Isolation, Structure Elucidation, and (Bio)Synthesis of Haprolid, a Cell-Type-Specific Myxobacterial Cytotoxin

Article abstract of DOI:10.1002/anie.201603288

Myxobacteria are well-established sources for novel natural products exhibiting intriguing bioactivities. We here report on haprolid (1) isolated from Byssovorax cruenta Har1. The compound exhibits an unprecedented macrolactone comprising four modified amino acids and a polyketide fragment. As configurational assignment proved difficult, a bioinformatic analysis of the biosynthetic gene cluster was chosen to predict the configuration of each stereocenter. In-depth analysis of the corresponding biosynthetic proteins established a hybrid polyketide synthase/nonribosomal peptide synthetase origin of haprolid and allowed for stereochemical assignments. A subsequent total synthesis yielded haprolid and corroborated all predictions made. Intriguingly, haprolid showed cytotoxicity against several cell lines in the nanomolar range whereas other cells were almost unaffected by treatment with the compound.

Full text of DOI:10.1002/anie.201603288

Angewandte
Communications
Chemie
International Edition: DOI: 10.1002/anie.201603288
German Edition: DOI: 10.1002/ange.201603288
Natural Products
Isolation, Structure Elucidation, and (Bio)Synthesis of Haprolid, a Cell-
Type-Specific Myxobacterial Cytotoxin
Heinrich Steinmetz, Jun Li, Chengzhang Fu, Nestor Zaburannyi, Birgitte Kunze,
Kirsten Harmrolfs, Viktoria Schmitt, Jennifer Herrmann, Hans Reichenbach, Gerhard Hçfle,
Markus Kalesse,* and Rolf Mꢀller*
Abstract: Myxobacteria are well-established sources for novel
natural products exhibiting intriguing bioactivities. We here
report on haprolid (1) isolated from Byssovorax cruenta Har1.
The compound exhibits an unprecedented macrolactone
comprising four modified amino acids and a polyketide frag-
ment. As configurational assignment proved difficult, a bioin-
formatic analysis of the biosynthetic gene cluster was chosen to
Figure 1. Haprolid (1).
predict the configuration of each stereocenter. In-depth anal-
ysis of the corresponding biosynthetic proteins established
a hybrid polyketide synthase/nonribosomal peptide synthetase
origin of haprolid and allowed for stereochemical assignments.
A subsequent total synthesis yielded haprolid and corroborated
all predictions made. Intriguingly, haprolid showed cytotox-
icity against several cell lines in the nanomolar range whereas
other cells were almost unaffected by treatment with the
compound.
determined without defining the configuration.[1d] We here
report on the structure, the determination of its stereochem-
istry, the biosynthesis, and the synthesis of this compound and
show haprolid to exhibit intriguing and selective cytotoxic
activity against eukaryotic cells.
Haprolid (1) was obtained as a white powder and was
assigned a molecular formula of C38H58N4O7 on the basis of
1
M
yxobacteria have shown a remarkable capacity to pro-
HR-ESI-MS data. Analysis of the H and 13C NMR spectra
duce structurally diverse natural compounds exhibiting
a broad spectrum of biological activities.[1] During a screening
campaign for novel metabolites, haprolid (1; Figure 1) was
isolated from the myxobacterium Byssovorax cruenta strain
Har1 at the HZI already in 2000 and its constitution was
(Table S1) revealed the presence of one N-H amide proton
(d(H) = 8.28 ppm), seven methyl groups (including one O-
methyl and two N-methyl groups), one phenyl ring, seven
methine groups (five of which were bound to heteroatoms),
nine methylene groups, and five carbonyl-type carbon atoms.
1
Interpretation of the H, 13C, and 2D NMR data suggested
a peptidic segment linked to a lipophilic C12 chain. The
detailed analysis of COSY, HMQC, and HMBC correlations
revealed one proline unit and three additional amino acids
including NMe-leucine (N-Me-Leu), sarcosine (Sarc), and
phenylalanine (Phe) identifying haprolid (1) as partially
constructed as a tetrapeptide (Figures S4–S7). Specifically,
the loss of Pro (m/z = 115), N-Me-Leu (m/z = 127), and Sarc
(m/z = 71) observed in the ESI-MS/MS spectrum (Figure S11)
confirmed the amino acid sequence. Attachment of the
polyketide chain to Phe was confirmed by HMBC correla-
tions between the Phe-a-Proton and C-1 of the polyketide
chain. The constitution of the lipophilic dodecyl chain,
including the position of unsaturation (C-3/4), was elucidated
by combining the analyses of COSY and HMBC correlations.
The geometry of the double bond between C-3 and C-4 was
assigned as Z based on the strong ROESY correlation
between H-3 and Me at C-4, and further supported by
ROESY correlation between H-2b (d(H) = 2.64 ppm) and H-
5b (d(H) = 2.20–2.14 ppm) (Figures S8 and S9). The cyclic
structure of haprolid (1) was implied by strong HMBC
correlation between H-7 (d(H) = 4.62 ppm) and Pro C-1. The
absolute configuration of the amino acids was determined
[*] H. Steinmetz, Dr. B. Kunze
Helmholtz Center for Infection Research (HZI)
Department Microbial Drugs
Inhoffenstrasse 7, 38124 Braunschweig (Germany)
Dr. J. Li, Prof. Dr. M. Kalesse
Institute for Organic Chemistry
Gottfried Leibniz Universitꢀt Hannover
and
Centre of Biomolecular Drug Research (BMWZ)
Schneiderberg 1B, 30167 Hannover (Germany)
E-mail: markus.kalesse@oci.uni-hannover.de
Dr. J. Li, Prof. Dr. H. Reichenbach, Prof. Dr. G. Hçfle,
Prof. Dr. M. Kalesse
Helmholtz Center for Infection Research (HZI)
Inhoffenstrasse 7, 38124 Braunschweig (Germany)
Dr. C. Fu, Dr. N. Zaburannyi, Dr. K. Harmrolfs, V. Schmitt,
Dr. J. Herrmann, Prof. Dr. R. Mꢁller
Helmholtz Institute for Pharmaceutical Research Saarland (HIPS),
Helmholtz Center for Infection Research and Pharmaceutical
Biotechnology
Saarland University, Campus Building E8.1
66123 Saarbrꢁcken (Germany)
E-mail: Rolf.Mueller@helmholtz-hzi.de
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under http://dx.doi.org/10.
1002/anie.201603288.
using
indicating the presence of l-Pro, d-NMe-Leu, and l-Phe.
a
chemical degradation/derivatization procedure[2]
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

Angewandte
Communications
Chemie
As the configurational assignment of the secondary
alcohols in the polyketide side chain proved to be challenging,
we decided to perform a bioinformatic ketoreductase analy-
sis,[3,4] for which the knowledge of the biosynthetic gene
cluster sequence was required. To identify the biosynthetic
gene cluster of haprolid the genome of the producer strain
Byssovorax cruenta Har1 was sequenced with the aid of the
Illumina sequencing technology and a draft genome compris-
ing ten scaffolds was achieved. According to the chemical
structure of haprolid, the biosynthetic gene cluster is sup-
posed to include a hybrid polyketide synthase/nonribosomal
peptide synthetase (PKS-NRPS) assembly line. After analyz-
ing all predicted PKS-NRPS hybrid gene clusters by anti-
SMASH analysis,[5] we found a candidate gene cluster with
two gaps which is in line with the “retro-biosynthesis” analysis
of haprolid. The gaps of this gene cluster were closed by
sequencing of the polymerase chain reaction (PCR)-ampli-
fied DNA fragments covering the gaps (Table S2). The
function of every gene in the predicted gene cluster was
analyzed by BLAST, indicating three giant genes (hapA,
hapB, and hapC) which encode two PKS megaenzymes and
one hybrid PKS-NRPS multienzyme complex. The detailed
annotations of the upstream and downstream genes of hapA–
hapC indicated that none of these adjacent genes have any
obvious function in haprolid biosynthesis. Haprolid is synthe-
sized starting from the multienzyme HapA containing
a mixed-loading and chain-extending module frequently
observed in biosynthesis of myxobacterial natural products
like soraphen,[6] melithiazol,[7] thuggacin,[8] and disciformy-
cin.[9] The polyketide chain is extended by three modules of
the megaenzyme HapB. The hybrid PKS-NRPS protein
HapC containing another PKS module and four NRPS
modules not only extends the polyketide chain but also adds
four amino acid residues.
domains and four adenylation (A) domains were analyzed for
substrate-activating specificity. The ATs were compared to
the crystallized malonyl-CoA:acyl carrier protein transacy-
lase from Escherichia coli (Table S3, Figure S12).[10] The
loading acyltransferase (ATL) domain which is expected to
transfer acetyl-CoA onto ACPL, surprisingly, bears the
arginine residue (R117), which is thought to be important
for the activation of dicarboxylic acid extender units (Fig-
ure S12).[11] However, there are already reported cases of this
protein modification carrying an arginine residue which can
shuttle either acetyl-CoA or malonyl-CoA.[11,12] The sub-
strate-specificity analysis of these AT domains revealed that
the collinear assembly line loads acetyl-CoA as a starter unit,
and then extends the polyketide chain by three subsequent
malonyl-CoAs, followed by one methylmalonyl-CoA, and
another malonyl-CoA incorporation, which is consistent with
the haprolid chemical structure (Scheme 1, Figure S12, and
Table S3). The A domains were analyzed for their putative
amino acid specificity (Table S4, Figure S13)[13] which was in
agreement with the haprolid structure (Scheme 1).
Based on the biosynthetic analysis, we proposed the
absolute configurations of all chiral centers by analyzing
ketoreductase (KR) domains, adenylation (A) domains, and
condensation (C) domains of the Hap cluster.[14] Configura-
tional analyses of amino acid residues were in agreement with
the results of chemical analysis (vide supra). The absolute
configuration of both oxymethine groups in the polyketide
chain was indicated by comparing the specific motif differ-
ence between the A-type and B-type stereospecificity KR
domains (Figure S14).[3,15] In addition, our statistical analysis
of both ketoreductases gave a ScoreDiff value of 27,64 for the
alcohol at C-18 and 47,39 for the C-22 methoxy group.[3d] Both
oxy functionalities at C-18 and at C-22 were therefore
predicted to be d-configured as they are generated by B-
type KR domains (KR1, KR3). This is only the second time
that the fidelity of a biosynthetic structure prediction has been
To further confirm the functionalities encoded in the
biosynthetic gene cluster, all the six acyltransferase (AT)
Scheme 1. Biosynthesis of haprolid. For details see text.
2
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
quantified using the absolute value of a statistical analysis.
catalyst which allowed selective epoxidation of the terminal
Formation of the Z-configured double bond in haprolid is
supposed to be catalyzed by module 5 which contains a B-
type KR domain. Although most cis alkenes in polyketides
are formed by dehydratase (DH) domain catalyzed elimina-
tion of water from (3S)-hydroxylacyl intermediates, which are
in turn generated by A-type KR domains,[16] exceptions to this
rule were discovered such as the B-type KR domains found to
be involved in the formation of cis double bonds in the
biosyntheses of borrelidin,[17] chivosazol,[4b] difficidin,[18] and
mupirocin.[19]
double bond without affecting the Z-configured internal
olefin. The construction of the secondary alcohols was
achieved through enzymatic kinetic resolution and selective
epoxide opening. For the transformation from 7 to 8, an
improved protocol for generating thermally stable iodome-
thyllithium was employed which traces back to the seminal
work of Kçbrich.[23]
The synthesis commenced with neryl acetate (Scheme 2),
which was treated with mCPBA leading to epoxidation of the
terminal double bond[24] followed by treatment with vinyl-
magnesium bromide in the presence of copper(I) to provide
compound 7.[25] The introduction of the monosubstituted
double bond as a synthon for the carboxylate was the result of
experimentation with other functional groups that failed to
provide polyketide 2. Consequently, compound 7 was first
converted to the corresponding aldehyde by treatment with
NaIO4 and then treated with iodomethyllithium generated in
situ from CH2I2 and MeLi in the presence of LiBr.[26] This
improved protocol for using a-halogenated organolithium
compounds (Kçbrich reagents)[27] allows reactions to take
place at moderate temperatures (À20 to 08C) without
The combination of the C-domain-type determinations
with the substrate-specificity prediction of the A domains
leads to collinear biosynthesis, which starts from the activa-
tion of one l-Phe by domain A6 and the linking of the
upstream polyketide acyl chain to the PCP-bound l-Phe by
C6. Next, Gly is activated, incorporated into the growing
hybrid PK-NRP chain, and N-methylated by the N-MT
domain of module 7 followed by l-Leu insertion, N-methyl-
ation, and epimerization. Subsequently, the C9 domain acts as
a
DCL-type condensation domain constructing the d–l pep-
tide linkage between the d-Leu and the l-Pro incorporated
by module 9 (Scheme 1). Finally, the TE
domain of module 9 performs the intramolec-
ular macrocyclization yielding haprolid.
Based on the structural assignment we
devised a retrosynthetic disconnection that
joined the peptidic and polyketide portion
through a peptide bond formation and either
a Shiina[20] or a Mitsunobu[21] lactonization,
potentially diverting one synthetic route into
two diastereomers at the final stage of the
synthesis.
In order to keep the synthesis as efficient as
possible we started with neryl acetate, which
had to be converted to segment 2 (Figure 2).[22]
The major challenges of this synthesis were the
Z-configured double bond that had to be kept
out of conjugation with respect to the acid
moiety and the construction of two relatively
isolated secondary alcohols. To overcome the
first challenge we used a terminal olefin as
a replacement for the acid moiety. Liberation
of the acid was later achieved with Strukulꢀs
Scheme 2. Synthesis of segment 10. CAL-B=lipase B from Candida antarctica,
mCPBA=m-chloroperoxybenzoic acid, DMAP=4-dimethylaminopyridine, DMS=di-
methyl sulfide, HKR=hydrolytic kinetic resolution, TEA=triethylamine, Ts=tosylate.
decomposition of the reagent. The resulting epoxide 8
was subjected to Jacobsenꢀs hydrolytic kinetic resolu-
tion using the chiral (R,R)(salen)CoIII complex.[28] With
chiral epoxide 4 in hands treatment with the Grignard
reagent derived from 5 provided the polyketide skel-
eton.
Transformation of the terminal double bond to the
carboxylate was achieved by a four-step sequence
involving an epoxidation of the terminal double bond
with the platinum catalyst reported by Strukul et al
(Scheme 3).[29] Condensation with the corresponding
tripeptide was done under standard peptide coupling
conditions and generated compound 15. Hydrolysis of
both the acetate moiety and the methyl ester provided
the seco acid, which was subjected to the Mitsunobu
Figure 2. Haprolid and its retrosynthetic disconnection.
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

Angewandte
Communications
Chemie
did not observe specific effects on the cytoske-
leton (actin, tubulin) or certain hallmarks of
apoptosis, such as mitochondrial membrane
depolarization, lysosomal rupture, and frag-
mentation of nuclei in KB-3.1 cells that were
treated with haprolid (Figure S15). Future
biochemical studies will address the intriguing
question of the mode of action of the haprolid
class of natural products.
In summary, we have isolated a new secon-
dary metabolite from myxobacteria, completed
its biosynthetic analysis, and verified its struc-
ture by total synthesis. This biosynthetic struc-
ture elucidation took advantage of established
protocols and a statistical refinement from
which a quantitative assessment of the struc-
ture prediction was possible. The synthesis
features under-represented protocols for the
transformation of epoxides, namely the use of
Strukulꢀs catalyst and a thermally stable iodo-
methyllithium reagent. Finally, the remarkable
biological activities in combination with its
short and efficient synthesis (12 steps in the
longest linear sequence) provide the back-
ground for potential biological applications of
this metabolite.
Scheme 3. Endgame of the haprolid (1) synthesis. DEAD=diethyl azodicarboxylate,
DIPEA=N,N-diisopropylethylamine, EDC=1-ethyl-3-(3-dimethylaminopropyl)carbo-
diimide, HOBt=hydroxybenzotriazole.
Keywords: cell-line-selective cytotoxin ·
haprolid · polyketide (bio)synthesis ·
retrosynthesis · structure elucidation
protocol to provide haprolid in 55% yield. All spectroscopic
data of the synthetic material were in full agreement with
those of the authentic sample (information on the organic
[1] a) H. Reichenbach, G. Hçfle in Drug Discovery from Nature
(Eds.: S. Grabley, R. Thiericke), Springer, Berlin, 1999; b) K. J.
Weissman, R. Mꢁller, Nat. Prod. Rep. 2010, 27, 1276 – 1295;
c) T. F. Schꢂberle, F. Lohr, A. Schmitz, G. M. Kçnig, Nat. Prod.
Rep. 2014, 31, 953 – 972.
synthesis in the Supporting Information).
Haprolid (natural and synthetic) was tested on a small
panel of human cancer cell lines (Table 1). Intriguingly, the
compound was highly active (IC50 values in low to mid
nanomolar range) on the cervix carcinoma cells lines Hep2
and KB-3.1 (HeLa derivatives), and THP-1 and HL-60
leukemia cell lines. The IC50 values on Huh7.5 hepatocarci-
noma and U-2 OS osteosarcoma cells were in the high
nanomolar range. In contrast, haprolid was not active on A-
549 lung carcinoma and SK-OV-3 ovarian adenocarcinoma
cells. The reason for the selective cytotoxicity is currently
unknown as is also the molecular target of the compound. We
[2] I. Abe, S. Kuramoto, S. Musha, J. High Resolut. Chromatogr.
1983, 6, 366 – 370.
[3] a) P. Caffrey, ChemBioChem 2003, 4, 654 – 657; b) P. Caffrey,
Chem. Biol. 2005, 12, 1060 – 1062; c) R. Reid, M. Piagentini, E.
Rodriguez, G. Ashley, N. Viswanathan, J. Carney, D. V. Santi,
C. R. Hutchinson, R. McDaniel, Biochemistry 2003, 42, 72 – 79;
d) A. Kitsche, M. Kalesse, ChemBioChem 2013, 14, 851 – 861;
e) The online profile Hidden Markov Model analysis tool can be
found at: https://akitsche.shinyapps.io/profileHMM_App.
[4] a) S. C. Wenzel, R. Mꢁller, Nat. Prod. Rep. 2009, 26, 1385 – 1407;
b) O. Perlova, K. Gerth, O. Kaiser, A. Hans, R. Mꢁller, J.
Biotechnol. 2006, 121, 174 – 191.
Table 1: Cytotoxic activity of haprolid.
[5] T. Weber, K. Blin, S. Duddela, D. Krug, H. U. Kim, R.
Bruccoleri, S. Y. Lee, M. A. Fischbach, R. Mꢁller, W. Wohlleben,
et al., Nucleic Acids Res. 2015, 43, W237 – 43.
Cell line (accession no.)
IC50 [mm]
1 (from B. cruenta)
1 (synthetic)
[6] C. J. Wilkinson, E. J. Frost, J. Staunton, P. F. Leadlay, Chem. Biol.
2001, 8, 1197 – 1208.
[7] S. Weinig, H.-J. Hecht, T. Mahmud, R. Mꢁller, Chem. Biol. 2003,
10, 939 – 952.
[8] K. Buntin, H. Irschik, K. J. Weissman, E. Luxenburger, H.
Blçcker, R. Mꢁller, Chem. Biol. 2010, 17, 342 – 356.
[9] F. Surup, K. Viehrig, K. I. Mohr, J. Herrmann, R. Jansen, R.
Mꢁller, Angew. Chem. Int. Ed. 2014, 53, 13588 – 13591; Angew.
Chem. 2014, 126, 13806 – 13809.
A-549 (ACC-107)
Hep2 (CCL-240)
HL-60 (ACC-3)
Huh7.5 (APC49)
KB-3.1 (ACC-16)
SK-OV-3 (HTB-77)
THP-1 (ACC-16)
U-2 OS (ACC-785)
>100
0.0042Æ0.0032
0.10Æ0.11
0.29Æ0.23
0.014Æ0.002
>100
>100
0.01
n.d.
0.37
n.d.
>100
0.06
n.d.
0.025Æ0.014
1.73Æ0.64
4
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!

Angewandte
Communications
Chemie
[10] G. Yadav, R. S. Gokhale, D. Mohanty, J. Mol. Biol. 2003, 328,
[18] X. H. Chen, J. Vater, J. Piel, P. Franke, R. Scholz, K. Schneider,
335 – 363.
A. Koumoutsi, G. Hitzeroth, N. Grammel, A. W. Strittmatter,
et al., J. Bacteriol. 2006, 188, 4024 – 4036.
[19] A. K. El-Sayed, J. Hothersall, S. M. Cooper, E. Stephens, T. J.
Simpson, C. M. Thomas, Chem. Biol. 2003, 10, 419 – 430.
[20] I. Shiina, M. Kubota, H. Oshiumi, M. Hashizume, J. Org. Chem.
2004, 69, 1822 – 1830.
[11] L. Serre, E. C. Verbree, Z. Dauter, A. R. Stuitje, Z. S. Dere-
wenda, J. Biol. Chem. 1995, 270, 12961 – 12964.
[12] S. F. Haydock, J. F. Aparicio, I. Molnꢃr, T. Schwecke, L. E.
Khaw, A. Kçnig, A. F. Marsden, I. S. Galloway, J. Staunton, P. F.
Leadlay, FEBS Lett. 1995, 374, 246 – 248.
[13] T. Stachelhaus, H. D. Mootz, M. A. Marahiel, Chem. Biol. 1999,
6, 493 – 505.
[21] O. Mitsunobu, Synthesis 1981, 1 – 28.
[22] For the use of terpenes in polyketide synthesis see: a) J.
Swatschek, L. Grothues, J. O. Bauer, C. Strohmann, M. Christ-
mann, J. Org. Chem. 2014, 79, 976 – 983; b) P. Winter, J.
Swatschek, M. Willot, L. Radtke, T. Olbrisch, A. Schꢂfer, M.
Christmann, Chem. Commun. 2011, 47, 12200 – 12202.
[23] G. Kçbrich, Angew. Chem. Int. Ed. Engl. 1967, 6, 41 – 52; Angew.
Chem. 1967, 79, 15 – 27.
[24] J. Germain, P. Deslongchamps, J. Org. Chem. 2002, 67, 5269 –
5278.
[25] P. Winter, C. Vaxelaire, C. Heinz, M. Christmann, Chem.
Commun. 2011, 47, 394 – 396.
[14] For other examples of structure prediction based on the
biosynthetic analysis see: a) O. Hartmann, M. Kalesse, Angew.
Chem. Int. Ed. 2014, 53, 7335 – 7338; Angew. Chem. 2014, 126,
7463 – 7466; b) C. Jahns, T. Hoffmann, S. Mꢁller, K. Gerth, P.
Washausen, G. Hçfle, H. Reichenbach, M. Kalesse, T. Mꢁller,
Angew. Chem. Int. Ed. 2012, 51, 5239 – 5243; Angew. Chem.
2012, 124, 5330 – 5334; c) S. Essig, S. Bretzke, R. Mꢁller, D.
Menche, J. Am. Chem. Soc. 2012, 134, 19362 – 19365; d) T.
Brodmann, D. Janssen, M. Kalesse, J. Am. Chem. Soc. 2010, 132,
13610 – 13611; e) M. Bock, K. Buntin, R. Mꢁller, A. Kirschning,
Angew. Chem. Int. Ed. 2008, 47, 2308 – 2311; Angew. Chem. 2008,
120, 2341 – 2344; f) D. Menche, F. Arikan, O. Perlova, N.
Horstmann, W. Ahlbrecht, S. C. Wenzel, R. Jansen, H. Irschik,
R. Mꢁller, J. Am. Chem. Soc. 2008, 130, 14234 – 14243; g) D.
Janssen, D. Albert, R. Jansen, M. Mꢁller, M. Kalesse, Angew.
Chem. Int. Ed. 2007, 46, 4898 – 4901; Angew. Chem. 2007, 119,
4985 – 4988.
[26] J. M. Concellꢄn, H. Cuervo, R. Fernꢃndez-Fano, Tetrahedron
2001, 57, 8983 – 8987.
[27] J. Aubꢅ, Comprehensive Organic Synthesis B. M. Trost, I.
Fleming, Eds. Pergamon: New York, 1999, 1, p. 260.
[28] a) M. Tokunaga, J. F. Larrow, F. Kakiuchi, E. N. Jacobsen,
Science 1997, 277, 936; b) L. P. C. Nielsen, C. P. Stevenson,
D. G. Blackmond, E. N. Jacobsen, J. Am. Chem. Soc. 2004, 126,
1360 – 1362.
[15] A. P. Siskos, A. Baerga-Ortiz, S. Bali, V. Stein, H. Mamdani, D.
Spiteller, B. Popovic, J. B. Spencer, J. Staunton, K. J. Weissman,
et al., Chem. Biol. 2005, 12, 1145 – 1153.
[29] M. Colladon, A. Scarso, P. Sgarbossa, R. A. Michelin, G. Strukul,
J. Am. Chem. Soc. 2006, 128, 14006 – 14007.
[16] D. H. Kwan, F. Schulz, Molecules 2011, 16, 6092 – 6115.
[17] C. Olano, B. Wilkinson, C. Sanchez, S. J. Moss, R. Sheridan, V.
Math, A. J. Weston, A. F. Brana, C. J. Martin, M. Oliynyk, et al.,
Chem. Biol. 2004, 11, 87 – 97.
Received: April 4, 2016
Published online: && &&, &&&&
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

Angewandte
Communications
Chemie
Communications
Natural Products
H. Steinmetz, J. Li, C. Fu, N. Zaburannyi,
B. Kunze, K. Harmrolfs, V. Schmitt,
J. Herrmann, H. Reichenbach, G. Hçfle,
M. Kalesse,* R. Mꢁller*
&&&& — &&&&
Isolation, Structure Elucidation, and
(Bio)Synthesis of Haprolid, a Cell-Type-
Specific Myxobacterial Cytotoxin
The macrolactone haprolid was isolated
from Byssovorax cruenta. Bioinformatic
analysis of the haprolid biosynthetic gene
cluster aided stereochemical assign-
ments based on the hybrid polyketide
synthase/nonribosomal peptide synthe-
tase origin of the compound. Total syn-
thesis corroborated all predictions. Hap-
rolid showed cytotoxicity against several
cell lines in the nanomolar range whereas
other cells were almost unaffected.
6
www.angewandte.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!