PDF herunterladen
Synlett 2014; 25(14): 2025-2029
DOI: 10.1055/s-0034-1378380
letter
© Georg Thieme Verlag Stuttgart · New York

A Metathesis–Acylation Approach to the Bicyclic Core of Polycyclic Poly­prenylated Acylphloroglucinols

Stefanie Schmitt
a   Institut für Organische Chemie II, Universität des Saarlandes, 66123 Saarbrücken, Germany
,
Eva Feidt
a   Institut für Organische Chemie II, Universität des Saarlandes, 66123 Saarbrücken, Germany
,
David Hartmann
a   Institut für Organische Chemie II, Universität des Saarlandes, 66123 Saarbrücken, Germany
,
Volker Huch
b   Institut für Anorganische Chemie, Universität des Saarlandes, 66123 Saarbrücken, Germany   Fax: +49(681)30264151   eMail: j.jauch@mx.uni-saarland.de
,
Johann Jauch*
a   Institut für Organische Chemie II, Universität des Saarlandes, 66123 Saarbrücken, Germany
› Institutsangaben
Weitere Informationen

Publikationsverlauf

Received: 21. März 2014

Accepted after revision: 03. Juni 2014

Publikationsdatum:
23. Juli 2014 (online)

 
 als PDF herunterladen Lizenzen und Reprints Alle Artikel dieser Rubrik


Abstract

An approach to a model compound for polycyclic polyprenylated acylphloroglucinols is developed using a ring-closing metathesis approach to give a substituted cyclooctene. This undergoes cyclization via an intramolecular acylation leading to a substituted bicyclo[3.3.1]nonan-9-one related to hyperforin, nemorosone, clusianone, garsubellin A and other members of the polyprenylated acylphloroglucinol.


#

Key words

bicyclic compounds - natural products - metathesis - alkene acylation - steric hindrance

Polycyclic polyprenylated acylphloroglucinols (PPAPs) are a class of polyketides which contains more than 210 members having interesting bicyclo[3.3.1]nonane, bicy­clo[3.2.1]octane or tricyclic core structures, with several prenyl or geranyl substituents and one acyl group at various positions.[1] Prominent members of the PPAPs (Figure [1]) are hyperforin (1), nemorosone (2), clusianone (3) and garsubellin A (4). Besides their challenging structural features, the PPAPs also show fascinating biological activities,[1] [2] which are compelling reasons to develop total syntheses of these natural products.

Synthetic endeavors and total syntheses of PPAPs have been reviewed in several publications,[1] [3] whilst some more recent total syntheses of PPAPs have also been described.[4] Furthermore methods to prepare the bicy­clo[3.3.1]nonan-9-one core have been reviewed.[5]

Each synthetic approach toward the PPAPs and precursors with a substituted bicyclo[3.3.1] framework published to date starts from a six-membered-ring compound (substituted cyclohexanones, substituted phenols, resorcinols or phloroglucinols), and install a suitable three-carbon unit to complete the bicyclo[3.3.1]nonan-9-one core of the PPAP.

Our idea was to synthesize a suitably substituted cyclooctene derivative and to install the bridging carbonyl group across the cyclooctene ring. Herein, we report our initial results along these lines.

For convenience, and as a proof-of-principle that our strategy was feasible, we first focused our efforts on the simpler model structure 5, which is structurally related to compounds 1–4.

Zoom Image
Figure 1 Structures of hyperforin (1), nemorosone (2), clusianone (3) and garsubellin A (4)
Zoom Image
Scheme 1 Retrosynthetic analysis of model compound 5; X = Cl (acid chloride) or OCOR (mixed acid anhydride); PG = protecting group

Retrosynthetic analysis (Scheme [1]) of 5 via cyclooctene derivative 6 is straightforward. Compound 5 should be accessible through intramolecular acylation of 6. The substituted cyclooctene 6 should be formed through ring-closing metathesis (RCM) from 7, which could be synthesized through aldol reaction of known compound 9 and commercially available 4-pentenal (8).

Methyl hexenoate 9 was synthesized in 84% yield according to the method of Santelli et al.,[6] which involved a titanium tetrachloride (TiCl4) catalyzed Sakurai reaction with 4-methyl-2-oxo-3-pentenenitrile (10).[6c] Aldol addition of 9 to aldehyde 8, following the procedure of Wei et al.,[7] gave diene 11 as a mixture of diastereomers (syn/anti = 3:1, in agreement with the results of Wei[7]), which were protected as methyl ether 7 using methyl triflate and Proton-sponge® as the base,[8] in 93% yield (Scheme [2]). After methylation of the hydroxy group in 11, the diastereomers were separated and all following reactions were performed with the syn diastereomer.

Zoom Image
Scheme 2 Preparation of intermediate 7. Reagents and conditions: (a) (i) TiCl4, CH2Cl2, allyl trimethylsilane, –78 °C to –30 °C; (ii) MeOH, –30 °C to r.t.; (b) LDA, THF, –78 °C, 8; (c) Proton-sponge®, MeOTf, CH2Cl2, r.t.

The next task was RCM of 7 to give 12. There are several literature reports on the successful application of the RCM for the synthesis of eight-membered rings.[9] In our case, the use of the Grubbs II catalyst, together with triphenylphosphine oxide as an additive,[10] led to the formation of 12 in a very good yield (Scheme [3]). Important for the success of the RCM is the presence of the quaternary carbon with the geminal dimethyl groups (Thorpe–Ingold effect[11]), since RCM reactions with similar compounds[12] under comparable conditions were sluggish and gave low yields.

Zoom Image
Scheme 3 RCM of 7 (syn isomer only) into cyclooctene 12 and conversion into acid chloride 6a. Reagents and conditions: (a) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux; (b) H2O, DME, LiOH (17 equiv), reflux; (c) (COCl)2, CH2Cl2, r.t.

Whereas ring closure is accelerated via the Thorpe–Ingold effect, the subsequent hydrolysis of carboxylic acid ester 12 was retarded by the steric congestion exerted by the geminal dimethyl grouping. After trying different methods for the hydrolysis of sterically hindered esters,[13] the method of Snider et al.[13c] was found to work best, giving acid 13 in 94% yield, although the reaction time was rather long (ca. 4–5 days; Scheme [3]). Crystallization from chloroform gave crystals of 13 suitable for X-ray analysis.[14] Transformation of the acid 13 into acid chloride 6a was straightforward using oxalyl chloride.[15]

The next part of our strategy was the transannular cyclization reaction of the acid chloride 6a across the eight-membered ring. Such Friedel–Crafts-like acylation reactions of double bonds are well known,[16] but are rarely applied to natural product syntheses due to competing side reactions, for example, polymerization of the olefin. Intramolecular acylations for ring-closing reactions have also been published,[16] and even the transformation of cyclooct-4-ene carboxylic acid 14 into cyclooct-4-ene carboxylic acid chloride 15 and subsequent cyclization to give 2-chlorobicyclo[3.3.1]nonan-9-one (16) have been reported.[17] Hence, we began our acylation studies (Scheme [4]) by repeating the conversion of acid chloride 15 into bicyclic 16. Whereas Kretschmar[17a] [b] reported a reaction time of 72 hours for the uncatalyzed cyclization in 1,2-dichloroethane at reflux temperature, and Kraus[17d] obtained similar results after reaction for 12 hours, we found that the cyclization was complete after seven to nine days. The product was obtained in quantitative yield as a mixture of diastereomers (endo/exo = 3:2, in agreement with Kraus[17d]). When we used aluminum chloride (AlCl3) or other Lewis acids as the catalyst under similar conditions, we obtained inferior results.[18] Therefore, the cyclization of acid chloride 6a was carried out in 1,2-dichloroethane as solvent at reflux temperature. The reaction was complete after only 21 hours to give 5a as a mixture of diastereomers (exo/endo = 18:1) in 40% yield after chromatography,[19] together with 9% of the elimination product (structure not shown). Similar to the results obtained in the RCM reactions (Scheme [3]), this increase in the rate of cyclization can be interpreted as a result of the Thorpe–Ingold effect of the geminal dimethyl arrangement.

Zoom Image
Scheme 4 Transannular cyclization of cyclooctenecarboxylic acid derivatives. Reagents and conditions: (a) (COCl)2, neat, r.t.; (b) DCE, reflux; (c) DCE, reflux; (d) TFAA, CHCl3, r.t.; (e) TFAA, CHCl3 (free of stabilizers), 0 °C.

Since increasing the reactivity of the acid chloride 6a was unsuccessful,[18] we reasoned that mixed anhydride 6b might also add across the eight-membered ring to the double bond, resulting in the construction of the carbonyl bridge and introduction of an oxygen next to a bridgehead atom (similar to the arrangement in PPAPs). Mixed trifluoroacetic anhydrides[20] have been described in the literature and are easily prepared from carboxylic acids and trifluoroacetic anhydride (TFAA).[20b] Thus, we attempted to synthesize the mixed anhydride 17 according to the literature,[20b] and found that cyclization into 2-trifluoroacetoxybicyclo[3.3.1]nonan-9-one (18) occurred almost instantaneously at room temperature (Scheme [4]). The product was obtained as a mixture of diastereomers (exo/endo = 3:4). Due to the observed Thorpe–Ingold effect[11] during cyclization of 6a, cyclization of acid 13 via mixed anhydride 6b was expected to be much faster and therefore was performed at 0 °C. Product 5b formed smoothly in almost quantitative yield as the exo diastereomer only. To complete the synthesis of bicyclic model compounds related to PPAPs, we cleaved the trifluoroacetoxy group in compounds 18 and 5b with saturated aqueous NaHCO3 solution at room temperature in 86% and 87% yields, respectively. Subsequent oxidation of resulting alcohols 19 and 20 with Dess–Martin periodinane (DMP) gave the diketones 21 (60%) and 22 (64%) (Scheme [5]). Compared to PPAPs, the last introduced carbonyl group is ‘on the wrong side’ of the bicyclic framework. Studies toward modification of the described synthesis to establish this carbonyl group ‘on the right side’ of the molecule, and the use of compound 22 as an intermediate for the synthesis of PPAP analogues, are currently underway in our group. Selected experimental procedures are given in the Supporting Information.

Zoom Image
Scheme 5 Synthesis of bicyclo[3.3.1]nonandiones. Reagents and conditions: (a) sat. aq NaHCO3, r.t.; (b) DMP, CH2Cl2, r.t.

In summary, we have developed a short and efficient synthesis of a substituted bicyclo[3.3.1]nonan-9-one related to the core structure of PPAPs. The method is based on a ring-closing metathesis to give a cyclooctene derivative and a simple transannular cyclization of a mixed trifluoroacetic anhydride. Adjusting this strategy to enable the syntheses of PPAPs in both racemic and enantiomerically pure form is currently underway, and the results will be published in due course.

Zoom Image
Scheme 6 Reagents and conditions: (i) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux, 30 min. (ii) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux, 2 d.
Zoom Image
Figure 2Structure representation for 13 with ellipsoids at 50% probability

#

Acknowledgment

We thank the Universität des Saarlandes for generous support and Prof. Dr. D. Volmer and T. Dier for recording the HRMS spectra. E.F. thanks Graduiertenförderung der Universität des Saarlandes for a PhD scholarship.

Supporting Information

    for this article is available online at http://www.thieme-connect.com.accesdistant.sorbonne-universite.fr/products/ejournals/journal/ 10.1055/s-00000083.
  • Supporting Information

  • References and Notes

    • 7a Wei Y, Bakthavatchalam R. Tetrahedron 1993; 49: 2373
      CrossrefSuche in Google Scholar
    • 7b In aldol adducts where the chiral carbon between the carbonyl group and the hydroxy group contains a bulky substituent, the syn isomer shows a larger coupling constant than the anti isomer, e.g., see: Heng KK, Simpson J, Smith RA. J, Robinson WT. J. Org. Chem. 1981; 46: 2032
      Suche in Google Scholar

      • See also:
    • 7c Kitamura M, Nakano K, Miki T, Okada M, Noyori R. J. Am. Chem. Soc. 2001; 123: 8939
      CrossrefSuche in Google Scholar
    • 7d Williamson RT, Marquez BL, Barrios Sosa AC, Koehn FE. Magn. Reson. Chem. 2003; 41: 379
      CrossrefSuche in Google Scholar
    • 7e Our stereochemical assignment is in full agreement with the X-ray structure depicted in ref. 15.

      • For other examples of aldol reactions of sterically congested ester enolates with aldehydes, see:
    • 7f Indium enolates: Hirashita T, Kinoshita K, Yamamura H, Kawai M, Araki S. J. Chem. Soc., Perkin Trans. 1 2000; 825
    • 7g Zinc enolates: Wei C.-Q, Zhao G, Jiang X.-R, Ding Y. J. Chem. Soc., Perkin Trans. 1 1999; 3531
    • 7h Samarium enolates: Nagano T, Motoyoshiya J, Kakehi A, Nishii Y. Org. Lett. 2008; 10: 5453
      CrossrefSuche in Google Scholar

      For successful RCM, an additive was required to prevent isomerization of the starting material and product. Triphenylphosphine oxide worked best in our case. For studies on additives in RCM, see:
    • 10a Burgeois D, Pancrazi A, Nolan SP, Prunet J. J. Organomet. Chem. 2002; 643-644: 247
      CrossrefSuche in Google Scholar
    • 10b Schiltz S, Ma C, Ricard L, Prunet J. J. Organomet. Chem. 2006; 691: 5438
      CrossrefSuche in Google Scholar
    • 10c Formentin P, Gimeno N, Steinke HG. J, Vilar R. J. Org. Chem. 2005; 70: 8235
      CrossrefSuche in Google Scholar
    • 10d Vedrenne E, Dupont H, Oualef S, Elkaim L, Grimaud L. Synlett 2005; 670
      Thieme Connect
    • 10e Hong SH, Sanders DP, Lee CW, Grubbs RH. J. Am. Chem. Soc. 2005; 127: 17160
      Suche in Google Scholar
    • 12a In a related RCM reaction with the Grubbs II catalyst (Scheme 6), we also observed acceleration exerted by a geminal dimethyl arrangement (Feidt, E. unpublished results).

      • For the RCM of 1,9-decadiene into cyclooctene, see:
    • 12b Yang H, Ma Z, Wang Y, Wang Y, Fang L. Chem. Commun. 2010; 46: 8659
      CrossrefSuche in Google Scholar
    • 12c Nelson DJ, Ashworth IW, Hillier IH, Kyne SH, Pandian S, Parkinson JA, Percy JM, Rinaudo G, Vincent MA. Chem. Eur. J. 2011; 17: 13087
      CrossrefSuche in Google Scholar
    • 14a Crystal structure representation of 13 (Figure 2): Crystals suitable for single crystal X-ray analysis were obtained from CHCl3. The data were collected at 133 K on a Bruker AXS X8Apex CCD diffractometer operating with graphite-monochromated MoKα radiation. Frames of 0.5° oscillation were exposed; deriving data in the θ range of 2–37° with a completeness of ~99%. Unit cell: triclinic, P-1, a = 7.1553(4) Å, b = 7.4403(4) Å, c = 11.4373(7) Å, α = 83.096(2)°, β = 77.747(2)°, γ = 74.477(3)°, V = 572.02(6) Å3. Structure solution and full least-squares refinement with anisotropic thermal parameters of all non-hydrogen atoms and free refinement of the hydrogens were performed using SHELX.14b The final refinement resulted in: R1 = 0.036. Crystallographic data for this structure have been deposited with the Cambridge Crystallographic Data Centre, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK. Copies of the data can be obtained free of charge on quoting the deposition number CCDC 990861 (www.ccdc.cam.ac.uk/data_request/cif).
    • 14b Sheldrick GM. Acta Crystallogr. A 2008; 64: 112
      CrossrefPubMedSuche in Google Scholar
  • 15 Ceccherelli P, Curini M, Marcotullio MC, Pisani E, Rosati O, Wenkert E. Tetrahedron 1997; 53: 8501
    CrossrefSuche in Google Scholar
    • 17a Erman WF, Kretschmar HC. J. Org. Chem. 1968; 33: 1545
      CrossrefSuche in Google Scholar
    • 17b For additional details, see: Kretschmar HC. US Patent 3524884, 1970 ; Chem. Abstr. 1970, 73, 87533.
    • 17c Fickers GN, Kemp KC. J. Chem. Soc., Chem. Commun. 1973; 84
    • 17d Heumann A, Kraus W. Tetrahedron 1978; 34: 405
      CrossrefSuche in Google Scholar
    • 17e Heumann A, Kolshorn H. Tetrahedron 1975; 31: 1571
      CrossrefSuche in Google Scholar
    • 17f For an analogous reaction giving a bicyclo[3.2.1]octan-8-one skeleton, see: Nelsen SF, Kapp DL. J. Am. Chem. Soc. 1986; 108: 1265
      CrossrefSuche in Google Scholar
    • 17g Cyclooct-4-ene carboxylic acid 14 was synthesized according to: Bloodworth AJ, Melvin T, Mitchell JC. J. Org. Chem. 1988; 53: 1078 ; with a modified hydrolysis procedure (sat. KOH in MeOH–H2O, 9:1, reflux, 3 d). The acid chloride 15 was prepared according to ref. 17f.
      CrossrefSuche in Google Scholar
  • 18 The elimination product, bicyclo[3.3.1]non-2-en-9-one, became the main product and the total yield decreased. Since acid bromides and acid iodides are more reactive than acid chlorides, we also added tetrabutylammonium bromide or tetrabutylammonium iodide to the reaction mixture to form the acid bromides and acid iodides in situ, but this also led to decreased yields of the corresponding cyclization products and larger amounts of elimination products.
  • 19 Interestingly, the 1H and 13C NMR spectra of 5a before and after chromatographic purification were identical; it seems that either 5a decomposes to some extent during chromatography, or is partially absorbed irreversibly.

  • References and Notes

    • 7a Wei Y, Bakthavatchalam R. Tetrahedron 1993; 49: 2373
      CrossrefSuche in Google Scholar
    • 7b In aldol adducts where the chiral carbon between the carbonyl group and the hydroxy group contains a bulky substituent, the syn isomer shows a larger coupling constant than the anti isomer, e.g., see: Heng KK, Simpson J, Smith RA. J, Robinson WT. J. Org. Chem. 1981; 46: 2032
      Suche in Google Scholar

      • See also:
    • 7c Kitamura M, Nakano K, Miki T, Okada M, Noyori R. J. Am. Chem. Soc. 2001; 123: 8939
      CrossrefSuche in Google Scholar
    • 7d Williamson RT, Marquez BL, Barrios Sosa AC, Koehn FE. Magn. Reson. Chem. 2003; 41: 379
      CrossrefSuche in Google Scholar
    • 7e Our stereochemical assignment is in full agreement with the X-ray structure depicted in ref. 15.

      • For other examples of aldol reactions of sterically congested ester enolates with aldehydes, see:
    • 7f Indium enolates: Hirashita T, Kinoshita K, Yamamura H, Kawai M, Araki S. J. Chem. Soc., Perkin Trans. 1 2000; 825
    • 7g Zinc enolates: Wei C.-Q, Zhao G, Jiang X.-R, Ding Y. J. Chem. Soc., Perkin Trans. 1 1999; 3531
    • 7h Samarium enolates: Nagano T, Motoyoshiya J, Kakehi A, Nishii Y. Org. Lett. 2008; 10: 5453
      CrossrefSuche in Google Scholar

      For successful RCM, an additive was required to prevent isomerization of the starting material and product. Triphenylphosphine oxide worked best in our case. For studies on additives in RCM, see:
    • 10a Burgeois D, Pancrazi A, Nolan SP, Prunet J. J. Organomet. Chem. 2002; 643-644: 247
      CrossrefSuche in Google Scholar
    • 10b Schiltz S, Ma C, Ricard L, Prunet J. J. Organomet. Chem. 2006; 691: 5438
      CrossrefSuche in Google Scholar
    • 10c Formentin P, Gimeno N, Steinke HG. J, Vilar R. J. Org. Chem. 2005; 70: 8235
      CrossrefSuche in Google Scholar
    • 10d Vedrenne E, Dupont H, Oualef S, Elkaim L, Grimaud L. Synlett 2005; 670
      Thieme Connect
    • 10e Hong SH, Sanders DP, Lee CW, Grubbs RH. J. Am. Chem. Soc. 2005; 127: 17160
      Suche in Google Scholar
    • 12a In a related RCM reaction with the Grubbs II catalyst (Scheme 6), we also observed acceleration exerted by a geminal dimethyl arrangement (Feidt, E. unpublished results).

      • For the RCM of 1,9-decadiene into cyclooctene, see:
    • 12b Yang H, Ma Z, Wang Y, Wang Y, Fang L. Chem. Commun. 2010; 46: 8659
      CrossrefSuche in Google Scholar
    • 12c Nelson DJ, Ashworth IW, Hillier IH, Kyne SH, Pandian S, Parkinson JA, Percy JM, Rinaudo G, Vincent MA. Chem. Eur. J. 2011; 17: 13087
      CrossrefSuche in Google Scholar
    • 14a Crystal structure representation of 13 (Figure 2): Crystals suitable for single crystal X-ray analysis were obtained from CHCl3. The data were collected at 133 K on a Bruker AXS X8Apex CCD diffractometer operating with graphite-monochromated MoKα radiation. Frames of 0.5° oscillation were exposed; deriving data in the θ range of 2–37° with a completeness of ~99%. Unit cell: triclinic, P-1, a = 7.1553(4) Å, b = 7.4403(4) Å, c = 11.4373(7) Å, α = 83.096(2)°, β = 77.747(2)°, γ = 74.477(3)°, V = 572.02(6) Å3. Structure solution and full least-squares refinement with anisotropic thermal parameters of all non-hydrogen atoms and free refinement of the hydrogens were performed using SHELX.14b The final refinement resulted in: R1 = 0.036. Crystallographic data for this structure have been deposited with the Cambridge Crystallographic Data Centre, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK. Copies of the data can be obtained free of charge on quoting the deposition number CCDC 990861 (www.ccdc.cam.ac.uk/data_request/cif).
    • 14b Sheldrick GM. Acta Crystallogr. A 2008; 64: 112
      CrossrefPubMedSuche in Google Scholar
  • 15 Ceccherelli P, Curini M, Marcotullio MC, Pisani E, Rosati O, Wenkert E. Tetrahedron 1997; 53: 8501
    CrossrefSuche in Google Scholar
    • 17a Erman WF, Kretschmar HC. J. Org. Chem. 1968; 33: 1545
      CrossrefSuche in Google Scholar
    • 17b For additional details, see: Kretschmar HC. US Patent 3524884, 1970 ; Chem. Abstr. 1970, 73, 87533.
    • 17c Fickers GN, Kemp KC. J. Chem. Soc., Chem. Commun. 1973; 84
    • 17d Heumann A, Kraus W. Tetrahedron 1978; 34: 405
      CrossrefSuche in Google Scholar
    • 17e Heumann A, Kolshorn H. Tetrahedron 1975; 31: 1571
      CrossrefSuche in Google Scholar
    • 17f For an analogous reaction giving a bicyclo[3.2.1]octan-8-one skeleton, see: Nelsen SF, Kapp DL. J. Am. Chem. Soc. 1986; 108: 1265
      CrossrefSuche in Google Scholar
    • 17g Cyclooct-4-ene carboxylic acid 14 was synthesized according to: Bloodworth AJ, Melvin T, Mitchell JC. J. Org. Chem. 1988; 53: 1078 ; with a modified hydrolysis procedure (sat. KOH in MeOH–H2O, 9:1, reflux, 3 d). The acid chloride 15 was prepared according to ref. 17f.
      CrossrefSuche in Google Scholar
  • 18 The elimination product, bicyclo[3.3.1]non-2-en-9-one, became the main product and the total yield decreased. Since acid bromides and acid iodides are more reactive than acid chlorides, we also added tetrabutylammonium bromide or tetrabutylammonium iodide to the reaction mixture to form the acid bromides and acid iodides in situ, but this also led to decreased yields of the corresponding cyclization products and larger amounts of elimination products.
  • 19 Interestingly, the 1H and 13C NMR spectra of 5a before and after chromatographic purification were identical; it seems that either 5a decomposes to some extent during chromatography, or is partially absorbed irreversibly.

   Lizenzen und Reprints Alle Artikel dieser Rubrik
Zoom Image
Figure 1 Structures of hyperforin (1), nemorosone (2), clusianone (3) and garsubellin A (4)
Zoom Image
Scheme 1 Retrosynthetic analysis of model compound 5; X = Cl (acid chloride) or OCOR (mixed acid anhydride); PG = protecting group
Zoom Image
Scheme 2 Preparation of intermediate 7. Reagents and conditions: (a) (i) TiCl4, CH2Cl2, allyl trimethylsilane, –78 °C to –30 °C; (ii) MeOH, –30 °C to r.t.; (b) LDA, THF, –78 °C, 8; (c) Proton-sponge®, MeOTf, CH2Cl2, r.t.
Zoom Image
Scheme 3 RCM of 7 (syn isomer only) into cyclooctene 12 and conversion into acid chloride 6a. Reagents and conditions: (a) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux; (b) H2O, DME, LiOH (17 equiv), reflux; (c) (COCl)2, CH2Cl2, r.t.
Zoom Image
Scheme 4 Transannular cyclization of cyclooctenecarboxylic acid derivatives. Reagents and conditions: (a) (COCl)2, neat, r.t.; (b) DCE, reflux; (c) DCE, reflux; (d) TFAA, CHCl3, r.t.; (e) TFAA, CHCl3 (free of stabilizers), 0 °C.
Zoom Image
Scheme 5 Synthesis of bicyclo[3.3.1]nonandiones. Reagents and conditions: (a) sat. aq NaHCO3, r.t.; (b) DMP, CH2Cl2, r.t.
Zoom Image
Scheme 6 Reagents and conditions: (i) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux, 30 min. (ii) Grubbs II (1.4 mol%), Ph3P=O (5 mol%), Et2O, reflux, 2 d.
Zoom Image
Figure 2Structure representation for 13 with ellipsoids at 50% probability