Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Cl HO Me NMe2 OH NMe2 The tetracycline family of antibiotics H H H H OH OH The natural products H N NH2 Cl HO Me NMe2 HO Me OH NMe2 H H H H OH OH OH OH OH O HO O O OH OH O HO O O clomocycline methacycline (rondomycin) NH2 NH2 (megaclor) 1963 1965 OH OH OH O HO O O OH O HO O O Me OH NMe2 NMe2 NMe2 chlortetracycline oxytetracycline H H H H OH OH (aureomycin) (terramycin) 1948 1949 NH2 NH2 HO Me NMe2 Cl HO NMe2 OH OH H H H H OH O HO O O OH O HO O O OH OH doxycycline minocycline NH NH (vibramycin) (minocin) 2 2 1967 1972 OH OH OH O HO O O OH O HO O O NMe2 NMe2 H H tetracycline demethylchlortetracycline OH (achromycin) (declomycin) O H 1953 1957 Me N NH2 N H OH Me Me OH O HO O O Semisynthetic derivatives on the market t-butylglycylamidominocycline (tigilcycline) 1993 HO Me OH NMe2 HO Me OH NMe2 H H H H (Phase III clinical trials in progress) OH OH

H H Cl HO Me OH NMe2 N N 5 7 6 5a 4a 4 3 OH OH OH 8 OH O HO O O N OH O HO O O HN D C B A 2 9 NH2 4 12a rolitetracycline limecycline 10 11 12 1 (reverin) (tetralysal) H2N CO2H OH O HO O O 1958 1961 Notation

I. Chopra, M. Roberts. Microbiol. Mol. Biol. Rev. 2001, 65, 232. 1 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Discovery and The Dawn of Semisynthetic Antibiotics HO Me OH NMe2 H H OH

NH2 OH OH O HO O O Cl HO Me NMe2 H H OH terramycin

NH2 OH OH O HO O O aureomycin

Bayer Pharmaceuticals Benjamin Duggar University of Missouri

The first tetracycline antibiotic discovered, aureomycin was isolated in 1948 from a Missouri The Nobel Prize Committee soil sample by Lederle R. B. Woodward Laboratories. The Lederle team was led by Benjamin Duggar - a About the same time as the Lederle discovery of aureomycin, Pfizer was consultant who was a 73-year-old scouring the globe for new antibiotics. Soil samples were collected from emeritus professor of botany who jungles, deserts, mountaintops, and oceans. But ultimately terramycin was had recently retired from the isolated in 1949... from a soil sample collected on the grounds of a factory in University of Missouri! As Jie Terre Haute, Indiana, owned by Pfizer! Jack Li cracks, "Your greatest discovery could happen in your retirement." From the beginning, terramycin was a molecule enveloped in controversy. It was the subject of the first mass-marketing campaign by a modern About Lederle Labs: pharmaceutical company. Pfizer advertised the drug heavily in medical Lederle Labs was founded in journals, eventually spending twice as much on marketing as it did to discover 1902 in an old farmhouse on the and develop terramycin. Still, it turned Pfizer - then a small company - into a Pearl River in New York. pharmaceutical giant. Aureomycin was one of many lifesaving products developed by Pfizer and R.B. Woodward collaborated to determine the structure of Lederle, including vaccines for terramycin, succeeding for the most part in 1952 (JACS 1953, 75, 5455). The polio and smallpox. It is now a stereochemistry at C was revised after X-ray crystallography and NMR part of Wyeth Pharmaceuticals. 4a studies in the 1960's (JACS 1965, 87, 134; JACS 1963, 85, 851). Ebay Ad for aureomycin as additive in cattle feed

J. J. Li. "A History of Drugs and Their Discoverers." Pfizer Intranet Magazine. March-April 2004. 2 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Big Pharma Behaving Badly: In 1955, Conover Now, Back to Actual Science discovered that hydrogenolysis of aureomycin gives a deschloro product that is just as active as the original Biosynthesis and Biological Activity: Tetracyclines are polyketide antibiotics, product. This proved for the first time that chemically- biosynthesized in a fashion similar to that of fatty acids, erythromycin and a host of modified antibiotics could have biological activity. Within a other antibiotics. Tetracyclines are produced naturally by Streptomyces few years, a number of semisynthetic tetracyclines had aureofaciens (T. Nakano, et al. Biosci. Biotechnol. Biochem. 2004, 68, 1345.). entered the market, and now most antibiotic discoveries are of novel active derivatives of older compounds. Tetracyclines bind to the bacterial ribosome, preventing the binding of aminoacyl- tRNA to the ribosomal A site. This prevents bacterial protein translation (I. Chopra, Conover's discovery, however, provoked further controversy M. Roberts. Microbiol. Mol Biol. Rev. 2001, 65, 232.). for tetracycline. Pfizer became embroiled in a patent dispute with American Cyanamid, which owned the rights to The Challenge to Synthetic Chemists: Muxfeldt and colleagues outline the basic aureomycin (the starting material for Conover's procedure obstacles to achieving a total synthesis of any of the natural tetracyclines: to make tetracycline). Pfizer and American Cyanamid eventually settled the dispute out of court when they Stereochemical Complexity. There are up to five contiguous asymmetric centers realized that neither company held truly exclusive rights to (terramycin) which must be established. the drug, and agreed to cooperate on selling the drug in order to drive off competitors trying to enter the tetracycline Chemical Sensitivity. For the 6-methyl-6-hydroxy tetracyclines, mild acid rapidly market. At one point, Pfizer employed a private decective catalyzes dehydration, ketalization and a retro-aldol to produce the lactone below. to tap the phones of Bristol-Meyers, a competitor seeking to Mildly basic conditions results in deprotonation of the C5 and C6 hydroxyls, enter the tetracycline market! Bristol-Meyers agreed to About.com initiating a cascade of events which leads to decomposition of the molecule. overlook this brazen act in exchange for a share of the Finally, the C4 stereocenter is readily epimerized upon exposure to acetic acid or Lloyd Conover aqueous buffers. tetracycline market. Eventually five companies colluded in Pfizer order to maintain artificially high prices for tetracycline. However, the Federal Trade Commission stepped in after several years, finding Pfizer and company guilty of patent HO Me OH NMe2 Me NMe2 fraud and anti-trust violations, and broke up the monopoly. H H H OH OH Legal issues aside, for this discovery Lloyd Conover is now in the American acid Inventors' Hall of Fame, alongside Thomas Edison and the Wright brothers. O NH2 NH2 U.S. Federal Trade Commission, "Anticipating the 21st Century: Competition OH OH Policy in the New High-Tech, Global Marketplace". OH O HO O O OH OH O O O base M. Mintz. "Golden Ox of Antitrust." The Nation 14 April 1969, Vol. 208, Issue 15. acid pp. 467-468. O

Cl HO Me NMe2 H HO Me NMe2 HO Me OH NMe2 H H H H H H OH OH OH COOH H2, Pd/C HO Me NH2 MeOH/ NH2 NH2 COOH dioxane OH OH OH OH O HO O O OH O HO O O OH O HO O O Conover, L.H. 1955. U.S. Patent No. 2,699,054. H. Muxfeldt, et al. J. Am. Chem. Soc. 1979, 101, 689.

3 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Woodward's First Total Synthesis of a Biologically-Active Tetracycline, 6- O O Demethyl-6-deoxytetracycline. Cl Cl MeO OMe CO Me L.H. Conover, et al. J. Am. Chem. Soc. 1962, 84, 3222. (Initial communication) CO Me (2 eq.) 2 R.B. Woodward. Pure Appl. Chem. 1963, 6, 561. (A personal account) 2 J.J. Korst, et al. J. Am. Chem. Soc. 1968, 90, 439. (Full article) then NaH (4 eq.), OH (i) O then MeOH (1 eq.), rt --> 80 oC OMe O OH OMe O OMe O O O 45% NaH CO2Me intermolecular condensation outcompetes intramolecular! DMF OMe OMe (ii) O OMe Triton-B MeOH is essential to suppressing the kinetically Cl dioxane-MeOH Cl OMe o favorable intramolecular condensation and OMe OMe O 50-70 C permitting the intermolecular condensation with NaH CO2Me DMF 88% the oxalate prior to formation of the desired 55% tricycle. In the absence of MeOH, Woodward observed formation of the intramolecular product OH O O first, followed by condensation onto the oxalate OMe OH O CO2Me (i) AcOH/ to form the five-membered ring shown: CO2Me CO2Me H2SO4 Cl O (ii) H2SO4 Cl CO Me CO Me 2 MeOH/CHCl3 2 CO2Me n OMe 44% OMe AcOH/HCl H CO2 Bu

(i) H2, 200 psi (i) NaOH H2O Mg(OMe)2 OH o O Pd/C CO2Me H2O 90 C toluene o AcOH, 30 oC 100 C OMe O OH 73% OMe O OH reflux 52% (ii) H SO (ii) I , AcOH; 2 4 CO2Me 2 MeOH/CHCl Cl CO nBu 3 then Cl2 in AcOH 2 Cl NMe2 93% OMe (iii) HF, neat o H H (i) NHMe2, -10 C; 63% over 3 steps o n (ii) NaBH4, -70 C CO2 Bu Cl Cl H SO O 69% CO2H 2 4 CO Me OH MeOH/CHCl 2 3 OMe O OH OMe O OH reflux 66% The thermodynamically more favorable diastereomer is formed OMe O OMe O exclusively in this step, with the carboxyamino substituent assuming an equatorial position and thus establishing the cis relationship of Chlorination blocks the para the bridgehead hydrogens. Ketone reduction is also stereoselective. position, forcing condensation onto the more hindered ortho position.

4 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

NMe2 Cl NMe2 Cl H H H H H H NMe2 NMe2 H H O n O 48% HBr CO2 Bu TsOH O 20 min CONH OH toluene O 2 CONHtBu reflux OH OH OH OH OMe O OH 90% OMe O OH OMe OH OH OH 15% bsm from A; 30% of A recovered Zn dust Cl NMe2 H , Pd/C formic acid 2 NMe2 H H Et N H H 1 min 3 O2 91% OH 81% CO2H CeCl3 DMF-MeOH glycine-NaOH buffer, CONH2 OH OMe O OH pH = 10 OH O OH O 15 min Mixture of epimers at C4 (i) O NMe2 NMe2 H H H H OH This was a difficult step to optimize - Woodward himself noted dryly that "the case CO2H O Cl at hand was by no means the smoothest we had encountered." Competitive hydroxylation at C was also observed, as well as rapid decomposition of the (ii) O O O 11a product under prolonged reaction conditions, forcing Woodward and colleagues EtO2C to halt the reaction prematurely. OMe O OH EtO NHtBu OMe O OH NH tBu Mg(OEt) A No acylation of the enols by the CaCl NMe chloroformate was observed. 2 2 6 H H OH NMe2 H H BuOH-H2O, pH = 8.5 NaH O ethanolamine buffer reflux, 10 min DMF/MeOH CONH2 O 6% over 2 steps, OH o OH O OH O 120 C EtO 10% recovered SM 15 min t OMe O O O N Bu Thermodynamic equilibration 6-desmethyl-6-deoxytetracycline to the desired epimer. NMe2 Observe the classic Woodward master H H O stroke. Despite the presence of four This was the first total synthesis of a tetracycline with all the requisite enolates, we observe only one of two functionality for full antibiotic activity. Note, however, that this is not the total plausible intramolecular condensation synthesis of a tetracycline natural product. Substituents at the C6 position are CONHtBu events. The other event is impossible since missing. the enolate double bond cannot rotate to OMe OH OH OH bring the amide into position for cyclization.

5 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Shemyakin: The First Total Synthesis of a Tetracycline Natural Product HO Me O + HO Me NO2 A.I. Gurevich, et al. Tet. Lett. 1967, 8, 132. H Et3NH H H M.N. Kolosov, S.A. Popravko, M.M. Shemyakin. Lieb. Ann. 1963, 668, 86. O2N OEt CO Et B.-M.G. Gaveby, J.C. Huffmann, P. Magus. J. Org. Chem. 1982, 47, 3779. 2 THF Note that the Lieb. Ann. reference cites a number of obscure Russian journals. The JOC reference, however, illustrates Shemyakin's approach to the tricyclic OBn O OH OBn O OH precursor produced below.

O O Me NO2 Me NH2 H H Zn H . t HCl CO Et dust CO Et BF3 OEt2 LiAl(O Bu)3H 2 2 + EtOH AcOH 86% 64% H OBn OH O OBn OH O OH O OAc OH O OAc

O O (i) O (i) 0.1 N KOH, THF-H O H H MeMgBr 2 BnBr 6 eq O (ii) O N O K2CO3 74% OEt Me NPhth N H H 54% H O OEt OH OH OAc BnO OH OAc CO Et (ii) MeI, Ag2O 2 O

OH OH Me H Me H OBn OMe O KOH/MeOH

85% PCl in DMF, then H H 5 BnO OH OAc BnO OH OH O O Me NPhth Me NPhth H H EtO NH2 OH HO Me CO2H H Mg(OEt) CONH Jones reagent EtO2C 2 60% OBn OMe O OBn OMe O

OBn O OH Notice Shemyakin adopting the Woodward approach to ring A.

6 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

HO2C Me NPhth O H Na+ OH S Me HN H OH DMSO CONH EtO2C 2 OBn OMe O CONH2 OBn OMe O OH

Me NMe2 H OH (i) HBr-AcOH O2 over Pt

(ii) MeI in THF Et N CONH 3 2 THF OH OH O OH rt, 8 hr This intercepts a degradation product which had A.I. Gurevich, M.G. Karapetyan, previously been elaborated into tetracycline. M.N. Kolosov. Khim. Prirodn. Soedin., Akad. Nauk Uz.SSR 1966, 141.

Me NMe2 (i) O , hν H 2 OH 3,4-benzopyrene (cat.) benzene

(ii) H2, Pd/C CONH2 OH OH OH O O Mechanism? Answer on Slide 16. M. Schach von Wittenau. J. Org. Chem. 1964, 29, 2746.

HO Me NMe2 H H OH

CONH2 OH OH O OH O tetracycline

7 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Ph Ph Muxfeldt's Total Synthesis of 6-Desmethyl-6-deoxytetracycline Cl N O Cl N H. Muxfeldt, W. Rogalski. J.Am. Chem. Soc. 1965, 87, 933. (Communication) O H. Muxfeldt, E. Jacobs, K. Uhlig. Chem. Ber. 1962, 95, 2901. (Prep of precursors) HCl O THF O Cl Cl (i) NaOH O O (ii) pyrolysis, MeO CO Me CO2Me MeO O Br 2 NaOMe o CO Me 160 C 2 Ph MeOH O O CO Me 85% MeO2C CO2Me 2 MeO2C Cl N over three steps NHtBu OMe OMe O taut. NaH (2 eq.) Cl Cl (i) H SO O 2 4 THF-Et2O t CO H CO H MeOH o MeO C CONH Bu 2 H PO 2 35 C, 24 hr 2 3 4 95% MeO O O 80 oC (ii) ethylene glycol CO2H TsOH, benzene quant. Ph Ph OMe OMe O 91% Cl N Cl N O O Cl Cl (i) MsCl O pyridine O CO2Me LiAlH4 OH 97% CONHtBu CONHtBu MeO2C MeO2C (ii) NaCN benzene-Et2O MeO O O MeO O O o NaI (cat.) O O 0 C O O MeO MeO DMF-H2O 94% 92% O O

Cl N Ph Cl HN Ph O O Cl Cl taut. O

CN CHO C6H5 N CO2H Li(EtO) AlH t t 3 H MeO C CONH Bu CONH Bu 2 MeO2C benzene-Et2O Pb(OAc)4 (cat.) MeO O O MeO O O o O O 0 C O O Ac2O MeO MeO 64% Now the stage is set for the second cyclization in this magnificent transformation. Only one equivalent of NaH used so far!

8 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

O

Cl HN Ph Cl NHBz Muxfeldt's Last Hurrah: Total Synthesis of Terramycin O O H. Muxfeldt, et al. J. Am. Chem. Soc. 1968, 90, 6534. NaH H. Muxfeldt, et al. J. Am. Chem. Soc. 1979, 101, 689.

CONHtBu CONHtBu MeO2C MeO Terramycin is a much more difficult target than the prototypical tetracyclines discussed previously - Woodward and Muxfeldt avoided many of the problems MeO O O MeO O O O outlined earlier with by targeting a structure without the troublesome C5 and C6 substituents, while Shemyakin targeted a tetracycline which did not have the Cl NHBz Cl NHBz C6 hydroxyl. Here Muxfeldt and colleagues (including a young Edwin Vedejs!) O OH tackle those problems head-on! Sadly, this is reported in a posthumous communication from the Muxfeldt laboratories.

CONHtBu CONHtBu

MeO O O O MeO O OH O O (i) Ac2O, H2SO4 O OAc (i) MeMgI, H o 82% from the starting aldehyde 0 oC -65 C isolated as mixture of epimers 83% 82% at C 4 (ii) 1-acetoxy- (ii) NaOH Muxfeldt thus effects the construction of the A and B rings in a single step! The only butadiene 84% problem, unfortunately, is the failure to control C stereochemistry. benzene, reflux H 4 OH O AcO O Cl NH2 60% Me Me OH (i) Me3OBF4 H2, Pd/C, H2CO OH OH O O KClO3 Me H (i) acetone, Me H (ii) HBr/AcOH, Et N, MeOH CuSO4 OsO4 (cat.) CONH2 3 100 oC 84% 50 oC OH O OH O (ii) Ac2O, 89% (i) deprotects the benzoyl amide; (ii) deprotects the remaining functional groups. NaOAc H H HO O 95% AcO O NMe2 NMe2 Me Me Me Me OH OH O O O O Me H Pb(OAc)4 Me H CONH CONH2 OH o O 2 OH 40 C OH O OH O OH O OH O 6-Demethyl-6-deoxytetracycline OH O H H AcO O AcO O Here they intercept an intermediate from the Woodward synthesis. They also report Mixture of cis-diols hydroxylation with O2 over platinum (Angew. Chem. Intl. Ed. Eng. 1962, 1, 157).

9 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Me Me Me Me Preparation of C: DBU-AcOH, O O O O O O O NH3 O NH2 O Me H piperidine Me H (i) O O 3 (cat.) o -50 C MeO OMe MeOH MeO NH 33% 2 xylenes (ii) H O reflux 2 O CHO 68% H 52% over H conc. HCl O O O AcO O two steps AcO O C CHCl3 60% MeO NH2 (i) H Me Me Me Me N Coupling of B and C: o 2 : 1 BuLi (1.0 eq), -78 C, then O O CH2Cl2 : O O Me Me Me H 0.5 N NaCO3 Me H Ph in H2O 91% O O N CHO CHO Me H S CHO 85% (ii) NaH, then MOMCl H O AcO O O HO O 90% Mixture of C epimers O O O 5 MOMO O

Me Me Me Me MeO NH2 THF, reflux 2 h O O O O Me H silica gel, Me H N deactivated Me Me S Me Me S CHO 70-80% O O HN Ph O O HN Ph Me H H Me H H OH OH MOMO O MOMO O + The thermodynamically more favorable epimer is obtained exclusively. CONH2 MeO C CONH2 H 2 Me Me MOMO O OH O MOMO O O O S Ph Ph O O N 27% Mixture of diastereomers at C4, C4a Me H N S Once again, Muxfeldt employs his beautiful method for forming the A and B Pb2(OH)(OAc)3 rings in a single step. And once again, there is little stereocontrol - all four O 77% possible epimers at C4 and C4a are formed in solution. Fortunately, the desired diastereomer readily crystallizes. The reason for employing the thiazolanone B MOMO O rather than the oxazolanone employed before will become clear shortly...

10 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Me Me S Me Me S S 9 : 1 O O HN Ph O O HN Ph AcOH : OH OH HN Ph (i) MeI in THF OH OH NH3 Cl Me H H Me H H Me H H (ii) 0.17 N HCl Me H H OH H2O OH OH OH in THF-H2O, 1.5 h reflux, CONH 6 min 2 CONH2 CONH2 CONH2 H 90% H OH OH MOMO O OH O HO O OH O HO O OH O HO O OH O

No epimerization at C4 observed! While the oxazolone substrate could also be carried to this step, the resulting benzoyl amide could not be deprotected at this stage, nor could any other S amide devised, without decomposition. By contrast, deprotection conditions for the thioamide proved to be sufficiently gentle. OH OH HN Ph Me H H OH 12% OH OH NMe2 Me H H desired C12a Me2SO4 OH CONH2 hydroxylated (i) P(OEt)3, NaH, O2 OH HO O OH O product DMF-THF-H2O (i-Pr)2NEt 15 min + 23% from CONH2 thioamide OH HO O OH O (ii) 0.01 N HCl Me Me S in MeOH terramycin rt, 1.5 h O O HN Ph Me H H 47% OH C11a hydroxylated This concludes an elegant synthesis which assembles the A and B rings in a byproduct single step. Unfortunately, Muxfeldt and colleagues never satisfactorily address CONH2 the issues of controlling the C4 and C4a stereocenters, nor do they improve OH HO O OH O upon Woodward's solution to the C12a hydroxylation problem. +

14% recovered SM (i) hydroxylates the molecule; (ii) cleaves the acetonide. Unfortunately, hydroxylation occurs principally at C11a. In a fortunate accident, however, it turned out that the acetonide could not be cleaved unless the C12a hydroxyl was present. Thus separation of the desired deprotected product from the undesired major product was quite facile by polyamide chromatography.

11 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Stork: Controlling the C , C Stereocenters O 4 4a O O O OAllyl G. Stork, et al. J. Am. Chem. Soc. 1996, 118, 5304. piperidine (11 eq) O OAllyl AcOH (40 eq) Me O Here Gilbert Stork and colleagues take a completely different approach in order mol. sieves Me CHO to achieve stereocontrol at the C4 and C4a centers. benzene O O OH 0 oC --> rt Me (i) MeMgBr 2.5 h OH O 78% 97% OH O 100% (ii) ∆ 45% overall yield from start 100% of the synthetic sequence! OH O OH O OH O OBn O This sets the C stereocenter. Now watch Stork use this 6 MeO C O OAllyl stereocenter to bootstrap his way through the molecule... 2 N , then NMe 4a 2 OEt OEt Me2N O OEt O Me 5a O O n-Bu SnH, O NaHMDS, then H Br Me 3 Me AIBN the above tricycle N Pd(PPh3)4 Br MeO2C Br N,N-dimethyl- ∆ 95% from OBn the tricycle aniline, ∆ 90% OH O 98% Here Stork exploits the stereochemistry of the tricycle to OH O OH O direct conjugate addition to the more accessible face. O O Observe that the C5a and C4a stereocenters are now set. , then OH HO O O Me OH NMe HS SH S 2 TFA anhydride, then NMe2 Me H H O 4a Bu SnOCH O . the dithiane Me 3 3 5a 4a BF3 OEt2 S 5a H O N o 92% o 0 C, 15 min N 60 C Stork postulates a ketene intermediate MeO2C MeO C 88% OH O 97% 2 MeO2C formed from the mixed anhydride. OH O OBn OBn Mild reagent for O O O O OH O lactone cleavage (i) PhI(OTFA)2, Me NMe2 O OAllyl MeOH O OAllyl H H This protects the C and Me 92% Me 6 S TMSCN O C10 hydroxyls, and sets (ii) 5% aq. HCl CHO O N the stage for the KCN remaining cyclization S quant. 18-crown-6 MeO C 2 MeO2C steps. Transketalization, followed OTMS OTMS OBn OH O by hydrolysis to aldehyde. OH O

12 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

KH Me NMe A Note on C12a Hydroxylation: This intercepts an intermediate which has Me NMe2 2 H H (25 eq) H H been hydroxylated at the C12a position according to literature reports, O O completing in principle the formal synthesis of tetracycline. However, Stork o O N -78 --> 0 C, O N and colleagues were unable to successfully apply any of the C12a 3 h, then hydroxylation methods previously reported. The presence of the C4 MeO C o MeO2C 2 MeO2C 0 --> 50 C, dimethylamino substituent seems to interfere with the hydroxylation. Clearly BnO OTMS OTMS O BnO OTMS OTMS 30 min a satisfactory solution to the C12a hydroxylation problem is still needed...

O NMe2 OH NMe2 Me H H Me H H O O N N

MeO BnO BnO O O O O OH O OH OH 59%

The protecting group scheme permits formation of the A ring first, followed by in situ deprotection and cyclization of the B ring to complete the basic tetracycline framework. Previous studies had indicated that failure to protect the C11 ketone resulted in formation of a BCD tricycle for which conditions to complete A ring cyclization could not be found.

OH NMe2 Me H H H2 Pd black OH 94% CONH2 OH O OH OH 12a-deoxytetracycline

13 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Tatsuta: Asymmetric Total Synthesis of Natural (-)-Tetracycline CrO3, H2SO4 K. Tatsuta, et al. Chem. Lett. , 647. OBn O NHCbz o 2000 H 0 C, 10 min OBn 85% Here Tatsuta and colleagues not only produce an asymmetric total synthesis, but OTMS they also take a very different approach to the synthetic problem, constructing OH the A and B rings first and exploiting the carbohydrate chiral pool for starting BnO tBu tBu H materials. And as a bonus, they solve the C12a hydroxylation problem as well! NHCbz OH O OBn

(i) DMSO 170 oC DCC, Py-TFA (i) HCl-MeOH OTBS OTBS Me 72% 97% 93% H C O 2 O (ii) Me HO (ii) Ph3PCH3Br NO BnO BuLi/THF, BnO 2 CbzHN o -78 C --> rt CbzHN NHCbz O OMe OMe Me OH NHCbz 91% SeCN H H H OBn OBn SOCl2 PBu3 (cat.) O Et3N 90% OMe LDA, -40 oC, -30 oC NO2 15 min 10 min (i) BnBr O OH OBn OMe OH O OH OBn . HO 80% 90% BH3 THF, BaO/Ba(OH)2 Se 0 --> 45 oC; CH2 84% O (i) BBr H2C then H O , Me NHCbz 3 Me NHBoc O 2 2 (ii) HgCl2 H o H NaOH/THF BnO -78 C THF-H2O OBn OH BnO 69% CbzHN 15 min OMe 67% CbzHN OMe (ii) H2, Pd/C Boc2O, Et3N 92% over (i) MsCl, Et N OMe OH O OH OBn OMe OH O OH OH BnO 3 two steps BnO 0 oC, 15 min CH2 O 82% OH Me NHBoc TMSCHN2 H BnO (ii) DBU, -30 oC OH BnO CHO i-Pr2NEt CbzHN CbzHN quant. OH 72% In addition to eliminating to the enone, (ii) also epimerizes to the thermodynamic diastereomer. OMe OH O OMe OH Attempts to directly oxidize this 1,3-diol to the 1,3-dicarbonyl failed, requiring the following detour of sequential alcohol oxidations.

14 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Me NHBoc Me NHBoc H Br2 H OH OH O (Bu3Sn)2O OH Br mol. sieves O N o N N -78 C, 15 min N N OMe OH O OMe OH OMe OH O OMe 85% O H H Me NHBoc H Dess-Martin O Zn, AcOH periodinane 2 min (i) H3PO4 Me NH2 Me NMe2 Br H 100 oC,45 min H 15 min OH OH 91% 68% OMe OH O OMe O (ii) H2CO, CN CONH2 OH HCOOH OH OMe OH O O 80 oC, 1 h OMe OH O O Me NHBoc Me NHBoc H Dess-Martin H 80% O periodindane O

Me NMe2 15 min OH H 62% over OH O2, hν H BBr3 OMe OH O OMe OH two steps OMe OH O OMe 0 --> rt meso-tetraphenyl- porphyrin (cat.) Cl 88% CONH2 OH 10 min B OH OH O O 75% TsN NTs O O Me NHBoc . H (i) H2NOH HCl OOH NMe OH 2 Ph Ph Et3N, 30 min Me H OH O (ii) O Et3N -78 oC, 30 min OH CONH2 60% OMe OH O O N N N OH N OH O O O

Here Tatsuta et al. employ DMDO to achieve the desired 80%over 2 steps hydroxylation. They also achieve enantioselectivity by Here Tatstuta et al apply a protocol developed by Wassermann, Lu and exploiting the chiral boron catalyst which Corey developed Mechanism? Scott for hydroxylating anhydrotetracyclines. Provide a mechanism for this for enantioselective aldols and Diels-Alder reactions. Note reaction, and rationalize the stereospecific nature of this reaction. the fantastic yield! H. Wassermann, T.-J. Lu, A.I. Scott. J. Am. Chem. Soc. 1986, 108, 4237.

15 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Myers' Rapid Asymmetric Access to Analogs of Tetracycline O O O H O H M.G. Charest, C.D. Lerner, J.D. Brubaker, D.R. Siegel, A.G. Myers. Science 2005, H 308, 395. H H H H H In an extraordinary report, Myers and colleagues present a highly efficient and NMe2 NMe2 O O O O O O enantioselective method for accessing the tetracyclines. H H H H OH O OH O H H mCPBA H O H O A. eutrophus B9 EtOAc N N 12a CO H O CO H H H 2 2 75%, >95% ee 83% CO2H OH OH Wassermann, Lu and Scott invoke a formal ene reaction. The orbital OH OH alignment requirements dictate that only the axial hydrogen can participate Notice how Myers begins with installation of the in the reaction, inducing hydroperoxidation on the upper face of the This bacterial-catalyzed reaction troublesome C hydroxyl group, and then molecule and thus ensuring the proper stereochemistry at C . can be run on a 90 g scale! 12a 6 proceeds to build the molecule around it!

1. TMSCHN NMe2 OOH NMe 2 o 2 OH NMe2 THF, -78 C Me H Me H H 2. TBSOTf, Et3N OH O H /Pt OH O 2 70% NMe2 O N TBSO TBSO 62% CO2Me O CONH2 OBn CONH2 OTBS OH OH N OTBS O OH O O O OH O OH O Li 73% (-)-tetracycline OBn

Me NMe 1. LiOTf, 2 There are many elegant features to this synthesis. Tatsuta and colleagues Me O H o N N mimic Stork's Diels-Alder approach to establishing stereochemistry, but employ toluene, 60 C O it to define the troublesome C stereocenter immediately. They construct the B A N 4a 2. TFA, CH Cl central tetracycline scaffolding in just three steps from simple precursors. And 2 2 BnO HO 60% TBSO OH they solve the C12a hydroxylation problem with a very mild oxidant in the OH OBn O TBSO O presence of a chiral catalyst, and introduce the C6 hydroxyl stereospecifically TBSO at a very late stage of the synthesis. 21% over 7 steps Here Myers closes the ring and sets the C4 amine stereochemistry. Myers compares this key ring-closing step to a Sommelet-Hauser rearrangement, where the amine initially undergoes an intramolecular SN-prime epoxide ring opening, followed by ylide formation and finally a [2,3] sigmatropic rearrangement. TFA selectively deprotects the allylic alcohol. Notice the remarkable yield so far!

16 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Now Myers takes his key intermediate A and converts it into two fragments: B, BnO2CO NMe2 which will go on to form C6-deoxy analogs of tetracycline, and C, which will go H on to form analogs with the normal C6-oxygenation. O N C NMe 2 NMe2 H H OBn O PPh3, DEAD; O O O N N OTBS then O HO 2 S NH With these fragments in hand, Myers now can install the C and D rings, and he OH OBn 2 OH TBSO O N OBn proceeds to do so in a fashion that allows for analogs of tetracycline with deep- H TBSO O seated structural modifications to the D ring. A NO2 74% O

Me OBn 1. HCl, MeOH NMe2 Me O NMe H LDA, TMEDA, 2 2. IBX, DMSO THF, -78 oC; H H 3. TBSOTf, 2,6-lutidine O O N B then C, N CO2Ph o o 66% -78 C -> 0 C OBn BocO OBn O O 79% OH O OH O OTBS OTBS

NMe2 H NMe2 H O 1. CBr , PPh 4 3 O Me OH NMe2 N 2. PhSH, Et3N H H N 1. HF, MeCN OH HO 2. H2, Pd, THF/MeOH 87% PhS OH OBn TBSO O OH OBn NH2 TBSO O 90% OH A OH O OH O O

Me Me Cl (-)-doxycycline OH NMe2 1. BnO CCl, DMAP 18 steps, 8.3% Cl H 2 1. N 2. TBAF, HOAc O 3. IBX, DMSO S O N O2 4. TBSOTf, Et3N

2. P(OMe)3, MeOH OH OBn TBSO O 85% 70 oC 76%

17 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

With this strategy, Myers and colleagues are able to synthesize a number of remarkable analogs of tetracycline:

Me NMe2 H H OH Me B

NH2 CO2Ph OH OBoc OH O OH O O (-)-6-deoxytetracycline

NMe2 Me Me H H B OH

N HN NH2 CO2Ph OH OBn O O OH O O

NMe2 H H N Me B N OH

NH2 CO2Ph OH O OH O O

NMe2 H H OH CH2Br B

NH2 CO Ph 2 OH O OH O O

NMe2 CH2Br H H B OH

CO2Ph NH2 OMe OH OMe O OH O O

18 Baran Lab T h e T e t r a c y c l i n e s D. W. Lin

Addendum: Tetracycline Tidbits Barton and colleagues (including a young Steven Ley!) also discovered the utility of phenylseleninic anhydride for the deprotection of dithianes. This led to their applying this reagent in a variety of transformations: D.H.R. Barton spent over a decade tinkering with tetracycline, but never completed a total synthesis of the molecule. Over the course of this work, however, he O O O O R discovered some interesting chemistry (naturally). Me R Me PhSe SePh O Photocyclizations of acetals onto enones: Me R' Me R'

O O

hν AcO AcO H H H H ArCO2H H D.H.R. Barton, D.J. Lester, S.V. Ley. J. Chem. Soc. Perkin Trans. I 1980, 2209. O O O O O O Ph Ph In his book Reason and Imagination, Barton concludes his chapter on the tetracyclines with the following perspective on the role of academic research in synthetic chemistry today:

"Just as the studies on the bitter principles [a class of natural products] O OH convinced me that X-ray crystallography was a superior procedure for structure determination, the major effort on tetracycline synthesis convinced me that this sort of work should be left to Industrial friends who have the money and the resources to finish any multi-step synthesis, if it is economically justified. So it H is the originality in the reactions and the reagents and any new principles O O that finally justify academic effort in synthesis. We are far away from the O O O O Woodwardian dogma of completely planned synthesis." Ph Ph

D.H.R. Barton, et al. J. Chem. Soc. Perkin Trans. I 1976, 503.

19