Champacyclin, a new cyclic octapeptide from Streptomyces strain C42 isolated from the Baltic Sea

Article abstract of DOI:10.3390/md11124834

New isolates of Streptomyces champavatii were isolated from marine sediments of the Gotland Deep (Baltic Sea), from the Urania Basin (Eastern Mediterranean), and from the Kiel Bight (Baltic Sea). The isolates produced several oligopeptidic secondary metabolites, including the new octapeptide champacyclin (1a) present in all three strains. Herein, we report on the isolation, structure elucidation and determination of the absolute stereochemistry of this isoleucine/leucine (Ile/Leu = Xle) rich cyclic octapeptide champacyclin (1a). As 2D nuclear magnetic resonance (NMR) spectroscopy could not fully resolve the structure of (1a), additional information on sequence and configuration of stereocenters were obtained by a combination of multi stage mass spectrometry (MSⁿ) studies, amino acid analysis, partial hydrolysis and subsequent enantiomer analytics with gas chromatography positive chmical ionization/electron impact mass spectrometry (GC-PCI/EI-MS) supported by comparison to reference dipeptides. Proof of the head-to-tail cyclization of (1a) was accomplished by solid phase peptide synthesis (SPPS) compared to an alternatively side chain cyclized derivative (2). Champacyclin (1a) is likely synthesized by a non-ribosomal peptide synthetase (NRPS), because of high content of (D)-amino acids. The compound (1a) showed antimicrobial activity against the phytopathogen Erwinia amylovora causing the fire blight disease of certain plants.

Full text of DOI:10.3390/md11124834

Mar. Drugs 2013, 11, 4834-4857; doi:10.3390/md11124834
OPEN ACCESS
marine drugs
ISSN 1660-3397
www.mdpi.com/journal/marinedrugs
Article
Champacyclin, a New Cyclic Octapeptide from Streptomyces
Strain C42 Isolated from the Baltic Sea
Alexander Pesic 1,†, Heike I. Baumann 2,†, Katrin Kleinschmidt 2, Paul Ensle 1, Jutta Wiese 2,
Roderich D. Süssmuth 1,* and Johannes F. Imhoff 2,*
1
Technical University of Berlin, Institute for Chemistry, Straße des 17. Juni 124, Berlin 10623,
Germany; E-Mails: pesic@chem.tu-berlin.de (A.P.); ensle@chem.tu-berlin.de (P.E.)
2
GEOMAR Helmholtz Center for Ocean Research Kiel, Marine Microbiology, Düsternbrooker
Weg 20, Kiel 24105, Germany; E-Mails: hbaumann@geomar.de (H.I.B.);
kkleinschmidt@geomar.de (K.K.); jwiese@geomar.de (J.W.)
†
These authors contributed equally to this work.
* Authors to whom correspondence should be addressed;
E-Mails: suessmuth@chem.tu-berlin.de (R.D.S); jimhoff@geomar.de (J.F.I.);
Tel.: +49-30-314-78774 (R.D.S.); Fax: +49-314-79651 (R.D.S.);
Tel.: +49-431-600-4450 (J.F.I.); Fax: +49-431-600-4452 (J.F.I.).
Received: 22 October 2013; in revised form: 14 November 2013 / Accepted: 15 November 2013 /
Published: 2 December 2013
Abstract: New isolates of Streptomyces champavatii were isolated from marine sediments
of the Gotland Deep (Baltic Sea), from the Urania Basin (Eastern Mediterranean), and
from the Kiel Bight (Baltic Sea). The isolates produced several oligopeptidic secondary
metabolites, including the new octapeptide champacyclin (1a) present in all three strains.
Herein, we report on the isolation, structure elucidation and determination of the absolute
stereochemistry of this isoleucine/leucine (Ile/Leu = Xle) rich cyclic octapeptide
champacyclin (1a). As 2D nuclear magnetic resonance (NMR) spectroscopy could not
fully resolve the structure of (1a), additional information on sequence and configuration of
stereocenters were obtained by a combination of multi stage mass spectrometry (MSn)
studies, amino acid analysis, partial hydrolysis and subsequent enantiomer analytics with
gas chromatography positive chmical ionization/electron impact mass spectrometry
(GC-PCI/EI-MS) supported by comparison to reference dipeptides. Proof of the
head-to-tail cyclization of (1a) was accomplished by solid phase peptide synthesis (SPPS)
compared to an alternatively side chain cyclized derivative (2). Champacyclin (1a) is likely
synthesized by a non-ribosomal peptide synthetase (NRPS), because of its high content of

Mar. Drugs 2013, 11
(D)-amino acids. The compound (1a) showed antimicrobial activity against the
4835
phytopathogen Erwinia amylovora causing the fire blight disease of certain plants.
Keywords: Streptomyces champavatii; antimicrobial activity; fire blight; cyclic peptides;
sequence ladder analysis; chiral GC-PCI/EI-MS; ESI-Ion-Trap-MSn; solid phase peptide
synthesis (SPPS); on-resin cyclization; 2D NMR
1. Introduction
The actinobacterial genus Streptomyces is widely distributed in both terrestrial and aquatic
ecosystems. These bacteria exhibit diverse physiological and metabolic properties, such as the
production of extracellular enzymes and the synthesis of a large variety of secondary metabolites, a
number of which are exploited by the pharmaceutical industries. Many of these antimicrobials are of
low molecular weights (≤2000 Da) and belong to macrolides, polyenes, tetracyclines, amino
glycosides, or other compound classes with heterogeneous chemical structures. Streptomycetes also
produce a large variety of oligo- and polypeptides including signaling pheromones and a range of
antimicrobial peptides with potential lead structures used for chemical synthesis of derivatives.
Biosynthesis of peptides occurs either ribosomally or non-ribosomally contributing to the chemical
diversity of secondary metabolites synthesised by these microorganisms. Ribosomally synthesized and
posttranslationally modified peptides (RiPPs) have an increased variation of chemical structures due to
post-translational modification. Structures of non-ribosomally synthesized peptides (NRPs) are
different from the ribosomal peptides by the frequent occurrence of (D)-amino acids and otherwise
modified non-proteinogenic amino acids. Furthermore, the structural complexity can be dramatically
increased if fatty acids and/or carbohydrate moieties are present in these molecules [1].
One characteristic property of peptides is their possibility to form cyclic structures. Notable important
cyclic (depsi)-peptides are cyclosporine (immunosuppressant), polymyxin
B
(antibiotic),
didemnin B (antitumor), gramicidin S (antibiotic), and valinomycin (ionophore), as reviewed by
Bockus et al. (2013) [2].
Due to their potential for use as drugs, there is a high demand not only to screen for new potent
cyclic peptides but also for fast and reliable methods to characterize them. In particular alternative
methods are needed, if analytical methods such as NMR are not applicable, due to low quantities of the
substances available. For peptides and proteins, de novo sequencing by tandem mass spectrometric
techniques is still an indispensable tool. These fragmentation techniques include collision induced
fragmentation (CID) and electron capture dissociation (ECD). Routinely b- and y-ions are annotated in
linear peptides, while for cyclic peptides additional fragmentation pathways in tandem mass
spectrometry (MS2) experiments are possible, yielding also non-direct sequence (NDS) ions through a
scrambling of the amino acid sequence, which can be difficult to interpret [3].
Kjeldsen et al. (2003) [4] used a variant of electron-capture dissociation (ECD), named “Hot ECD”
(HECD), which induces w-type fragment ions through a high collision energy secondary fragmentation
process to distinguish between Leu and Ile. However, HECD is not applicable to most mass
spectrometers and requires state-of-the-art mass spectrometers. To overcome this problem

Mar. Drugs 2013, 11
4836
Armirotti et al. (2007) [5] introduced a low-energy ESI-Ion-TRAP-MSn method to distinguish between
Leu and Ile. It was demonstrated that in a series of fragmentation steps, Ile, unlike Leu, generates a
69 Da ion by fragmentation of its 86 Da immonium ion. However, with regard to sensitivity, every
MSn step dramatically decreases the ion abundance.
An alternative to sequencing of peptides by tandem mass spectrometry constitutes partial hydrolysis
or partial enzymatic digestion to yield overlapping amino acid sequences (sequence ladder) of smaller
peptides. This method proved an excellent tool, when coupled to a mass spectrometer equipped with
an electron impact (EI) ionization source [6]. Mass spectrometric sequence analyses of oligopeptide
fragments (up to nine amino acids) were performed after derivatization yielding the corresponding
N-trifluoroacetyl peptide alkyl esters and direct injection into the EI ion-source [6]. In some cases, in
addition to the sequence information, the distinction between Xle residues was possible due to
different ion fragments formed in the EI-ion source (Supplementary Figure S47).
As mentioned above, cyclic peptides may also contain (D)-amino acids making enantiomer
analytics indispensable. Data about the content of (D)- and (L)-amino acids in a peptide can be
routinely generated by total hydrolysis of the protein (6 M HCl, 110 °C, 24 h, vacuum) and subsequent
chiral gas-chromatography mass spectrometry analysis (GC-MS). To assign the configuration at
ambiguous positions, again partial hydrolysis, preferably yielding dipeptides, combined with chiral
GC-MS analysis has to be carried out. Accordingly, Pätzold et al. (2006) [7] have extensively studied
gas chromatographic separation of stereoisomers of dipeptides. It should, however, be pointed out, that
it is impossible to separate all possible 392 dipeptides (1520 stereoisomers plus the achiral Gly-Gly) on
a chiral column.
In the present work, a cyclic antimicrobial octapeptide was isolated from new isolates of
Streptomyces closely related to Streptomyces champavatii. Streptomyces champavatii was first isolated
from Champavathi river mud (Andhra Pradesh) and Bangalore virgin soil (Mysore State) in India
(Narasimha Rao and Uma, 1958) [8]. Three isolates of this bacterium were identified by paper
chromatography, where two compounds are antifungal heptaene antibiotics, which were termed
afterwards champamycin A and champamycin B with heptane group characteristic absorption curves
(λmax = 360, 380, 405 nm), and a third antifungal compound named champavatin with no selective
absorption in the region of 250–400 nm [8,9], of which chemical structure was not, thus far,
determined. Here we report on the structure and biological activities of a cyclic octapeptide named
champacyclin (1a) (Figure 1), which is considered to correspond to the third previously unknown
component reported by Rao and Narasimha Rao (1967) [9]. Recent database searches (Dictionary of
Natural Products, DNP) for compound (1a) revealed no structural identity to known compounds.
Therefore, a full structure determination of (1a) was performed as described in the sections below. The
assignment of the isoleucine/leucine (Ile/Leu = Xle) residues as well as the absolute stereochemistry to
distinct positions in the cyclic peptides was a challenging task, which was solved by a combination of
mass spectrometry, 2D NMR spectroscopy and synthetic approaches. Champacyclin (1a) shows
activity against the fire blight plant pathogen Erwinia amylovora in the µM -range.

Mar. Drugs 2013, 11
Figure 1. Structure of head-to-tail-cyclized octapeptide champacyclin (1a) isolated from
4837
Streptomyces spp. C42.
2. Results and Discussion
2.1. Isolation and Identification of Streptomyces Strains
Spore forming actinobacteria were isolated from sediments of the Gotland Deep (Baltic Sea), from
the eastern Mediterranean (Urania Basin) and from the Kiel Bight, respectively. Three strains were
identified as members of the genus Streptomyces and were further characterized. Strain C42 was
isolated from a sediment core of the Gotland Deep, strain i6a from the Urania Basin, and strain XX19
from the Kiel Bight.
The 16S rRNA gene sequences of the three strains displayed highest similarity to the closely related
type strains of Streptomyces coelicolor NBRC 12854T (99.7%/100%), Streptomyces albidoflavus
NBRC 13010T (99.7%/100%), Streptomyces champavatii NRRL B-5682T (99.6%/99.9%),
Streptomyces sampsonii ATCC 25496T (99.7%/99.9%), and Streptomyces daghestanicus NRRL
B-5418T (99.4%/99.8%). Strains C42 and XX19 had identical 16S rDNA sequences. All three strains
were found in a clade together with S. coelicolor, S. champavatii, S. sampsonii and S. albidoflavus
(Figure 2). These species have highly similar 16S rRNA gene sequences of >99.9% similarity,
indicating a difference of less than one out of 100 nucleotides. Based on the genotypic data of
multilocus sequence analysis (MLSA) and DNA-DNA hybridization (DDH), combined with key
phenotypic properties, it was proposed that these Streptomyces species of the S. albidoflavus clade,
including S. champavatii, should be merged into a single genomic species, for which the name
S. albidoflavus was proposed [10].
Despite the phylogenetic considerations, according to their metabolic profiles, our isolates clearly
grouped with S. champavatii. Three main compounds were produced by these strains, which were
previously mentioned as metabolites of S. champavatii [9,10]: the champamycin A and
champamycin B with absorption maxima (λmax = 362, 384 and 407 nm) characteristic for the heptaene
group and a third compound, which we named champacyclin (1a).

Mar. Drugs 2013, 11
Figure 2. Phylogenetic tree of Streptomyces sp. strains C42, XX19, and i6a including type
4838
strains of closest relatives, further members of the genus Streptomyces and Micrococcus
lylae as outgroup. The calculation was performed using the maximum-likelihood estimation
(MLE). Bootstrap values are given in percent (only numbers above 50% are shown).
Streptomyces rimosus subsp. rimosus JCM 4667T (AB045883)
Streptomyces sporocinereus NBRC 100766T (AB249933)
Streptomyces ribosidificus NBRC 13796T (AB184487)
53
Streptomyces thermolineatus DSM 41451T (Z68097)
Streptomyces morookaensis NBRC 13416T (AB184878)
T
63
Streptomyces globisporus NRRL B-2872 (EF178686)
100
Streptomyces tanashiensis IFO 12919T (AY999856)
70
Streptomyces griseus subsp. griseus NBRC 15744T (AB184699)
Streptomyces mirabilis DSM 40553T (AY999730)
Streptomyces acidiscabies ATCC 49003T (D63865)
Streptomyces galbus DSM 40089T (X79852)
52
81
Streptomyces durhamensis NRRL-B3309T (AY999785)
Streptomyces mashuensis DSM 40221T (X79323)
Streptomyces albidoflavus NBRC 13010T (AB184255)
53 98
C42
XX19
i6a
79
50
83
Streptomyces coelicolor NBRC 12854T (AB184196)
Streptomyces sampsonii ATCC 25496T (D63871)
71
68
T
75
Streptomyces champavatii NRRL B-5682 (DQ026642)
Streptomyces koyangensis VK-A60T (AY079156)
Streptomyces rutgersensis subsp. rutgersensis DSM 40077T (Z76688)
Streptomyces variegatus LMG 20315T (AJ781371)
Streptomyces mexicanus CH-M-1035T (AF441168)
Streptomyces thermodiastaticus DSM 40573T (AB018096)
Streptomyces ambofaciens ATCC 23877T (M27245)
Streptomyces caelestis NRRL 2418T (X80824)
84
91
Streptomyces speibonae PK-BlueT (AF452714)
Streptomyces coeruleofuscus ISP 5144T (AJ399473)
Streptomyces somaliensis DSM 40738T (AJ007403)
Streptomyces carpaticus NRRL B-16359T (DQ442494)
Streptomyces cheonanensis VC-A46T (AY822606)
100
Micrococcus lylae DSM 20315T (X80750)
0.01
The strains C42 and XX19 showed a positive PCR amplification product of the adenylation domain
(A) from a NRPS complex, which is an indicator of non-ribosomal peptide synthesis. Every NRPS
module contains an adenylation domain (A), which activates the corresponding amino acids, and
results after severe reactions in amino acid assembly. Therefore, the correlation between the positive
PCR product of the A domain and the isolated compound (1a) from the fermented culture broth is not
possible without extensive investigations by sequencing of the biosynthesis gene cluster.
Champacyclin (1a) was found in extracts of both, culture supernatant (using ethyl acetate) and
biomass (using methanol). Higher amounts were present in the cells compared to the culture medium.
The methanolic extracts of S. champavatii strain C42 were separated by reversed-phase HPLC on a
C18 column, and the elution profiles were studied with HPLC-DAD-ESI-(+)-MS. On the basis of
similar absorption properties, we assume that (1a) corresponds to the third component found but not
characterized by Narasimha Rao and Uma (1958) [9].
Strain C42 was selected for growth experiments in order to increase the production of (1a) and to
decrease that of co-produced metabolites (Figure 3). Growth parameters varied in these experiments
were the growth medium (tryptic soy broth (TSB), oak flakes medium (OAK), Micromonospora

Mar. Drugs 2013, 11
4839
medium (MIC), actinomycetes medium (ACT), and GYM medium (GYM)), the salt concentration
(0%–4% NaCl in 1% intervals), the incubation temperature (20 °C, 24 °C , and 28 °C), the incubation
time (one day–eight days in one-day intervals) and supplementation with a mixture of amino acids
((L)leucine (L), (L)isoleucine (I), (L)phenylalanine (F), and (L)valine (V)). Champacyclin (1a) was
produced in all five growth media, though the metabolite profiles varied with the medium. The best
production was found in the actinomycetes medium (5.6 g wet weight biomass of 100 mL culture),
where at the same time the production of the heptaenes was lowest. The highest yield of (1a) was
obtained in media without NaCl and after incubation for seven days at 24 °C. A further increase in the
production of compound (1a) was obtained by addition of a mixture of the amino acids leucine,
isoleucine, and phenylalanine, in concentrations of 1 mg/mL each to the medium (Figure 3).
Figure 3. Optimization of growth conditions for champacyclin (1a) biosynthesis of strain
C42. In each experiment base parameters were maintained except for one parameter, which
was variable in the experiment: 100 mL culture volume, actinomycetes medium except in
the salt experiments GYM medium was used, 1.5% salt (NaCl), six days incubation time,
28 °C incubation temperature, 125 rpm shaking, and methanolic extracts. The results are
comparable within each experiment series. Amino acids (1 mg/mL supplement): no = no
supplementation, LIFV = (L)leucine (L), (L)isoleucine (I), (L)phenylalanine (F) and
(L)valine (V). Temp. = Temperature. Media abbreviations see text.
The Streptomyces isolates showed antimicrobial activities, which resemble those reported for the
Indian strains of S. champavatii. Antimicrobial activity was found with crude extracts and purified
substance. The crude extracts of all isolated strains (C42, XX19, and i6a) inhibited the yeast Candida
glabrata. The purified substance (1a) showed a 40% inhibition of the fire blight pathogen Erwinia
amylovora at a concentration of 25 µM. There was no activity observed against the test strains Bacillus

Mar. Drugs 2013, 11
4840
subtilis, Escherichia coli K-12, Staphylococcus lentus, Pseudomonas syringae pv. aptata,
Pseudomonas fluorescens, Xanthomonas campestris, and Ralstonia solanacearum (DSM 9544).
2.2. Structure Elucidation of Champacyclin (1a)
The octapeptide champacyclin (1a) was isolated in amounts of 2.2 mg/L. UV-DAD-data were in
support of the presence of a peptide with an absorption at λmax = 258 nm (ε 1523 in MeOH) and the
occurrence of aromatic amino acids within the structure. Subsequent High-Resolution Electrospray
Orbitrap Mass Spectrometry (HR-ESI-(+)-Orbitrap-MS) showed
a
molecular mass at
m/z 912.62574 [M + H]+ (relative mass error Δm = −2.572 ppm, calcd. 912.62809 [M + H]+), which
corresponds to the calculated tentative molecular formula of [C48H82N9O8] [M + H]+ with 13 DBE for
the uncharged molecule.
2.2.1. Quantitative Chiral Amino Acid Analysis
Chiral Gas Chromatography Positive Chemical Ionization Mass Spectrometry (GC/PCI-MS) of the
total hydrolyzate of (1a) on a Chirasil®-(L)-Val column was performed to quantify the amino acids and
to evaluate the presence of (D)-amino acids. Derivatization of the amino acids to the
N-pentafluoropropionic acid 2-propyl esters, yielded the following analytical results with GC-MS
(Figure 4): (D)Ala, (AlloD)Ile, (L)Ile, (D)Leu, (D)Phe, and (L)Lys. The ratio of the peak areas of (AlloD)Ile and
(L)Ile is 1:2.8, corresponding to a ratio of 1:3 (Supplementary Table S6). Additionally, small amounts
of racemic (L)Leu (6% relative to (D)Leu), (D)Lys (5% relative to (L)Lys), and (L)Phe (3.2% relative to
(D)Phe) were detected (Figure 4 and Supplementary Table S6). By comparison of the peak areas
(Supplementary Figure S13 and Table S6) of GC-MS chromatograms we deduced the amino acid
content: (D)Ala, (AlloD)Ile, (L)Ile, (D)Leu, (D)Phe, and (L)Lys as 1:1:3:1:1:1. Summing up all molecular
masses of these amino acids combined to the restriction to only 13 DBE (as derived from the HR-MS
data), the molecular mass of the peptide (1a) is only compatible with a cyclic peptide structure
(Figure 1). Such a cyclic peptide still has various possibilities for the amino acid sequence as well as
for cyclization. For the latter there exist two possibilities, i.e., through a head-to-tail ring formation
between Nα(L)Lys1-CO-(AlloD)Ile8 as in (1a) (Figure 1) or a head-to-side-chain ring formation as
represented in (2) (synthetic derivative, Supplementary Figure S3) between Nζ-(L)Lys-CO-(AlloD)Ile8
(Supplementary Figure S3).
2.2.2. Mass Spectrometry
In order to obtain the primary sequence of the peptide, HR-ESI-(+)-Orbitrap-In-Source-fragmentation
(SID) spectra (Figure 5) were measured, which yielded preferably the y- and b-ion fragments
according to the Roepstorff nomenclature for peptide fragmentation [11]. In combination with the
results from the amino acid analysis (Figure 4) the primary sequence of the peptide could be established
as: [(L)Lys1-(x)Xle2-(x)Xle3-(D)Phe4-(x)Xle5-(x)Xle6-(D)Ala7-(x)Xle8] with x = D, L, or AlloD leaving the
identity and the stereochemistry of five amino acids of the Leu/Ile-(Xle)-type undetermined.
In order to discriminate between Ile/Leu (Xle) in the y2/b7y3 185 Da ion (Figure 5) we performed
SID-MSn-experiments (Supplementary Figure S2) on the Orbitrap ESI-(+)-mass spectrometer using an

Mar. Drugs 2013, 11
4841
ion-trap analyzer. The subsequent MSn-studies (transition 185 Da → 86 Da → 69 Da) showed the
69 Da ion characteristic of Ile, as reported by Armirotti et al. (2006) [5]. For both y2/b7y3 ions,
fragmentation mechanisms have been formulated (Supplementary Figure S2C) indicating the presence
of an (x)Ile6,8 at position 6 and/or 8.
Figure 4. Chiral GC-PCI-MS analysis of the amino acid hydrolyzate of champacyclin (1a)
on a Chirasil®-(L)-Val column. Amino acids were analyzed as their N-pentafluoropropionic
acid 2-propyl esters and identified by their [M + H]+-ions.
Figure 5. HR-ESI-(+)-SID-Orbitrap-MS of champacyclin (1a) with assignment of
characteristic fragments according to the Roepstorff nomenclature [11].

Mar. Drugs 2013, 11
4842
2.2.3. Chiral GC-PCI/EI-MS Analysis of the Partial Hydrolyzate of (1a)
Due to the high abundance of the isobaric amino acids (D)Leu (1x), (L)Ile (3x), and (AlloD)Ile (1x) in
(1a), the assignment of the configuration of amino acids in distinct positions of the peptide was a
challenging task. The problem was solved by performing partial hydrolysis of (1a) to generate an array
of overlapping sequences based on dipeptides. These dipeptides were analyzed by means of chiral
GC-PCI/EI-MS and identified by their [M + H]+ and [M + C2H5]+ molecular ions, respectively.
According to Pätzold et al. (2006) [7], the best separations for dipeptides were obtained for
N-trifluoroacetyl methyl esters on a Chirasil®-(L)-Val column (max. operating temperature 195 °C).
From various hydrolysis times and reaction temperatures we established 6 M HCl (200 μL) for 7 h at
110 °C to yield satisfying amounts of the desired dipeptides of (1a). In order to avoid diketopiperazine
formation mild derivatization conditions were used as previously reported by Pätzold et al. (2006) [7].
Due to the high-energy fragmentation in the PCI- and the EI-mode, we observed complex but
characteristic fingerprint fragmentation patterns. Additionally, the ion intensities in the milder
PCI-mode allowed distinguishing between isobaric dipeptides (Supplementary Information).
At this point of the structure elucidation a series of dipeptides as reference compounds was
synthesized to give proof of all possible retention times and fragmentation patterns of the individual
dipeptides. The sequences of the synthetic dipeptides are as follows: Xle-Ala/Ala-Xle (Supplementary
Figures S31–S38), Xle-Phe/Phe-Xle (Supplementary Figures S39–S46), and Xle-Xle (Supplementary
Figures S23–S30). With the careful analysis of the chromatograms and mass spectra in analogy to
Weygand et al. (1965) [6], we could distinguish between all isobaric dipeptides and even predict
whether Xle within the dipeptide is located at the N-terminal or C-terminal position. These
interpretations are based on the fact that only dipeptides with Xle at the C-terminus show a
characteristic fragment peak at m/z 86 in EI mode. For Xle at the N-terminal position of the dipeptides
the fragment ion m/z 183 is indicative and subsequent fragmentation of this ion allows the
differentiation between Ile (183 Da ion → 154 Da ion) and Leu (183 Da ion → 140 Da ion).
Figure 6 shows the single ion monitoring (SIM) chromatogram for the chromatographic separation
of dipeptide derivatives from in the partial hydrolyzate of (1a) on a Chirasil®-(L)-Val column. The
analysis of the fragmentation patterns proved the presence of the following dipeptides in the partial
hydrolyzate of the natural product (1a): (x)Ile-(D)Phe, (x)Ile-(D)Ala (x)Ilex-(x)Ilex, and (D)Leux-(x)Ilex
(Supplementary Figures S15–S22). The derivatized dipeptides (L)Lys1-(x)Xle2/(x)Xle8-(L)Lys1 were not
detected from the partial hydrolyzate (1a), perhaps they are not eluting from the column at the
operating temperature (195 °C) of the GC column or derivatization of lysine side chain amino group
NζH2 remained incomplete under the conditions applied. Combining the analysis from chiral
GC/PCI-EI-MS of the partial hydrolyzate of (1a) with the synthetic reference dipeptides, the partial
sequence (L)Lys1-(x)Xle2-(x)Ile3-(D)Phe4-(x)Xle5-(x)Ile6-(D)Ala7-(x)Xle8 could be derived leaving the
assignment of the dipeptides (x)Ilex-(x)Ilex and (D)Leux-(x)Ilex within the sequence unclear.

Mar. Drugs 2013, 11
Figure 6. GC-SIM-Chromatogram on a Chirasil®-(L)-Val column of champacyclin (1a)
4843
after N-trifluoroacetyl methyl ester derivatization. Rt (min): 19.97 (L)Ile-(D)Ala, 22.53
(L)Ile-(L)Ile, 24.61 (D)Leu-(L)Ile, 46.68 (L)Ile-(D)Phe. Assignment of stereochemistry with
reference dipeptides summarized in Table 1, based on the most abundant peaks in partial
hydrolyzate of (1a). Additional peaks correspond to side products from racemization of the
dipeptides during partial hydrolysis and derivatization, respectively.
Subsequently we focused on the elucidation of the remaining absolute stereochemistry of the amino
acids in dipeptides of the partial hydrolyzate of (1a) (Figure 6). To minimize the amount of dipeptides
necessary for stereochemical elucidation and to reduce difficulties in chiral separation on the
Chirasil®-(L)-Val column, we focused on the observed configurations of the amino acids according to
the previous chiral amino acid analysis after hydrolysis and N-pentafluoropropionic acid 2-propylester
derivatization (Figure 4) and considered the sequence, estimated by MS fragmentation pattern analysis
(Figure 5) and partial hydrolysis approach. Table 1 shows the synthetic and the observed
N-trifluoroacetyl dipeptide methyl ester in partial hydrolyzate of (1a) after GC-MS derivatization.
Based on Table 1, the assignment of the correct stereochemistry in Figure 6 was established only for
the most abundant peaks in partial hydrolyzate of (1a).
Dipeptides with identical retention times are shown in bold in Table 1. Unequivocal assignment
was achieved for the dipeptide derivatives [Rt (min), synthetic reference/partial hydrolysis of (1a)]:
(L)Ile-(D)Phe (46.75/46.68) as well as (L)Ile-(D)Ala (20.00/19.97). Consequently one additional building
block is (D)Leu-(L)Ile (24.75/24.61) and the last building block is (L)Ile-(L)Ile (22.55/22.53). As a
consequence we postulated (AlloD)Ile8 at position 8.

Mar. Drugs 2013, 11
Table 1. Gas-chromatographic enantioseparation of synthetic peptides and of peptides
4844
from the partial hydrolyzate of champacyclin (1a) as N-trifluoroacyl dipeptide methyl
esters on a Chirasil®-(L)-Val column.
Synthetic
Dipeptides
Retention Times Rt (min)
of Synthetic Dipeptides Partial Hydrolyzate (1a)
(L)Ile-(D)Ala
Dipeptides from the
Retention Times Rt (min) of Dipeptides
from the Partial Hydrolyzate (1a)
19.97
(L)Ile-(D)Ala
20.00
20.35
20.40
20.45
22.55
23.40
22.65
22.90
24.30
23.75
24.90
25.05
24.75
24.00
23.30
22.05
25.90
26.20
24.25
24.75
46.75
44.80
40.90
49.30
(AlloD)Ile-(D)Ala
(L)Ile-(L)Ala
(AlloD)Ile-(L)Ala
(L)Ile-(L)Ile
(L)Ile-(L)Ile
22.53
24.61
46.68
(AlloD)Ile-(L)Ile
(L)Ile-(AlloD)Ile
(AlloD)Ile-(AlloD)Ile
(L)Ile-(D)Leu
(L)Ile-(L)Leu
(AlloD)Ile-(D)Leu
(AlloD)Ile-(L)Leu
(D)Leu-(L)Ile
(D)Leu-(AlloD)Ile
(L)Leu-(AlloD)Ile
(L)Leu-(L)Ile
(D)Leu-(L)Ile
(D)Leu-(D)Leu
(D)Leu-(L)Leu
(L)Leu-(L)Leu
(L)Leu-(D)Leu
(L)Ile-(D)Phe
(AlloD)Ile-(D)Phe
(L)Ile-(L)Phe
(L)Ile-(D)Phe
(AlloD)Ile-(L)Phe
2.2.4. Nuclear Magnetic Resonance (NMR) Spectroscopy
The partial hydrolysis approach could not unambiguously resolve the positions of (L)Ilex-(L)Ilex and
(D)Leux-(L)Ilex in (1a). To overcome the uncertainty in the assignment of dedicated amino acids of the
(x)Xle-type to the amino acid sequence we performed 1D and 2D-NMR experiments in d6-DMSO on a
1
600 MHz NMR spectrometer for (1a). A full spectral data set was acquired comprising H-NMR,
1H-1H-COSY, 1H-1H-NOESY (mixing time 100 ms), 1H-1H-TOCSY (spin lock 60 ms), 1H-13C-HMBC,
1H-13C-HMQC, HMQC-TOCSY (spin lock 80 ms), and DEPT-180-1H-13C-HMQC. Interpretation of
data begun with assignment of the individual amino acids from the eight signals attributed to the eight
Hα protons in the DEPT-180-1H-13C-HMQC (Figure 7) and assignment of their correlations in the
1
1H-1H-COSY (Supplementary Figure S9) and H-1H-TOCSY spectra (Supplementary Figure S11).
The presence of eight Hα protons in the amino acid building blocks, indicates the absence of symmetry
and according to the dispersion of the chemical shift of their signals (not random coil chemical shifts),
(1a) is an unsymmetrical cyclopeptide, as already discussed previously.

Mar. Drugs 2013, 11
Figure 7. Phase-sensitive DEPT-180-1H-13C-HMQC NMR data for champacyclin (1a)
4845
measured in d6-DMSO on a 600 MHz spectrometer. Correlations from protons with
1J coupling are visible. Black: CH and CH3-groups show positive signals, Red: CH2-groups
show negative signals.
The phase-sensitive DEPT-180-1H-13C-HMQC data provided additional information on the CH3-,
CH2- and CH-groups (Figure 7). To avoid overlap of signals, as five out of eight amino acids are
(x)Xle-type amino acids, HMQC-TOCSY data were used for individual assignment (Supplementary
Figure S7). Only one single Leu-residue ((D)Leu5) at position 5 could be assigned, meaning that the
remaining four (x)Xle amino acids are (x)Ile residues, which is also in accordance with the quantitative
GC-MS results (Figure 4, Supplementary Table S6).
In the 1H-1H-COSY spectrum (Supplementary Figure S9) the (D)Ala7 Hα proton (4.28 ppm) shows a
correlation to the Hβ/β′/β″ protons of the CH3 group (1.25 ppm), while three Hα protons, (L)Lys1
(4.34 ppm), (D)Phe4 (4.33 ppm), (D)Leu5 (4.22 ppm), show correlations to the Hβ/β′ protons of their
corresponding CH2-groups (Hβ/β′ (L)Lys1 (1.55/1.40 ppm), Hβ/β′ (D)Phe4 (3.29/2.70 ppm), Hβ/β′ (D)Leu5
(1.93/1.49 ppm)). The remaining four Hα protons, (x)Ilex (4.15 ppm), (x)Ile3 (3.78 ppm),
(x)Ile6 (4.05 ppm), and (x)Ilex (4.14 ppm) show correlations to the Hβ protons of their corresponding
CH-groups (Hβ (x)Ilex (1.85 ppm), Hβ (x)Ile3 (1.47 ppm), Hβ (x)Ile6 (1.82 ppm), Hβ (x)Ilex (1.85 ppm)).
1
Due to the low amount of (1a), H-13C-HMBC data could not be used for sequence-specific
assignment of residues but helped in the unambiguous assignment of (D)Leu to position 5 (Figure 8).
Further, sequential data were assigned using the sequential walk along the peptide backbone with
intra-residual and inter-residual HN(i) ↔ Hα(i) and Hα(i) ↔ HN(I + 1) couplings derived from
1
1H-1H-COSY and H-1H-NOESY spectra, respectively (Supplementary Figures S9 and S10). Hence,

Mar. Drugs 2013, 11
4846
only a stretch of amino acid sequence on the basis of chiral analysis and the sequential walk could be
established: [(L)Ile3-(D)Phe4-(D)Leu5-(L)Ile6-(D)Ala7], leaving the complete assignment of (x)Xxx1,2,8
(Xxx = Xle, (L)Lys) by NMR within the sequence undetermined.
Figure 8. NMR-spectroscopic structure assignment of champacyclin (1a) based on
correlations between nuclei. Bold lines represent 1H-1H-COSY-, black arrows
1
1H-13C-HMBC- and blue arrows H-1H-NOESY-contacts. Proposed structure of (1a) is
also based on data from chiral GC/PCI-EI-MS and HR-ESI-(+)-MSn analysis.
The results of NMR interpretation are summarized in Table 2. The NMR results are not complete
but helped in the assignment of (D)Leu5 at position 5. Therefore, the necessity of a (x)Ile at position 8 of
the peptide sequence is a consequence of quantitative GC-MS results and the 2D NMR data. Through
exclusion, the stereochemistry of the remaining (x)Ile8 was deduced to be (AlloD)Ile8, as mentioned
before. Taking all results so far the final cyclic sequence and stereochemistry of champacyclin (1a) is
deduced as: [(L)Lys1-(L)Ile2-(L)Ile3-(D)Phe4-(D)Leu5-(L)Ile6-(D)Ala7-(AlloD)Ile8].
1
13
Table 2. H- and C-NMR data for champacyclin (1a) in d6-DMSO on a 600 MHz NMR
spectrometer at 30 °C.
Amino
Acid/Position
Lys(1)
C=O
1H NMR
δ(1H) in ppm
13C NMR
δ(13C) in ppm (mult.) *
Amino
Acid/Position
Leu(5)
C=O
1H NMR
δ(1H) in ppm
13C NMR
δ(13C) in ppm (mult.) *
−
7.40
n.d.
−
−
n.d.
−
NH
NH
7.77
Hα
Hβ/β′
Hγ/γ′
Hδ/δ′
4.34
51.5, CH
31.6, CH2
n.d.
Hα
Hβ/β′
Hγ
Hδ/δ′
4.22
52.7, CH
39.7, CH2
24.1, CH
22.9/21.4, CH2
1.55/1.40
n.d.
1.93/1.49
1.72
n.d.
n.d.
0.96/0.89
Hε/ε′
n.d.
n.d.
NH2
n.d.
n.d.

Mar. Drugs 2013, 11
4847
Table 2. Cont.
Ile(x)
C=O
NH
Hα
Hβ
Hγ
Ile(6)
−
n.d.
−
C=O
NH
Hα
Hβ
Hγ
Hγ′
Hδ
−
6.91
n.d.
−
n.d.
4.15
1.85
0.84
n.d.
n.d.
57.1, CH
35.6, CH
14.7, CH3
n.d.
4.05
57.7, CH
35.5, CH
n.d.
1.82
0.80
Hγ′
Hδ
1.23/1.13
n.d.
23.5, CH2
n.d.
n.d.
Ile(3)
C=O
NH
Hα
Hβ
Hγ
Ala(7)
C=O
NH
Hα
−
171.4
−
−
172.4
−
7.93
3.78
1.47
0.45
7.78
4.28
1.25
57.7, CH
35.5, CH
14.9, CH
24.1, CH2
10.7, CH3
47.7, CH
18.4, CH3
Hβ/β′/β′′
Hγ′
Hδ
1.22/0.83
0.72
Phe(4)
C=O
NH
Hα
Hβ/β′
Hγ
Hδ/δ′
Hε/ε′
Hζ
Ile(x)
C=O
NH
Hα
Hβ
Hγ
−
8.40
4.33
3.29/2.70
−
171.0
−
−
n.d.
n.d.
−
54.8, CH
36.6, CH2
138.3, CH
129.4, CH
128.1, CH
126.3, CH
4.14
57.9, CH
35.6, CH
n.d.
1.85
n.d.
7.20
7.26
7.36
Hγ′
Hδ
1.34/1.22
n.d.
25.3, CH2
n.d.
n.d. = not determined; * = multiplicities were deduced from DEPT-HMQC.
2.2.5. Solid Phase Peptide Synthesis as a Proof of Head-to-Tail Cyclization of (1a)
Finally we performed synthesis of head-to-tail-cyclized (1a) and head-to-side-chain-cyclized (2)
(Supplementary Figure S3) by solid phase peptide synthesis with a hydrazide linker [12] to prove the
correct cyclization topology in champacyclin (1a). Interestingly, the side chain derivative (2) shows
significant lower ion abundance for the m/z 129 b1 ion compared to the synthetic compound (1b) and
to the champacyclin (1a) (Supplementary Figures S3). This data independently supports a head-to-tail
cyclization between Nα(L)Lys1-CO-(AlloD)Ile8 in the natural product champacyclin (1a) rather than
a head-to-side-chain between NζLys1-CO-(AlloD)Ile8 (synthetic compound (2)). Very recently,
Takada et al. (2013) [13] reported the isolation and structure elucidation of a series of cyclic
octapeptides, termed surugamides A-E, from marine a Streptomyces sp. JAMM992 with similarities to
champacyclin (1a) described in this article. Interestingly, surugamide A and champacyclin (1a) share
the same amino acid residues but differ from each other in the stereochemistry of (D)Ile (surugamide A)
and (AlloD)Ile in (1a) and the amino acid sequence.

Mar. Drugs 2013, 11
4848
3. Experimental Section
Unless otherwise noted, dry solvents were purchased from Fisher Scientific-ACROS (Schwerte,
Germany). All other reagents were used as obtained (ABCR, Karlsruhe, Germany; Bachem,
Bubendorf, Switzerland; Fisher Scientific, Schwerte, Germany; IRIS Biotech, Marktredwitz, Germany;
Merck, Darmstadt, Germany; Sigma-Aldrich, Taufkirchen, Germany).
3.1. Sampling Sites
The Streptomyces strains were isolated from the following marine locations.
Streptomyces strain C42. Samples of sediment cores were taken in an anoxic water layer
(ca. 90 µmol/L H2S in the central Gotland Deep) at 241 m water depth at station AL 93 (coordinates:
57° 20.00′ N and 20° 03.00′ E). Samples were taken during the ALKOR cruise AL156 from 23 March
to 2 April 2000, by using a Rumohr lot attached to a sediment corer. The samples were kept frozen on
board at −20 °C until analysis [14].
Streptomyces strain i6a. Sediment samples were taken in 3457 m water depth with a multiple corer
at station #294 in the Urania Basin (coordinates: 35° 14.00′ N and 21° 28.60′ E) during the METEOR
cruise M44/4 to the Eastern Mediterranean, in May 1999. The samples were kept frozen at −20 °C
until analysis [15].
Streptomyces strain XX19. Samples from the sediment surface in 5 m water depth were taken by
scuba diving close to the shore line in Strande of Kiel Bight (Baltic Sea), in 2009 (coordinates:
54° 26.00′ N and 10° 10.00′ E). The sample was cooled in a freezer box until arriving in the lab on the
same day and frozen at −20 °C until analysis.
3.2. Isolation Procedure
The sediment samples (1 g wet weight each) were dissolved in sterile-filtered (0.2 µm pore size)
Baltic Sea water (5 mL) containing 1.5%–3% sodium chloride in dependence of sample origin (sodium
chloride was additionally added to achieve 3%). The undiluted (100 µL ) and the 10-fold diluted
sample (100 µL ) were transferred to agar plates and incubated at 28 °C for 3 month. The growth
medium (GYM) contained 4 g yeast extract (BD-Bacto™, Becton Dickinson and Company, Sparks,
MD, USA), 4 g D-(+)-glucose-monohydrate (VWR, Darmstadt, Germany), 4 g malt extract
(BD-Bacto™) and 15 g agar (BD-Bacto™) in 1 L deionized water; after autoclaving the pH was 7.2.
Single colonies were transferred to sterile agar plates and grown at 28 °C for 7–14 days. This
procedure was repeated until pure cultures were obtained. Subsequently, purified colonies were
phylogenetically identified by 16S rDNA gene sequencing.
3.3. Phylogenetic Classification
The phylogenetic classification of the strains C42, i6a, and XX19 was performed by sequencing of
the 16S rRNA gene to determine the closest phylogenetic relative. For DNA analyses the DNA Kit
PuReTaq™Ready-To-Go™ PCR Beads containing Taq-polymerase 0.5 U; 2.5 µL buffer and 0.5 µL
dNTPs for each reaction were used. Amplification of the 16S rRNA gene (1500 bp) was performed

Mar. Drugs 2013, 11
4849
with the primer pairs (concentration: each 10 pmol/µL ): Eub27f with the sequence: 5′-ATT GGA TCC
GTT TGA TCM TGG CTC AG-3′ (primer 1) (Noda et al., 2006) [16]/Univ1492r with the sequence
5′-CGG TTA CCT TGT TAC GAC TT-3′ (primer 2) (Reysenbach et al., 2000) [17]. The code “M”
represents the bases C and A. From the total 25 µL PCR product 2.5 µL was transferred on a 1%
agarose gel and separated at 200 mV for 20 min. The fragment length was about 1500 bp as
determined with the DNA marker X (0.07–12.2 kbp). The sequencing procedure was performed
according to Wiese et al. (2009) [18]. Storage of the PCR products was at −20 °C. The 16S rRNA
sequences of the three isolates C42, i6a, and XX19 were edited using BioEdit (Hall, 1999) [19]. NCBI
database (BLAST tool) and the Ribosomal Database Project (RDP). The Ribosomal Database Project
database was used to select the type strains most closely related to our new isolates (Cole et al.,
2009) [20]. Sequence similarity values were determined with the “bl2seq” tool of the NCBI database
(Tatsuova and Madden, 1999) [21]. The alignment of the 16S rRNA gene sequences (1375 bp) of the
three isolates, of representatives of the Streptomyces albidoflavus group, of further type strains of
the genus Streptomyces, and of Micrococcus lylae as outgroup was performed with Clustal X
(Larkin et al., 2007) [22] and refined manually. The phylogenetic tree was calculated with Molecular
Evolutionary Genetics Analysis (MEGA5) by maximum likelihood analysis assuming the general time
reversal (GTR) model and applying 500 bootstrap replicates (Tamura et al., 2011) [23]. NJplot was
used to display the phylogenetic tree (Perrière & Goug, 1996) [24].
3.4. PCR for Non-Ribosomal Peptide Synthetases (NRPS)
The amplification of the NRPS gene cluster was performed with the primers of the adenylation
domain A2f (5′-AAG GCN GGC GSB GCS TAY STG CC-3′) and A3r (5′-TTG GGB IKB CCG GTS
GIN CCS GAG GTG-3′) (Doekel and Marahiel, 2001) [25]. The codes “N”, “S”, “B”, and “Y”
represent “unknown base”, G and C, “not A”, and pyrimidine C and T, respectively. The expected
fragment length was about 300 bp of the PCR product, which was determined with the DNA Marker
XIV (100 bp–1500 bp). From the total 25 µL PCR product, 5 µL was transferred on a 1.5% agarose
gel and separated at 130 mV for 35 min. The permanent storage of the PCR products was −20 °C.
3.5. Optimization Experiments to Increase Champacyclin (1a) Synthesis
In order to increase the yield of champacyclin, growth conditions of strain C42 were optimized. In
each experiment all basic parameters were maintained except the single variable parameter in the
experiment: 100 mL culture volume, actinomycetes medium, except in the salt experiments, GYM
medium was used, 1.5% salt (NaCl), 6 days incubation time, 28 °C incubation temperature, 125 rpm
shaking and methanolic extraction.
Five different media were used for cultivation of the strain.
I. TSB—Trypticase Soy Broth medium contained 3 g soybean casein digest broth (BD-BBL™
Trypticase™ Soy Broth, Dickinson and Company, Sparks, MD, USA), 10 g tropic marine salt
in 1 L deionized water.
II. OAK—Oak Flakes medium contained 20 g oak flakes and 3 mL trace element solution SL12
(3 g EDTA sodium salt, 42 mg ZnCl2, 50 mg MnCl2-tetrahydrate, 300 mg H3BO3, 190 mg

Mar. Drugs 2013, 11
4850
CoCl2-hexahydrate, 1100 mg FeSO4 heptahydrate, 2 mg CuCl2-dihydrate, 24 mg
NiCl2-hexahydrate, and 18 mg Na2Mo4-dihydrate in 1 L bidestilled water). The trace element
solution was sterile filtered and added to the medium after autoclaving. The pH was 6.3 after
autoclaving without adjustment.
III. MIC—Micromonospora medium contained 10 g starch (VWR, Darmstadt, Germany), 5 g
D-(+)-glucose-monohydrate (VWR), 4 g casein sodium salt from bovine milk (Sigma-Aldrich
Chemie GmbH, Munich, Germany), 4 g yeast extract (BD-Bacto™, Dickinson and Company,
Sparks, MD, USA), and 1 g calcium carbonate (Sigma-Aldrich Chemie GmbH) in 1 L
bidestilled water. The pH was adjusted to 7.2 before autoclaving.
IV. ACT—Actinomycetes medium: contained 2 g casein sodium salt from bovine milk
(Sigma-Aldrich Chemie GmbH, Munich, Germany), 4 g propionate acid sodium salt
(Sigma-Aldrich Chemie GmbH), 0.1 g magnesium sulphate heptahydrate (VWR, Darmstadt,
Germany), L-asparagine monohydrate (Roth, Karlsruhe, Germany), 0.5 g di-potassium
hydrogenphosphate (VWR), 0.001 g ferric(III)-sulphate hydrate (Sigma-Aldrich Chemie
GmbH), and 5 g glycerol (VWR) in 1 L sterile filtered Baltic Sea water (1.5% salt content).
The pH was adjusted to 8.1 prior to autoclaving.
V. GYM—GYM medium contained 4 g glucose, 4 g yeast extract, 4 g malt extract, 2 g CaCO3,
30 g tropic marine salt in 1 L bidestilled water. The pH of 7.2 was measured after autoclaving.
Salt concentration. Different salt contents in the medium were added to obtain the best growth
conditions for the strain C42 to produce champacyclin. Experiments with 0%, 1%, 2%, 3%, and 4%
NaCl, respectively, in the GYM medium were performed at an incubation temperature of 28 °C.
Incubation time. A time series was performed to determine the optimal growth time of strain C42
for champacyclin synthesis. The time series were performed for eight days with 100 mL culture
aliquots taken for analysis from day four to eight as the first analysis in actinomycetes medium.
Temperature. The temperature optimum was determined in relation to the biomass at 20 °C, 24 °C ,
and 28 °C in actinomycetes medium.
Influence of amino acids. Addition of the amino acids of L-leucine, L-isoleucine, L-phenylalanine,
and L-valine (each 1 mg/mL) in actinomycetes medium was performed to upgrade the
champacyclin product.
Cultivation, extraction, and isolation of champacyclin (1a). The culture conditions that yielded
both, the highest amounts of champacyclin and the lowest yield of heptaenes, were used for production
and purification of the peptide as described as follows. Erlenmeyer flasks (2000 mL) with four baffles
were filled with 1000 mL actinomycetes medium in 1 L bidestilled water. Sterile filtered (0.2 µm pore
size, non-pyrogenic) L-leucine, L-isoleucine, and L-phenylalanine (1 mg/mL each) were added into the
medium to increase the champacyclin production. For the extraction of champacyclin, the cells were
separated from the medium by centrifugation (11,000× g, 20 min) and extracted with methanol. Cell
lysis was performed with a cell disruption apparatus (Ultra Turrax, T25 basic, IKA®-Werke GmbH &
Co. KG, Staufen, Germany) at 13,000 rpm for 30 s and shaken for 30 s to dissolve the oligopeptides.
The methanol extract was evaporated in a rotary evaporator at 40 °C. The dry extract was dissolved in
2 mL methanol and filtered through a 0.2 µm PTFE filter and chromatographed on a reversed-phase

Mar. Drugs 2013, 11
4851
HPLC with a C18 column (250 × 4.6 mm, 5 μm, 120 Å, ODS-AQ, YMC Europe, Dinslaken,
Germany). A linear gradient of acidified (0.1% formic acid) of deionized H2O (A), and CH3CN (B)
was used for separation (30% to 35% B in 10 min, 35% to 70% B in 15 min, 70% to 100% B in 2 min)
at 40 °C and a flow rate of 1 mL/min. Champacyclin (1a) was eluted under these conditions at a
retention time of 19 min on an Elite LaChrom HPLC (VWR, Hitachi, Darmstadt, Germany). The
culture medium was extracted with ethyl acetate through phase separation and champacyclin (1a) was
purified as described above.
Physicochemical Properties of Champacyclin (1a). UV/VIS (MeOH) λmax (ε): 258 nm (1523);
1
HR-ESI-(+)-Orbitrap-MS m/z 912.62574 [M + H]+ (relative mass error Δm = −2.572 ppm); H and
13C-NMR data from 2D NMR experiments in d6-DMSO, see Supplementary Table S5.
3.6. Determination of the Molar Absorption Coefficient
The molar absorption coefficient of champacyclin (1a) (50 μg in methanol) was determined by
14
quantitative analysis of the compound constituent L-12C9 N-phenylalanine after acidic hydrolysis
15
(6.0 M HCl at 110 °C for 20 h). The internal standard L-13C9 N-phenylalanine (1 μg/µL ) (Isotec,
Miamisburg, Ohio, OH, USA) was added before acidic hydrolysis. After derivatization of
the dry hydrolyzate with N-methyl-N-(tert-butyldimethylsilyl)-trifluoroacetamide (MTBSTFA) in
tetrahydrofuran (THF) and trifluoroacetic acid (TFA) (50/50/0.1; v/v/v), the solution was analyzed on
a GC/EI-MS (Agilent Technology, Waldbronn, Germany). Analyses were conducted on a capillary
column (30 m DB-1301, 0.32 mm i.d., 0.25 μm film thickness) under the following separation
conditions: 1 min at 120 °C, 120 to 250 °C at a rate of 10 °C/min. The integrals of the fragment ions at
m/z 234 (30%), m/z 308 (15%), m/z 336 (11%), and m/z 243, m/z 317, and m/z 346 of L-12C9
15
14N-phenylalanine and L-13C9 N phenylalanine, respectively, were applied to determine the amount of
non-labeled amino acids. This value was correlated to the UV absorption of champacyclin at 258 nm
in methanol measured before acidic hydrolysis.
3.7. Antimicrobial Activity
Antimicrobial bioassays were performed against Gram-positive and Gram-negative bacteria and
yeast using the following test organisms: Bacillus subtilis (DSM 347), Escherichia coli K-12 (DSM
498), Staphylococcus lentus (DSM 6672), Pseudomonas syringae pv. aptata (DSM 50252),
Pseudomonas fluorescens (NCIMB 10586), Xanthomonas campestris (DSM 2405), Ralstonia
solanacearum (DSM 9544), Candida glabrata (DSM 6425), and Erwinia amylovora (DSM 50901).
Assay mixtures were prepared by transferring 10 µL aliquots of methanolic solutions of extracts or
pure substances into three wells of a 96-well microtiter plate and evaporating the solvent in a vacuum
centrifuge. Overnight cultures of the test organisms in TSB medium (BD-BBL™ Trypticase™ Soy
Broth (Dickinson and Company, Sparks, MD, USA) 12 g/L; NaCl, 5 g/L) were diluted to an optical
density at 600 nm (OD600) of 0.001, and 200 µL aliquots of the resulting suspensions were added to the
wells. After incubation for 15 h at 600 rpm and 28 °C, 10 µL of a resazurin solution (0.2 mg/mL
phosphate-buffered saline, pH 7.2) was added to each well and the plate was incubated at 28 °C until
the color of the negative control changed from blue to pink. For evaluation of the cell viability, the

Mar. Drugs 2013, 11
4852
transformation of resazurin to resorufin was assayed by measuring the fluorescence at 560 nm after
excitation at 590 nm. The resulting values were compared to those for a positive control (100 µg/well
chloramphenicol for the bacteria, 200 µg/well cycloheximide for yeast) and a negative control (no
compound) on the same plate.
3.8. Mass Spectrometry (MS)
HPLC-HR-ESI-(+)-Orbitrap-(SID)-MSn was performed on an Orbitrap LTQ XL mass spectrometer
(Thermo Fisher Scientific, Waltham, MA, USA) coupled to an Agilent 1260 HPLC-system (Santa
Clara, CA, USA). In Source Fragmentation (SID) and subsequent MSn-analyses were performed using
direct infusion with a concentration 0.1 mg/mL The flow was adjusted between 1 and 20 µL /min to
obtain high abundance. Further parameters for the analyses are described in the supporting material.
Low-Resolution-MS data of the partial hydrolyzate (1a) were performed on an
ESI-Triple-Quadrupol-MS 6460 (Agilent Technologies, Waldbronn, Germany) coupled to an
Agilent 1290 UHPLC-system (Santa Clara, CA, USA), with the parameters described in the
Supplementary Information.
3.9. Nuclear Magnetic Resonance (NMR) Spectroscopy
1H-NMR and 2D-NMR experiments data were acquired on a 600 MHz NMR spectrometer (Bruker,
Karlsruhe, Germany) in d6-DMSO (500 μL), equipped with a TXI probe head with Z-Gradient. All
chemical shifts were calibrated on the residual solvent peak (d6-DMSO, 2.50 ppm (1H) and 39.5 ppm
(13C)). The chemical shifts, reported in delta (δ) units, in parts per million (ppm) are referenced
relatively to TMS. Additional spectra and parameter are described in NMR section in the
Supplementary Information.
3.10. Hydrolysis of (1a)
3.10.1. Partial Hydrolysis
Partial hydrolysis of (1a) (0.1 mg) was performed at 110 °C in aqueous 6 M hydrochloric acid
solution for amino acid analysis (200 µL ) (Sigma Aldrich, Schnelldorf, Germany) for 7 h in
glass-ampoules. After 7 h, the sample was freeze-dried prior derivatization for GC/MS analysis. The
status of the reaction was measured by mass spectrometry.
3.10.2. Total Hydrolysis
Total hydrolyses of (1a) (0.1 mg) were performed at 110 °C in aqueous 6 M hydrochloric acid
solution (200 µL ) for amino acid analysis (Sigma Aldrich, Schnelldorf, Germany) under vacuum for
24 h in glass-ampoules. After 24 h, the hydrochloric acid was removed in a gentle stream of nitrogen.
3.11. Chiral GC/EI-PCI-MS Analysis
GC/EI-PCI-MS analyses were performed on a GC/MS 5975C (Agilent Technologies, Waldbronn,
Germany). Aliquots of derivatized amino acids/dipeptide of 0.2 to 5 µL were injected to the GC/MS

Mar. Drugs 2013, 11
4853
5975C (Agilent Technologies, Waldbronn, Germany), equipped with a Chirasil®-(L)-Val column.
Parameters and conditions are described in supporting information. Further EI- and PCI-MS spectra of
all dipeptides analyzed of partial hydrolyzate and synthetic dipeptides are also shown in the
Supplementary Information.
3.11.1. Amino Acid Analysis as N-Pentafluoropropionic 2-Propyl Amino Acid Derivatives
Amino acid standards (<0.1 mg/mL) and dried total hydrolyzate of (1a) were suspended in a total
volume of 200 µL acetylchloride:2-propanol (v:v, 150:50 µL ) in Reacti-Vials sealed and derivatized
for 30 min at 110 °C. Excess of solvents was removed in a gentle steam of nitrogen. To the dry
residues 50 µL pentafluoropropionic anhydride (PFPA) and 100 µL dichloromethane were added and
derivatized for 15 min at 110 °C. Excess of reagents was evaporated in a stream of nitrogen and
dissolved in 100 µL DCM.
3.11.2. Dipeptide Analysis as N-Trifluoroacetyl Methyl Dipeptide Derivatives
Partial hydrolyzed (1a) and synthetic dipeptides were dried before derivatization. The protocols
were mainly adapted by Pätzold et al. (2006) [7].
To the samples a total volume of 360 µL acetyl chloride:methanol (v:v, 60:300 µL ) were added in
Reacti-Vials sealed and derivatized for 3 h at room temperature (21 °C). The excess of reaction
mixture was evaporated in a gentle steam of nitrogen. To the dry residues 50 µL trifluoroacetic
anhydride (TFAA) and 100 µL dichloromethane (DCM) were added and derivatized for 10 min at
room temperature (21 °C) and then 5 min at 110 °C. The samples were cooled down to room
temperature (21 °C) and again dried in a steam of nitrogen and finally dissolved in 100 µL DCM.
3.12. Solid Phase Peptide Synthesis (SPPS)
Standard Fmoc-protected amino acids were purchased from Orpegen (Heidelberg, Germany).
Fmoc-(AlloD)Ile-OH and Boc-(L)Lys(Fmoc)-OH were bought from Iris Biotech (Marktredwitz,
Germany). The coupling reagents were obtained from Iris Biotech (Marktredwitz, Germany). The TCP
resin was bought from Intavis (Reutlingen, Germany) and the 4-Fmoc-hydrazinobenzoyl AM Novagel
resin for cyclizing cleavage was bought from Merck (Darmstadt, Germany). DMF (99.8%) was
purchased from VWR (Darmstadt, Germany). The chemicals for peptide cleavage were obtained from
Sigma-Aldrich (Schnelldorf, Germany) and ABCR (Karlsruhe, Germany). Automated solid phase
Fmoc-synthesis was performed on a Prelude parallel synthesizer by Protein Technologies (Tucson,
AZ, USA).
3.12.1. Synthesis of Dipeptides
Synthesis protocol for automated solid phase peptide synthesis: Automated solid-phase peptide
synthesis was performed in 50 µmol scale. Loading: To a 10 mL syringe reactor with frit and cap were
added 1 g of tritylchloride (TCP) resin (1.56 mmol/g) and 7 mL dry DCM. The resin was pre-swollen
for 10 min and the solvent was removed by evaporation in vacuum. A mixture of the amino acid
(0.6 mmol) and 3 equivalents of DIPEA dissolved in 5 mL dry DCM was added to the resin. The

Mar. Drugs 2013, 11
4854
syringe was agitated for 30 min at room temperature. The solution was removed and the resin was
washed (2 × 5 mL DMF, 2 × 5 mL DCM). Capping of non-reacted functional groups of the resin was
performed with DCM, methanol and DIPEA 80:15:5 (2 × 10 mL, 10 min). After washing (5 × 5 mL
DMF), Fmoc-removal was achieved with DMF/piperidine (4:1, 5 mL, 1 × 2 min, 1 × 20 min). After
final washing (2 × 5 mL DMF, 1 × 5 mL methanol, 3 × 5 mL DCM), the resin was dried in vacuo.
Coupling of Fmoc/tBu-protected amino acids: To 100 mg of the resin (~0.5 mmol/g), a 0.15 M
solution of the amino acid in DMF (3 eq relative to resin loading) was added. After addition of a 0.3 M
solution of DIPEA in DMF (3 eq) and a 0.15 M solution of HATU in DMF (3 eq), the reaction
solution was mixed for 60 min. A second coupling was performed for 60 min. Finally, the resin was
washed with DMF (6 × 2.5 mL).
Fmoc removal: DMF/piperidine (4:1, 2.5 mL) was added to the resin and mixed for 2.5 min. The
procedure was repeated 4 times. The resin was washed with DMF (5 × 2.5 mL), then DCM (5 × 2 mL).
Global deprotection: The resin was transferred to a 5 mL syringe with frit and cap. After addition
of the cleavage cocktail (TFA, H2O 90:10), the syringe was shaken for 2 h. The cleaving solution was
collected and the resin was washed with MeOH (2 × 3 mL). The combined fractions were concentrated
in vacuo.
3.12.2. Synthesis of Head-to-Tail (1b) between Nα(L)Lys1-CO-(AlloD)Ile8 and Head-to-Side-Chain (2)
between Nζ-(L)Lys-CO-(AlloD)Ile8 Cyclized Peptides for Mass Spectrometry
Synthesis protocol for automated solid-phase peptide synthesis: Automated solid-phase peptide
synthesis was performed in 64 µmol scale. Loading: 100 mg of the hydrazine resin was deprotected
(DMF/piperidine 4:1, 2.5 mL, 3 × 3 min), and then washed with DMF (6 × 2 mL). The first amino acid
was loaded with standard coupling procedure.
Coupling of Fmoc/tBu-protected amino acids: To 100 mg of the resin (0.64 mmol/g), a 0.25 M
solution of the amino acid in DMF (4 eq relative to resin loading) was added. After addition of a 0.5 M
solution of DIPEA in DMF (4 eq) and a 0.25 M solution of TBTU in DMF (4 eq), the reaction solution
was mixed for 30 min. A second coupling was performed for 30 min. Finally the resin was washed
with DMF (6 × 2.5 mL).
Fmoc removal: DMF/piperidine (4:1, 2.5 mL) was added to the resin and mixed for 2.5 min. The
procedure was repeated 4 times. The resin was washed with DMF (6 × 2.5 mL). After the final
coupling cycle, the resin was washed with DCM (5 × 2 mL).
Cyclising cleavage: The resin was transferred to a 5 mL syringe with frit and cap. A mixture of
2,6-O-dichlorobenzene and pyridine (1:1, 3 mL) was saturated with Cu(OAc)2 and added to the resin.
The syringe was shaken for 3 days. The solution was poured into 1 M aqueous HCl (5 mL) and the
resin was washed with DCM (3 × 3 mL). The layers were separated and the aqueous layer was
extracted with DCM (3 × 5 mL). The combined organic layers were concentrated in vacuo.

Mar. Drugs 2013, 11
4855
4. Conclusions
The fermentation, isolation and structure elucidation of the marine derived cyclooctapeptide
champacyclin (1a) was described in this contribution. According to NMR spectroscopy measurements,
partial and total hydrolysis followed by derivatization and GC-MS analytics as well as
ESI-IonTrap-MSn experiments, the sequence and the absolute stereochemistry of the Xle rich cyclic
peptide could be elucidated as [(L)Lys1-(L)Ile2-(L)Ile3-(D)Phe4-(D)Leu5-(L)Ile6-(D)Ala7-(AlloD)Ile8] (1a) with a
suggested head-tail cyclization (Nα(L)Lys1-CO-(AlloD)Ile8). The use of different analytical methods
including synthesis of dipeptides was necessary to unambiguously assign the sequence and
configuration of the peptide (1a). While the linear sequence could be partially elucidated by the
2D-NMR data, the interpretation of MSn-experiments was a challenge due to the high number of Leu
and Ile residues. Only the synthesis of dipeptides combined with the use of GC-PCI/EI-MS and
total/partial hydrolysis studies finally validated the NMR results and further enabled the stereospecific
assignment within the sequence.
Acknowledgments
This study was supported by the BMBF program BIOKATALYSE 2021 (to J.I.) and by the Cluster
of Excellence (“Unifying concepts in catalysis”—UniCat coordinated by TU Berlin, Germany) granted
by the German Research Council (DFG) (to R.D.S.). The authors are grateful to T. Rahn for excellent
laboratory work and A. Erhard for performing the bioactivity assays. We thank M. B. Schilhabel
(Institute for Clinical Molecular Biology, University Hospital Schleswig-Holstein, Kiel, Germany) and
his team for sequencing. Peter Schmieder (Leibniz-Institute of Molecular Pharmacology, Berlin,
Germany) and his team are greatly acknowledged for performing NMR experiments.
Conflicts of Interest
The authors declare no conflict of interest.
References
1. McIntosh, J.A.; Donia, M.S.; Schmidt, E.W. Ribosomal peptide natural products: bridging the
ribosomal and non-ribosomal worlds. Nat. Prod. Rep. 2009, 26, 537–559.
2. Bockhus, A.T.; McEwen, C.M.; Lokey, R.S. Form and function in cyclic peptide natural products:
a pharmacokinetic perspective. Curr. Top. Med. Chem. 2013, 13, 821–836.
3. Harrison, A.G.; Young, A.B.; Bleiholder, C.; Suhai, S.; Paizs, B. Scrambling of sequence
information in collision-induced dissociation of peptides. J. Am. Chem. Soc. 2006, 128,
10364–10365.
4. Kjeldsen, F.; Haselmann, K.F.; Sørensen, E.S.; Zubarev, R.A. Distinguishing of Ile/Leu amino
acid residues in the PP3 protein by (hot) electron capture dissociation in Fourier transform ion
cyclotron resonance mass spectrometry. Anal. Chem. 2003, 75, 1267–1274.
5. Armirotti, A.; Milo, E.; Damonte, G. How to discriminate between leucine and isoleucine by low
energy ESI-TRAP MSn. J. Am. Soc. Mass Spectrom. 2007, 18, 57–63.

Mar. Drugs 2013, 11
4856
6. Weygand, F.; Prox, A.; Fessel, H.H.; Kun Sun, K. Sequenzanalyse von Peptiden als
N-Trifluoracetyl-peptid-ester. Z. Naturforschg. 1965, 20, 1169–1182.
7. Pätzold, R.; Theis, C.; Brückner, H. Gas-chromatographic separation of steroisomers of
dipeptides. Chirality 2006, 18, 551–557.
8. Narasimha Rao, P.L.; Uma, B.N. Formation of antifungal antibiotic by a new Streptomyces
species. Nature 1958, 182, 115–116.
9. Rao, U.K.; Narasimha Rao, P.L. Actinomycetes. II. Purification and pharmacological properties
of champamycin-B from Streptomyces champavatii. Indian J. Exp. Biol. 1967, 5, 39–42.
10. Rong, X.; Guo, Y.; Huand, Y. Proposal to reclassify the Streptomyces albidoflavus clade on the
basis of multilocus sequence analysis and DNA-DNA hybridization, and taxonomic elucidation of
Streptomyces griseus subsp. solvifaciens. Syst. Appl. Microbiol. 2009, 32, 314–322.
11. Roepstorff, P.; Fohlman, J. Proposal for a common nomenclature for sequence ions in mass
spectra of peptides. Biomed. Mass Spectrom. 1984, 11, 601.
12. Rosenbaum, C.; Waldmann, H. Solid phase synthesis of cyclic peptides by oxidative cyclative
cleavage of an aryl hydrazide linker—synthesis of stylostatin 1. Tetrahedron Lett. 2001, 42,
5677–5680.
13. Takada, K.; Ninomiya, A.; Naruse, M.; Sun, Y.; Miyazaki, M.; Nogi, Y.; Okada, S.; Matsunaga, S.
Surugamides A–E, Cyclic Octapeptides with Four D-Amino Acid Residues, from a Marine
Streptomyces sp.: LC-MS-Aided Inspection of Partial Hydrolysates for the Distinction of D- and
L-Amino Acid Residues in the Sequence. J. Org. Chem. 2013, 78, 6746–6750.
14. Fan, L.; Liu, Y.; Li, Z.; Baumann, H.I.; Kleinschmidt, K.; Ye, W.; Imhoff, J.F.; Kleine, M.;
Cai, D. Draft genome sequence of the marine Streptomyces sp. strain PP-C42, isolated from the
Baltic Sea. J. Bacteriol. 2011, 193, 3691–3692.
15. Pätzold, J.; Halbach, P.E.; Hempel, G.; Weikert, H. METEOR-Berichte 00-3. Östliches
Mittelmeer—Nördliches Rotes Meer 1999. Cruise No. 44. 22 January–16 May 1999; Leitstelle
Meteor: Hamburg, Germay, 1999.
16. Noda, S.; Inoue, T.; Hongoh, Y.; Kawai, M.; Nalepa, C.A.; Vongkaluang, C.; Kudo, T.; Ohkuma,
M. Identification and characterization of ectosymbionts of distinct lineages in Bacteroidales
attached to flagellated protists in the gut of termites and a wood-feeding cockroach. Environ.
Microbiol. 2006, 8, 11–20.
17. Reysenbach, A.L.; Longnecker, K.; Kirshtein, J. Novel bacterial and archaeal lineages from an
in situ growth chamber deployed at a Mid-Atlantic Ridge hydrothermal vent. Appl. Environ.
Microbiol. 2000, 66, 3798–3806.
18. Wiese, J.; Thiel, V.; Nagel, K.; Staufenberger, T.; Imhoff, J.F. Diversity of antibiotic-active
bacteria associated with the brown alga Laminaria saccharina from the Baltic Sea. Mar.
Biotechnol. 2009, 11, 287–300.
19. Hall, T.A. BioEdit: A user-friendly biological sequence alignment editor and analysis program for
Windows 95/98/NT. Nucleic Acids Symp. Ser. 1999, 41, 95–98.
20. Cole, J.R.; Wang, Q.; Cardenas, E.; Fish, J.; Chai, B.; Farris, R.J.; Klam-Syed-Mohideen, A.S.;
McCarell, D.M.; Marsh, T.; Garrity, G.M.; et al. The Ribosomal Database Project: improved
alignments and new tools for rRNA analysis. Nucleic Acids Res. 2009, 37, D141–D145.

Mar. Drugs 2013, 11
4857
21. Tatsuova, T.A.; Madden, T.L. BLAST 2 Sequences, a new tool for comparing protein and
nucleotide sequences. FEMS Microbiol. Lett. 1999, 174, 247–250.
22. Larkin, M.A.; Blackshields, G.; Brown, N.P.; Chenna, R.; McGettigan, P.A.; McWilliam, H.;
Valentin, F.; Wallace, I.M.; Wilm, A.; Lopez, R.; et al. Clustal W and Clustal X version 2.0.
Bioinformatics 2007, 23, 2947–2948.
23. Tamura, K.; Peterson, D.; Peterson, N.; Stecher, G.; Nei, M.; Kumar, S. MEGA5: Molecular
evolutionary genetics analysis using maximum likelihood, evolutionary distance and maximum
parsimony methods. Mol. Biol. Evol. 2011, 28, 2731–2739.
24. Perrière, G.; Goug, M. WWW-Query: An on-line retrieval system for biological sequence banks.
Biochemie 1996, 78, 364–369.
25. Doekel, S.; Marahiel, M.A. Biosynthesis of natural products on modular peptide synthetases.
Metab. Eng. 2001, 3, 64–77.
© 2013 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article
distributed under the terms and conditions of the Creative Commons Attribution license
(http://creativecommons.org/licenses/by/3.0/).