Short chemoenzymatic total synthesis of ent-hydromorphone: An oxidative dearomatization/intramolecular [4+2] cycloaddition/amination sequence

Article abstract of DOI:10.1002/anie.201400286

A short synthesis of ent-hydromorphone has been achieved in twelve steps from β-bromoethylbenzene. The key transformations involved the enzymatic dihydroxylation of the arene to the corresponding cis-dihydrodiol, Mitsunobu coupling with the ring?A fragmen

Full text of DOI:10.1002/anie.201400286

Angewandte
Chemie
DOI: 10.1002/anie.201400286
Natural Products
Short Chemoenzymatic Total Synthesis of ent-Hydromorphone: An
Oxidative Dearomatization/Intramolecular [4+2] Cycloaddition/
Amination Sequence**
Vimal Varghese and Tomas Hudlicky*
Abstract: A short synthesis of ent-hydromorphone has been
achieved in twelve steps from b-bromoethylbenzene. The key
transformations involved the enzymatic dihydroxylation of the
arene to the corresponding cis-dihydrodiol, Mitsunobu cou-
pling with the ring A fragment, oxidative dearomatization of
the C3 phenol, and the subsequent [4+2] cycloaddition to form
ring B of the morphinan. The synthesis was completed by
intramolecular amination at C9.
synthesis, at the stage of phenol 7 (Figure 1). This approach to
ent-hydromorphone (4) is enantiodivergent and is also
applicable to the natural enantiomer. In several previous
approaches to morphinans we have demonstrated that the
configuration at C5 in intermediates of type 1 controls all
subsequent stereochemical events in either enantiomeric
series and that the natural configuration in 1 is attainable by
double Mitsunobu inversion.[2a] Herein we report a short
synthesis of 4 by a new design employing an oxidative
dearomatization/intramolecular [4+2] cycloaddition/amina-
tion strategy.
The Diels–Alder reaction has been used only once in
a direct construction of ring B of the morphine skeleton,
namely in an intermolecular [4+2] approach by Kerr and
Tius.[3] Two model studies utilizing an intramolecular Diels–
Alder reaction to construct ring B, leading to morphinan
substructures, have been reported by us[4,5] and by Rodrigo
and co-workers.[6] Thus our current report constitutes the first
use of the intramolecular Diels–Alder strategy for ring B
closure in a total synthesis of a morphinan and also the first
total synthesis of ent-hydromorphone (4).
A
truly practical synthesis of morphine and its congeners has
not yet appeared in spite of focused efforts and many creative
approaches having been published.[1] We have been involved
in the design and synthesis of morphinans for many years and
published several total syntheses of codeine and congeners,
the shortest, at 12 steps, still far removed from reaching the
realm of practicality.
We recently designed an advanced strategy to construct
the morphine skeleton by an intramolecular [4+2] cyclo-
addition of dienone 2 produced by oxidative dearomatization
of a phenol such as 1 (Figure 1). The homochiral portion of 1,
prepared by toluene-dioxygenase-mediated dihydroxylation
of an appropriate arene, is coupled with the phenolic frag-
ment by a Mitsunobu reaction. The functionalities at C16 and
C9 are appropriate for the incipient closure of the ethylamino
bridge in 3 (or its aromatized equivalent). Deprotection and
oxidation will lead directly to ent-hydromorphone (4).
To test the feasibility of the cycloaddition of a species such
as 2, we pursued a model study of the cycloaddition of 5
(without the nucleophilic group Y) to 6 followed by the
known hydroamination methodology[2] to set C9 late in the
In 1992[4] and 1998[5] we published two simple model
studies on the intramolecular Diels–Alder cycloadditions of
homochiral dienes with unsaturated tethers to furnish tricyclic
systems with five contiguous stereogenic centers, representing
rings B, C, and E of morphine. The dienes were derived in
three steps from toluene[4] or b-bromoethylbenzene[5] by
enzymatic dihydroxylation. In 1998, a Diels–Alder/Cope
sequence, similar in concept to our model studies, was
published by Rodrigo and co-workers.[6e] The oxidative
dearomatization of phenols[7] and subsequent cycloaddition
was exploited by Rodrigo and co-workers on structurally
different substrates in their approaches to various natural
products, including the synthesis of a partial morphine
skeleton, that is, rings A, B, C, and E.[6a,e,f]
[*] V. Varghese, Prof. T. Hudlicky
Department of Chemistry and Centre for Biotechnology
Brock University, St. Catharines, ON L2S 3A1 (Canada)
E-mail: thudlicky@brocku.ca
[**] We thank the following agencies for financial support of this work:
Natural Sciences and Engineering Research Council of Canada
(NSERC) (Idea to Innovation and Discovery Grants), Canada
Research Chair Program, Canada Foundation for Innovation (CFI),
TDC Research, Inc., TDC Research Foundation, the Ontario
Partnership for Innovation and Commercialization (OPIC), and The
Advanced Biomanufacturing Centre (Brock University). We also
appreciate the skillful assistance of Jef de Brabander, Jr. and Stuart
Williamson for the preparation of several starting materials. We
thank Razvan Simionescu and Tim Jones for their help with NMR
and mass spectrometry analyses. Last but not least we are indeed
very grateful to Christie and Eric Boros, Gabor Butora, Andrew Gum,
and Khalil Abboud for their efforts in the model studies some two
decades ago.
The above precedents bode well for a successful approach,
which we began with the synthesis of the two subunits to be
joined to the precursor required for a compound such as 5.
The syntheses of the homochiral subunits 13 and 14 were
accomplished as shown in Scheme 1 and as previously
described.[2a,8] Dihydroxylation of 8 by whole-cell fermenta-
tion with E. coli JM 109 (pDTG601A)[9] yielded 9, which was
immediately subjected to a selective reduction with potassium
azodicarboxylate and subsequent protection of the diol to
give 10. The displacement of bromine in 10 with methylamine
produced 11 (the tosylation of this compound could be used in
the future to provide 14 and hence 19b, in a more direct way;
see Scheme 3). Hydrolysis of the acetonide with HCl and
aubsequent protection of the secondary amine as a Boc
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201400286.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
phenol with MOMCl under mild basic con-
ditions afforded aldehyde 17, and the hydrol-
ysis of the acetyl group completed the syn-
thesis of the required phenol 18.
The two units were coupled, by a Mitsu-
nobu protocol, to give either 19a (R = Boc) or
19b (R = Ts, obtained from 14), as shown in
Scheme 3. After the aldehyde was converted
into a vinyl group by a Wittig reaction, the
MOM group in 20 was removed under mild
reaction conditions[11] to produce the free
phenol 21. Exposure of this material to lead
tetraacetate in refluxing dichloroethane pro-
vided, via dienone 5, the [4+2] adduct 6 in
50% yield upon isolation. It is possible that
only one diastereomer of 5 underwent the
cycloaddition, thus further optimization of this
step will be required.
Figure 1. Advanced strategy to access morphinans by cycloaddition to construct ring B.
Dienone 5 underwent the cycloaddition exclusively at the
site of the exocyclic diene, and no other cycloaddition
products or dimerizations were observed in this reaction
(only once, about 3–4% of the product of the endocyclic
cycloaddition was detected). Even though the endocyclic
diene is quite reactive, the exclusive formation of 6 can be
attributed to steric reasons, which deny the dienophile the
proximity of the endocyclic diene. The 13C NMR spectrum of
6 exhibited three carbonyl signals, a signal at d = 188.2 ppm
corresponding to the enone, a signal at d = 170.9 ppm
corresponding to the ester carbonyl group of the acetyl
group, and a signal at d = 155.4 ppm corresponding to the
amide carbonyl group. The assignment of the stereochemistry
at C4/C12 in 6, obtained as a single stereoisomer, was
complicated by the presence of rotamers and was not made
(H12 should be b, assuming the exo transition state). Rodrigo
and co-workers showed in their synthesis of indolinocodeine
that the quinoid portion acts as both diene and dienophile,
thus leading to mixtures of different products.[6a,e]
Scheme 1. a) E. coli JM 109 (pDTG601A), 10–15 gLÀ1; b) 1. potassium
azodicarboxylate, AcOH, MeOH, 08C, 83%; 2. 2,2-dimethoxypropane,
acetone, pTsOH, 80%; c) MeNH2, K2CO3, THF, sealed tube, 93%;
d) 1. 3m HCl, EtOH; 2. Boc2O, NaHCO3, EtOH, 74% (2 steps);
e) TBSCl, imidazole, CH2Cl2, À788C to RT, 92%; f) 1. TFA, CH2Cl2,
08C; 2. TsCl, Et3N, CH2Cl2, RT, 86% over two steps. Boc=tert-
butoxycarbonyl, TBS=tert-butyldimethylsilyl, TFA=trifluoroacetic acid,
Ts =4-toluenesulfonyl.
Dienone 6 was treated with trifluoroacetic acid to afford
phenol 22 by re-aromatization and the concomitant hydrolysis
of the Boc carbamate. We then turned our attention to the
construction of the D ring by hydroamination of 22. However,
all our attempts failed to install the ethylamino bridge
through either an aminomercuration used previously by
us[2a] or through a photostimulated addition of lithium
amide reported Trost and Tang.[2c] Analysis of the crude
reaction mixture suggested some evidence of hydroamination
under the aforementioned conditions but the isolation of pure
products from these reactions was not possible. The failure of
the aminomercuartion was likely due to the instability of 22.
This compound (22) was therefore taken to the next step
without purification to provide the tosyl amide 23 upon
treatment with excess tosyl chloride (with concomitant
tosylation of the phenolic hydroxyl) in 45% yield, over the
two steps. Treatment of 6 with trifluoroacetic acid for longer
period of time resulted in the formation of 7, which was
converted into the corresponding amide 24 by treatment with
excess tosyl chloride. The establishment of the ethylamino
bridge in 25 was accomplished by a nitrogen-centered radical
cyclization enabled by a dissolved-metal reduction of the tosyl
carbamate provided 12 in a one-pot operation. Regioselective
silylation of the distal hydroxy group then provided alcohol
13, and subsequent deprotection and tosylation gave 14.
The arene coupling partner 18 was synthesized from 3,4-
dihydroxybenzaldehyde by adjustment of a known proto-
col,[10] as shown in Scheme 2. The synthesis of 18 started with
the regioselective acetylation of 3,4-dihydroxybenzaldehyde
(15) to the monoacetylated derivative 16. Protection of the
Scheme 2. a) Ac2O, NaOH, THF, 08C, 82–85%; b) MOMCl, K2CO3,
DMF, 08C to RT, 76–80%; c) K2CO3, MeOH, RT, 88–90%. DMF=N,N-
dimethylformamide.
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
The successful oxidative dearomatization and
cycloaddition strategy presented here bodes
well for the pursuit of the next generation of
this approach and further improvements in the
synthesis of other morphinans in either or both
enantiomeric series.
Received: January 10, 2014
Published online: && &&, &&&&
Keywords: alkaloids · cycloaddition ·
.
natural products · oxidation · total synthesis
[1] For reviews of morphine alkaloid syntheses and
discussion of strategies, see: a) U. Rinner, T.
Hudlicky, Top. Curr. Chem. 2012, 309, 33 – 66;
b) J. Zezula, T. Hudlicky, Synlett 2005, 388 – 405;
c) D. F. Taber, T. D. Neubert, M. F. Schlecht in
Strategies and Tactics in Organic Synthesis,
Vol. 5 (Ed.: H. Michael), Elsevier, London,
2004, pp. 353 – 389; d) T. Hudlicky, J. Hetero-
cycl. Chem. 2000, 37, 535 – 539; e) B. H. Novak,
T. Hudlicky, J. W. Reed, J. Mulzer, D. Trauner,
Curr. Org. Chem. 2000, 4, 343 – 362; f) T.
Hudlicky, G. Butora, S. P. Fearnley, A. G.
Gum, M. R. Stabile in Studies in Natural
Products Chemistry, Vol. 18, Part K (Ed.: Atta-
ur-Rahman), Elsevier, Amsterdam, 1995,
pp. 43 – 154; g) M. Maier in Organic Synthesis
Highlights II (Ed.: H. Waldmann), VCH, Wein-
heim, 1995, pp. 357 – 369.
Scheme 3. a) TMAD, PBu3, 81–85%; b) CH3PPh3Br, nBuLi, THF, À788C to 08C then
reflux for 4 h, 82–88%; c) ZnBr2, CH3(CH2)10CH2SH, CH2Cl2, RT, 10 min, 92%; d) Pb-
(OAc)4, DCE, reflux, 4 h, 50%; e) TFA, CH2Cl2, 08C, 15 min; f) TsCl, Et3N, CH2Cl2, 08C to
RT, 45% over two steps; g) TBAF, THF, RT, 86%; h) Li, tBuOH, NH3(liq), THF, À788C,
10 min [82–86% for 23 to 25; 93% for 24 to 26]; i) tBuOK, PhCOPh, PhCH3/DME, 858C,
8 h, 44%; j) 1. Hg(OCOCF3)2, CH3CN RT; 2. NaBH4, THF, À788C to RT. DCE=1,2-
dichloroethane, DME=dimethoxyethane, TBAF=tert-n-butylammonium fluoride,
TMAD=N,N,N’,N’-tetramethylazodicarboxamide.
[2] See for example: a) H. Leisch, A. T. Omori,
K. J. Finn, J. Gilmet, T. Bissett, D. Ilceski, T.
Hudlicky, Tetrahedron 2009, 65, 9862 – 9875; b) A. T. Omori,
K. J. Finn, H. Leisch, R. J. Carroll, T. Hudlicky, Synlett 2007,
2859 – 2862; c) B. M. Trost, W. Tang, J. Am. Chem. Soc. 2002, 124,
14542 – 14543; d) B. M. Trost, W. Tang, F. D. Toste, J. Am. Chem.
Soc. 2005, 127, 14785 – 14803.
amide according to conditions adapted from the work of
Parker and Fokas[12] and Chida and co-workers,[13] and similar
in concept to Trostꢀs lithium amide cyclization[2c,d] noted
above. The reductive cyclization of 23 to 25 was superior to
the hydroamination conditions [Hg(OAc)2, followed by
LiAlH4] we have used previously,[2a,b] and gave 25 in 82–
86% yield. In addition, the reductive cyclization was also
performed with the free alcohol 24 to give 26 in 93% yield.
Oxidation of 26 to ent-hydromorphone (4) was accom-
plished with benzophenone and tBuOK in 44% yield (83%
based on recovered starting material) using a modified
procedure from Woodward and co-workers and Rapoport
and co-workers.[14–16] Because of poor solubility of 26 in most
of the solvents, the reaction did not proceed to completion,
and the starting material (53%) was recovered from the
reaction. Another explanation for the low yield is the trans
relationship between the C5 and C6 positions in the ent-
dihydroisomorphine 26. Rapoport has provided a reasonable
explanation based on the pseudo-six-membered ring confor-
mation involved in such oxidations.[15]
[3] M. A. Tius, M. A. Kerr, J. Am. Chem. Soc. 1992, 114, 5959 – 5966.
[4] T. Hudlicky, C. H. Boros, E. E. Boros, Synthesis 1992, 174 – 178.
[5] G. Butora, A. G. Gum, T. Hudlicky, K. A. Abboud, Synthesis
1998, 275 – 278.
[6] a) J. Gao, J. Orso Simon, R. Rodrigo, A. Assoud, J. Org. Chem.
2013, 78, 48 – 58; b) Y. Lang, F. E. S. Souza, X. Xu, N. J. Taylor,
A. Assoud, R. Rodrigo, J. Org. Chem. 2009, 74, 5429 – 5439;
c) H. S. Sutherland, K. C. Higgs, N. J. Taylor, R. Rodrigo,
Tetrahedron 2001, 57, 309 – 317; d) F. E. S. Souza, R. Rodrigo,
Chem. Commun. 1999, 1947 – 1948; e) R. Carlini, K. Higgs, R.
Rodrigo, N. Taylor, Chem. Commun. 1998, 65 – 66; f) R. Carlini,
K. Higgs, C. Older, S. Randhawa, R. Rodrigo, J. Org. Chem.
1997, 62, 2330 – 2331; g) R. Carlini, K. Higgs, N. Taylor, R.
Rodrigo, Can. J. Chem. 1997, 75, 805 – 816.
[7] a) Q. Ding, Y. Ye, R. Fan, Synthesis 2013, 1 – 16; b) C.-X. Zhuo,
W. Zhang, S.-L. You, Angew. Chem. 2012, 124, 12834 – 12858;
Angew. Chem. Int. Ed. 2012, 51, 12662 – 12686; c) S. P. Roche,
J. A. Porco, Jr., Angew. Chem. 2011, 123, 4154 – 4179; Angew.
Chem. Int. Ed. 2011, 50, 4068 – 4093; d) L. Pouysꢁgu, D.
Deffieux, S. Quideau, Tetrahedron 2010, 66, 2235 – 2261; e) S.
Quideau, L. Pouysꢁgu, D. Deffieux, Synlett 2008, 467 – 495; f) F.
Lꢂpez Ortiz, M. J. Iglesias, I. Fernꢃndez, C. M. Andffljar Sꢃn-
chez, G. Ruiz Gꢂmez, Chem. Rev. 2007, 107, 1580 – 1691; g) J. M.
Keane, W. D. Harman, Organometallics 2005, 24, 1786 – 1798;
h) C.-C. Liao, Pure Appl. Chem. 2005, 77, 1221 – 1234; i) D.
Magdziak, S. J. Meek, T. R. R. Pettus, Chem. Rev. 2004, 104,
1383 – 1429; j) C.-C. Liao, R. K. Peddinti, Acc. Chem. Res. 2002,
In summary, the most effective route to the first total
synthesis of ent-hydromorphone (4) (NMR matched with an
authentic sample prepared from oripavine)[17] described in
this paper was accomplished in just seven steps (five
operations) from 19a (it can be also attained from 19b
through six steps or 12 steps from b-bromoethylbenzene),
thus making it one of the shortest synthesis of a morphinan.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
35, 856 – 866; k) P. L. Smith, M. D. Chordia, W. Dean Harman,
Tetrahedron 2001, 57, 8203 – 8225; l) A. R. Pape, K. P. Kaliap-
pan, E. P. Kunding, Chem. Rev. 2000, 100, 2917 – 2940; m) S.
Quideau, L. Pouysegu, Org. Prep. Proced. Int. 1999, 31, 617 –
680; n) V. Singh, Acc. Chem. Res. 1999, 32, 324 – 333; o) T. Bach,
Angew. Chem. 1996, 108, 795 – 796; Angew. Chem. Int. Ed. Engl.
1996, 35, 729 – 730; p) L. N. Mander, Synlett 1991, 134 – 144.
[8] J. Duchek, T. G. Piercy, J. Gilmet, T. Hudlicky, Can. J. Chem.
2011, 89, 709 – 729.
[12] a) K. A. Parker, D. Fokas, J. Am. Chem. Soc. 1992, 114, 9688 –
9689; b) K. A. Parker, D. Fokas, J. Org. Chem. 2006, 71, 449 –
455.
[13] M. Ichiki, H. Tanimoto, S. Miwa, R. Saito, T. Sato, N. Chida,
Chem. Eur. J. 2013, 19, 264 – 269.
[14] R. B. Woodward, N. L. Wendler, F. J. Brutschy, J. Am. Chem.
Soc. 1945, 67, 1425 – 1429.
[15] H. Rapoport, R. Naumann, E. R. Bissell, R. M. Bonner, J. Org.
Chem. 1950, 15, 1103 – 1107.
[9] G. J. Zylstra, D. T. Gibson, J. Biol. Chem. 1989, 264, 14940 –
14946.
[10] A. M. Sawayama, H. Tanaka, T. J. Wandless, J. Org. Chem. 2004,
69, 8810 – 8820.
[16] R. Csuk, G. Vasileva, A. Barthel, Synthesis 2012, 2840 – 2842.
[17] M. A. E. Arias, T. Hudlicky, unpublished observations, United
States Provisional Patent Application No. 61/937,126, Methods
for the preparation of Hydromorphone. Filed: February 7, 2014.
[11] a) J.-H. Sohn, N. Waizumi, H. M. Zhong, V. H. Rawal, J. Am.
Chem. Soc. 2005, 127, 7290 – 7291; b) J. H. Han, Y. E. Kwon, J.-
H. Sohn, D. H. Ryu, Tetrahedron 2010, 66, 1673 – 1677.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Natural Products
V. Varghese, T. Hudlicky*
&&&& — &&&&
Short Chemoenzymatic Total Synthesis of
ent-Hydromorphone: An Oxidative
Dearomatization/Intramolecular [4+2]
Cycloaddition/Amination Sequence
A short synthesis of ent-hydromorphone
has been achieved in twelve steps from b-
bromoethylbenzene. The key transforma-
tions involved the enzymatic dihydroxy-
lation of the arene to the corresponding
cis-dihydrodiol, oxidative dearomatiza-
tion, subsequent [4+2] cycloaddition to
form ring B of the morphinan, and intra-
molecular amination at C9. Boc=tert-
butoxycarbonyl, TBS=tert-butyldimethyl-
silyl.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!