The Total Synthesis of Discodermolide
Me Me Me HO Me
O O OH O NH2 Me Me OH O Me Me OH The Journey from 7 mg to 60 g
Literature Group Meeting by Dan Caspi
Monday June 28, 2004 • 147 Noyes, 8 PM
Outline of Topics for Discodermolide
I. Introduction: Partial Bibliography: a. Structure and Bioactivity Gunasekera, S. P. J. Org. Chem. 1990, 55, 4912-5. b. Historical Outline (-) Schreiber, S. L. J. Am. Chem. Soc. 1993, 115, 12621-2. II. Smith Synthesis (-) Smith, A. B. J. Am. Chem. Soc. 1995, 117, 12011-2. (+) Schreiber, S. L. J. Am. Chem. Soc. 1996, 118, 11054-80. a. First-Generation Route (-) Myles, D. C. J. Org. Chem. 1997, 62, 6098-99. b. Second-Generation Route (+) Marshall, J. A. J. Org. Chem. 1998, 63, 7885-92. c. Third-Generation Route (+) Smith, A. B. Org. Lett. 1999, 1, 1823-6. (+) Smith, A. B. J. Am. Chem. Soc. 2000, 122, 8654-64. III. Paterson Synthesis (+) Paterson, I. Angew. Chem. Int. Ed. 2000, 39, 377-80. (+) Paterson, I. J. Am. Chem. Soc. 2001, 123, 9535-44. a. First-Generation Route b. Second-Generation Route (+) Kinder, F. R., Jr. Org. Lett. 2003, 5, 1233-6. (Formal Total Synth.) (+) Smith, A. B. Org. Lett. 2003, 5, 4405-8. IV. Novartis Synthesis (+) Myles, D. C. J. Org. Chem. 2003, 68, 6646-60. (+) Paterson, I. Org. Lett. 2003, 5, 35-8. a. Hybrid Smith-Paterson Approach (+) Mickel, S. J. Org. Proc. Res. Dev. 2004, 8, 92-130 (5 papers). V. Conclusion (+) Loiseleur, O. Org. Proc. Res. Dev. 2004, ASAP.
Reviews:
Chemtracts 2000, 13, 229-36. Paterson, I. Eur. J. Org. Chem. 2003, 2193-2208. C&E News 2004, 82(9), 33-35. The Structure of Discodermolide
• Structure determined by extensive spectroscopic studies as well as single-crystal X-ray.
• Linear polypropionate chain containing 13 stereocenters, 6 of which bear hydroxyls, remaining 7 bear methyls.
• Contains a tetrasubstituted !-lactone, two di- and one trisubstituted (Z)-alkene.
• Contains a carbamate moiety and a terminal (Z)-diene.
• Discodermolide adopts a U-shaped conformation, where the internal (Z)-alkenes act as conformational locks by minimizing A(1,3) strain and syn-pentane interactions along the backbone.
• The !-lactone is held in a boat-like conformation.
(Hydrogens omitted for clarity) 24 Me Me Me 23 8 HO 17 19 22 9 Me 7 21 O O OH O NH2 Me Me 1 5 13 OH O Me Me OH
Why is Total Synthesis of Discodermolide Important?
Me Me Me HO Me
O O OH O NH2 Me Me OH O Me Me OH
Bioactivity:
• Initial studies revealed that (+)-discodermolide was an apparent immunosuppresant.
• Later studies revealed that Discodermolide stabilizes microtubes faster and more potently than any of the other known MTS agents, and is a potent inhibitor of tumor cell growth in vitro including Taxol-resistant cells. It is also more water soluble than Taxol.
• Promising synergy observed when used in combination with Taxol.
• Licensed by Novartis in 1998 to develop it as a new-generation anticancer drug. Currently, in Phase I Clinical Trials.
Production:
• From deep-water marine sponge, Discodermia dissoluta, which must be harvested by manned submersibles off the Bahamas at a depth in excess of 33 m.
•!Natural material is scarce: 0.002% w/w from frozen sponge (7 mg isolated from 454 g of crude)
• Fermentation has not been succesful to date. Discodermolide "Highlights" Timeline
Me Me Me HO Me
O O OH O NH2 Me Me OH O Me Me OH
Harbor Branch Oceanographic Institution, Inc., FL Led by S. Guanasekera. Isolation Paper. A. B. Smith. Gram-Scale Synthesis. S. Schreiber. Chem. Biol. 1994. J. Org. Chem. 1990, 55, 4912-5. Org. Lett. 1999. Correction: J. Org. Chem. 1991, 56, 1346. First Natural Enantiomer Synthesis.
1990 1995 2000 2005
S. Schreiber. First Total Synthesis. Licensed by Novartis. S. Mickel. Novartis Process Group.
J. Am. Chem. Soc. 1993. Org. Res. Proc. Dev. 2004.
Amos B. Smith, III: First Generation Approach
Wittig Olefination Me Me Me The absolute HO stereochemistry was Me Yamamoto unknown when they Olefination O O OH O NH2 began the synthesis. Me Me
OH Negishi Cross- O Me Me Coupling OH
H Me Me Me TBSO PMBO O Me I I A EtS O B C Me Me TBSO O O OTBS Me Me PMP OTBS
Me Me PMBO N Can these fragments be derived O from a common precursor? TBSO O Denotes the repeating stereochemical triad. J. Am. Chem. Soc. 1995, 117, 12011-2. Amos B. Smith, III: Constructing the Common Precursor
NH
1. PMBO CCl3
PPTS, CH2Cl2/Cyclohexane Me Me Me 2. LiAlH4, THF HO PMBO N CO2Me 3. Swern [O] O 4. Bn OH O "Roche Ester" (59% overall) N O Common Precursor
O O Can be executed on 50g scale n-Bu2BOTf, Et3N, CH2Cl2
5. HN(Me)OMe•HCl, AlMe3, THF
This was Schreiber's SM as well.
!
PMB =
OMe
Amos B. Smith, III: Constructing Fragment A
Me Me 1. TBSOTf, 2,6-lut., CH2Cl2 Me Me 2. H2, Pd(OH)2/C, EtOH PMBO N N O S O 3. SO •Py, Et N, DMSO 3 3 S OH O 4. TMSS(CH2)3STMS, ZnCl2, Et2O TBSO O (71%)
1. DIBAL, THF Me Me Me Me 2. MeOH, CH(OMe) , 3 BnO 1. PhI(CF3CO2)2 TsOH•H2O OMe MeOH, H O OMe 2 BnO HO 3. t-BuLi, THF, HMPA; S S OTBSOMe 2. Me4NBH(OAc)3 OH OH OTBSOMe anti:syn > 99:1 BnO O (78%) (71%)
HO 1. TBSOTf, 2,6-lut., CH2Cl2 OBn H 2. H2, Pd/C, EtOH, EtOAc MeO O TBSO TsOH•H2O 3. EtSH, MgBr2, Et2O O (!:" = 6:1) PhH EtS O Me Me 4. Swern [O] !:" = 2:1 OTBS A (87%) (70%) Me Me OTBS anomers were separable, anomers separated but carried forward at the thioacetal stage Amos B. Smith, III: Constructing Fragment B
Br Me Me 1. 1. TBSOTf, 2,6-lut., Me Me Ph3P CO2Et PMBO N CH2Cl2 O PMBO O PhH, ! HO O 2. DIBAL, THF 2. DIBAL, CH2Cl2 TBSO H (88%) 3. MsCl, Et3N, CH2Cl2 4. LiBHEt3, PhH I n-BuLi, I - + 2 (Z:E = 8.5:1, 64%) I Ph3P Et I-Ph P+ 3 or NaHMDS, THF
(Z:E = 6:1, 41%)
PMBO Me Br PMBO Zn, NiBr2, KI Me I Me Me B HMPA/DMF OTBS Me Me (86%, 55% over 5 steps) OTBS
olefin geometry checked by halogen-metal exchange, only t-BuLi in Et2O completely prevented a retro-[1,5]-Brook rearrangement.
Amos B. Smith, III: Constructing Fragment C
Bn Me Me H N O Me Me PMBO N 1. DDQ, CH2Cl2 O O O O O O 2. LiAlH4, THF HO O n-Bu2BOTf, Et3N, CH2Cl2 (62%) (80%)
OMe
Bn Me Me Me Me Me Me Me Me Me N O I 1. TBSOTf, 2,6-lut., CH2Cl2 I O O OH O O TBSO O O 2. LiBH4, EtOH, THF O O OTBS 3. Ph3P, I2, Imid., PhH/Et2O PMP PMP (77%) PMP C Amos B. Smith, III: Fragment Coupling B and C Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
postulated that a mixed t-butyl-alkyl zinc intermediate is reactive in the process
Me Me Me Me Me Me Me I t-BuLi (3 eq.), Et2O, -78 °C PMBO TBSO O O then Pd(PPh3)4, Et2O, rt TBSO O O Me Me PMP PMBO "intimate Me I mixture" OTBS PMP + ZnCl2 (premixed) (66%) Me Me OTBS
Me Me Me Me 1. DDQ (1.05 eq.) CH2Cl2/H2O - + I PPh3 2. PPh3, I2, Imid. PhH/Et2O
3. PPh (15 eq.), i-PrNEt (3 eq.) TBSO O O 3 Me Me 80 °C neat OTBS PMP
(87% for DDQ, 37% for next 2 steps) problems with competing cyclization, esp. with more polar solvent
Amos B. Smith, III: Fragment Coupling BC and A Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me Me Me Me Me 1. NaHMDS, THF TBSO - + Me I PPh3 H EtS O OTBSOPMB OH Me Me TBSO O O TBSO Me Me O OTBS Me Me OTBS PMP EtS O OTBS (2 steps, 67% Z:E >49:1)
Me Me OTBS
2. DIBAL, CH2Cl2
Me Me Me TBSO Me 1. SO •Py, Et N, DMSO, CH Cl 3 3 2 2 EtS O OTBSOPMB Me Me 2. Ph2PCH2CH=CH2 t-BuLi, Ti(Oi-Pr) , MeI, THF OTBS 4 Me Me Yamamoto Olefination OTBS (2 steps, 67% Z:E = 16:1) Amos B. Smith, III: Completion (First Generation)
Me Me Me 1. HgCl2, MeCN, THF; pH 7 buffer TBSO 2. DMSO, Ac2O Me 3. DDQ, CH Cl , H O EtS O OTBSOPMB 2 2 2 Me Me (66%) OTBS Me Me OTBS
Kokovsky Protocol Me Me Me 1. Cl3CCONCO, DCE TBSO K CO , MeOH Me 2 3 O O OTBSOH 2. 48% HF/MeCN (1:9) Me Me (53%) OTBS Me Me OTBS Me Me Me HO Me
O O OH O NH2 Me Me OH O Me Me
(–)-Discodermolide (note: opposite enatiomer pictured)
Amos B. Smith, III: Second Generation Approach
Wittig New Goals: Olefination Me Me Me • Prepare 1 g of Discodermolide • Use Similar Strategy HO Me Yamamoto Olefination Problems from G1 Synthesis: O O OH O NH2 Me Me • Long Synthesis of A OH O • Poor Yield on Phosphonium Salt Negishi Cross- Me Me Coupling OH
H Me Me Me TBSO PMBO O Me I I A O O B C Me Me TBSO O O OTBS Me Me PMP OTBS
Me Me PMBO N O TBSO O Denotes the repeating Org. Lett. 1999, 1, 1823-6. stereochemical triad. Amos B. Smith, III: Constructing the Common Precursor
Changed chiral auxilliary to eliminate the need for chromatography at this stage. NH
1. PMBO CCl3 Me Ph PPTS, CH Cl /Cyclohexane Me 2 2 Me Me 2. LiAlH , THF HO 4 PMBO N O CO Me 2 3. Swern [O] 4. OH O O Me Ph "Roche Ester" 55% yield, 4 steps N O highly crystalline adduct (confirmed with X-ray) O O
n-Bu2BOTf, Et3N, CH2Cl2
Me Me PMBO N HN(Me)OMe•HCl, AlMe3 O • Auxilliary was easily THF OH O recovered (80-90%) (98%)
Common Precursor
Can be executed on 60-70g scale
Amos B. Smith, III: Constructing Fragment A
Me Me Me Me 1. TBSOTf, 2,6-lut., CH2Cl2 O N PMBO N 2. H2, Pd(OH)2/C, EtOH O O 3. SO •Py, i-PrNEt , DMSO OH O 3 2 TBSO O (85-93%) crystalline
HO O O O 1. TiCl4, CH2Cl2 O O K-Selectride 2. OTMS PhMe/THF Me Me Me Me , -78 °C (dr 9:1; 50-79%) OTBS OTBS 3. TFA/Hexanes, rt crystalline ds > 20:1 (70-77%) note: allylmetal addition followed by asymmetric dihydroxylation failed to give higher than 3:1 selectivity. carried through directly Me4NB(OAc)3H also gave a 3:1 ratio as the lactone H TBSO O 1. TBSCl, Imid., CH Cl O O 2 2 • 7 Steps Eliminated from First-Generation Route!
2. O3, CH2Cl2; Ph3P A Me Me (80-82%) OTBS Amos B. Smith, III: Constructing Fragment B
Me Me 1. TBSOTf, 2,6-lut., Me Me CH Cl PMBO N 2 2 PMBO O O 2. DIBAL, THF HO O TBSO H
1. Ph3PEtI, n-BuLi, 0 °C PMBO 2. 0.1 M I in THF, -78 °C Me I 2 B 3. NaHMDS, -23 °C Me Me • Single Chromatography Required 4. Add substrate, -33 °C OTBS
(40-46% overall, ds 8-17:1)
Amos B. Smith, III: Constructing Fragment C
Bn Me Me H N O Me Me PMBO N 1. DDQ, CH2Cl2 O O O O O O 2. LiAlH4, THF HO O n-Bu2BOTf, Et3N, CH2Cl2
OMe
Bn Me Me Me Me Me Me Me Me Me N O I 1. TBSOTf, 2,6-lut., CH2Cl2 I O O OH O O TBSO O O 2. LiBH4, EtOH, THF O O OTBS 3. Ph3P, I2, Imid., PhH/Et2O PMP 60-71%, 3 steps PMP PMP (69-78%) C
No Major Changes Here... Amos B. Smith, III: Fragment Coupling B and C Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me Me Me Me Me 1. DDQ PMBO 2. TrCl • DMAP, Py I 1. ZnCl2 (1.15 eq.), Et2O TBSO O O 3. DIBAL, 0 °C TBSO O O 2. t-BuLi (3 eq.), Et2O, -78 °C Me Me 4. DMP [O], NaHCO3 OTBS PMP PMP then 5% Pd(PPh3)4, Et2O, rt 1.15 eq. PMBO (66%) (usually 1.5 eq. required) Me I Me Me OTBS Yamamoto Olefination done on BC fragment, vs. fully coupled core (Z:E = 4:1)
Me Me Me Me Me Me Me Me 1. Ph2P Ti(Oi-Pr)4Li HO TrO O H then MeI TBSO OPMB TBSO OPMB Me Me Me Me 2. Cl-catecholborane/MeOH (3:1) OTBS OTBS (Z/E: 8-12:1, 85-98%)
Me Me Me Amos B. Smith, III: Fragment Coupling BC and A HO Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me Me Me Me Me Me - + HO 1. PPh3, I2, PhH/Et2O I Ph3P 2. PPh , i-Pr NEt, PhH/Tol TBSO OPMB 3 2 TBSO OPMB Me Me Me Me Dauben's High-Pressure OTBS Method: 185,000 psi OTBS 10-14 days, 2 g max! (82%) Me Me Me TBSO 1. NaHMDS, THF, -20 °C Me reported to be unstable -- H O O OTBSOH 1. Cl CONCO, Al O Me Me 3 2 3 just residue from CDCl3 TBSO (+)–Discodermolide O OTBS 2. 3 N HCl/MeOH, rt Me Me O O (88%) OTBS O Me Me (note: DDQ cycloaddtion takes Cl CN care of other olefin isomer!) OTBS DDQ Cl CN 2. DDQ, CH2Cl2, H2O O (60%) Ian Paterson: Retrosynthetic Analysis
Lithium Aldol Coupling
Me Me Me Boron Aldol Coupling OH Me
O O OH O NH2 Me Me OH O Me Me OH
ArO O Me Me PMBO Me Me MeO Me H O OTBSO Me Me A B C O OPMB Nozaki- OTBS Hiyama-Kishi
BnO PMBO OBz O O O
Angew. Chem. Int. Ed.. 2000, 39, 377-80.
Ian Paterson: Fragment A Construction
O
Me H Me HO 1. BnOC(=NH)CCl3, cat. TfOH, Et2O a. c-Hex2BCl, Et3N, Et2O, 0 °C CO2Me BnO 2. i-PrMgCl, HN(Me)OMe • HCl, THF b. LiBH4 "Roche Ester" 3. EtMgBr, THF, 0 °C O c. H2O2 (30%), MeOH, NaOH, 0 °C (71%) (86%, >97% ds) axial H- delivery BnO Me Me Me H L H L BnO Me Me H B B Me O L BnO Me O Me O Me L O O O R B H OH OH H L L
1. (COCl)2, DMSO, CH2Cl2 Me Me A 1. TBSOTf, 2,6-lut., CH Cl 2. NaClO , NaH PO , 2 2 Me Me 2 2 4 2. CSA, MeOH/CH2Cl2, 0 °C 2-methyl-2-butene MeO HO t-BuOH, H2O 3. Pd(OH)2/C, H2, EtOH O OTBS O 3. CH2N2, Et2O (67%) OTBS OH (93%) Ian Paterson: Fragment B Construction O
H Me Me HO 1. PMBOC(=NH)CCl3, cat. TfOH, Et2O a. c-Hex2BCl, Et3N, Et2O, -78 ! -20 °C CO2Me PMBO 2. i-PrMgCl, HN(Me)OMe • HCl, THF b. H2O2 (30%), MeOH, 0 °C "Roche Ester" 3. EtMgBr, THF, 0 °C O (94%, >97% ds) (77%)
Me Me Me Me Me4NBH(OAc)3 PhSeCH CH(OEt) PMBO PMBO 2 2 MeCN, AcOH PhMe, PPTS (cat.), reflux O OH OH OH (94%, >97% ds) (94%)
Me Me Me Me PMBO a. NaIO4, NaHCO3, MeOH/H2O Me Me PMBO O O O b. DBU, H2C=C(OMe)OTBS O O xylenes, reflux PMBO Me (82%) O PhSe
ArO O ArO O 1. KOH, MeOH, reflux PMBO TBSOTf PMBO Me Me 2. 2,6 dimethylphenol 2,6-lut., CH2Cl2 DCC, DMAP, CH Cl 2 2 Me Me B Me Me (97%) (83%) OH OTBS
Ian Paterson: Fragment C Construction
TBSO H (1.5 - 2 eq.)
Me O EtO 3 steps a. c-Hex2BCl, Me2NEt, Et2O, -78 ! -20 °C OH OBz O (65%) b. H2O2 (30%), MeOH, 0 °C O (99%, >97% ds)
Me Me Me Me PMBOC(=NH)CCl , TfOH (cat.) 3 a. NaBH , MeOH, H O TBSO TBSO 4 2 OBz OBz Et2O b. K2CO3, MeOH OH O PMBO O (93%) (86%) TMS a. Me Me Me Me NaIO4 Br TBSO TBSO O OH MeOH, H2O CrCl2, THF PMBO OH PMBO H b. KH, THF, 0 °C (98%)
H H Me Me R O CrLn 1. CSA (cat.) Me Me TMS MeOH, CH Cl TBSO 2 2 H H 2. Dess-Martin [O] OPMB C O OPMB (84%) Ian Paterson: Coupling B and C Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
ArO O ArO O Me Me PMBO a. LiTMP, LiBr, THF Me PMBO LiAlH4 Me b. Me Me Me Me OH OPMB THF Me Me H OTBS (62%, >97% ds) (2 equiv.) OTBS one pot aldol/red'n O OPMB
HO Me Me Me Me 1. Et N, CH Cl PMBO 3 2 2 PMBO Me Me OH OPMB OTBSOPMB Me Me SO2Cl Me Me OTBS OTBS
2. LiAlH4, THF 3. TBSOTf, Et3N, CH2Cl2 (91%)
Ian Paterson: Elaboration of the BC fragment Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me Me HO PMBO DDQ Me Me OTBSOH OTBSOPMB CH2Cl2, pH 7 buffer Me Me Me Me (91%) OTBS OTBS
Me Me 1. TEMPO, PhI(OAc)2, CH2Cl2 a. Cl3CC(O)NCO, CH2Cl2
MeO2C Me 2. 18-Cr-6, K2CO3 b. K2CO3, MeOH OTBSOH F CCH O (98%) 3 2 OMe Me Me P OTBS F3CCH2O O O (87%)
Me Me Me Me 1. DIBAL, CH2Cl2 O MeO2C Me Me 2. Dess-Martin [O] OTBSO NH2 Me Me H OTBSO NH2 (88%) Me Me OTBS O OTBS O Me Me Me Ian Paterson: Completion of the Synthesis HO Me O O OH O NH Me Me 2 OH O Me Me OH
1. (+)-Ipc2BCl, Et3N, Et2O Me Me Me Me MeO O Me O OTBS O
H OTBSO NH2 Me Me 2. H2O2 (30% aq.), MeOH, 0 °C This step generates an 85:15 mixture of OTBS O (77% ds, 52% desired) open chain:lactone form which is carried on.
Me Me
HO 1. Me4NBH(OAc)3, MeCN, AcOH Me Me Me 2. HF • Py, THF MeO OTBSO NH2 Me Me (84%) O OTBSO OTBS O
Me Me Me HO Me
O O OH O NH2 Me Me OH O Me Me OH
Ian Paterson: Second Generation Retrosynthesis
Lithium Aldol Coupling Second Generation Approach:
• Reduce Number of Steps Me Me Me Boron Aldol • Eliminate Chiral Reagents and Coupling OH Auxillaries Me
O O OH O NH2 Me Me OH O Me Me OH
ArO O Me Me PMBO Me Me MeO H Me H O OTBSO Me Me A B C O OPMB Nozaki- OTBS Hiyama-Kishi
Me Me PMBO OH
OH
Org. Lett. 2003, 5, 35-8. Ian Paterson: Common Precursor
O
Me H H Me Me Me 3 steps a. c-Hex BCl, Et N, Et O HO 2 3 2 PMBO OH CO2Me PMBO (84%) "Roche Ester" b. H2O2 (30%) O O 95:5 dr
NaBH(OAc)3 Me Me PMBO OH THF, AcOH
(2 steps, 62%) OH
Ian Paterson: Fragment A Construction
1. NaClO Me Me Me Me 2 TEMPO 2. MeI, K2CO3 PMBO OH PMBO O PhI(OAc)2 3. TBSOTf, 2,6-lut. CH Cl OH 2 2 OH H (85%, 4 steps)
Me Me 1. DDQ, CH2Cl2 Me Me A pH 7 buffer MeO OPMB MeO H 2. Swern [O] O OTBS O OTBS O (2 steps, 90%) Ian Paterson: Fragment B Construction
1. 18-Cr-6, KHMDS
F3CCH2O Me Me OMe Me Me P TEMPO O OMe PMBO F3CCH2O PMBO O O O Me Me PhI(OAc) 2 PMBO OH OH CH2Cl2 OH H Me 2. TBSOTf, 2,6-lut. OH (3 steps, 72%)
I 1. DIBAL Me Me ArO O LiHMDS, MeCO2Ar 2. I2, Imidazole, PPh3 PMBO Me THF PMBO Me (90%) OH H (72%) B Me Me OTBS
Ar =
MeO
Ian Paterson: Fragment C Construction
Me Me Me Me Me Me 1. TBSCl, Imidazole OTBS DIBAL OTBS PMBO OH 2. DDQ, mol. sieves O O THF OH OPMB OH (82%) (92%) PMP
1. CSA (cat.) Me Me 1. Dess-Martin Me Me MeOH, CH Cl 2 2 H TBSO TMS 2. Dess-Martin [O] 2. C O OPMB OPMB Br (84%)
CrCl2, THF; then, KH
(84%) Ian Paterson: Coupling B and C Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
ArO O ArO O Me Me PMBO a. LiTMP, LiBr, THF Me PMBO LiAlH4 Me b. Me Me Me Me OH OPMB THF Me Me H OTBS (62%, >97% ds) (2 equiv.) OTBS one pot aldol/red'n O OPMB
HO Me Me Me Me 1. Et N, CH Cl PMBO 3 2 2 HO Me Me OH OPMB OTBSOH Me Me SO2Cl Me Me OTBS OTBS
2. LiAlH4, THF 3. TBSOTf, Et3N, CH2Cl2 4. DDQ (83%)
Ian Paterson: Endgame (2nd Generation) Me Me Me HO Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me Me 1. TEMPO, PhI(OAc)2, CH2Cl2 HO Me 2. 18-Cr-6, K2CO3 Me OTBSOH F CCH O O OTBSOH Me Me 3 2 Me Me P OTBS F3CCH2O OTBS O O
(12:1 Z:E, 90%)
Me Me Me Me MeO H R L Me Me O OTBS O L Cl3CC(O)NCO O OTBSO NH2 H H Me Me c-Hex2BCl, Et3N B neutral Al O O L 2 3 O OTBS O Et2O (97%) R (> 95:5 dr, 64%) 1,6 stereoinduction Me Me O Me Me Me 1. AcOH, aq. THF (+)-Discodermolide MeO OTBSO NH2 Me Me 2. K-Selectride, PhMe, THF 3. HCl, MeOH • Boron-Aldol works on gram scale O OTBSOH OTBS O • Li-enolate produces opposite config. (97:3 dr, 66%) (Felkin-Anh induction from aldehyde) Novartis Process Group: Scale-Up Route (Phase I Clinical Trial Campaign)
Org. Proc. Res. Dev. 2004 Suzuki Coupling Me Me Me HO Boron Aldol Me Nozaki-Hiyama- Kishi O O OH O NH2 Me Me OH O Me Me OH
Me Me Me O OTBSO PMBO Me I I O N Me Me B TBSO O O Me Me A C OTBS PMP
Me Me PMBO N O Hybrid Novartis-Smith-Paterson Route TBSO O
Novartis Process Group: Scale-Up Route
NH -replaced LiAlH4 with LiBH4 (slow filtration) 1. PMBO CCl3 -replaced Swern with PPTS, CH2Cl2/Cyclohexane (100%) Me Me TEMPO/Bleach (DMS 2. LiBH , THF/EtOH (100%) HO 4 PMBO H not environmentally CO2Me friendly) 3. TEMPO/Bleach, CH2Cl2 (100%) O "Roche Ester" -not stable for extended storage - subjected to syn-aldol without further purification Bn
Bn N O Me Me Me Me 1. LiOH, H2O2 (71%) O O PMBO N O PMBO N 2. CDMT, N-Me-Morpholine, O n-Bu2BOTf, Et3N, OH O O HN(Me)OMe (85%) CH2Cl2 OH O 20-50 g, 75% able to crystallize (80%)! 20-25 kg, 46-55% AlMe3 could not be used (process safety) so a mixed-anhydride Quality of Bu2BOTf is really important. approach was employed. Cl
CDMT = N N
MeO N OMe Novartis: Constructing Fragment A
Me Me Me Me 1. TBSOTf, 2,6-lut., toluene O N PMBO N 2. H2, Pd/C, t-BuOH O O 3. TEMPO, PhI(OAc) OH O 2 TBSO O (85-93%)
Me Me 1. MeMgBr O N O 2. SO , Py, DMSO 3 TBSO O (66% for 5-step sequence, 'routine several kg scale')
Novartis: Constructing Fragment B
Me Me 1. TBSOTf, 2,6-lut., Me Me Switched to toluene in TBS Toluene, 0 °C (90%) protection PMBO N PMBO O O 2. Red-Al, Toluene Switched to Red-Al for Hydride HO O -20 °C (68%) TBSO H addition (-20 °C vs. -78 °C)
Ph3P I PMBO Me I B NaHMDS, THF Brutal scale up, 2.5 kg max. Me Me Complex workup, unstable product (30%) OTBS Z:E = 15:1 Novartis: Constructing Fragment C
anhydrous cond. Bn are critical Me Me H N O Me Me PMBO N 1. DDQ, 4ÅMS, Toluene (61%) O O O O O O
HO O 2. LiAlH4, THF (91%) n-Bu2BOTf, Et3N, CH2Cl2 (85%)
OMe
Bn Me Me Me Me Me Me Me Me Me N O I 1. TBSOTf, 2,6-lut., CH2Cl2 I O O OH O O TBSO O O 2. LiBH4, EtOH, THF O O OTBS 3. Ph3P, I2, Imid., Toluene PMP PMP (50%) PMP C
light sensitive
Novartis: Key Suzuki Coupling
Me Me Me Me Me Me I 1. t-BuLi, 9-MeOBBN, THF, -78 °C PMBO Me TBSO O O 2. Cs2CO3, DMF, Pd(dppf)2Cl2, 20 °C OTBSO O Me Me PMP PMBO Me I OTBS PMP
Me Me Marshall's Suzuki approach was superior, as OTBS the Negishi coupling (Smith) resulted in the formation of some inseparable side products. (73%; some des-iodo of SM obs. also)
Me Me Me 1. CrCl2, PMBO O Br Me Me Me 1. DIBAL (92%) Me PMBO TMS Me OTBSOPMB H 2. SO / Py 3 Me Me OTBSOPMB DMSO (93%) OTBS 2. KOH Me Me OTBS (81%) slightly basic workup critical Paterson's Protocol:
Nozaki-Hiyama-Kishi (2 diastereomers), then Peterson syn-elimination Novartis: Further Elaboration
Me Me Me Me Me Me Me PMBO 1. DDQ (88%) O H Me OTBSOH OTBSOPMB 2. PhI(OAc)2, TEMPO, H2O (91%) Me Me Me Me OTBS OTBS
0.1 eq. H2O necessary for reaction to proceed
F CCH O 3 2 OMe P Me Me Me F3CCH2O O O MeO2C Me CCl3CONCO 18-Cr-6, KHMDS OTBSOH then Na2CO3, MeOH (76%, 2 steps) Me Me OTBS (100%) Still-Gennari variation of HWE
Me Me Me Me Me Me 1. DIBAL O MeO2C Me Me 2. PhI(OAc)2, TEMPO OTBSO NH2 H OTBSO NH2 Me Me Me Me (80%) OTBS O OTBS O
Me Me Me HO Novartis: Critical Endgame Maneuvers Me O O OH O NH Me Me 2 OH O Me Me OH
(6.6 equiv!) Me Me N O O Me Me Me O TBSO DISASTER!!! O Me BCl H OTBSO NH2 , Et N Me Me 3 OTBS O 2 (+)-(DIP-Cl)
• Quality of commercial DIP-Cl was "capricious". Solved by using 70% DIP-Cl in hexane
• 4:1 epimer mixture at 250 mg scale (45%), but 23% on 50g! The problem was identified as the direct reduction of the aldehyde.
• They determined the cause of reduction was incomplete enolization. Changing to 24 hrs worked better on small scale (50% yield), but again disaster on larger scale.
• They determined that the workup (peroxide, silica gel) were lowering the yield. A quick aqueous workup was performed instead, and purification by reverse-phase silica gel led to reproducible 50-55% yields.
• Epimer can be recycled and used to make the final product. Me Me Me HO Novartis: The Grand Finale! Me O O OH O NH Me Me 2 OH O Me Me OH
Me Me Me HO Me Me4N(OAc)3BH
O OTBSO NH2 Me Me Me (75%) N OTBS O O Me O OTBS
Panic broke out on the final step when they realized that the > 99% HPLC pure final product in solution registered an impurity around 8% after Me Me Me crystallization! HO Me HCl / MeOH Discodermolide, 60 g! HO OTBSO NH2 Me Me Me (70%) Totals: N OTBS O O Me 43 chemists O OTBS 17 chromatographic purifications 20 months to complete (~1 step / 2 weeks)
The Afterglow of the Novartis Synthetic Campaign
Ian Paterson Amos Smith, III Stuart Schreiber "Spectacular...It's probably the best piece of synthetic work from an industrial company" -Steven Ley (Cambridge, UK)
"Some 3,000 kg of the sponge, a quantity that probably does not exist, would have been needed to deliver 60 g" -Stuart Mickel
"Clearly, the Novartis synthesis is a wonderful accomplishment, demonstrating that if a new drug Stuart Mickel candidate is sufficiently valuable, Sarath Gunasekera (Isolation) synthetic chemists will rise to the challenge of developing a viable synthetic approach no matter how complex the structure" -Amos Smith, III
"On a positive note, this project was a first for Novartis, and its progress was avidly followed by the entire department who were all interested in the disco". -Novartis Summary of Routes Used
Suzuki Coupling Me Me Me Boron Aldol Coupling HO Me Nozaki-Hiyama- Novartis: Kishi O O OH O NH2 Me Me OH O Me Me OH Lithium Aldol Coupling
Me Me Me Boron Aldol Coupling OH Paterson: Me Nozaki-Hiyama- Kishi O O OH O NH2 Me Me OH O Me Me OH Wittig Olefination Me Me Me HO Me Yamamoto Smith: Olefination O O OH O NH2 Me Me OH Negishi Cross- O Me Me Coupling OH
Me Me Me HO What Makes A Great Total Synthesis? Me O O OH O NH Me Me 2 OH O Me Me OH
Amos B. Smith, III: Smith G1: Smith G2:
-Clever Use of 'Common Precursor' A: 14%, 19 steps A: 25%, 13 steps -Prepared 1 g as proof-of-concept B: 19%, 8 steps B: 25%, 8 steps C: 22%, 11 steps C: 30%, 11 steps
Common Steps: 11 Common Steps: 11 Total for Fragments: 27 Total for Fragments: 21
Finish: 3%, 13 steps Finish: 20%, 13 steps
Ian Paterson: Paterson G1: Paterson G2: A: 46% 10 steps A: 40%, 11 steps -Common Precursor Approach (G2) B: 43% 10 steps B: 24%, 11 steps -Effective use of boron aldol reaction to C: 43% 12 steps C: 28%, 13 steps assemble molecule -Substrate-controlled creation of Common Steps: 0 Common Steps: 11 stereocenters (no auxilliaries) Total for Fragments: 32 Total for Fragments: 24 Finish: 21%, 13 steps Finish: 19%, 12 steps
Novartis: Novartis: A: 22%, 11 steps -Preparation of 60 grams! B: 6%, 9 steps C: 8%, 12 steps
Common Steps: 12 Total for Fragments: 20
Finish: 7%, 15 steps