Rep. 2006, 23, 321 – 334; c) M. Gonzalez, Curr. Bioact. Compd. 2007, 3, 1 – 36; d) L. Mayol, V. Piccialli, D. Sica, Tetrahedron 1986, 42, 5369 – 5376; e) V. Piccialli, D. Sica, J. Nat. Prod. 1987, 50, 915 – 920; f) L. Mayol, V. Piccialli, D. Sica, Tetrahedron Lett. 1985, 26, 1253 – 1256; g) L. Mayol, V. Piccialli, D. Sica, Tetrahedron Lett. 1985, 26, 1357 – 1360. [2] M. E. Rateb, W. E. Houssen, M. Schumacher, W. T. Harrison, M. Diederich, R. Ebel, M. Jaspars, J. Nat. Prod. 2009, 72, 1471 – 1476. [3] L. Mayol, V. Piccialli, D. Sica, Tetrahedron Lett. 1987, 28, 3601 – 3604. [4] a) G. Ma, M. Zancanella, Y. Oyola, R. D. Richardson, J. W. Smith, D. Romo, Org. Lett. 2006, 8, 4497 – 4500; b) H. Nguyen, G. Ma, T. Gladysheva, T. Fremgen, D. Romo, J. Org. Chem. 2011, 76, 2 – 12; c) H. Nguyen, G. Ma, D. Romo, Chem. Commun. 2010, 46, 4803 – 4805; d) G. Ma, H. Nguyen, D. Romo, Org. Lett. 2007, 9, 2143 – 2146. [5] a) G. S. Cortez, R. L. Tennyson, D. Romo, J. Am. Chem. Soc. 2001, 123, 7945 – 7946; b) S. H. Oh, G. S. Cortez, D. Romo, J. Org. Chem. 2005, 70, 2835 – 2838; c) H. Henry-Riyad, C. Lee, V. C. Purohit, D. Romo, Org. Lett. 2006, 8, 4363 – 4366; d) R. J. Duffy, K. A. Morris, R. Vallakati, W. Zhang, D. Romo, J. Org. Chem. 2009, 74, 4772 – 4781; e) K. A. Morris, K. M. Arendt,

&

&

Chem. Eur. J. 2014, 20, 1 – 5

www.chemeurj.org

[12] [13] [14] [15] [16] [17]

[18]

S. H. Oh, D. Romo, Org. Lett. 2010, 12, 3764 – 3767; f) C. A. Leverett, V. C. Purohit, D. Romo, Angew. Chem. Int. Ed. 2010, 49, 9479 – 9483; Angew. Chem. 2010, 122, 9669 – 9673; g) G. Liu, M. E. Shirley, D. Romo, J. Org. Chem. 2012, 77, 2496 – 2500. a) M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279 – 3301; b) T. Bçttcher, S. A. Sieber, Angew. Chem. Int. Ed. 2008, 47, 4600 – 4603; Angew. Chem. 2008, 120, 4677 – 4680. a) H. Ren, G. Dunet, P. Mayer, P. Knochel, J. Am. Chem. Soc. 2007, 129, 5376 – 5377; b) G. Dunet, P. Mayer, P. Knochel, Org. Lett. 2008, 10, 117 – 120. G. Stork, R. L. Danheiser, J. Org. Chem. 1973, 38, 1775 – 1776. J. L. Luche, J. Am. Chem. Soc. 1978, 100, 2226 – 2227. B. H. Lipshutz, C. Hackmann, J. Org. Chem. 1994, 59, 7437 – 7444. CCDC-970423 (()-13) contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_ request/cif. L. F. Fieser, M. Fieser, Reagents for Organic Synthesis Vol. 1, Wiley, New York, 1967, p. 1276. A. Metzger, M. A. Schade, P. Knochel, Org. Lett. 2008, 10, 1107 – 1110. G. Picotin, P. Miginiac, J. Org. Chem. 1987, 52, 4796 – 4798. M. Valencic, T. van der Does, E. de Vroom, Tetrahedron Lett. 1998, 39, 1625 – 1628. See Supporting Information for details. Optically active spongiolactone ((+)-3) and its regioisomer ((+)-16) were obtained through a related but unoptimized NCAL-based, kinetic resolution. Details of these studies will be described in a full account of this work elsewhere. A. F. Kluge, R. C. Petter, Curr. Opin. Chem. Biol. 2010, 14, 421 – 427.

Received: November 6, 2014 Published online on && &&, 0000

4

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

ÝÝ These are not the final page numbers!

Communication

COMMUNICATION & Total Synthesis N. L. Harvey, J. Krysiak, S. Chamni, S. W. Cho, S. A. Sieber, D. Romo* && – && Protein acylating potential! The first total synthesis of spongiolactone was accomplished by a late-stage, nucleophile-catalyzed aldol lactonization (see scheme). While several simplified tricyclic b-lactones did not exhibit antiproliferative activity, an unnatural isomer

Chem. Eur. J. 2014, 20, 1 – 5

showed greater potency than the natural product. The described synthesis and structure–activity studies set the stage for activity-based protein profiling to identify cellular targets modified by these protein acylating agents.

Synthesis of ()-Spongiolactone Enabling Discovery of a More Potent Derivative

5

 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.chemeurj.org

These are not the final page numbers! ÞÞ

&

&

Synthesis of (±)-spongiolactone enabling discovery of a more potent derivative.

An eleven-step synthesis of (±)-spongiolactone from 1,3-cyclohexanedione is reported that relies on a diastereoselective, nucleophile-catalyzed aldol ...
527KB Sizes 0 Downloads 10 Views