University of Groningen
Asymmetric Cu-catalyzed 1,2 and 1,4-additions of Grignard reagents
Calvo González, Beatriz
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Calvo González, B. (2018). Asymmetric Cu-catalyzed 1,2 and 1,4-additions of Grignard reagents: Development of new substrates and their application in total synthesis. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Asymmetric Cu-catalyzed
1,2 and 1,4-additions of
Grignard reagents
Development of new substrates and their application in
total synthesis
PhD thesis
to obtain the degree of PhD at the
University of Groningen
on the authority of the
Rector Magnificus Prof. E. Sterken
and in accordance with
the decision by the College of Deans.
This thesis will be defended in public on
Friday 7 September 2018 at 09.00 hours
by
Beatriz Carmen Calvo Gonzalez
born on 9 February 1986
in Madrid, Spanje
Supervisor
Prof. A.J. Minnaard
Assessment Committee
Prof. W.R. Browne
Prof. A.S.S. Dömling
Prof. E.V. van der Eycken
Index
1. Chapter 1: State of the art in steroid synthesis
1.1 Introduction pag. 5 1.2 Total synthesis of Estrone and Estrone derivatives pag. 5 1.3 Radical cyclization as the key step pag. 11 1.4 Cholesterol and Desogestrel syntheses pag. 13 1.5 Other steroid-like compounds • Non-natural occurring steroid-type compounds pag.16 • Heterocyclic steroids pag. 25 1.6 Conclusion pag. 29 1.7 References pag. 312. Chapter 2: Enantioselective conjugate additions followed
by enolate trapping. Construction of quaternary centers
2.1 Introduction pag. 34 2.2 Results and discussion pag.36 2.3 Conclusion pag. 44 2.4 Experimental part pag.46 2.5 References pag. 543.
Chapter 3: Broadening the scope of the Cu-catalyzed 1,2-additions of Grignard reagents
3.1 Introduction pag. 58 3.2 Results and discussion pag. 59 3.3 Conclusion pag. 66 3.4 Experimental part pag. 67 3.5 References pag. 77
4. Chapter 4: Total synthesis of steroid hybrids employing
asymmetric synthesis
4.1 Introduction pag. 79 4.2 Results and discussion pag. 83 4.3 Conclusion pag. 92 4.4 Experimental part pag. 95 4.5 References pag. 1115. Outlook and perspectives
pag. 1136. Summary
6.1English summary pag. 115
6.2
Nederlandse samenvatting pag. 118
7. Acknowledgements
pag. 1218. List of abbreviations
pag. 123Chapter 1: State of the art in steroid total synthesis
Introduction
The field of steroid synthesis has been a playground for chemists since its bloom in the 1930´s and 1940´s, due to their great potential as contraceptives and anti-inflammatory agents.1 Benchmark total syntheses of steroids are the syntheses of estrone by the Torgov2,3 and the Vollhardt4 groups and the Woodward synthesis of cholesterol.5 A plethora of synthesis methodologies has been employed through the years in the synthesis of steroids, either to study their pharmacological properties, with the aim to produce steroid drugs on industrial scale, or to illustrate and test novel reactions. In particular carbon-carbon bond forming reactions using transition metal catalysis,6-7 pericyclic reactions8-9,7 and (organo-catalyzed) aldol cyclizations8, have been used. As steroids are chiral and contain quaternary stereocenters, stereoselectivity is at the heart of steroid total synthesis as well.10,11 Due to the enormous number of literature reports, even when limited to total synthesis, we focus in this review on steroid syntheses reported from 2000 to 2015. Even with these constraints, we had to focus on reports dealing with the construction of the carbon skeleton.
Total syntheses of Estrone and Estrone derivatives
(+)-Estrone (1) (Figure 1) is an estrogen hormone found in the ovaries and in adipose tissue. It is responsible for the development and function of the female secondary sexual characteristics. Estrone was discovered and isolated by Adolf Butenandt, by its extraction from urine.
Figure 1: (+)-Estrone (1) Metal catalysis as the key step
In 2007, Knochel and co-workers published an enantioselective formal synthesis of (+)-estrone (1).12 Hydroboration followed by B/Zn exchange of Dane´s diene 2 afforded organozinc reagent 3 that underwent a copper(I)-mediated anti-SN2’-allylic substitution with enantiopure cyclopentene diol 4 to yield iodide 5 in 66% yield. Conversion of the iodide into the corresponding ketone 6 followed by acid catalysed ring closure, alcohol deprotection and oxidation to the ketone afforded Torgov´s diene 7. Conversion of the diene into (+)-estrone (1) had already been reported in 3 steps by Quinkert and Ogasawara.13,14
Scheme 1: Knochel’s formal synthesis of (+)-estrone (1).
The Linclau group developed a strategy for the enantioselective synthesis of (+)-estrone via a C-ring closing metathesis and a B-ring Heck cyclization.15 From commercially available 3-methoxytoluene (8) 6 steps
HO O H H H (+)-Estrone (1) 1) Et2BH 2) Et2Zn Zn 2 MeO MeO OC(O)C6F5 I TBSO 98% (trans-4: 99% ee) CuCN•2LiCl THF MeO 66%, 97% ee I OTBS
tBuLi (2 eq), -78 °C then B(OMe)3, 78 °C, then NaBO3•H2O MeO 45% O OTBS MeO 66% O HO O H H H 1) p-TsOH, benzene 2) TBAF, THF, 3) CrO3, Celite, DCM: Et2O (+)-Estrone (1) 2 3 4 5 6 7 Steps
lead to bromide 9, which can undergo diastereoselective conjugate addition followed by alkylation to furnish a 11 : 5 : 69 mixture of 10, 11 and 12, respectively. Compound 14 was obtained after ring-closing metathesis employing the Hoveyda-Grubbs 2nd generation catalyst (13). Heck reaction of bromide 14 furnished intermediate 15, which upon reduction of the double bond, or double bond isomerization followed by reduction, yielded a 3 : 7 mixture of products in favor of the desired estrone methylether (17), that was obtained pure after crystallization. Deprotection of the methyl ether gave (+)-estrone (1) in high yield (Scheme 2).
Scheme 2: Linclau´s strategy for the enantioselective synthesis of (+)-estrone (1).
In 2011, Kotora and co-workers published a route to (−)-estrone (ent-1) in 12 steps from commercially available methoxytetralone (18).16 Introducing homochirality in the construction of the B ring, epoxide (19) was synthesized. This underwent Lewis acid catalyzed rearrangement to afford (−)-estrone methyl ether (ent-17) together with tertiary alcohol (20) (Scheme 3). Br MeO (X)2P O MeO Steps BuLi, THF, -78 ºC then then Br MeO (X)2P O H O Br MeO (X)2P O H O Br MeO (X)2P O H O + + 1) HCl 3 M, THF, 93% 2) Hoveyda-Grubbs 2nd (51) (35 mol%), toluene, 58% Br MeO H O H O H H MeO PdOAc P (oTol)2 2 (2 mol%) Bu4NOAc, MeCN, DMF, H2O 115 ºC, 3 h H O H H MeO H O H H MeO + Pd/ C, cyclohexadiene EtOH 98% 16 : 17= 3 : 7 AlCl3, NaI, 92% H O H H HO 10 :11 : 12 = 11% : 5% : 69% quantitative 8 9 10 11 12 14 15 1 16 13 17 N N RuCl Cl O (X)2= Me N N Me Br O
Scheme 3: Kotora´s synthesis of (−)-estrone methyl ether (ent-17).
The group of Tietze developed a method for the enantioselective synthesis of (+)-estradiol (21) via multiple palladium-catalyzed transformations.17 Heck reaction of bromophenyl vinyl bromide (22) and the enantiopure indene 23 using Pd(OAc)2 as catalyst led to (24) with excellent regio- and stereoselectivity. Subsequent intramolecular Heck reaction of (24) employing palladacene catalyst (25) furnished steroid intermediate (26) in excellent yield. The synthesis of estradiol (21) from (26) was subsequently accomplished in a few steps (Scheme 4). Scheme 4: Tietze´s synthesis of estradiol (21). Diels-Alder reactions as the key step MeO O Steps MeO H H O H MeO H H H O + MeO H H OH BF3•Et2O (4 eq) toluene, -78 °C ent-17 18 19 20 Steps MeO H H MeO H H H O Co2CO8 toluene MeO Br Br OtBu H
+ Pd(OAc)2,PPh3, nBu4OAc DMF/ MeCN/ H2O, 66% OtBu H H Br MeO P Pd PdP (oTol)2 O O (oTol)2 O O 25
25 (2 mol%), nBu4NOAc, CH3CN, DMF, H2O, 115 °C, 99% OtBu H H MeO H OH H H HO H Steps Estradiol (21) 24 22 23 26
In 2004, Corey and co-workers published an enantioselective synthesis of (+)-estrone (1),18 whereas a few years later another enantioselective approach employing a CBS reduction as the key step was reported by the same laboratory.19 Enantioselectivity is introduced in the C ring via a Diels-Alder reaction employing a chiral oxazaborolidinium salt as the catalyst (27). The synthesis started with a Diels-Alder reaction between Dane´s diene (2) and dienophiles (28a) and (28b). Intermediates (29a) and (29b) were obtained in high yield and high enantiomeric excess (94% ee) that was improved to 100% ee by recrystallization. Treatment with 1 equivalent of MeMgBr afforded a γ-hydroxyester, which was subsequently reduced and then oxidized via Swern oxidation to furnish ketoesters (30a) and (30b). Base-catalyzed aldol cyclization followed by acid treatment provided Torgov´s diene (7) and diene (31). Finally, reduction of the double bonds and methylether deprotection yielded (+)-estrone (1) (Scheme 5).
Scheme 5: Corey´s enantioselective synthesis of (+)-estrone (1).
A few years later, Göbel and co-workers reported an enantioselective synthesis of (+)-estrone (1) via a hydrogen bond-promoted Diels-Alder reaction.20 Employing Dane’s diene (2) and methylcyclopentenedione (32), intermediate (33) could be synthesized in 91% total yield. The Diels-Alder reaction between (2) and (32) employing the amidinium catalyst (34) afforded the product in high yield and good enantioselectivity. Removal of MeO + R O EtO O BO N H H Ph H Tf2N 27 27 (20 mol%) MeO H R H OEt O O 92%, 94% ee,
100% after recrist. MeO
H R H O O 1) MeMgBr 2) LiAlH4, THF Me2SO, 78% Cl O O Cl MeO R O HO O H H H 1) KOH, MeOH, Δ 2) HCl, AcOH, rt 1) H2, Pd/C, Et3SiH
2) Et3SiH, CF3COOH, TBAI
3) HBr (cat.) 80% 1 2 28a; 28b; 3) R=Me
R=Et R=MeR=Et
R=Me R=Et 7; R=Me 31; R=Et 29a; 29b; 30a; 30b;
the hydroxyl group and double bond isomerization gave (35) in 99% ee after recrystallization. Torgov´s diene (7) was formed in good yield by reduction and posterior acid treatment. Stereoselective double bond reductions yielded (17) in 82% and final methyl ether deprotection followed by HPLC purification afforded (+)-estrone (1) in high yield and more than 99.9% ee (Scheme 6).
Scheme 6: Göbel´s hydrogen bond-promoted Diels-Alder approach to estrone.
Schotes and Mezzetti reported the use of dicationic ruthenium PNNP complexes (36) as catalysts in the asymmetric Diels-Alder reaction of unsaturated b-ketoesters.21 This approach was used in the synthesis of estrone precursor (37) reacting Dane´s diene (2) with dienophile (38), which successfully afforded the product in very good yield and good enantioselectivity (Scheme 7). MeO O O + HO OH O H N H H N H TFBP 34 H H O MeO OH H O MeO O MeO O MeO H H H O HO H H H 1) Tf2O 2,6-Lutidine 2) Et3SiH, PdCl2(dppf), 71%, 81% ee (99% ee after recristallization) 1) LiHMDS THF, -78 °C 2) AcOH 74% 1) H2, Pd/C, benzene 2) TFA, Et3SiH 82% 1) BBr3, DCM, 85% 2) HPLC purification 87%, > 99.9% ee 2 32 33 35 7 17 1 91%, 81% ee
Scheme 7: Asymmetric Diels-Alder reaction catalysed by dicationic ruthenium PNNP complexes.
Radical cyclization as the key step
In 2004, Pattenden´s group developed novel cascade radical cyclizations that could be employed for the synthesis of rac-estrone (1).22 Starting with aldehyde (39), Horner-Wittig reaction followed by Heck arylation and Z-selective Wittig reaction afforded intermediate ester (40). Reduction of the ester functionality, cyclopropanation and again oxidation gave aldehyde (41). Grignard reaction followed by oxidation and Horner-Wittig reaction yielded (42) in 90%. TBAF deprotection and Appel reaction gave rise to iodide (43), which underwent a radical cascade cyclization to form (44). Lastly, Cr(VI) oxidation to the ketone and BBr3 induced ether deprotection to furnish rac-estrone (1) in good yield (Scheme 8). MeO + O O O (Et3O)PF6, DCM:Et2O N N P P Ru Cl Ph2 Ph2 Cl 36 36 (5 to 10 mol%) O H H MeO O O quantitative, 86% ee (85%, 99% ee after recrystallization) 2 38 37
Scheme 8: rac-estrone (1) synthesis via a cascade radical cyclizations Organocatalysis; Lewis acid-catalyzed cyclization
The List group published in 2014 a very elegant variation23 of Torgovs route,2 by making the acid-catalyzed cylization step enantioselective with the use of a chiral Brønsted acid. Grignard addition to methoxytetralone (18) afforded tertiary allylic alcohol (49), which underwent base-catalyzed alkylation to give diketone (50). Here, enantioselection was induced in Torgov´s cyclization by chiral disulfonimide (51), which furnished Torgov´s diene (7) in 95% yield and 94% ee. Stereoselective double bond hydrogenation and subsequent methyl ether deprotection yielded (+)-estrone (1) in 75% (Scheme 9). MeO I O MeO I O OEt MeO O OEt O MeO O OEt OTBS MeO OH OTBS MeO O OTBS MeO O OTBS MeO OTBS OMe MeO I OMe MeO O H H H HO O H H H MeO OMe H H H EtO2CCH2PO(OEt)2, BuLi THF 72% yield CH2=CHCH2OH, Pd(OAc)2, n-Bu4NCl DMF 62–53% yield IPh3P(CH2)3OTBS, KHMDS THF 80% yield iBu2AlH, DCM 82% yield 1) Et2Zn, CH2I2, 70% yield 2) PDC, 84% yield 1) MePPh3Br, n-BuLi, 86% yield 2) Et2Zn, CH2I2, 70% yield MeOCH2P(O)Ph2, LDA, THF,
then NaH, THF 90%yield 1) TBAF, THF 98% yield 2) I2, imidazole, PPh3, 80% yield Bu3SnH, AIBN toluene 12% yield CrO3, H2SO4 (cat.), Me2CO 94% yield BBr3 THF 79% yield rac-Estrone (rac-1) 39 45 46 40 47 41 48 42 43 44 rac-17 MeO OMe MeO OMe MeO MeO OMe OMe Macrocyclisation 1) Transannulation
Scheme 9: List´s synthesis of (+)-estrone (1).
Cholesterol and Desogestrel syntheses
Cholesterol (52) (Figure 2) is an essential component of eukariotic cell membranes due to its regulatory function of their fluidity.steroid hormones such as progesterone, testosterone, estradiol and cortisol. Figure 2: Cholesterol (52). Rychnovsky´s laboratory developed a 16-step synthesis of ent-cholesterol (52) starting from commercially available (S)-citronellol (53) in 2% overall yield.24 Conversion of (53) into the corresponding α-diazo-β-ketoester (54) was achieved in good yield in 3 steps. Compound (54) underwent a diastereoselective C-H insertion employing the valine-derived phthalate ligand (55). The product was obtained in a 3.6 : 1 ratio in favour of the desired diastereomer, which upon column chromatography and recrystallization afforded the pure diastereomer (56) with 99% ee. Compound (56) could be transformed into ketone (57) after double bond O MeO MgBr THF, 40 °C OH MeO 92% O O
Triton B (BnNMe3OH)
xylene/tBuOH MeO O O 45% O MeO O2 S S O2 NH NO2 NO2 SF5 SF5 SF5 SF5 O HO H H H 95%, 73% over 3 steps 51 (5 mol%) toluene, 4 Å MS, -50 °C (2d) then -20 °C (2d) 51 18 49 50 7 1 94% ee 3 steps (See scheme 6) H H H HO cholesterol 52 H
hydrogenation, alkylation and ester removal. Ketone (57) was converted into thio-ether (58) in 4 steps. Annulation of 58, with β-ketoester (59) provided enone (60) in 73% yield. Enone reduction followed by alkylation provided (61), which underwent acetal deprotection, aldol cyclization and finally selective reduction of the ketone with Li(OtBu)3AlH to furnish ent-cholesterol (52) in 80% yield (Scheme 10).
Scheme 10: The synthesis of ent-cholesterol (52).
In 2004, in parallel with their enantioselective synthesis of estrone, Corey and co-workers published a modified synthesis25 of the third generation birth control compound desogestrel (67).18 Up to (31), the route is the same as previously described for estrone (1) (vide supra), with the exception that dienophile (28b) contains an ethyl instead of a methyl group. Acid catalyzed transposition of (31) gave (68). Following known procedures,26 intermediate (70) was synthesized and finally desogestrel (67) was obtained via their previous route (Scheme 11).25 OH (S)-citronellol O O O N2 O H O O + O H O O N O O O O 4 Rh2 55 DCM, 23 °C 76%, 99% ee after crist. of 54 diast-56:56=1:3.6 O H SPh H O H H O O H O H H O O H H H HO 1) PhSO2Cl, Et3N 2) OLi O ONa 65% 3) p-ABSA, Et3N, MeCN, 23 °C 96% 1) NaOMe, 59 2) NaOH 3) Δ, 0.2 Torr 73% O O O O 1) Li, NH3, THF, -78 °C 2) MeI, THF, -33 °C 81% 1) HCl, MeOH, reflux, 90% 2) KOtBu,tBuOH, then AcOH, H2O, 23 °C
3) Li(OtBu)3AlH, THF, 23 °C 80% ent-cholesterol 52 62 54 55 diast-56 56 95 58 59 60 61 53 H H O H2, Pd/BaSO4, MeOH, 23 °C 99% O H O O MeI, acetone K2CO3 95% O H O O 1) NaCN, HMPA, 80 °C 2) KOH, MeOH 93% O H O NaOMe, 23 °C, 65% 2) TsOH, PhMe, 110 °C 84% 1) O H 65 MeO OMg(OMe) O DMF, 125 °C 90% O H COOH 1) H2, Pd/BaSO4, MeOH, -5 °C 2) CH2O, DMSO, piperidine, 23 °C 3) PhSH, TEA 57% 57 63 64 66
Scheme 11: Corey´s synthesis of desogestrel (67).
A few years later, Tietze´s laboratory reported on a synthesis of desogestrel (67) using double Heck arylations as the key steps.27 Alkene (72) is obtained in 3 steps from enone (71). Heck arylation between iodide (73) and (72) affords a 7 : 1.7 : 1 mixture of isomers in favour of desired (74). Purified bromide (74) could undergo intramolecular Heck arylation using palladacene (25) to successfully yield intermediate (75). Platinum catalysed double bond reduction of (75) and subsequent double bond isomerization produced (76). Alcohol (77) was made in 4 steps in 55% overall yield from (76). Dess-Martin oxidation followed by Peterson olefination yielded alkene (78) in 95%. Ether deprotection afforded (79) together with a minor isomer that could not be purified by means of recrystallization. Therefore, the mixture underwent another Dess-Martin oxidation and acetylene addition to form pure desogestrel (67) in 83% yield after column chromatography and recrystallization (Scheme 12). MeO O MeO O H MeO OR H H HO MeO OTBS H H OH H H Desogestrel (67) 1) H2, Pd (C), Et3SiH, 0 °C Ref. 17 31 68 69 70 Ref. 18 Ref. 18 2) H+
Scheme 12: Tietze´s synthesis of desogestrel (67).
Other steroid-like compounds
Non-natural occurring steroid-type compounds
In 2006, it was reported by the Dyker group, that a suitable gold catalysed domino process leads to the synthesis of the steroid skeleton28 (Scheme 13). Alkyne (80), made in 3 steps from methylcyclopentadione, underwent a Sonogashira coupling followed by TMS deprotection to afford aldehyde (81) in very good yields. This intermediate was employed in a gold catalysed cyclisation, yielding (82), (83), and (84) in a 3:1:6 ratio. These could be unified via ester hydrolysis of (84) leading to (85) which, together with its diastereomer, was dehydrated to form diketone (82). Finally, hydrogenation and hydrogenolysis afforded partially reduced (86) and the racemic target compound (87) (Scheme 13). It is somewhat disappointing O OtBu HO OtBu O OtBu O OtBu H DIBAL-H 95% 97% N N O Pd(OAc)2, Pn-Bu3 dioxane, 90 °C 97% MeO I OtBu H + Pd(OAc)2,PPh3, AgNO3 77% 74 :75 : 76 = 7 : 1.7 : 1 OtBu H OtBu H OtBu H Br MeO Br MeO H + + H Br OMe
25 (2 mol%), nBu4NOAc,
AcCN, DMF, H2O 94% OtBu H H MeO H OtBu H H MeO H OtBu H H MeO OtBu H H MeO HO H OtBu H H HO H S S H OtBu H H HO H H OtBu H H H H OH H H H H O H H H H OH H H H H PtO2•H2O, H2 91% KOtBu, DMSO, 89% 1) BH3•THF, 2) aq. H2O2, NaOH, 90% 76 1) Li, NH3, iPrOH 2) HCl 3) ethane dithiol, BF3•Et2O Li, NH3, THF, -40 °C 55% over 4 steps BF3•Et2O Dess-Martin periodinane 96% desogestrel Br 71 72 73 72 74 75 76 77 78 79 67 Steps LiCCH, TMEDA 83%
that the keto function in the C-ring is removed in the final step, as functionalisation at this point is considered difficult. Scheme 13: Steroid framework synthesis via a gold catalysed cyclization. As well in the same year, the group of Kotora published a strategy based on the repetitive use of the Negishi reagent to obtain the steroid skeleton (93), (94) and (95) with cis-fused C and D rings (Scheme 14a).29 The synthesis started with iodide (88), which after benzylation and Stille coupling afforded alkene (89). Oxidative addition of the benzyl ether to Cp2ZrBu2 and subsequent CuCl-catalyzed reaction with dichlorobutene afforded chloroallyl-ene (90), that was immediately converted into methoxy derivative (91). A second Cu-catalysed reaction with Cp2ZrBu2 and isobutenyl chloride led to (92). Treatment of (92) with Cp2ZrBu2 and CO afforded a mixture of cis-fused C and D rings steroids (93), (94) and (95). Compound (94) was converted into (95) upon reductive dehalogenation. An alternative route was employed to obtain trans-C,D-rings fused steroids (Scheme 14b). Starting from (96), bis-alkene (97) was obtained following the same route as for (92). Carbonylation of (97) under thermodynamic control afforded the C,D-trans-fused steroid (98), in low yield due to thermal decomposition. Br O O AcO TMS O AcO O PdCl2(PPh3)2 (2 mol%),
CuI (2 mol%), TEA, 88% 2) KF, THF, MeOH, rt, 92% AuCl3 (3 mol%) MeCN, 80 °C O O OAc O O OH O O + 82 : 83 : 84 = 22%: 7%: 62% + KHSO4, Ac2O, 71% H2, Pd/C, EtOAc K2CO3, THF, MeOH, 86% O + H O O H rac-87 57% 36% 1) O O OH 80 81 82 83 84 86 O O Steps 85
Scheme 14: Kotora´s first strategy based on the repetitive use of the Negishi reagent.
Two years later the authors published an alternative route30 for the formal total synthesis of estrone (1). Based on the same strategy, but now performing a ring closing metathesis at the end of the synthesis to obtain intermediate (100), that could be converted into rac-estrone (1) according to previous literature31 (Scheme 15). Fluorine derivative (99) was employed in that case to favour the cyclization of the double bonds on the zirconium instead of the oxidative addition. I OH BnBr, NaH, Bu4NI Bu3SnCH=CH2 Pd(PPh3)4 OBn Cp2ZrBu2 Cl Cl CuCl Cl NaOMe/ MeOH 62% 83% 66% OMe Cp2ZrBu2 Cl CuCl H H 59% Cp2ZrBu2 CO, I2 O H H H + O H H H I + OH H H H 13% 12% 11% Bu3SnH, AIBN Br MeO COOH Steps H H MeO Cp2ZrBu2 CO, H+ 20 °C 80 °C Cp2ZrBu2 CO, H+ O H H H + O H H H O H H H 49% 11% 88 89 90 91 95 94 93 96 97 92 a) b) zirconacycle equilibrated 4 h at 80 °C before carbonylation 3 : 1 MeO MeO MeO 98 OMe Cp2ZrBu2 Cl F CuCl F H H MeO 75% H H H H H H MeO MeO MeO ( )-estrone (1) Cp2ZrBu2 Cl CuCl 80% Grubbs 2nd gen. toluene, 90 °C 124 99 100 O HO H H H +-Ref. [15]
Scheme 15: Kotora´s second strategy based on the repetitive use of the Negishi reagent.
Another strategy based on an intramolecular alkylative arylation of an oxabicylic alkene was envisioned for the synthesis of racemic estrone analogues.32 A Diels-Alder reaction between diene (101) and dienophile (102) gave oxabicycle (103). Reduction of the anhydride formed diol (104) which underwent acid catalyzed intramolecular alkylative arylation to form estrone analogues (105) and (106) in a 3 : 1 ratio, respectively (Scheme 16).
Scheme 16: Intramolecular alkylative arylation of an oxabicylic alkene. In 2008, Taber and Sheth reported on an efficient three-step route to a tricyclic steroid precursor.33 Wittig reaction of aldehyde (107) and phosphonium salt (108) gave alkenyl cyclopropane (109). UV irradiation of (109) in the presence of Fe(CO)5 afforded 2-substituted cyclohexenone (110), that upon acid catalyzed cyclization delivered tricyclic steroid precursor (111) (Scheme 17).
Scheme 17: Synthesis of a tricyclic steroid precursor.
Kitagaki et al. reported the synthesis of estrone derivative (112) employing sequential pericylic reactions of ene-diallenes.34 Starting from methylcyclopentenone (113), following a previously published procedure,35 rac-(114) was obtained. Protection via formation of a thioacetal and reduction of the ester group lead to (115), which was subsequently oxidised and underwent Wittig reaction to form aldehyde (116), coupled with (117) to afford (118), and deprotected to obtain diol MeO O O O O MeO O O O O MeO O OH OH LiAlH4 THF, 0 ºC to rt 40% O MeO OH H O OH H OMe + Lewis acid or H+ 3 : 1 70% 101 102 103 104 105 106 MeO H O + Ph3P Br MeO Fe(CO)5 hν MeO O MeO O BF3•Et2O KOtBu rt 107 108 109 110 111
(119). Finally, sequential pericylic reaction of (119) afforded (112) after desulfoxidation (Scheme 18). Scheme 18: Synthesis of estrone derivative (112) by sequential pericyclic reactions of ene-diallenes. An interesting example of a route to “hybrid” steroids has been published by the group of De Groot applying Mukaiyama reactions.36 Two years later, this strategy was used by the same group in the development of new approaches towards the synthesis of (D-homo) steroids.37 The synthesis starts with carbocation formation of tertiary alcohol (49) with ZnBr2. Reaction of TMS enolether (120) with this carbocation, gave (121) in good yield. As such, the silylenol ether is formed by conjugate addition to 2-methyl cyclopentenone followed by trapping of the (regioselectively formed!) enolate. Acid catalysed cyclization of (121) and stereoselective double bond reduction provided rac-(122) in 90% yield over the last two steps (Scheme 19). A strategy to carry out the conjugate addition in an enantioselective fashion has recently been disclosed.38 O Ref. 17 O MeO2C 1) 1,3-propanedithiol, BF3•OEt2, DCM 2) LiAlH4, Et2O HO S S S S 1) Dess-Martin periodinane, DCM 2) Ph3P=CHOMe, THF 3) 10% HCl, acetone HO S S OTHP nBuLi, THF -78 °C 89% 1) TsOH, MeOH 2) NCS, AgNO3, CH3CN, H2O HO OH H H H O 1) PhSCl, Et3N, THF, rt to reflux 32% 113 rac-114 115 116 117 118 119 OTHP 112 2) Raney-Ni, THF, reflux O OSPh PhS [2,3]-sigmatropic rearrangement Ph(O)S S(O)Ph 6π electrocyclization S(O)Ph S cycloaddition[4+2] H H H S S(O)Ph O O O O O O Ph(O) Ph(O) 77% 50% 75%
Scheme 19: De Groot´s synthesis of steroid hybrid (122).
Chung and co-workers published in 2006 a route towards the basic steroid skeleton employing a Pauson-Khand reaction.39 Starting from commercially available β-tetralone, intermediate (123) could be made in a few steps. Intramolecular Pauson-Khand reaction of (123) followed by addition of NMO gave after 5 days steroid derivative (124) (Scheme 20). Although efficient, the product is just reminiscent of the steroid skeleton.
Scheme 20: A Pauson-Khand reaction in the construction of the steroid skeleton.
Another novel strategy to construct terpenes and steroids was developed by Tang and co-workers, in which the key step is an enantioselective palladium-catalyzed dearomative cyclization.40 Ketal (125) was reacted with bromide (126) to give (127) in a low 17% yield. Vinyl triflate formation and benzyl ether deprotection gave (128). This underwent palladium-catalyzed dearomative cyclization in presence of P-chiral biaryl monophosphine ligand (129) to afford product (130) in high yield and very good enantioselectivity (Scheme 21). Scheme 21: Palladium-catalyzed dearomative cyclization by Tang et al. MeO OH + OTMS ZnBr2 DCM, - 10 ºC MeO O 83% pTsOH MeO 79% MeO H H H 1) Pd/ CaCO3 2) Et3SiH, CF3COOH 90% (two steps) 120 121 122 49 O Steps OH H H H O OH 96% 1) Co2(CO)8 toluene, 1 h 2) NMO, 80 ºC, 4 h 123 124 O O O Br BnO + NaH DMSO 17% (80% brsm) O O O BnO 1) Tf2O 2) BBr3 70% O O TfO HO O O O [{Pd(cinnamyl)Cl}2], 129, K2CO3 toluene, 90 ºC N P O Ph Ph 157 125 126 127 128 129 130 90% >99:1 dr
Malacria´s group reported on the diastereoselective synthesis of (rac)-(11)-aryl steroid skeletons via cobalt(I)-mediated [2 + 2 + 2] cyclizations of allenediynes.41 trans-allene-diyne (132) was obtained in 2 steps from alkyne (131) Cyclization gave η4-cobalt-complexed compound (133) in 60% yield as a single diastereomer, which was crystallized from a mixture of pentane/DCM. Finally, treatment of (133) with silica gel gave free (134) in 90% yield (Scheme 22).
Scheme 22: Cobalt(I)-mediated [2 + 2 + 2] cyclization of allenediynes. Lu and Ma reported on a Rh-catalyzed triple allene approach for the stepwise synthesis of steroid-like tetracyclic skeletons.42 trans-RhCl(CO)(PPh3)2-catalyzed cyclization of 1,5-bisallene (135) and monoallene (136) in toluene provided (137) in a moderate yield. Allylation and Diels-Alder reaction of (137) and dienophile (138) furnished tetracyclic compound (139) in good yield and diastereoselectivity. Compound (140) could be synthesized in 3 steps from (137) (Scheme 23).
Scheme 23: A Rh-catalyzed triple allene approach to the steroid skeleton. Sünnemann and de Meijere developed a strategy towards steroids and steroid-like molecules employing Stille coupling – Heck reaction
• O O TMS CpCo(CO)2 xylenes, hν, Δ O CoCp SiO2, DCM, rt O 60% 90% 131 132 133 134 1) KHMDS, THF, -15 °C 2) Ph MsO TMS 4 • 3) K2CO3, MeOH PhO2S SO2Ph • • • OMe O MeO O + trans-RhCl(CO)(PPh3)2 (5 mol%) toluene, 80 ºC PhO2S SO2Ph O OMe O MeO 46% TBAB, K2CO 3 DMF , 35 ºC St eps H H H PhO2S PhO2S O OEt O OMe MeO O 85% Br O OEt PhO2S MeO OMe O O 135 136 137 138 139 140
sequences.43 Bromides (142) and (143), obtained in 4 steps (including a Stille coupling) from enone (141), underwent Heck reaction to furnish (144) and (145) in good yields. Heating in decalin yielded the cyclized products (146) and (147) in 71-75% (Scheme 24). This is an interesting route since it provides unnatural cis-regular-ring fused steroids.
Scheme 24: Steroids and steroid analogues by Stille coupling – Heck reaction sequences.
Another elegant route from de Meijere and co-workers, leading to a great variety of steroids, was to combine Stille coupling and Diels-Alder reaction.44 Preparing diene (148) via a Stille coupling, this could subsequently undergo different high yielding Diels-Alder reactions to afford steroid derivatives (Scheme 25). O OtBu Steps CO2tBu R1 R2 OtBu H Br R1 R2 OtBu H 25 (8 mol%)
nBu4NOAc, tBu acrylate
DMF, MeCN, H2O decalin 215 °C OtBu H CO2tBu R1 R2 (74-79)% (71-75)% 142; R1= OMe, R2= H 143; R1=R2= OCH2CH2O 144; R1= OMe, R2= H 145; R1=R2= OCH2CH2O 146; R1= OMe, R2= H 147; R1=R2= OCH2CH2O 141
Scheme 25: A Stille coupling – Diels-Alder sequence in the synthesis of steroid derivatives.
Gagné´s laboratory published a biomimetic steroid synthesis based on an alkene- terminated cation-olefin cascade reaction. This is initiated by the dicationic platinum complex (PPP)PtI2 (PPP = bis-(2-diphenylphosphanylethyl)phenylphosphane).45 The use of a polar solvent (EtNO2) together with Ph2NMe or a resin bound piperidine base afforded successfully the cyclization of the trienes (149), (150) and (151) to yield (152), (153) and (154) in 89-97% (Scheme 26). O OtBu Steps O O OtBu H benzene, 100 °C O O O CN NC toluene, 100 °C N O O benzene, 95 °C CO2Me MeO2C benzene, 100 °C benzene, 100 °C Cl CN DCM, rt N N N O O Ph O O H H OtBu H O O O O O H H OtBu H CN CN O O H H OtBu H N O O O O H H OtBu H CO2Me CO2Me O O H H OtBu H ClCN O O N N H H OtBu H N O O Ph 77% 67% 65% 79% 84% 80% 141 148
Scheme 26: Platinum-catalyzed cyclisation reactions
Nörret and Sherburn developed a strategy for the synthesis of tetracycles reminiscent to the steroid skeleton. Employing a domino intramolecular Diels-Alder reaction (IMDA), the tetracyles could be obtained in a single step in a stereoselective fashion.46 The domino IMDA precursor (156) was obtained in 6 steps from aldehyde (155). IMDA reaction afforded three stereoisomers; the cis-fused C/D ring system in (157) and (158), resulting from an endo-docking mode, and the trans-fused C/D moiety in (159) from a minor exo-pathway (Scheme 27).
Scheme 27: Domino IMDA cyclization in the synthesis of steroid-like compounds.
Heterocyclic steroids:
The synthesis of heteroatom containing steroids is an interesting field due to the unknown properties of these non-natural steroids. The group of Otto published several syntheses of azasteroids based on Diels-Alder
R R H H [Pt]+ OMe H H [Pt]+ OMe H (PPP)PtI2 AgBF4 EtNO2 N N (PPP)PtI2 AgBF4 EtNO2 149: R=H 150: R=OMe 152, 89%153, 97% 154, 95% 151 O Steps MeO2C O O H H Co2HMe H H H O H H Co2HMe H H H O H H Co2HMe H H H 155 156 158 159 Et2AlCl (1.9 eq) DCM 157: 158: 159 = 72:14:14. 79% 157
reactions.47,48 The combination of Dane´s diene analogue (160) and maleimide (161) yielded TMSO-substituted azasteroids (162) and (163) in a 87 : 13 ratio, respectively. Acid catalyzed hydrolysis afforded azasteroids in very good yield (Scheme 28). Scheme 28: Otto´s azasteroid synthesis.
Ibrahim-Ouali et al. published on the synthesis of aza-thia, 49 aza-seleno and aza-telluro steroids50 prepared via intramolecular Diels-Alder reaction. Intermediates (1659 and (166) were obtained in 3 steps from so-called BISTRO (164) and chloroacetic anhydride. These compounds underwent alkylation with iodides (167) and (168) to generate the corresponding aza-thio and aza-seleno intermediates in moderate yields. Finally, intramolecular Diels-Alder reaction of afforded 5 : 1 and 9 : 1 mixtures of the corresponding heterocyclic steroids in favour of the all-trans-fused ones (Scheme 29).
Scheme 29: Synthesis of aza-thio and aza-seleno steroids.
An elegant extension of this strategy for the synthesis of more than 250 unnatural steroids was recently reported by the Santelli group.51 Starting with BISTRO (164), which was obtained via reductive dimerization of buta-MeO SiO N O O + neat N O O SiO MeO MeOH/H+ N O O O MeO 98% 91% 160 161 162 OMe O N O O SiO MeO 163 OMe O OMe O H H H H H H + N O O O MeO + N O O O MeO N O O O MeO + OMe O OMe O OMe O OMe O H H H H H H H H H H H H H H H H (syn-cis) (anti-cis) 162:163= 87:13 Cl O O O Cl SiMe3 SiMe3 + Step s St eps AcS HO NCSe HO N S HO N S OH H H H N S OH H H H + 5 : 1 Δ ο-xylene 130 ºC, 82% N H 1) , MeCN 2) K2CO3, N I 70% 1) NaBH4, THF/EtOH 2) 62% MeO I Se HO Se OH H H H Se OH H H H + 9 : 1 Δ ο-xylene 130 ºC, 80% MeO MeO 164 165 167 166 168 MeO
1,3-diene in presence of TMSCl, 1,1-disubstituted 2,5-divinylcyclopentanes could be obtained via reaction with various electrophiles e.g. anhydrides and 1,2-diones. The 2,5-divinylcyclopentanes could be coupled with cyclobutenes. Heat-induced ring opening of the benzocyclobutenes and subsequent intramolecular cycloaddition with a vinyl substituent forms the steroid skeletons (169), in overall yields exceding 25% (Scheme 30).
Scheme 30: The BISTRO-strategy for the synthesis of unnatural steroids. Natural occurring steroids
In 2008, the Sherburn group reported the formal total synthesis of triptolide (170),52 an anti-tumour and anti-inflammatory natural product isolated for the first time in 1972 from the Asian vine Tripterygium wilfordii.53 The key steps in this synthesis are two Diels–Alder reactions and a new deoxygenative aromatisation reaction. Two Diels-Alder reactions between (171), (172) and (173) formed (174), that could be transformed into (175). Triptolide was obtained in 8 subsequent steps from (175) (Scheme 31).
Scheme 31: Sherburn´s formal synthesis of triptolide (170).
In 2000, Stoltz, Kano, and Corey published the enantioselective synthesis of the nicandrenones,54 a family of steroid-derived natural products with insect repellent and antifeedant properties.55 Exo-selective Diels-Alder reaction between diene (176) and dienophile (177) provided the
+ Li Me3SiCl + 70% SiMe3 SiMe3 Bistro X1 R3 R2 R1 O O O R1 or O O or... X1 X3 X2 X4 H H H R2 around 250 unnatural steroids R3=I, OMs, COCl
X3= CH 2, O, C=O X1= R4C, N X2= R5C, O, C=O, S, R6N X4= OH, Me, CF 3, Ph 164 169 O O OH O OMe + + Two Diels-Alder reactions O O O O H H H H O OMe O O OH O steps steps H O O H O 171 172 173 174 175 Triptolide170
tetracyclic compound (178). In 16 steps from (178), (179) was obtained. Stille coupling with vinylstannane (180) furnished (181), which upon reduction and epoxidation, formed (182). Conversion into epoxylactone (183) was achieved in 90% yield over two steps. (262) was obtained from (261) by replacement of the Me2PhSi group by a hydroxyl function and β-elimination by acetylation followed by treatment with DBU. Finally, reduction of both carbonyls, selective acetylation of the more reactive lactol hydroxyl, Dess-Martin oxidation, and deacetylation furnished nicandrenone (185) (Scheme 32).
Scheme 32: Corey´s group synthesis of the nicandrenones.
Jung and Yoo published in 2011 the first total synthesis of the cardiac glycoside rhodexin A (186),56 which had been isolated in 195157 from the Japanese evergreen Rhodea japonica. Silyl-protected enynone (187) underwent metathesis reaction with Grubbs 1st generation catalyst to afford diene (188). Inverse-electron-demand Diels-Alder reaction between (188) and vinyl silylenol ether 189 yielded the adduct in 87% forming 4 contiguous stereocenters in 1 step. Dihydroxylation and protection of the diol was accomplished to give (190), which could be transformed into (191) in several steps. Cross metathesis followed by cyclization furnished (192) in good yield. In seven additional steps (193) could be obtained.
PhMe2Si O + O OTES O OTES O PhMe2Si H H Steps ONf O PhMe2Si OH H H O MeAlCl2 DCM, -78 °C + Bu3Sn OH TBSO O PhMe2Si OH H H O OH OTBS O PhMe2Si OH H H O OH OTBS O O PhMe2Si OH H H O O O O O OH H H O O O O O OH H H O O OH O Pd(PPh3)4, CuCl, LiCl, DMSO 1) H2 (1 atm), Rh(nbd)Id-ppb)BF4 2) t-BuOOH, VO(acac)2 1) Bu4NF
2) NaOCl, cat. TEMPO, cat. KBr
1) Hg(OAc)2, AcOOH, AcOH, 23 °C
2) Ac2O, Et3N, DMAP, DCM, 23 °C 3) DBU, DCM, 23 °C 1) DIBAL-H 2) Ac2O, Et3N 3) Dess-Martin periodinane 4) K2CO3, MeOH 176 177 178 179 180 181 182 183 184 185
Hydrolysis of the TBS ether and subsequent reaction with the Bestmann reagent formed the butenolide in 70% yield. Selective removal of the C-3 acetate, in the presence of the C-11 acetate, employing HCl in methanol gave diol (194) in good yield. Finally, glycosidation and deprotection provided rhodexin A (186) (Scheme 33). Scheme 33: The first synthesis of rhodexin A (186) by Jung and Yoo. Conclusion This review compiles the recent developments in the field of steroid total synthesis from 2000 on. Many different approaches have been employed, transition metal catalysis as well as Diels-Alder and other pericyclic reactions reaction being dominant. As for the latter, this is not surprizing, as these are the strategies par excellence to form cyclic compounds. Nevertheless, organocatalysis, more precise asymmetric organocatalysis is an important novel approach, and in need, as many of the reported syntheses reported earlier are racemic. In general one can state that several steroid skeletons and a series of closely related derivatives are now readily available by total synthesis. Where steroid synthesis was once considered equivalent with lengthy synthesis routes, this is no longer the case. However, even slight deviations in carbon skeleton and substitution pattern can change this picture and add a series of additional
O
TESO TESO
O Grubbs 1st gen. cat. benzene, 93% + TESO Tf2NH, DCM, -78 °C 87% H O TESO OTES H O TESO OTES O O 1) OsO4, NMO, 73% 2) Me2C(OMe)2, CSA (cat.),88% H O O OTES O O H H O OTES O O H O H AcO OH O H TBSO Steps OTBS H H AcO OH H HO H O O H HO OH H O H O O O HO HO HO B O O Grubbs 2nd gen. cat.
NaBO3•4H2O, 57% 2) 10% aq. KOH, MeOH, 60 °C, 70% 1) 1) Glycosidation 2) Deprotection 1) pTsOH, MeOH, 91% 2) Ph3P=C=C=O, benzene, Et3N, 70% 3) 10% aq. HCl, MeOH 187 188 189 190 193 192 194 Rodhexin A186 H O O OTES O O H O O OTES 1) Dess-Martin periodinane 2) Pb(OAc)4 PhH, 61% 1) CH2N2 CHCl3, 0 °C 72% 2) 1,2-C6H4Cl2 180 °C, 52% 1) Li/NH3 THF, -78 °C 2) allyl bromide THF, 52% 191 O O
transformations. As the production of the steroids used in medicine has been optimized and streamlined for years, there is currently limited interest of the pharmaceutical industry in novel steroid synthesis. In chemical biology, however, interest in steroids and steroid-derivatives is increasing as steroids play important and multifaceted roles. Contrary to lipids, that are difficult to label without disturbing their structure, steroids can be labelled and their fate in the cell can be studied with spectroscopic techniques.
In this thesis, we will show our approach to the enantioselective synthesis of steroids derivatives. Therefore, we considered that a review of the current state of art in steroids synthesis would be an interesting addition to the topic. In further chapters, methodoly to perform enantioselective Cu-catalyzed 1,4 and 1,2-additions will be investigated. And finally, the application of this methodoly to the steoid synthesis will be shown.
References
(1) Gardner, J. N.; J.; Gnoj, O.; Watnick, A. S.; Gibson, J.; Steroids, 1964, 4, p.801 (2) Ananchenko, S. N.; Limanov, V. Y.; Leonov, V. N.; Rzheznikov, V. N.; Torgov, I. V., Tetrahedron, 1962, 18, p.1355 (3) Zhdanov, R. I.; Corey, E. J.; Steroids, 2009, 74, p.723 (4) Funk, R. L.; Vollhardt, P. C.; J. Am. Chem. Soc., 1979, p.215 (5) Woodward, R. B.; Sondheimer, F.; Taub, D.; J. Am. Chem. Soc., 1951, 73, p.3547 (6) Skoda-Földes, R.; Kollár, L.; Chem. Rev.; 2003, 103, p.4095 (7) Singh, R.; Panda, G.; Tetrahedron, 2013, 69, p.2853 (8) Sarabèr, F. C. E.; de Groot, A.; Tetrahedron, 2006, 62, p.5363 (9) Mackay, E.; Sherburn, M.; Synthesis (Stuttg)., 2014, 47, p.1 (10) Biellmann, J. F.; Chem. Rev.; 2003, 103, p.2019 (11) Chapelon, A. S.; Moraléda, D.; Rodriguez, R.; Ollivier, C.; Santelli, M.; Tetrahedron, 2007, 63, p.11511 (12) Soorukram, D.; Knochel, P.; Org. Lett.; 2007, 9, p.1021(13) Quinkert, G.; Del Grosso, M.; Döring, A.; Döring, W.; Schenkel, R. I.; Bauch, M.; Dambacher, G. T.; Bats, J. W.; Zimmermann, G.; Dürner, G.;
Helv. Chim. Acta, 1995, 78, p.1345
(14) Sugahara, T.; Ogasawara, K.; Tetrahedron Lett.; 1996, 37, p.7403 (15) Foucher, V.; Guizzardi, B.; Groen, M. B.; Light, M.; Linclau, B.; Org.
Lett.; 2010, 12, p.680
(16) Betík, R.; Kotora, M.; Eur. J. Org. Chem.; 2011, 2011, p.3279
(17) Tietze, L. F.; Wiegand, J. M.; Vock, C. J.; Organomet. Chem.; 2003, 687, p.346
(18) Hu, Q.-Y.; Rege, P. D.; Corey, E. J.; J. Am. Chem. Soc.; 2004, 126, p.5984 (19) Yeung, Y. Y.; Chein, R. J.; Corey, E. J.; J. Am. Chem. Soc.; 2007, 129,
(20) Weimar, M.; Dürner, G.; Bats, J. W.; Göbel, M. W.; J. Org. Chem.; 2010, 75, p.2718 (21) Schotes, C.; Mezzetti, A.; J. Am. Chem. Soc.; 2010, 132, p-3652 (22) Pattenden, G.; Gonzalez, M. A; McCulloch, S.; Walter, A.; Woodhead, S.; J. Proc. Natl. Acad. Sci. U. S. A., 2004, 101, p.12024 (23) Prévost, S.; Dupré, N.; Leutzsch, M.; Wang, Q.; Wakchaure, V.; List, B.; Angew. Chem. Int. Ed. Engl.; 2014, 53, p.8770 (24) Belani, J. D.; Rychnovsky, S. D.; J. Org. Chem.; 2008, 73, p.2768 (25) Corey, E. J.; Huang, A. X.; J. Am. Chem. Soc., 1999, 121, p.710 (26) Schwa, S.; Ring, S.; Weber, G.; Teichmiiller, G.; Tetrahedron, 1994, 50, p.10709 (27) Tietze, L. F.; Krimmelbein, I. K.; Chemistry, 2008, 14, p.1541 (28) Hildebrandt, D.; Dyker, G.; J. Org. Chem., 2006, 71, p.6728 (29) Herrmann, P.; Kotora, M.; David, S.; Org. Lett., 2006, 8, p.1315
(30) Herrmann, P.; Budesínský, M.; Kotora, M.; J. Org. Chem., 2008, 73, p.6202
(31) Bartlett, P. A.; Johnson, W. S.; J. Am. Chem. Soc., 1973, 95, p.7501 (32) Li, W. D. Z.; Wei, K.; Org. Lett., 2004, 6, p.1333
(33) Taber, D. F.; Sheth, R. B.; J. Org. Chem., 2008, 73, p.8030
(34) Kitagaki, S.; Ohdachi, K.; Katoh, K.; Mukai, C.; Org. Lett.; 2006, 8, p.95 (35) Ito, Y.; Nakatsuka, M.; Saegusa, T.; J. Am. Chem. Soc.; 1982, 104,
p.7609
(36) Sarabèr, F. C. E.; de Groot, A.; Tetrahedron Lett., 2004, 45, p.9431 (37) Sarabèr, F. C. E.; Baranovsky, A.; Jansen, B. J. M.; Posthumus, M. A.; de
Groot, A.; Tetrahedron, 2006, 62, p.1726
(38) Calvo, B. C.; Madduri, A. V. R.; Harutyunyan, S. R.; Minnaard, A. J.;
Adv. Synth. Catal.; 2014, 356, p.2061
(39) Kim, D. H.; Kim, K.; Chung, Y. K.; J. Org. Chem.; 2006, 71, p.8264 (40) Du, K.; Guo, P.; Chen, Y.; Cao, Z.; Wang, Z.; Tang, W.; Angew. Chemie
Int. Ed., 2015, 54, p.3033
(41) Petit, M.; Aubert, C.; Malacria, M.; Tetrahedron, 2006, 62, p.10582 (42) Lu, P.; Ma, S.; Org. Lett.; 2007, 9, p.5319
(43) Sünnemann, H. W.; de Meijere, A.; Angew. Chem. Int. Ed. Engl.; 2004, 43, p.895
(44) Sunnemann, H. W.; Hofmeister, A.; Magull, J.; Banwell, M. G.; De Meijere, A.; Org. Lett.; 2007, 9, p.517
(45) Sokol, J. G.; Korapala, C. S.; White, P. S.; Becker, J. J.; Gagné, M. R.;
Angew. Chemie Int. Ed., 2011, 50, p.5658
(46) Norret, M.; Sherburn, M. S.; Angew. Chem. Int. Ed.; 2001, 40, p.4074 (47) Bodtke, A.; Stubbs, M. T.; Otto, H. H.; Monatshefte für Chemie, 2006,
137, p.83
(48) Sultani, A.; Dietrich, H.; Richter, F.; Otto, H. H.; Monatshefte fur
Chemie, 2005, 136, p.1651
(49) Oumzil, K.; Ibrahim-Ouali, M.; Santelli, M.; Tetrahedron Lett., 2005, 46, p.5799
(50) Ibrahim-Ouali, M.; Tetrahedron Lett., 2009, 50, p.1607
(51) Maurin, P.; Moraleda, D.; Pellissier, H.; Rodriguez, R.; Santelli, M.;
Synlett, 2015, 26, p.725
(52) Miller, N. a; Willis, A. C.; Sherburn, M. S.; Chem. Commun.; 2008, p.1226
(53) Kupchan, S. M.; Court, W. A.; Dailey, R. G.; Gilmore, C. J.; Bryan, R. F.; J.
Am. Chem. Soc., 1972, 502, p.7194
(54) Stoltz, B. M.; Kano, T.; Corey, E. J.; J. Am. Chem. Soc., 2000, 122, p.9044
(55) Nalvandov, O.; Yamamoto, R. T.; Fraenkel, G. S.; J. Agric. Food Chem., 1964, 12, p.55
(56) Jung, M. E.; Yoo, D.; Org. Lett., 2011, 13, p.3766 (57) Nawa, H., Proc. Jpn. Acad., 1951, 27, p.436
Chapter 2: Cu-catalyzed conjugate addition of Grignard
reagents to 2-methylcyclopentenone and sequential
enolate alkylation
Introduction
The enantioselective Cu-catalysed conjugate addition of organometallic reagents has over the years become a well-established tool for asymmetric C-C bond formation.1–10 Various organometallic reagents are employed in this transformation and the substrate scope comprises a variety of Michael acceptors, both cyclic and acyclic. In our recent work,11–13 we focus on the use of Grignard reagents, because of their straightforward preparation from readily available alkyl bromides. In the cyclic series, the use of Grignard reagents in the copper-catalyzed asymmetric Michael addition has been reported for cyclopentenone,14–16 cyclohexenone,5,14,16–22 cycloheptenone,14–17 β-substituted cyclic enones,22,23 and lactones.15,17,18,24,25 In these studies, however, α-substitution (2-substitution) is lacking, a situation no different from reports with other organometallic reagents. Until very recently, the sole report in this field came from Vuagnoux-d’Augustin and Alexakis, and comprized the enantioselective addition of Me3Al and Et3Al to 2-methyl cyclohexenone23 (Scheme 1). This knowledge was subsequently used by Helmchen et al. in a synthesis of pumiliotoxin C.26 Neither the use of Grignard reagents nor 2-methyl cyclopentenone had been used until during the preparation of this manuscript Mauduit, Alexakis et al. reported the successful application of Cu(I)-N-heterocyclic carbene complexes in the asymmetric addition of Grignard reagents to 2-
(B. C. Calvo; A. V. R. Madduri; S. R. Harutyunyan and A. J. Minnaard; 2014,
methyl cyclopentenone and –hexenone.27 The resulting magnesiumenolates were subsequently alkylated to provide a quaternary stereocenter vicinal to the initially formed tertiairy stereocenter.
Here we report the asymmetric conjugate addition of Grignard reagents to 2-methyl cyclopentenone with a copper catalyst based on Rev-JosiPhos (L1) and the subsequent enolate alkylation with a variety of electrophiles. Scheme 1. Previous and current work. O Et3Al O CuTC (2 mol%) L11 (4 mol%) Et2O, -30 °C 82% yield, dr = 80 : 20 er (trans) = 92 : 8 O OP N L11 Previous work (Alexakis' group):
Our work: DMPU (10 eq) (3.5 eq) O 90% yield dr = 90:10 Br 20 h, rt O CuBr•SMe 2 (5 mol%) L1 (6 mol%) tBuOMe, -78 ˚C, 3 h OMgBr MgBr (1.7 eq) . Fe Cy2P Ph2P L1 (S,RFe)-Rev.Josiphos 2 1 3 O 1 Cu(OTf)2 (0.75 mol%) L12 (1 mol%) Et2O, -30 ˚C, 4 h MgBr (1.2 eq) OMgBr HMPA (10 eq) (2 eq) Br -30 ˚C to rt, 12 h er = 91:9 O 52% yield dr = 98:2 er = 92:8 N N HO PF6 L12 11 18 Recent work (Alexakis' group):
Results and Discussion
As a starting point, 2-methyl cyclohexenone (2) was chosen as the model substrate. Various copper salts, and chiral ligands (Table 1, Figure 1) were studied in the conjugate addition of ethylmagnesium bromide. Table 1. Conjugate addition of ethylmagnesium bromide to (2); variation in copper salt and ligandEntry Ligand Copper salt 3: 3a b) er(trans)d) e)
1 - CuBr•SMe2 38:56 - 2 L1 CuBr•SMe2 96:3 70:30 3 L2 CuI 73:26 60:40 4 L3 CuCl 29:62c) 50:50 5a) L3 CuCl 42:56c) 50:50 6 L3 CuBr•SMe2 35:64c) 50:50 7a) L3 CuBr•SMe2 21:79c) 50:50 8 L4 CuBr•SMe2 61:37 55:45 9 L5 CuBr•SMe2 94:5 52:48 10 L6 CuBr•SMe2 91:9 55:45 11 L7 CuBr•SMe2 65:32 54:46 O
Copper salt (5 mol%) Ligand (6 mol%) tBuOMe (0.1 M), -78 °C, 5-10 h EtMgBr (1.3 eq) O + + HO OH 2 3 3a 3b
12 L8 CuBr•SMe2 91:6 63:39
13 L9 CuBr•SMe2 94:4 52:48
14 L10 CuBr•SMe2 95:5 53:47
a) Reactions were performed at –10 °C. b) Selectivity determined by GC-MS. c) No ee found for 3a. d) Enantioselectivities determined by chiral GC, see SI. e) Diastereomeric ratios between 60 : 40 and 80 : 20.
Figure 1. Chiral ligands used in this study.
The combination of L1 and CuBr•SMe2 resulted in a very high selectivity for the conjugate addition product (3), considerably higher than the reaction with just catalytic copper. A moderate but distinct enantioselectivity was obtained as well (entry 2). With (L2) only low enantioselectivity was
O P N O L6 Ph Ph (Ra,Sc,Sc)-Phosphoramidite O P N O L5 Ph Ph (Sa,Sc,Sc)-Phosphoramidite Fe PPh2 PCy2 L4 (R,SFe)-Josiphos P(Tol)2 P(Tol)2 L2 Fe Cy2P Ph2P L1 (S,RFe)-Rev-JosiPhos (R)-TolBinap Fe Ph2P Me2N Ph2P L3 (R,RFe)-TaniaPhos Fe PPh2 P(tBu) 2 L7 (R,SFe)-Josiphos-type MeO MeO PPh2 PPh2 L8 (R)-Meo-Biphep-type Fe Ph2P Me2N L9 (R,R,SFe)-MandiPhos-type Ph2P Me2N Fe L10 (R,RFe)-WalPhos-type P Ph2P F3C CF3 CF3 F3C
obtained whereas (L3) provided the racemate, as did (L4 – L7). Notably, only when employing (L3), no catalyst control was observed, while on the contrary, (L5 – L6) did provide an excellent regioselectivity in favour of (3). (L8) provided a small enantioselectivity whereas (L9) and (L10) gave racemates, though with very high regioselectivity. With all ligands, the
trans:cis ratio in (3) varied as expected23,28 between 60 : 40 and 80 : 20.
The situation improved when 2-methyl cyclopentenone (1) was studied in the conjugate addition of methyl- and ethylmagnesium bromide, employing the combination of CuBr•SMe2 and (L1) as the most successful catalyst thus far (Scheme 2).
Scheme 2. Cu/L1 catalyzed asymmetric conjugate addition reactions. a dr =
trans:cis ratio. b er of the major trans diastereomer. The absolute
configuration of (5), and in analogy (6), was established by comparison of the optical rotation with the literature value.29
The addition of methylmagnesium bromide to (2) showed like in the case of EtMgBr a high selectivity for the formation of the conjugate addition product, however without enantioselectivity. The same reaction with
2-O O O RMgBr + CuBr•SMe2 (5 mol%) L1 (6 mol%) MTBE, -78 °C, 5-10 h (1.3 eq) . O O 89% dra= 66:34 erb= 50:50 dra= 81:19 erb= 70:30 96% 76% 98% dra= 82:18 erb= 78:22 dra= 90:10 erb= 92:8 3 4 5 6 n 1, n=0 2, n=1 R= Et, Me
methyl cyclopentenone (1), however, resulted in a significant er (78 : 22). Rewarding results were obtained in the Cu/(L1) catalyzed addition of EtMgBr to (1); virtually full regioselectivity and an er of 92 : 8! Therefore we decided to focus on (1) as the substrate, and determine the scope of Grignard reagents that could be added enantioselectively (Table 2).
Table 2. Grignard reagent scope of the conjugate addition to 2-methyl cyclopentenone (1).
Entry R Product Conversion a)/ Yield erc) d) e)
1 6 98%/ n.db) 92:8 2 7 99%/ 95% 91:9 3 8 95%/ n.db) 50:50 4 9 94%/ n.db) 92:8 5 10 76%/ 70% 85:15 a Conversions determined by GC-MS. b Not determined due to the volatility of the product. c Enantioselectivities determined by chiral GC, see SI. d er of the major trans diastereomer. e Diastereomeric ratio’s between 85:15 and 90 : 10.
Table 2 shows that high er’s are obtained with a set of linear saturated and unsaturated Grignard reagents. Linear alkyl and alkenyl substituents, or
RMgBr + CuBr•SMe2 (5 mol%) L1 (6 mol%) tBuOMe, -78 °C, 3-4 h (1.7 eq) O O R . Fe Cy2P Ph2P L1 (S,RFe)-Rev.Josiphos 1 (6-10)
substituents branched at the δ-position, perform very well. This makes the catalyst system highly complementary to the aforementioned Mauduit-Alexakis system that performs best with sterically more hindered α-branched Grignard reagents. Unfortunately, when the substrate was changed for a more sterically hindered one, 2-pentyl cyclopentenone, the conjugate addition resulted in very poor conversions. Conjugate addition to cyclic enones results, in addition to a C–C bond and a stereocenter, also in regioselective formation of an enolate. Especially in the current case, subsequent reaction with a carbon electrophile is attractive, as a second, quaternary, stereocenter is formed. Reaction of in situ formed enolates has been reported a number of times,20,30–33 however, unlike lithium enolates, magnesium enolates react only sluggishly and additives or co-solvents are mostly used to accelerate the reaction.
Protonation of the enolate formed, gave already reasonable diastereomeric ratio’s in the range of 6 : 4 to 8 : 2 in favour of the trans compound.29 Benzyl bromide was selected to study the alkylation under various conditions34 (Table 3). In the absence of additives and cosolvents no reaction took place (entry 1). Table 3. Conjugate addition followed by benzylation Entry Additive (No. of eq. A) Co-solvent (MTBE:Co-solvent) Conversion a) drb) 1 - - - - O OMgBr MgBr(1.7 eq) Additive BnBr (3.5 eq) O Solvent, -78 ˚C to rt, 20 h . CuBr•SMe2 (5 mol%) L1 (6 mol%) tBuOMe, -78 ˚C, 3 h Full conversion, er = 91:9 1 11
2 HMPA(5) - 95% 90:10 3 TMEDA(3) - 20% 89:11 4 - THF (1:2) 20% 98:2 5 HMPA(5) THF (2:3) 51% 90:10 6 - DME (1:2) - - a Conversions determined by GC-MS. b Diastereoselectivity determined by GC-MS. The absolute configuration was established by comparison of the optical rotation with analogues in the literature.27 The relative configuration was determined by NMR (NOESY), see SI.
When HMPA was added, the reaction reached almost full conversion with high diastereoselectivity (entry 2), whereas with TMEDA the conversion decreased dramatically (entry 3). The addition of coordinating solvents was also studied. The use of THF as a co-solvent did not improve the conversion although it afforded a nearly complete diastereoselectivity. Addition of DME resulted in no reaction. In order to get a more complete picture, a set of additional electrophiles was studied as well, under mostly identical conditions. Table 4. Conjugate addition followed by reaction with allyl iodide. Entry Additive (No. of eq. A) Co-solvent (MTBE:Co-solvent) Conversion d) dre) 1a) - - - - 2a) HMPA(3) - 40% 80:20 3b HMPA(5) - 65% 70:30
4c HMPA(5) - 43% 70:30
5 - THF (1:2) 60% 90:10
6 HMPA(5) THF (2:3) 89% 80:20
7 - DME (1:2) 93% 90:10
a Allyl bromide was used as electrophile. b Reaction performed at 42 °C. c Reaction performed at 70 °C. d Conversions determined by GC-MS. e Determined by GC-MS.
As in the benzylation, the allylation reaction with no additives showed no conversion (Table 4, entry 1). Remarkably, however, addition of HMPA now only showed moderate improvement whereas DME increased the conversion to 93%! (entry 7). Tert-butyl bromoacetate (Table 5), turned out to be rather unreactive in this reaction and also the diastereomeric ratio dropped. Reaction with propargyl bromide gave upon addition of 3 eq HMPA incomplete conversion (75%) and a dr of 85 : 15.
Table 5. Alkylation reaction followed by reaction with tert-butyl bromoacetate. Entry Additive (No. of eq. A) Co-solvent (MTBE:Co-solvent) Conversionc) drd) 1a) HMPA (5) - 25% 60:40 2b) HMPA(5) - 50% 60:40 3 HMPA(5) THF (2:3) 53% 50:50 4 - DME (1:2) 52% 50:50
determined by GC-MS. d Determined by GC-MS.
The picture that arose from this study is that although high conversion and diastereoselectivities could be obtained for most electrophiles (tert-butyl bromoacetate being an exception), the optimal reaction conditions are highly dependent on the electrophile. The addition of HMPA was in general most effective but, on the other hand, is toxic. DMPU (1,3-dimethyltetrahydropyrimidine-2(1H)-one) can often be used as a versatile alternative, and is mostly applied in somewhat larger amounts.35 The addition of 10 eq of DMPU was therefore studied (Scheme 3). We were pleased to see that this procedure gave consistently excellent conversions, good isolated yields and diastereomeric ratio’s for the various electrophiles.36 Also tert-butyl bromoacetate performed well now, and also methyl iodide (not used earlier) gave close to complete conversion (Scheme 3). The diastereoselectivity is not strongly dependent on substitution at the 3-position. The similar reaction sequence with EtMgBr and benzyl bromide afforded a dr of 84 : 16, whereas the combination of isobutylmagnesium bromide and tert-butyl bromoacetate afforded a dr of 85 : 15.
44
Scheme 3. Conjugate addition followed by α-alkylation in
tBuOMe/DMPU. (3.5 eq of the electrophile and 10 eq of DMPU were
used.)37 a Determined by GC-MS. b Determined by chiral GC, see SI. c
cis/ trans mixture
Conclusions
An efficient catalyst system has been identified for the conjugate addition of Grignard reagents to 2-methyl cyplopentenone (1). The products are obtained in good yields, and in high enantioselectivities employing CuBr•SMe2 and Rev-JosiPhos (L1) in t-BuOMe at –78 °C. The method is nicely complimentary with the one recently reported by Mauduit and Alexakis. Linear Grignard reagents perform well, whereas O CuBr•SMe2 (5 mol%) L1 (6 mol%) tBuOMe, -78 ˚C, 3 h OMgBr MgBr (1.7 eq) . er (trans)b = 90:10 er (cis)b = 93:7 DMPU 20 h, rt O 75% yieldc dra = 82:18 I DMPU O 90% yieldc dra = 90:10 Br 20 h, rt 20 h, rt Br O 60% yieldc dra = 82:18 DMPU MeI 20 h, rt O 67% yieldc Br O 70% yieldc dra = 80:20 O O O O 20 h, rt DMPU 1 11 12 13 15 16 DMPU
45
with their reported Cu-NHC catalyst system, branched Grignard reagents perform superior. A thorough study on the subsequent alkylation of the regioselectively formed enolate identified DMPU as an essential and superior additive. This sequential conjugate addition – enolate alkylation leads in a one-pot reaction to the formation of two vicinal stereocenters one of which is quaternary. The products made this way are highly useful building blocks in the synthesis of natural products.
