Synthesis of tyrosine derivatives for saframycin MX1 biosynthetic studies

Article abstract of DOI:10.1016/j.tetlet.2004.03.112

Saframycin MX1 and structural relatives are natural anticancer agents isolated from bacteria and marine invertebrates. For biosynthetic studies and to make a library of modified natural products, a series of tyrosine derivatives were synthesized in a conc

Full text of DOI:10.1016/j.tetlet.2004.03.112

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 3921–3924
Synthesis of tyrosine derivatives for saframycin MX1
biosynthetic studiesq
Eric W. Schmidt,* James T. Nelson and John P. Fillmore
Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT 84112, USA
Received 5 November 2003;revised 1 December 2003;accepted 19 March 2004
Abstract—Saframycin MX1 and structural relatives are natural anticancer agents isolated from bacteria and marine invertebrates.
For biosynthetic studies and to make a library of modified natural products, a series of tyrosine derivatives were synthesized in a
concise manner.
Ó 2004 Elsevier Ltd. All rights reserved.
Saframycin MX1 (1)1 is a member of a group of iso-
quinoline natural products, including renieramycins,2
jorumycin,3 and ecteinascidin 743 (Et-743),4 that show
promise in the treatment of some cancers.5 Because of
the potential clinical efficacy of Et-743, several synthetic
and semisynthetic routes to this compound class have
been reported.6–8 In addition, combinatorial strategies
have been employed to increase structural diversity of
saframycins and their relatives.9 Combinatorial biosyn-
thesis provides a complimentary method to produce a
large number of saframycin analogs, but an under-
standing of the saframycin biosynthetic pathway is
required before such a method can be used. Here, we
report on our preliminary efforts to de-convolute the
biosynthesis of 1.
Figure 1. Biosynthesis of saframycin MX1 (1) and potentially incor-
porated Tyr derivatives. Depicted amino acid incorporation is based
Saframycins are composed of the amino acids glycine,
alanine, and tyrosine, or derivatives thereof, as demon-
strated by feeding studies (Fig. 1) and further supported
upon studies with saframycin A.10
by biosynthetic gene sequencing.10–13 Analysis of the
genes by us and by the Shen group14 revealed that a
ing the biochemistry for drug design, we are exploring
single adenylation domain is likely responsible for acti-
the biosynthesis of saframycin MX1 (1).
vation and incorporation of two modified Tyr deriva-
tives into 1. We propose that the known safA and safB
Eight
L-Tyr derivatives, three of which can be pur-
genes probably also catalyze heterocyclization via novel
Pictet–Spengler- or Bischler–Napieralski-type conden-
sations (Scheme 1).15;16 Because of the novel mechanism
of heterocycle formation and the possibility of exploit-
chased, were identified as candidates for incorporation
by SafA2 (Fig. 1). The p-quinone derivatives of Tyr are
not likely precursors because of their instability and
because the oxidation state of Tyr in saframycin deriv-
atives is highly variable. We sought to design a simple
synthesis that would be flexible enough to provide the
other five intermediates using parallel chemistry and
that would allow the later synthesis of compound
libraries. We chose aromatic iodination and formylation
to synthesize intermediates 10, 12, and 16, which could
q Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.tetlet. 2004.03.112
* Corresponding author. Tel.: +1-801-585-5234;fax: +1-801-585-9119;
e-mail: ews1@utah.edu
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.03.112

3922
E. W. Schmidt et al. / Tetrahedron Letters 45 (2004) 3921–3924
Scheme 1. Proposed biogenesis of saframycin heterocycle. Route A: Pictet–Spengler route;and Route B: Bischler–Napieralski route.
be further elaborated by Stille coupling, oxidation, and
etherification. All compounds were prepared from the
readily available starting material, N-Boc-Tyr.
conditions, but can be readily reduced by metal hydrides
(data not shown) or cleaved by the Dakin reaction.
Using the formylation/oxidation protocol,17 5 was syn-
thesized from N-Boc-formyl-Tyr 10 in 60% over three
steps (Scheme 2). The two methyl groups of 11 were
introduced in a single step,18 but carboxyl-protected Tyr
derivatives proved extremely resistant to Dakin or
Baeyer–Villiger oxidation. Compound 11 was depro-
tected with LiOH/DMF/H2O, oxidized using the Dakin
reaction with H2O2/(PhSe)2, and treated with 1:1 TFA/
DCM to give Dopa derivative 5 in 68% overall yield.
Key steps in the synthesis of Dopa derivatives 5, 8, and 9
were the C-acylation of tyrosine and eventual oxidative
cleavage of the acyl group to afford a phenol. A variety
of acylation/Baeyer–Villiger oxidation protocols were
explored, but ultimately we settled on the method of
Jung and Lazarova,17 which employs a Riemer–
Tiemann reaction followed by a modified Dakin oxi-
dation. While the Riemer–Tiemann reaction affords
formylated tyrosine 10 in low yield (25–35% in our
hands), it is simple to perform and does not racemize the
With an efficient route to Dopa derivatives in hand, we
investigated the efficiency of iodination and Stille cou-
pling of various Tyr derivatives. Appropriate iodination
substrate (>95% L-isomer recovered). The resulting
formyl group is stable to acid, base, and oxidative
Scheme 2. Synthesis of Dopa derivatives. Reagents and conditions: (a) BocTyr/NaOH/CHCl3/H2O, 65–70 °C, 16 h (25–35%);(b) DCC/DMAP/
EtOH, rt, 36 h (quant);(c) I 2/H2O2/MeOH, rt, 16 h (77%);(d) (CH 3)4Sn/NMP, Pd2dba2ÆCHCl3/TPP/CuI (cat), 70 °C, 1 d (94%);(e) MeI/K 2CO3/
acetone, 50 °C, 16 h (96%);(f) i. LiOH/DMF/H 2O, rt, 16 h, ii. Ph2Se2/H2O2/DCM, rt, 18–20 h, iii. TFA/DCM, rt, 20 min (77%);(g) i. LiOH/DMF/
H2O, rt, 16 h, ii. Ph2Se2/H2O2/18-crown-6/K2HPO4/DCM, rt, 18–20 h, iii. TFA/DCM, rt, 20 min (50%);(h) as in (e) (86%);(i) as in (f) (53%).

E. W. Schmidt et al. / Tetrahedron Letters 45 (2004) 3921–3924
3923
protocols were developed, and Stille reactions were
performed following the method of Hudgens and
Turnbull.19 C-Methylated Tyr derivatives were synthe-
sized by coupling tetramethyltin and N-Boc-3-iodo-
treatment with TFA, the Dopa product 9 was produced
in 52% isolated yield.
The chiral purity of the amino acid products was
assessed by Marfey’s method.21 Surprisingly, despite
suggestions that harsh, basic reaction conditions such as
those employed by Jung and Lazarova will lead to
substantial racemization,22 only a single isomer was
detected for each compound. Based upon HPLC of
Marfey’s derivatives, Tyr and Dopa compounds 5–9 are
at least 90% enantiopure, since only a single peak was
detected at the appropriate wavelength and elution time
in each HPLC run. By comparison, an authentic stan-
20
tyrosine ethyl ester (16;Scheme 3). The reaction was
slow on this substrate, taking approximately 6 d at
70 °C. By contrast, C-methylated Dopa derivatives were
more efficiently produced from key intermediate 12,
providing the C-methyl compound 13 in 94% yield in
24 h. O-Methylation and deprotection of these inter-
mediates proceeded in a similar manner to that used for
11.
Dakin oxidation of 13 followed by Boc deprotection
cleanly afforded the Dopa derivative 8 (Scheme 2).
Unexpectedly, upon LiOH deprotection the O-methyl
derivative 14 formed a stable hydrate or acylal (Fig. 2),
which was resistant to oxidative cleavage. The hydrate
structure was proposed because the CHO proton singlet
at ꢀd 10.3 ppm in the parent compound 14 was replaced
by a doublet at ꢀd 8.0 ppm after hydrolysis, while all
other protons exhibited their expected chemical shifts.
Treatment of this intermediate with acids or bases
transiently yielded the aldehyde compound, but upon
workup only acetal was retrieved. Therefore, a tandem
dehydration/Dakin oxidation reaction was performed
by adding 18-crown-6 and dipotassium phosphate to the
standard Dakin conditions. After 72 h and subsequent
dard of D,L-tyrosine gave rise to two sharp HPLC
peaks, which eluted after 37.2 and 39.8 min. Because
complete inversion of configuration is an unlikely event
under our reaction conditions, the derivates all adopt
the L-configuration. Chemical shift data for 5–9 are
given below.23
Supplementary data
Detailed experimental procedures and 1H and 13C NMR
data for all compounds.
Acknowledgements
This research was supported by a startup grant from the
University of Utah. Stipend support to J.T.N. was
provided by an NIH Training Grant in Biological
Chemistry GM08537, and J.P.F. was funded in part by
the University of Utah Biosciences Undergraduate
Research Program.
References and notes
€
1. Irshik, H.;Trowitzsch-Kienast, W.;Gerth, K.;H ofle, G.;
Reichenbach, H. J. Antibiot. 1988, 41, 993–998.
2. He, H. Y.;Faulkner, D. J. J. Org. Chem. 1989, 54, 5822–
5824.
3. Fontana, A.;Cavaliere, P.;Wahidulla, S.;Naik, C. G.;
Cimino, G. Tetrahedron 2000, 56, 7305–7308.
4. Rinehart, K. L.;Holt, T. G.;Fregeau, N. L.;Stroh, J. G.;
Keifer, P. A.;Sun, F.;Li, L. H.;Martin, D. G.
Chem. 1990, 55, 4512–4515.
5. Aune, G. J.;Furuta, T.;Pommier, Y. Anticancer Drugs
2002, 13, 545–555.
6. Corey, E. J.;Gin, D. Y.;Kania, R. S. J. Am. Chem. Soc.
1996, 118, 9202–9203.
7. Martinez, E. J.;Owa, T.;Schreiber, S. L.;Corey, E. J.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3496–3501.
8. Cuevas, C.;Perez, M.;Martin, M. J.;Chicharro, J. L.;
Fernandez-Rivas, C.;Flores, M.;Francesch, A.;Gallego,
P.;Zarzuelo, M.;de la Calle, F.;Garcia, J.;Polanco, C.;
Rodriguez, I.;Manzanares, I. Org. Lett. 2000, 2, 2545–
2548.
J. Org.
Scheme 3. Synthesis of Tyr derivatives. Reagents and conditions: (a) i.
NaI/NaOH/MeOH, ii. NaOCl (4% aq), 4 °C, 2 h (quant);(b) DCC/
DMAP/EtOH, rt, 1 h (68%);(c) (CH 3)4Sn/NMP, Pd2dba2ÆCHCl3/TPP/
CuI (cat), 70 °C, 6 d (90%);(d) MeI/K 2CO3/acetone, 50 °C, 16 h (59%);
(e) i. LiOH/DMF/H2O, rt, 16 h, ii. Ph2Se2/H2O2/DCM, rt, 18–20 h, iii.
TFA/DCM, rt, 20 min (88%);(f) as in (e) (80%).
9. Myers, A. G.;Plowright, A. T. J. Am. Chem. Soc. 2001,
123, 5114–5115.
10. Mikami, Y.;Takahashi, K.;Yazawa, K.;Arai, T.;
Namikoshi, M.;Iwasaki, S.;Okuda, S.
1985, 260, 344–348.
J. Biol. Chem.
Figure 2. Proposed hydrate structure.

3924
E. W. Schmidt et al. / Tetrahedron Letters 45 (2004) 3921–3924
11. Palaniswamy, V. A.;Gould, S. J. J. Am. Chem. Soc. 1986,
108, 5651–5652.
12. Pospiech, A.;Bietenhader, J.;Schupp, T. Microbiology––
UK 1996, 142, 741–746.
13. Pospiech, A.;Cluzel, B.;Bietenhader, J.;Schupp, T.
Microbiology––UK 1995, 141, 1793–1803.
2.81 (1H, dd, J ¼ 13:8, 8.0 Hz, bH0); 13C NMR (D2O,
CD3OD int. std., 125 MHz) d 174.3, 147.0, 145.3, 128.3,
121.7, 116.4, 113.0, 56.1, 54.0, 35.8;Compound 6: 1H
NMR (D2O, 400 MHz) d 6.92 (1H, d, J ¼ 1:2 Hz, C2-H),
6.84 (1H, dd, J ¼ 8:0, 1.2 Hz, C6-H), 6.68 (1H, d,
J ¼ 8:0 Hz, H-5), 3.74 (1H, dd, J ¼ 7:0, 5.6 Hz, aH),
2.99 (1H, dd, J ¼ 14:8, 5.6 Hz, bH), 2.83 (1H, dd,
J ¼ 14:8, 8.0 Hz, bH0), 2.02 (3H, s, C–CH3); 13C NMR
(D2O, CD3OD int. std., 125 MHz) d 153.0, 132.0, 128.0,
126.9, 125.5, 115.4, 53.8, 35.6, 15.2. Note: carboxyl C not
observed;Compound 7: 1H NMR (D2O, 400 MHz) d 6.91
(1H, d, J ¼ 9:2 Hz, C6-H), 6.90 (1H, s, C2-H), 6.79 (1H, d,
J ¼ 9:2 Hz, C5-H), 3.64 (3H, s, O–CH3), 3.63 (1H, dd,
J ¼ 7:6, 3.2 Hz, aH), 2.94 (1H, dd, J ¼ 14:4, 3.2 Hz, bH),
2.77 (1H, dd, J ¼ 14:4, 7.6 Hz, bH0), 1.99 (3H, s, C–CH3);
13C NMR (D2O, CD3OD int. std., 125 MHz) d 157.5,
132.6, 128.9, 128.8, 128.1, 112.4, 57.3, 54.8, 37.2, 16.3.
ꢀ
14. Du, L.;S anchez, C.;Chen, M.;Edwards, D. J.;Shen, B.
Chem. Biol. 2000, 7, 623–642.
€
15. Schumacher, H.-M.;R uffer, M.;Nagakura, N.;Zenk,
M. H. Planta Med. 1983, 48, 212–220.
16. Fodor, G.;Nagubandi, S. Tetrahedron 1980, 36, 1279–
1300.
17. Jung, M. E.;Lazarova, T. I. J. Org. Chem. 1997, 62, 1553–
1555.
18. Bigot, A.;Elise, M.;Dau, T. H.;Zhu, J.
1999, 64, 6283–6296.
J. Org. Chem.
19. Hudgens, T. L.;Turnbull, K. D. Tetrahedron Lett. 1999,
40, 2719–2722.
1
Note: carboxyl C not observed;Compound 8: H NMR
20. Edgar, K. J.;Falling, S. N. J. Org. Chem. 1990, 55, 5287–
5291.
21. Marfey, P. Carlsberg Res. Commun. 1984, 49, 591–596.
22. Chen, C.;Zhu, Y.-F.;Wilcoxen, K. J. Org. Chem. 2000,
65, 2574–2576.
23. Compound 5: 1H NMR (D2O, 400 MHz) d 6.83 (1H, d,
J ¼ 8:4 Hz, C6-H), 6.64 (1H, s, C2-H), 6.63 (1H, d,
J ¼ 8:4 Hz, C5-H), 3.73 (1H, dd, J ¼ 8:0, 5.6 Hz, aH),
3.67 (3H, s, OCH3), 2.97 (1H, dd, J ¼ 13:8, 5.6 Hz, bH),
(D2O, 400 MHz) d 6.51 (1H, s, ArH), 6.50 (1H, s, ArH),
3.88 (1H, dd, J ¼ 8:0, 5.6 Hz, aH), 2.99 (1H, dd, J ¼ 14:4,
5.6 Hz, b H), 2.84 (1H, J ¼ 14:4, 8.0 Hz, bH0), 2.04 (3H, s,
1
C–CH3);Compound 9: H NMR (D2O, 400 MHz) d 6.59
(2H, s, ArH), 3.80 (1H, dd, J ¼ 8:0, 5.2 Hz, aH), 3.65 (3H,
s, OCH3), 3.03 (1H, dd, J ¼ 14:8, 5.2, bH), 2.85 (1H, dd,
J ¼ 14:8, 8.0 Hz, bH0), 2.13 (3H, s, C–CH3); 13C NMR
(D2O, CD3OD int. std., 125 MHz) d 174.9, 149.7, 145.7,
134.0, 132.7, 124.5, 115.9, 61.3, 56.9, 36.8, 15.9.