Biomimetic total synthesis and antimicrobial evaluation of anachelin H

Article abstract of DOI:10.1021/jo701402b

(Chemical Equation Presented) The first biomimetic total synthesis of the iron chelator anachelin H isolated from the cyanobacterium Anabaena cylindrica is reported. A first generation approach delivered one enantiomeric series of the polyketide fragment. Comparison of the ¹H NMR data suggested the relative configuration of this anachelin fragment. The relative and absolute configuration of anachelin H was then established by total synthesis. A second generation approach involved the enzymatic conversion of N,N-dimethyltyramine to the anachelin chromophore. It was demonstrated that the enzyme tyrosinase is activated by the product during this reaction, the anachelin chromophore can serve as a tyrosinase activator. Anachelin H was evaluated against a panel of eleven bacterial and fungal pathogens, and moderate antibiotic activity (32 μg/mL) against Moraxella catarrhalis was found.

Full text of DOI:10.1021/jo701402b

Biomimetic Total Synthesis and Antimicrobial Evaluation of
Anachelin H
Karl Gademann,*,†,‡ Yann Bethuel,‡ Hans H. Locher,§ and Christian Hubschwerlen§
Chemical Synthesis Laboratory, Swiss Federal Institute of Technology (EPFL), SB-ISIC-LSYNC, CH-1015
Lausanne, Switzerland, Laboratorium fu¨r Organische Chemie, Swiss Federal Institute of Technology
(ETH), CH-8093 Zu¨rich, Switzerland, and Actelion Pharmaceuticals, Gewerbestrasse 16,
CH-4123 Allschwil, Switzerland
karl.gademann@epfl.ch
ReceiVed July 3, 2007
The first biomimetic total synthesis of the iron chelator anachelin H isolated from the cyanobacterium
Anabaena cylindrica is reported. A first generation approach delivered one enantiomeric series of the
polyketide fragment. Comparison of the 1H NMR data suggested the relative configuration of this anachelin
fragment. The relative and absolute configuration of anachelin H was then established by total synthesis.
A second generation approach involved the enzymatic conversion of N,N-dimethyltyramine to the anachelin
chromophore. It was demonstrated that the enzyme tyrosinase is activated by the product during this
reaction, the anachelin chromophore can serve as a tyrosinase activator. Anachelin H was evaluated
against a panel of eleven bacterial and fungal pathogens, and moderate antibiotic activity (32 µg/mL)
against Moraxella catarrhalis was found.
Introduction
dramatic threat for all life forms present, and those unable to
cope became extinct. In addition, the oxidative atmosphere
resulted in the conversion of soluble Fe(II) ions to Fe(III) salts,
of which the dominant iron oxide hydrates are insoluble at
physiological pH (poor bioavailability).3 Therefore, iron acquisi-
tion became crucial for every organism, and many sophisticated
strategies were developed by evolution. It is interesting to note
that although cyanobacteria probably caused this shortage of
iron by performing oxygenic photosynthesis,4 little was known
Cyanobacteria (or blue-green algae) are considered to be
among the oldest life forms still present on earth, populating
this planet since 3.5 billion years.1 These species are thought
to have modified the atmosphere of earth by performing
oxygenic photosynthesis.2 The release of oxygen caused a
† EPFL.
‡ ETH Zu¨rich.
§ Actelion.
(1) Schopf, J. W. Science 1993, 260, 640-646.
(2) Review: (a) Alexander, M. Annu. ReV. Microbiol. 1971, 25, 361-
392. (b) Howard, D. H. Clin. Microbiol. ReV. 1999, 12, 394-404.
(3) Bergeron, R. J.; McManis, J. S.; Phanstiel, O.; Vinson, J. R. J. Org.
Chem. 1995, 60, 109-114.
10.1021/jo701402b CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/29/2007
J. Org. Chem. 2007, 72, 8361-8370
8361

Gademann et al.
concerning their mechanism of iron uptake. In particular, no
complex siderophores, that is, small molecules secreted for iron
binding, transport, and uptake, have been isolated from cyano-
bacteria until recently. In 2000, the first complex metabolites
postulated to serve as siderophores from the freshwater bacte-
rium Anabaena cylindrica were isolated: Budzikiewicz, Walsby,
and co-workers isolated mixtures of anachelin H (1) and
anachelin-1 (2)5 and determined the constitution of the former.
Later, Murakami and co-workers reported the isolation and
constitution of anachelin-1 and anachelin-2 (3) together with
two related esters.6 In these isolation reports, the absolute and
relative configuration of four stereogenic centers could not be
determined.7 As this question could be addressed by total
synthesis, we chose to initiate a research program on the
synthesis of anachelin.
FIGURE 1. Galantin I (4a, 4b), galantinic acid (5), and its originally
proposed structure (6).
of galantin) were later shown to be incorrect and revised by
synthesis10 to 5. As galantinic acid features the same constitution
as the polyketide fragment of anachelin, we compared the
reported 1H NMR chemical shifts of both fragments. However,
the spectral data did not provide evidence that the relative
configuration of the polyketide fragment in anachelin was (3S*,
5S*, 6S*) as in galantinic acid (5). Therefore because of the
unknown relative and absolute configuration of this fragment
in anachelin, a stereodivergent synthetic strategy allowing for
the access to all possible eight stereoisomers of the polyketide
fragment was chosen.
The â-hydroxy-γ-amino ester 9 with the relative configuration
3,4-unlike (i.e., 3R, 4S) was targeted first, as this relative
configuration occurs in several natural products (so-called
statins) such as tandamarin, didemnin,11-13 and others. There-
fore, we followed well-established protocols used in the context
of their synthesis for the preparation of the new compound 9
(Scheme 1). A biomimetic C2-homologation according to
Masamune14 resulted in the clean formation of â-keto ester 8
on up to 40 g scale. Reduction of this compound by NaBH4
resulted in the formation of a mixture of separable diastereoi-
somers 9 and 10, of which both compounds are of use in our
stereodivergent synthesis.15 The major unlike-isomer 9 was
O-silylated with more reactive TBSOTf (standard conditions
using TBSCl gave poorer yields) and saponified. Another C2-
homologation then resulted in the aminohydroxyketoester 11
in good yield.
Herein, we describe the biomimetic total synthesis of anach-
elin H (1) in full detail.8 In addition, we present an optimized,
second generation route, which utilizes an enzyme-mediated
oxidative heterofunctionalization/cyclization cascade. The bio-
logical activity of anachelin H as antimicrobial agent was also
evaluated and moderate antibiotic activity against Moraxella
catarrhalis was determined.
Results and Discussion
Retrosynthetic analysis of anachelin is straightforward. As
anachelin is composed of different modules of polyketide,
peptide, and alkaloid biosynthetic origin, the preparation of each
fragment precedes their coupling. Moreover, as the four ste-
reogenic centers on the polyketide and alkaloid fragment were
not described in the original isolation reports, a stereodivergent
synthetic strategy allowing for the (possible) preparation of all
sixteen diastereoisomers of anachelin was mandatory.
The polyketide fragment of anachelin resembles galantinic
acid (5), constituent of the peptide antibiotic galantin I (4)
isolated from the culture broth of Bacillus pulVifaciens (Figure
1).9 The originally reported constitution of galantinic acid 6 (and
Another stereodivergent step was carried out next in produc-
ing both isomers 13 and 15. The 3,5-syn-reduction of the
hydroxyketoester 11 was addressed first (Scheme 2). After a
literature survey, we realized that the manifold of different
functional groups present in 11 could pose challenges to the
(10) (a) Sakai, N.; Ohfune, Y. Tetrahedron Lett. 1990, 29, 4151-4154.
(b) Sakai, N.; Ohfune, Y. J. Am. Chem. Soc. 1992, 114, 998-1010.
(11) Rinehart, K. L., Jr.; Gloer, J. B.; Cook, J. C.; Miszak, S. A.; Scahill,
T. A. J. Am. Chem. Soc. 1981, 103, 1857-1859.
(4) Hoehler, T. M.; Bebout, B. M.; Marais, D. J. Nature 2001, 412, 324-
327.
(5) (a) Beiderbeck, H.; Taraz, K.; Budzikiewicz, H.; Walsby, A. E. Z.
Naturforsch. C 2000, 55, 681-687. (b) Beiderbeck, H. Dissertation,
Universita¨t Ko¨ln, Germany, 2000.
(6) Itou, Y.; Okada, S.; Murakami, M. Tetrahedron 2001, 57, 9093-
9099.
(12) Rinehart, K. L., Jr.; Kishore, V.; Bible, K. C.; Sakai, R.; Sullins,
D. W.; Li, K.-M. J. Nat. Prod. 1988, 51, 1-21.
(13) Rinehart, K. L., Jr.; Kishore, V.; Nagarajan, S.; Lake, R. J.; Gloer,
J. B.; Bozich, F. A.; Li, K.-M.; Maleczka, R. E., Jr.; Todsen, W. L.; Munro,
M. H. G.; Sullins, D. W.; Sakai, R. J. Am. Chem. Soc. 1987, 109, 6846-
6848.
(14) Brooks, D. W.; Lu, L. D.-L.; Masamune, S. Angew. Chem., Int.
Ed. 1979, 18, 72-74. Mansour, T. S.; Evans, C. A. Synth. Commun. 1990,
20, 773-781.
(15) The relative configuration of 9 and 10 was assigned after deriva-
tization to the N,O-acetal using the method of Joullie´: Harris, B. D.; Joullie´,
M. M. Tetrahedron 1988, 110, 3560-3578.
(7) Three years after the initial report, the configuration was assigned
by degradation experiments: Ito, Y.; Ishida, K.; Okada, S.; Murakami, M.
Tetrahedron 2004, 60, 9075-9080.
(8) For preliminary communications, see (a) Gademann, K.; Bethuel,
Y. Angew. Chem., Int. Ed. 2004, 43, 3327-3329. (b) Gademann, K.;
Bethuel, Y. Org. Lett. 2004, 6, 4707-4710.
(9) Shoji, J.; Sakajaki, R.; Wakisaka, Y.; Koizumi, K.; Mayama, M.;
Matsurra, S. J. J. Antibiot. 1975, 28, 122-125.
8362 J. Org. Chem., Vol. 72, No. 22, 2007

Total Synthesis of Anachelin H
SCHEME 1. Preparation of 5,6-unlike-Hydroxyketone 11
SCHEME 3. Preparation of Both Isomers with the Relative
Configuration 5,6-like-21 and -22
SCHEME 2. Stereodivergent Synthesis of
Trihydroxyaminoesters 13 and 15
particular the carbamate protected amine, were found to be
incompatible with the standard reaction conditions. In fact, the
addition of pivalic acid in the course of the elaboration of the
borane reagent was found to be crucial for a high-yielding and
stereoselective reaction. Therefore, Et3B was allowed to react
with pivalic acid and MeOH before addition of the substrate,18
which was subsequently, after chelation, reduced by NaBH4.
TBS protection of the secondary alcohols resulted in the
preparation of fully protected 12, which was isolated in 75%
yield over two steps. Liberation of the 6,7-aminoalcohol by
hydrogenolysis followed by acylation of the amino group by a
protected salicylate19 (EDC, HOBt) and reprotection of the
primary alcohol (TBSOTf, 2,6-lutidine) gave the building block
13 in very good yield. The 3,5-anti reduction was carried out
using the method of Evans and co-workers20 utilizing triac-
etoxyborohydride, which resulted, after O-protection, in the
(17) Chen, K.-M.; Hardtmann, G. E.; Prasad, G. E.; Repic, O.; Shapiro,
M. J. Tetrahedron Lett. 1987, 28, 155-158. See also: Sletzinger, M.;
Verhoeven, T. R.; Volante, R. P.; McNamara, J. M.; Corley, E. G.; Liu, T.
M. H. Tetrahedron Lett. 1985, 26, 2951-2954.
(18) Lee, H. T.; Woo, P. W. K. J. Labelled Compd Radiopharm. 1999,
42, 129-133.
usual protocols of Narasaka et al.16 or Prasad et al.,17 as
substrates containing this dense array of functional groups, in
(16) Narasaka, K.; Pai, F.-C. J. Org. Chem. 1978, 43, 2923-2925.
J. Org. Chem, Vol. 72, No. 22, 2007 8363

Gademann et al.
FIGURE 2. Proton chemical shifts in natural anachelin H (1, 300 MHz, D2O, rt) and in the polyketide fragments (13, 15, 21, and 22, 300 MHz,
CDCl3, rt).
SCHEME 4. Preparation of the
N,N-Dimethylaminomethyldopamine 26
intermediate dioxanone by heating in the presence of MeOH in
toluene thus affording directly the Me ester 18.22 This amino-
hydroxyketoester was then reduced following chelate-controlled
via either the modified Sandoz conditions (f 19) or Evans’
conditions (f 20). Elaboration of both intermediates (Z depro-
tection and coupling to the protected salicylic acid) yielded both
diastereoisomers 21 and 22. By all these synthetic operations,
all possible diastereoisomers 13, 15, 21, and 22 of one
enantiomeric series (i.e., from L-serine) have been prepared. By
following the same chemical transformations starting from D-Ser,
ent-21 and ent-22 were also prepared.
Comparison of the chemical shifts of these protected frag-
ments in the 1H NMR spectrum in CDCl3 to those of the natural
product 1 in D2O revealed an interesting pattern (Figure 2). Of
all isomers, the shape of the peaks and in particular the chemical
shifts of the relevant protons of 21 matched best to those of 1.
In the spectrum of isomer 21, the signals of C(3)-H and C(5)-H
both appear at around 4.2 ppm, whereas C(6)-H resonates at
4.27 ppm. This pattern (superposition of C(3)-H and C(5)-H,
separation of C(6)-H) appears slightly shifted also in the natural
product. Other isomers such as 15 display the C(3)-H and
C(5)-H resonances separated in their chemical shifts. In addition,
the anachelin H isomer derived from 15 has been previously
prepared by us8a and has been shown to display clear differences
in the 1H NMR spectrum when compared to the natural product.
Therefore, we decided to advance the fragment having relative
configuration shown in 21, namely, (3R*, 5S*, 6S*).
intermediate 14. As for its diastereoisomer, elaboration of this
fragment was carried out by hydrogenation, amide coupling and
reprotection to give the target fragment 15.
The building blocks carrying the 5,6-like configuration were
accessed by performing the vinylogous aldol reaction catalyzed
by Eu(III) developed by Campagne and co-workers (Scheme
3).21 In contrast to their approach, we chose to transform the
Preparation of the Peptide Alkaloid Fragment. The tet-
rahydroquinolinium fragment of anachelin, which serves as a
fluorescent chromophore, is unprecedented in other natural
(19) Protection of the salicylic acid was carried out following a procedure
by Farkas and coworkers: Farkas, L.; Vermer, B.; No`gra`di, M. Tetrahedron
1967, 23, 741-744.
(20) Evans, D. A.; Chapman, K. T.; Carreira, E. M. J. Am. Chem. Soc.
1988, 110, 3560-3578.
(22) Sato, M.; Sunami, S.; Sugita, Y.; Kaneko, C. Chem. Pharm. Bull.
1994, 42, 839-845. See also Snider, B. B.; Song, F. Org. Lett. 2001, 3,
1817-1820.
(21) Moreau, X.; Campagne, J.-M. Tetrahedron Lett. 2001, 42, 4467-
4469.
8364 J. Org. Chem., Vol. 72, No. 22, 2007

Total Synthesis of Anachelin H
SCHEME 5. Preparation of the Anachelin Chromophore
Tripeptide Fragment 29
SCHEME 7. Completion of the Synthesis
to this strategy is the cyclization of protected diamine 26, of
which we developed a synthesis earlier. However, this synthesis
suffered from poor yields (10% over five steps from L-DOPA),23
which prompted us to develop the following, higher-yielding
route.
Boc-L-DOPA-dimethylamide (23) was OBn protected, the
Boc group removed and the amide reduced (Scheme 4). The
removal of the Boc group proved to be critical to prevent side
reactions such as carbamate reduction to the N-CH3 taking
place. Introduction of the Boc group and hydrogenolytic
cleavage of the Bn ethers delivered target diamine 26 in 40%
overall yield over five steps.
The cyclization to the anachelin chromophore was investi-
gated next. Departing from the diamine 26, two routes were
examined. The first route (Scheme 5) consisted in the direct
oxidative aza annulation of 26 using dianisyltellurium oxide,24
resulting in the formation of the tetrahydroquinolinium ring.
Protection of the catecholic OH groups allowed for the purifica-
tion of 27 on silical gel. This heterocyclic structure could then
be coupled, after removal of the Boc group, to a protected serine
derivative. The intermediate 28 was thus prepared from diamine
26 in 29% yield over four steps.
SCHEME 6. Alternative Route to the Tripeptide Anachelin
Chromophore Fragment 28
An improved route (Scheme 6) was found, by first attaching
the protected Ser acid to the diamine resulting in 30 (58% yield).
Analogous cyclization of compound 30 furnished the intermedi-
ate 28 in an improved yield (39% over four steps). Elaboration
of this compound by attaching the dipeptide Boc-L-Thr(OBn)-
D-Ser(OBn)-OH to 28 (49% yield over two steps) followed by
Boc-deprotection gave target fragment 29 (Scheme 5).
Merging of the peptide alkaloid part 29 with the polyketide
fragment was carried out by the mixed anhydride method
(Scheme 7), which proved superior to alternative coupling agents
products. We postulated a biogenetic hypothesis for this
fragment featuring an oxidative aza annulation reaction.23 Key
(24) Ley, S. V.; Meerholz, C. A.; Barton, D. H. R. Tetrahedron 1981,
37, 213-233. For a related application, see: Clews, J.; Cooksey, C. J.;
Garratt, P. J.; Land, E. J.; Ramsden, C. A.; Riley, P. A. J. Chem. Soc.,
Pekin Trans. 2000, 1, 4306-4315.
(23) Gademann, K. Chem. Bio. Chem. 2005, 6, 913-919.
J. Org. Chem, Vol. 72, No. 22, 2007 8365

Gademann et al.
SCHEME 8. Second-Generation Route to the Target
Diketide 34
SCHEME 10. Preparation of Anachelin Chromophore 37
Starting from Boc-L-Tyr-OH (35)
SCHEME 11. The Tyrosinase-Mediated Oxidative Cascade
for the Preparation of 37
SCHEME 9. Biogenetic Hypothesis for the Generation of
the Anachelin Chromophore
such as BOP-Cl, HBTU or EDC with respect to yield and ease
of purification. No epimerization was observed in the 1H NMR
spectrum after the coupling reaction. Thus, the acid resulting
from saponification of the ester 21 was activated with iBuOCOCl
in the presence of tertiary amine base and the resulting anhydride
coupled to fragment 29 resulting in protected anachelin H 31
(54% yield over two steps). Deprotection of the Bn groups was
carried out by hydrogenolysis in acidic medium catalyzed by
Pd/C. The removal of the Si-protecting groups by standard
F-based reagents proved cumbersome, as slow and only partial
deprotection was observed. We were successful by using acidic
MeOH (0.5 or 1% aq HCl in MeOH) in deprotecting the TBS
groups. These reactions resulted in synthetic anachelin H (1),
which was found to be identical to the natural product as judged
(17 steps longest linear sequence) and 4.9% overall yield. To
streamline the synthesis, we optimized the routes for both
polyketide and alkaloid fragments.
The key fragment for the polyketide part is the hydroxy keto
ester 18, which was prepared via the vinylogous Mukaiyama
aldol (Scheme 3) reaction in seven steps overall from com-
mercially available Z-Ser(OH)-OH.26 We found that the corre-
sponding t-Bu ester 32 can be prepared in only one step from
10 employing a direct Claisen condensation according to
Heathcock27 using an excess of tert-butyl lithium enolate (6
equiv) in 75% (Scheme 8).28 This reduces the number of overall
steps for this key fragment from seven to three from com-
mercially available starting materials, in particular as 10 can
be obtained as major isomer from the ketone 8 by using different
reagents (see ref 15 for a detailed discussion). Elaboration of
this fragment by similar methods as outlined above resulted in
the preparation of 34. Noteworthy is the fast, clean, and high-
yielding cleavage of the t-Bu ester by TMSBr to 34.
1
by the H NMR and MS data.25 The NMR spectra of this
cationic peptide alkaloid 1 are dependent on concentration and
other parameters. Therefore the final proof of identity was
delivered by mixing samples of synthetic and natural anachelin
H and observing their identity in the NMR spectrum of the
mixture and, independently, by HPLC co-injection of natural
and synthetic samples. Therefore, the absolute and relative
configuration of anachelin H (1) was confirmed to be as shown
in Scheme 7 by total synthesis.
Optimizing the synthetic route for the alkaloid fragment was
addressed next. We have postulated earlier23 that the biosynthesis
(26) The aldehyde precursor according to: Kang, M.; Park, J.; Konfradi,
A. W.; Pedersen, S. F. J. Org. Chem. 1996, 61, 5528-5531.
(27) Hubbs, J. L.; Heathcock, C. H. J. Am. Chem. Soc. 2003, 125,
12836-12843.
(28) We employed this reaction in the stereoselective total synthesis of
galantinic acid, see: Bethuel, Y.; Gademann, K. Synlett 2006, 1580-1582.
Second-Generation Synthesis. The preparation of anachelin
H (1) as presented above delivered synthetic material in 33 steps
(25) The HPLC traces and 1H NMR and MS spectra of the synthetic
anachelin H can be found in the Supporting Information.
8366 J. Org. Chem., Vol. 72, No. 22, 2007

Total Synthesis of Anachelin H
SCHEME 12a
a The anachelin chromophore 37 shortens the lag phase and induces greater initial velocity in the conversion of L-Tyr (38) to cyclo-DOPA (39). Thus, the
anachelin chromophore 37 serves as a tyrosinase actiVator.
of the anachelin chromophore originates from a dihydroxyphe-
nylalanine derivative, which is cyclized after reductive amina-
tion. One of us (K.G.) demonstrated that this hypothesis can be
chemically realized in the laboratory by the enzyme tyrosinase,
and mechanistic studies have provided insight into these
cyclizations.23 Herein, we would like to expand this biogenetic
hypothesis and corroborate it by enzyme-mediated biomimetic
synthesis of the anachelin chromophore.
It can be postulated that a mono-hydroxylated diamine derived
from tyrosine such as A can serve as precursor for the enzyme-
mediated aza annulation (Scheme 9). Oxidation (either first by
hydroxylation and then quinone oxidation or direct quinone
oxidation) would give quinone B which allows for an intramo-
lecular 1,4-addition providing either spiro-five-membered com-
pound or directly the six-ring annulated product C. Tautomer-
ization would then furnish the anachelin chromophore D.
Tyrosinase (E.C. 1.14.18.1) is well-known to serve as dual
function of both polyphenol oxidase and catechol oxidase,
therefore, we set out to test this hypothesis on a preparative
scale.
The precursor was prepared from Boc-tyrosine in four steps
as shown in Scheme 10. Thus, activation and amidation gave
Boc-Tyr(OH)-NMe2, which was reduced via a three-step
sequence to the diamine 36 (Scheme 10). The substrate 36 was
subjected to oxidation by molecular oxygen in the presence of
commercial mushroom tyrosinase (phosphate buffer, pH ) 6.8
at 24 °C). The reaction was monitored by taking aliquots, which
were characterized both by NMR and UV spectroscopy, and
therefore the product formation of 37 was measured. Initial
experiments were encouraging: Up to 15% conversion was
observed using 100 units29 of tyrosinase (Scheme 11, entry 1).30
However, when the initial enzyme concentration was doubled,
only a small increase in conversion to 18% could be observed
(entry 2).31 In general, varying enzyme concentration had only
a very small effect on conversion, as the use of 50 units resulted
in roughly equal conversion after 2 h. The breakthrough was
achieved when the initial amount of enzyme was split, and the
second portion added after the reaction stopped (entry 4).
Surprisingly, conversion around 50-60% was measured. Fi-
nally, splitting the enzyme in three portions added after 0, 2,
and 4 h led to full conversion noted by the disappearance of
starting material (entry 5). An explanation for this unusual
kinetic behavior could be that the product 37 of the transforma-
tion in Scheme 11 is actiVating the tyrosinase.
To evaluate whether the anachelin chromophore is indeed a
tyrosinase activator, we examined the action of tyrosinase on
its natural substrate L-Tyr resulting in cyclo-DOPA (Scheme
12).32 The control experiment (black line in Scheme 12) displays
the usual kinetics of this transformation featuring a lag phase,
where the inactive met-tyrosinase is converted into the active
(29) Unit definition: One unit will cause an increase in A280nm of 0.001
per minute at pH 6.5 at 25° in a 3 mL reaction mix containing L-tyrosine.
Enzyme activity as expressed in units was measured according to the Sigma
Quality Control Test Procedure, SPTYR001.001, T7755, Rev. 02/22/94.
Sigma Biochemicals, 3050 Spruce Street, Saint Louis, Missouri 53103.
(30) Tyrosinase is slowly decomposing in the course of the experiments
at 24 °C. Generally, there was no enzyme activity observed after 2 h
anymore.
(31) This experimental evidence can be correlated to the literature, where
it was shown that tyrosinase in its resting native form (met-Tyrosinase) is
catalytically inactive. See ref 32 for reviews.
(32) Reviews: (a) Duckworth, H. W.; Coleman, J. E. J. Biol. Chem.
1970, 245, 1613-1625. (b) Sanchez-Ferrer, A.; Rodriguez-Lopez, J. N.;
Garcia-Canovas, F.; Garcia-Carmona, F. Biochim. Biophys. Acta 1995, 1247,
1-11. (c) Land, E. J.; Ramsden, C. A.; Riley, P. A. Acc. Chem. Res. 2003,
36, 300-308.
J. Org. Chem, Vol. 72, No. 22, 2007 8367

Gademann et al.
form.32 The addition of a 0.64 mM solution of the anachelin
chromophore derivative 37 (orange line in Scheme 12) led to
the observation of both a shortened lag phase as well as higher
initial rates. Therefore, one can conclude that the anachelin
chromophore 37 displays biological activity as a tyrosinase
activator (AC50 ≈ 1 mM).
further support to the biogenetic proposal of anachelin. The
anachelin chromophore was shown to actiVate tyrosinase
thus facilitating its own formation via this enzyme-mediated
process. Finally, biological evaluation of anachelin H (1)
and its chromophore 37 demonstrated moderate antibacterial
activity against Moraxella catarrhalis, a respiratory tract
pathogen.
These results support the biosynthetic proposal outlined in
Scheme 9 that a tyrosine precursor could be involved in the
biogenesis of anachelin. The feature of autoactivation is
mechanistically interesting, where the product of this enzymatic
transformation is activating the enzyme of its own formation.
In addition, the transformation from 36 to 37 demonstrates the
power of enzymatic reactions in total synthesis further featuring
a formal aminohydroxylation/cyclization cascade.
Experimental Section
General. Reagents, abbreviations, instruments, and standard
methods are reported in the Supporting Information.
(6S)-6-Benzyloxycarbonylamino-7-(tert-butyl-dimethyl-silany-
loxy)-(5S)-5-hydroxy-3-oxo-heptanoic Acid Methyl Ester (18).
To a solution of [1-(tert-butyl-dimethyl-silanyloxymethyl)-3-(2,2-
dimethyl-6-oxo-6H-[1,3]dioxin-4-yl)-((3S)-2-hydroxy-propyl]-(2S)-
carbamic acid benzyl ester (175 mg, 0.36 mmol) in toluene (3 mL)
was added MeOH (58 µL, 1.44 mmol, 4 equiv), and the reaction
mixture was refluxed for 4 h. Then it was cooled to rt, and the
solvents were removed under reduced pressure. FC (EtOAc/hexane
2.5:7.5) gave the title compound 18 (141 mg, 0.31 mmol, 85%). Rf
) 0.5 (EtOAc/hexane 4:6). [R]D ) +0.65 (c 2.00, CHCl3, T )
Antimicrobial Evaluation. Many iron chelators display
antimicrobial activity by sequestering free Fe ions and thus
preventing bacterial iron uptake. Such so-called sideromycins
are very effective antibiotics with the drawback that resistance
quickly develops. We therefore wanted to profile the antimi-
crobial activity of anachelin H (1). The antibacterial activities
of anachelin H (1) and its chromophore (37) were tested against
a panel of well-characterized bacterial strains (Staphylococcus
aureus ATCC 29213, Staphylococcus aureus A-798, Entero-
coccus faecalis ATCC 29212, Enterococcus faecium A-949,
Streptococcus pneumoniae ATCC 49619, Streptococcus pneu-
moniae A-70, Haemophilus influenzae A-921, Moraxella ca-
tarrhalis A-894, Escherichia coli ATCC 25922, Pseudomonas
aeruginosa ATCC 27853, and Saccharomyces cereVisiae A-136).
Anachelin H (1) and the anachelin chromophore (37) displayed
moderate activity (MIC, 32 µg/mL) against Moraxella catarrha-
lis, a respiratory tract pathogen, but were not active against the
other strains at the highest concentration tested (64 µg/mL). To
make the assay more sensitive to iron chelators,33,34 we also
used iron limited medium (M9 medium with and without iron
supplement) for determination of MICs for E. coli and P.
aeruginosa. However, also under these conditions no antibacte-
rial activity was observed for these strains. Whether anachelin
H (1) and its chromophore fragment 37 act upon Moraxella
via a mechanism of iron deprivation is currently not known.
Another possible explanation concerning the mode of action
of 1 and 37 would reside in the presence of a quaternary
1
27.2 °C). H NMR (CDCl3, 300 MHz): δ 0.051 (s, 3H), 0.06 (s,
3H), 0.87 (s, 9H), 2.65 (dd, 1H, J1 ) 4.35, J2 ) 17.44), 2.79 (dd,
1H, J1 ) 8.4, J2 ) 17.43), 3.34 (s, 1H), 3.47 (s, 2H), 3.62-3.63
(m, 1H), 3.72 (s, 3H), 3.8 (d, 2H, J ) 4.35), 4.44-4.49 (m, 1H),
5.1 (s, 2H), 5.32 (br. d, J ) 9.34), 7.31-7.37 (m, 5H). 13C NMR
(CDCl3, 75 MHz): -5.4, 18.4, 26.0, 47.0, 49.8, 52.6, 54.5, 65.0,
67.1, 68.1, 128.4, 128.4, 128.8, 136.6, 156.7, 167.5, 202.5. IR:
3647-3170m, 1744m, 1716s. MS 476.2 (100, [M+Na]+), 444.2
(79, [M-MeOH+Na]+. HRMS calcd for C22H35NO7SiNa (M+Na)+,
476.2075; found, 476.2069.
(6S)-6-Benzyloxycarbonylamino-(3R,5S)-3,5,7-tris-(tert-butyl-
dimethyl-silanyloxy)-heptanoic Acid Methyl Ester (19). A solu-
tion of (6S)-6-benzyloxycarbonylamino-7-(tert-butyl-dimethyl-
silanyloxy)-(3R,5S)-4,5-hydroxy-3-oxo-heptanoic acid methyl
ester (19.0 mg, 41.7 µmol) in dry CH2Cl2 was cooled to -20 °C,
and 2,6-lutidine (19.4 µL, 0.17 mmol, 4 equiv) was added drop-
wise followed by the addition of TBSOTf (29 µL, 0.125 mmol, 3
equiv). The reaction mixture was stirred at -20 °C for 1 h 15 min.
Then CH2Cl2 was added and the organic phase was washed 3×
with citric acid (10%) and 2× with H2O, dried over Na2SO4, filtered,
and evaporated under reduced pressure. FC (EtOAc/hexane 1:9)
gave 19 (21.0 mg, 30.7 µmol, 75%). Rf ) 0.68 (EtOAc/hexane
2:8). [R]D ) -1.3 (c 1.05, CHCl3, T ) 25.5 °C). 1H NMR (CDCl3,
300 MHz): δ 0.056 (s, 6H), 0.062 (s, 3H), 0.07, 0.083(s, 6H),
0.098 (s, 3H), 0.86(s, 9H), 0.88 (s, 9H), 0.89 (s, 9H), 1.63-1.84
(m, 2H), 2.43 (d, 2H, J ) 6.22), 3.52 (dd, 1H, J1 ) 7.79,
J2 ) 9.03), 3.62 (s, 3H), 3.66 (d, 1H, J ) 5), 3.69 (s, 1H), 3.73
(m, 1H), 4.99 (br.d, 1H, J ) 8.7), 5.1 (s, 2H), 7.3-7.39 (m, 5H).
13C NMR (CDCl3, 75 MHz): -5.1, -4.5, -4.0, 18.0, 18.2,
18.6, 26.0, 26.1, 26.2, 42.1, 43.1, 51.6, 55.1, 62.0, 66.7, 66.9, 67.4,
128.3, 128.4, 128.7, 156.4, 171.6. IR: 1727m. MS 706.3 (100).
HRMS calcd for C34H65NO7Si3Na (M+Na)+, 706.3961; found,
706.3960.
(6S)-6-(2-Benzyloxy-benzoylamino)-(3R,5S)-3,5,7-tris-(tert-bu-
tyl-dimethyl-silanyloxy)-heptanoic Acid Methyl Ester (21). To
a solution of (6S)-6-benzyloxycarbonylamino-(3R,5S)-3,5,7-tris-
(tert-butyl-dimethyl-silanyloxy)-heptanoic acid methyl ester (19)
(70 mg, 102 µmol) in MeOH (1 mL) previously 3× flushed with
argon was added Pd/C. Then the flask was flushed with H2, and
the reaction mixture was stirred at rt for 4 h. Pd/C was filtered
over celite, and the solvent was removed under reduced pressure.
This residue was added to a solution of O-Bn-salicylic acid (27.9
mg, 122 µmol, 1.2 equiv) dissolved in dry CHCl3 (1.5 mL) with
EDC.HCl (23.4 mg, 122 µmol, 1.2 equiv), HOBt (18.9 mg, 122
µmol, 1.2 equiv), and NMM (34 µL, 306 µmol, 3 equiv). The
reaction mixture was stirred at rt for 20 h. The solvent was then
ammonium group,
activity.
a fragment known for antibacterial
Conclusion
We have described the total synthesis of the cyanobacterial
polyketide alkaloid anachelin H 1. This synthesis delivered
the structural proof for anachelin H by mixing samples of
synthetic and natural anachelin H and detecting identity by NMR
and HPLC. The first generation synthesis was characterized
by a stereodivergent approach resulting in the identification of
the relative configuration. The second generation approach
is based on a biomimetic oxidative heterofunctionalization/
cyclization cascade that shortened the synthesis and added
(33) For articles on the role of iron in infectious diseases see: (a)
Weinberg, E. D.; Weinberg, G. A. Curr. Opin. Infect. Diseases 1995, 8,
164-169. (b) Weinberg, E. D. Eur. J. Cancer PreV. 1996, 5, 19-36. (c)
Weinberg, E. D. Emerging Infect. Dis. 1999, 5, 346-352.
(34) For articles on the effect of iron deprivation see: (a) Archibald, F.
S.; DeVoe, I. W. J. Bacteriol. 1978, 136, 35-48. (b) Al-Younes, H. M.;
Rudel, T.; Brinkmann, V.; Szczepek, A. J.; Meyer, T. F. Cell. Microbiol.
2001, 3, 427-437.
8368 J. Org. Chem., Vol. 72, No. 22, 2007

Total Synthesis of Anachelin H
1
removed under reduced pressure, and the residue was taken up with
EtOAc. It was washed 3× with citric acid (10%) and 3× with
saturated NaHCO3. The organic phase was dried over MgSO4,
filtered, and evaporated under reduced pressure. FC (EtOAc/hexane
1:9) gave 21 (45 mg, 59 µmol, 58%). Rf ) 0.43 (EtOAc/hexane
2:8). [R]D ) -4.2 (c 2.235, CHCl3, T ) 26.5 °C). 1H NMR (CDCl3,
300 MHz): δ 0.061 (s, 6H), 0.071 (s, 3H), 0.091 (s, 3H), 0.094 (s,
3H), 0.105 (s, 3H), 0.84 (s, 9H), 0.86 (s, 9H), 0.89 (s, 9H), 1.67-
1.78 (m, 2H), 2.35-2.48 (m, 2H), 3.49 (s, 3H), 3.6 (dd, 1H, J1 )
8.4, J2 ) 9.33), 3.71 (dd, 1H, J1 ) 5.6, J2 ) 9.6), 4.1-4.21 (m,
2H), 4.23-4.28 (m, 1H), 5.29 (s, 2H), 6.91 (dd, 1H, J1 ) 0.6, J2
) 8.4), 7.0-7.1 (m, 1H), 7.28-7.37 (m, 6H), 8.0 (d, 1H, J ) 8.4),
8.16 (dd, 1H, J1 ) 1.87, J2 ) 7.8). 13C NMR (CDCl3, 75 MHz):
-5.1, -4.5, -4.2, -4.0, 17.9, 18.1, 18.5, 25.9, 26.0, 26.1, 42.1,
43.2, 51.4, 53.8, 61.2, 66.5, 67.1, 70.7, 113.2, 121.3, 122.3, 126.7,
128.1, 128.7, 132.2, 132.3, 135.9, 156.3, 164.9, 171.3. IR: 2953m,
2928m, 2856m, 1740m, 1654m. MS 760 (100, [M+H]+). HRMS
calcd for C40H69NO7Si3Na (M+Na)+, 782.4274; found, 782.4285.
(S,S)-{2-Benzyloxy-1-[1-(3,4-dihydroxy-benzyl)-2-dimethy-
lamino-ethylcarbamoy]-ethyl}-carbamic Acid tert-Butyl Ester
(30). The Boc-diamine (26) (108 mg, 0.348 mmol) was dissolved
in CH2Cl2 (1 mL) and cooled to 0 °C under an Ar atmosphere.
Trifluoroacetic acid (1 mL) was added dropwise, and the resulting
solution was stirred for 1 h at 0 °C and for 1 h at rt. The solvent
was removed under reduced pressure. The residue was twice
dissolved in toluene, and the solvent again was removed under
reduced pressure. The residue was dried under high vacuum and
used without further purification.
Colorless oil. Rf ) 0.11 (CH2Cl2/MeOH 10:1). H NMR (CDCl3,
300 MHz): 1.41 (s, 9H), 2.94 (dd, J1 ) 16.5, J2 ) 5.3, 1H), 3.24
(br. dd, J1 ) 16.5, J2 ) 10.6, 1H), 3.61 (s, 3H), 3.67 (s, 3H), 3.69-
3.83 (m, 2H), 3.96-4.01 (m, 1H), 4.40-4.53 (m, 1H), 4.53 (d, J
) 3.7, 2H), 5.11 (s, 2H), 5.37 (d, J ) 3.4, 2H), 5.58 (d, J ) 7.5,
1H), 6.58 (s, 1H), 7.22-7.35 (m, 16H), 7.37 (d, J ) 4, 1H), 7.39
(br. s, 1H), 8.30 (br. s, 1H). MS 666.3 (43, M+), 610.3 (100,
[M-tBu+H]+), 566.3 (28, [M-Boc]+). HRMS calcd for C40H48N3O6
(M+), 666.3538; found, 666.3531.
(3R, 5S, 6S)-6-(Benzyloxy-benzoylamino)-3,5,7-tris-(tert-butyl-
dimethyl-silanoxy)-heptanoic acid-L-Thr(OBn)-D-Ser(OBn)-L-
Ser(OBn)-(3S)-THQ(OBn) (31). Boc-L-Thr(OBn)-D-Ser(OBn)-L-
Ser(OBn)-(3S)-THQ(OBn) (36 mg, 33.6 µmol) was dissolved
in dioxane (0.6 mL) and HCl in dioxane (4 M, 100 µL, 400 µmol)
was added at 0 °C. The reaction mixture was stirred 1 h at
this temperature and 3 h at rt. The solvent was then removed under
reduced pressure, and the product 29 was dried under high vac-
uum 2 h. The resulting residue was used without further purifi-
cation.
To a solution of (3R,5S,6S)-6-(2-benzyloxy-benzoylamino)-3,5,7-
tris-(tert-butyl-dimethyl-silanyloxy)-heptanoic acid (21) (30 mg,
40.3 µmol, 1.2 equiv) in DMF (0.5 mL) at -18 °C was added -
i
BuOCOCl (5.3 µL, 40.3 µmol, 1.2 equiv) and NMM (4.4 µL, 40.3
µmol, 2 equiv). The reaction mixture was stirred at this temperature
for 20 min. Then L-Thr(OBn)-D-Ser(OBn)-L-Ser(OBn)-(3S)-THQ-
(OBn) (33.8 mg, 33.6 µmol, 1 equiv) dissolved in DMF (0.5 mL)
and NMM (4.4 µL, 40.3 µmol, 2 equiv) was added dropwise at
-18 °C, and the reaction mixture was allowed to warm over 4 h.
The solvent was evaporated under reduced pressure, and the residue
taken up in EtOAc. The organic layer was washed 3× with H2O,
and the resulting water phase extracted 3× with EtOAc. The
combined organic phases were dried over Na2SO4, filtered, and
evaporated under reduced pressure. FC (CH2Cl2/MeOH 10:1) gave
the title compound 31 (31 mg, 18.3 mmol, 54%). Colorless oil. Rf
) 0.18 (CH2Cl2/MeOH 10:1). [R]D ) +7.9 (c 0.55, CHCl3, T )
26 °C). 1H NMR (CDCl3, 300 MHz): 0.013 (s, 3H), 0.032 (s, 6H),
0.061 (s, 6H), 0.079 (s, 3H), 0.80 (s, 9H), 0.83 (s, 9H), 0.86 (s,
9H), 1.15 (d, 3H, J ) 6.54), 1.74-1.9 (m, 8H), 2.24 (dd, 1H, J1 )
6.54, J2 ) 14.6), 2.35 (dd, 1H, J1 ) 3.1, J2 ) 14.6), 2.86 (dd, 1H,
J1 ) 5.0, J2 ) 16.2), 3.18 (dd, 1H, J1 ) 11.5, J2 ) 16.2), 3.41 (s,
3H), 3.44 (s, 3H), 3.56-3.74 (m, 6H), 3.86-3.94 (m, 2H), 4.0-
4.08 (m, 1H), 4.1-4.18 (m, 1H), 4.18-4.28 (m, 2H), 4.41 (s, 2H),
4.42 (s, 2H), 4.48 (s, 2H), 4.52-4.62 (m, 2H), 4.64-4.74 (m, 1H),
5.13 (d, 2H, J ) 2.5), 5.21 (d, 2H, J ) 1.9), 5.26 (s, 2H), 6.6 (s,
1H), 6.86-6.94 (m, 3H), 7.4-7.46 (m, 36H), 7.84 (d, 1H, J )
7.5), 7.97 (d, J ) 8.7), 8.07 (dd, J1 ) 1.9, J2 ) 7.5), 8.38 (m, 2H).
IR: 1653m, 1520m, 1216s. MS 1662.8 (100, [M+H]+).
HRMS calcd for C95H129N6O14Si3 (M+H)+, 1661.8869; found,
1661.8897.
Anachelin H (1). (3R,5S,6S)-6-(Benzyloxy-benzoylamino)-3,5,7-
tris-(tert-butyl-dimethyl-silanoxy)-heptanoic acid-L-Thr(OBn)-D-
Ser(OBn)-L-Ser(OBn)-(3S)-THQ(OBn) (31) (10 mg, 5.89 µmol)
was dissolved in CH3OH (400 µL), and glacial acetic acid (30
drops) was added. The reaction flask was flushed three times with
Ar, and then Pd/C (10%, 8 mg) was added. The reaction flask was
three times flushed with H2, and the reaction mixture was stirred
2.5 h at rt. The reaction mixture was then filtered over celite, and
the solvent was removed under reduced pressure. The residue was
dissolved in cold MeOH (HCl 1%) (1 mL), and the solution stirred
for 4 h at 0 °C. Then the solvent was evaporated under reduced
pressure, and the residual acid was coevaporated three times with
chloroform. The residue was dissolved in water and lyophilized to
give anachelin H (1) (3.3 mg, 70%). Light yellow fluffy powder.
1H NMR (D2O, 300 MHz): 1.18 (d, 3H, J ) 6.2), 1.6-1.64 (m,
1H), 1.65-1.74 (m, 1H), 1.99-2.08 (m, 1H), 2.27-2.35 (m, 1H),
2.89-3.04 (m, 1H), 3.05-3.15 (m, 1H), 3.36-3.44 (m, 1H), 3.53
(s, 6H), 3.66-4.1 (m, 7H), 4.1-4.17 (m, 2H), 4.2-4.33 (m, 2H),
4.38-4.45 (m, 3H), 4.53-4.7(m, 1H), 6.75 (s, 1H), 6.96-7.07 (m,
Boc-L-Ser(OBn)-OH (102.9 mg, 0.348 mmol) was dissolved in
THF (1 mL) under Ar and cooled to -20 °C (bath temp.). To this
i
solution, Et3N (48.5 mL, 0.348 mmol) and BuOCOCl (45.3 mL,
0.348 mmol) were added dropwise, and a precipitate began to form
after 5 min. The reaction mixture was stirred for 15 min at -20 °C.
In the meantime, the diamine trifluoroacetate prepared above was
dissolved in THF (1 mL) and Et3N (0.97 mL, 0.696 mmol) was
added. This solution of the free base was slowly added to the above
mixture of the mixed anhydride at -20 °C. The reaction mixture
was allowed to warm to 10 °C in three h. The solvent was
evaporated under reduced pressure, and the residue taken up in
EtOAc. The organic layer was twice washed with 10% citric acid
solution, twice washed with brine, dried (MgSO4), and filtered, and
the solvent was evaporated under reduced pressure. FC (CH2Cl2/
MeOH 10:1) gave the title compound (30) (98.0 mg, 0.20 mmol,
58%). Colorless oil. Rf ) 0.12 (CH2Cl2/MeOH 10:1). [R]D ) -12.3
(c 1.00, MeOH). 1H NMR (CD3OD, 300 MHz): 1.45 (s, 9H), 2.63-
2.76 (m, 2H), 2.82 (br. s, 6H), 3.14-3.25 (m, 2H), 3.57 (d, J )
5.6, 2 H), 4.15 (t, J ) 5.3, 1 H), 4.36-4.40 (m, 1H), 4.47 (s, 2H),
6.53 (dd, J1 ) 8.1, J2 ) 2.0, 1H), 6.66 (d. J ) 2.0, 1H), 6.70 (d,
J ) 8.1, 1H), 7.24-7.35 (m, 5H). 13C NMR (CD3OD, 75 MHz):
27.5, 37.6, 43.2, 55.6, 60.6, 69.4, 73.1, 80.3, 115.4, 116.3, 120.6,
127.7, 127.9, 128.3, 137.9, 144.3, 145.3, 157.3, 172.4. IR: (thin
film): 3313s, 2947m, 1674s, 1529m. MS 488.3 (100, [M+H]+),
432.2 (31, [M-tBu+H]+), 388.2 (17, [M-Boc]+). HRMS calcd for
C26H38N3O6 (M+H)+, 488.2755; found, 488.2750.
(2S, 3S)-6,7-Bis-benzyloxy-3-(3-benzyloxy-2-tert-butoxycar-
bonylamino-propionylamino)-6,7-dihydroxy-1,1-dimethyl-1,2,3,4-
tetrahydro-quinolinium (28). (2S, 3S)-3-(3-benzyloxy-2-tert-
butoxycarbonylamino-propionylamino)-6,7-dihydroxy-1,1-dimethyl-
1,2,3,4-tetrahydro-quinolinium (149.2 mg, 0.3 mmol) was dissolved
under Ar in dry acetone (3 mL). To this solution, Cs2CO3 (300
mg, 0.92 mmol) and BnBr (110 mL, 0.93 mmol) were added, and
the resulting reddish solution was heated to reflux for 4 h. The
color changed to white, and the solvent was removed under reduced
pressure. The residue was taken up in EtOAc, and the organic layer
was washed three times with 1 N HCl solution and twice with brine.
The organic layer was dried (MgSO4) and filtered, and the solvent
was removed under reduced pressure. FC (5 g SiO2, CH2Cl2/MeOH
10:1) gave the title compound (28) (169.0 mg, 0.253 mmol, 85%).
J. Org. Chem, Vol. 72, No. 22, 2007 8369

Gademann et al.
3H), 7.42-7.53 (m, 1H), 7.79 (d, 1H, J ) 7.5). MS 779 (100,
[M]+). HRMS calcd for C35H51N6O14 (M)+, 779.3463; found,
779.3471.
acknowledged for skillful technical assistance. We thank Prof.
Dr. H. Budzikiewicz for an authentic sample of anachelin H.
Supporting Information Available: Experimental procedures,
characterization data, and copies of spectra are provided as
Supporting Information. This material is available free of charge
via the Internet at http://pubs.acs.org.
Acknowledgment. K.G. thanks Prof. Dr. Erick M. Carreira
for generous financial support in the context of his habilitation
(2002-2006) as well as the Schweizerischer Nationalfonds zur
Fo¨rderung der wissenschaftlichen Forschung (Grant No. 200021-
101601/1, Ph.D fellowship to Y.B.). Tobias Bru¨tsch is gratefully
JO701402B
8370 J. Org. Chem., Vol. 72, No. 22, 2007