Homotyrosine-containing cyanopeptolins 880 and 960 and anabaenopeptins 908 and 915 from Planktothrix agardhii CYA 126/8

Article abstract of DOI:10.1021/np800557m

Two homotyrosine-bearing cyanopeptolins are described from Planktothrix agardhii CYA 126/8. The compounds feature a common homotyrosine-containing cyclohexadepsipeptide and differ by sulfation of an exocyclically located 2-O-methyl-D-glyceric acid residue. In addition we describe two anabaenopeptins, which contain two homotyrosine residues, one of which is N-methylated. The anabaenopeptins have a common cyclopentapeptide portion and differ in the amino acid linked to it via an ureido bond, arginine and tyrosine, respectively.

Full text of DOI:10.1021/np800557m

172
J. Nat. Prod. 2009, 72, 172–176
Homotyrosine-Containing Cyanopeptolins 880 and 960 and Anabaenopeptins 908 and 915 from
Planktothrix agardhii CYA 126/8
Hilary S. Okumura,† Benjamin Philmus,† Cyril Portmann,†,‡ and Thomas K. Hemscheidt*,†,§
Department of Chemistry, UniVersity of Hawaii, 2545 McCarthy Mall, Honolulu, Hawaii 96822, and Natural Products & Cancer Biology
Program, Cancer Research Center of Hawaii, 651 Ilalo Street, Honolulu, Hawaii 96813
ReceiVed September 3, 2008
Two homotyrosine-bearing cyanopeptolins are described from Planktothrix agardhii CYA 126/8. The compounds feature
a common homotyrosine-containing cyclohexadepsipeptide and differ by sulfation of an exocyclically located 2-O-
methyl-D-glyceric acid residue. In addition we describe two anabaenopeptins, which contain two homotyrosine residues,
one of which is N-methylated. The anabaenopeptins have a common cyclopentapeptide portion and differ in the amino
acid linked to it via an ureido bond, arginine and tyrosine, respectively.
We recently isolated two homotyrosine-containing microcystins,
MC-HtyY and MC-HtyHty, from laboratory cultures of a unique
isolate of the limnic cyanobacterium Planktothrix rubescens from
Lake Schwarzensee in Austria.1 We also reported a comparison of
the predicted amino acid sequence of the adenylation domain of
McyC of the microcystin biosynthetic gene cluster of this isolate
of P. rubescens with the McyC adenylation domain of the cluster
from Planktothrix agardhii CYA 126/8, a strain only known to
produce microcystin-RR and a minor amount of microcystin-LR
in laboratory culture. We observed that the amino acid sequences
of the adenylation domains of these two species are identical with
the exception of one amino acid residue.1 This change is located
in a region of the protein backbone that is part of the amino acid
specificity-conferring region of the adenylation domain in the
accepted model.2,3 However, given the chemical difference between
homotyrosine (Hty) and arginine (R), the difference in the McyC
sequences appears to be an insufficient molecular basis for the
incorporation of arginine in position 4 of the MCs in P. agardhii
versus homotyrosine/tyrosine in the same position in P. rubescens.
We considered several possible explanations for the observation
that the MC metabolite spectrum differed very strongly between
these two species despite the high degree of genetic identity of the
gene clusters. We hypothesized that P. agardhii CYA 126/8 might
be unable to synthesize the nonproteinogenic amino acid homoty-
rosine in variance to this specific isolate of P. rubescens. Literature
review in mid-2004 using Medline and SciFinder using the search
term “homotyrosine” returned only three hits in conjunction with
either Planktothrix or Oscillatoria sp., of which only two described
defined compounds.4,5 We therefore elected to feed synthetic
homotyrosine to laboratory cultures of P. agardhii 126/8 in an
attempt to alleviate this putative biosynthetic deficiency. Extracts
of cultures of P. agardhii to which homotyrosine had been added
were screened by LCMS for production of homotyrosine-containing
microcystins. Since strain 126/8 produces MC-RR and MC-LR
under standard laboratory culture conditions, one would therefore
expect the production of MC-RHty or MC-LHty if McyC were
capable of activating Hty in a successful precursor-directed bio-
synthesis experiment. However, using selected ion monitoring we
did not obtain any evidence for the production of these two putative
microcystins. Instead, HPLC analysis indicated increases in the
concentrations of metabolites more polar than would be expected
for these putative MCs. These compounds were shown on the basis
of more detailed NMR-based screening to contain para-disubstituted
phenols, as expected for homotyrosine- or tyrosine-containing
compounds. In this report we describe several new natural products
belonging to the cyanopeptolin and anabaenopeptin groups of
cyanobacterial peptides that were isolated.
DL-Homotyrosine was prepared by a modification of a known
route starting from aspartic acid.6 The somewhat cumbersome and
messy final deprotection step of the literature procedure using HBr
in acetic acid was replaced by one employing iodotrimethylsilane
generated in situ from NaI and TMS-Cl in anhydrous acetonitrile.
The product was isolated as the zwitterion after ion exchange
chromatography on Dowex-50.
This compound was supplied to a growing culture of P. agardhii
over the course of two weeks, and the crude peptide fraction was
isolated from the resulting cell mass by extraction with 50% aqueous
MeOH followed by C18 flash chromatography and HPLC. The
major noticeable difference in the HPLC chromatogram of the
extract of this culture when compared to a control was a substantial
increase in the intensity of one peak eluting significantly after a
MC-LR standard. 1H NMR analysis of this peak indicated the
presence of peptidic substances containing a para-disubstituted
aromatic ring, consistent with the presence of either homotyrosine
or tyrosine. Further purification of this fraction on a different C18
stationary phase resulted in two chromatographically homogeneous,
structurally closely related compounds. Structure elucidation by
interpretation of high-field NMR data resulted in two new members
of the cyanopeptolin group of compounds as shown below.
Cyanopeptolin 880 (1)7,8 was isolated as an optically active solid,
which yielded a [M + H]+ pseudomolecular ion at m/z 881.4645
upon HRESITOFMS analysis, consistent with a molecular formula
of C45H64N6O12 for 1. A prominent fragment ion at m/z 863.4533
[M + H+ - H2O]+ corroborated this molecular formula (calcd
863.4508, ∆ ) 2.5 mmu). Analysis of 13C NMR and gHSQC data
suggested the presence of seven methyl, eight methylene, 16
methine, and 10 quaternary carbon atoms. The 1H NMR data
indicated the presence of a para-disubstituted and a monosubstituted
phenyl ring. These structural features as well as seven carbonyl
resonances accounted for 15 of the 17 degrees of unsaturation and
suggested the presence of two rings.
The planar structure of 1 was established from the NMR data
listed in the Experimental Section. Particularly useful was the
* To whom correspondence should be addressed. Tel: 808-956-6401.
Fax: 808-956-5908. E-mail: hemschei@hawaii.edu.
† University of Hawaii.
1
analysis of 1D-TOCSY spectra in conjunction with H and 13C
NMR data. This readily led to the identification of two isoleucine
residues, one glyceric acid, a ꢀ-substituted alanine, and a γ-sub-
stituted aminobutyric acid. Analysis of gHMBC data allowed the
latter two to be expanded to an N-methyl phenylalanine and a
‡ Permanent address: Chemical Synthesis Laboratory, SB-ISIC-LSYNC,
Swiss Federal Institute of Technology (EFPL), CH-1015 Lausanne,
Switzerland.
§ Cancer Research Center of Hawaii.
10.1021/np800557m CCC: $40.75
2009 American Chemical Society and American Society of Pharmacognosy
Published on Web 12/30/2008

Notes
Journal of Natural Products, 2009, Vol. 72, No. 1 173
Figure 1. Individual spin systems as identified by 1D TOCSY
(bold) and important gHMBC (arrows) correlations for 1, 2, 3, and
4.
homotyrosine, respectively. The most important correlations are
depicted in Figure 1. It is noteworthy that the threonine residue
did not yield to straightforward 1D-TOCSY analysis. This amino
acid was identified on the basis of data from two 1D-TOCSY and
from gHMBC experiments. Apparently, the dihedral angle between
the aminomethine and the hydroxymethine protons is such that
TOCSY magnetization transfer is inefficient. This explanation is
borne out by coupling constant analysis, which yields 3JH-2/H-3 e 1
Hz. The OMe resonance at 3.54 ppm was assigned as an O-methyl
ether function attached to C-2 of the glyceric acid residue on the
basis of gHMBC data. The last residue to be identified was the
aminohydroxypiperidone carboxylic acid (Ahp), which is part of
the strictly conserved motif of this compound class. Here again
two 1D-TOCSY experiments yielded two fragments (N-H to H-4
and OH to H-5, respectively), and gHMBC data were required to
establish the atom connectivity unambiguously.
The second, minor cyanopeptolin, 2, was determined to be the
sulfated version of the major one. Insufficient material was isolated
for a detailed analysis of gHMBC data. However, inspection of
the 1D TOCSY and the 1H NMR spectra indicated that 1 and 2 are
constitutionally almost identical, the chemical shift differences being
smaller than 0.05 ppm (1H) for all but three resonances. The only
substantial chemical shift differences affect the H-3 protons as well
as the methoxymethine proton H-2 of the methoxy glyceric acid
residue (∆δ ) 0.5 ppm). HRESITOFMS analysis gave a pseudo-
molecular ion cluster [M + H]+ at m/z 961.4223, suggesting a
molecular formula for 2 of C45H64N6O15S. This molecular formula
differs from that of 1 by the element of SO3 and suggested 2 to be
a sulfated version of 1. Corroboration of this interpretation came
from the observation of the typical sulfate stretch at 1206 cm-1 in
the FTIR spectrum of 2. Cyanopeptolin 960 is very closely related
to micropeptin KT946,9 differing only in the methoxyl group in
the glyceric acid moiety.
phenylalanine was synthesized by an adaptation of a literature
method.11-14 The stereochemical analysis was performed on two
aliquots of 1. The first sample was oxidized with excess PCC in
CH2Cl2 solution before hydrolysis in order to convert the aminal
of the 3-amino-6-hydroxypiperid-2-one (Ahp) residue to the imide.
Marfey analysis of the acid hydrolysate of the oxidized sample using
LCMS in selected ion mode suggested that the configuration of
the N-methylphenylalanine, the threonine, and the glutamic acid
derived from the Ahp residue is S. The configuration of the
isoleucine residues was determined by chromatography on chiral
stationary phase (Chirex 3126, Phenomenex) using aqueous cop-
per(II) sulfate/acetonitrile as an eluent. Under these conditions the
diastereoisomers of the L-series elute approximately 5 min apart.
Upon analysis of the hydrolysate of 1, only one peak was observed,
which coeluted with the L-isoleucine standard. This suggested that
L-isoleucine was the amino acid present in 1 in both positions
occupied by isoleucine. The homotyrosine did not give any
detectable Marfey derivative, presumably as a result of oxidative
degradation of the phenol during chromate oxidation. However,
since the oxidation could reasonably be expected to destroy the
2-methoxy glyceric acid moiety as well, a second aliquot was
hydrolyzed without prior oxidation. The Marfey derivatives of a
portion of this hydrolysate were again analyzed by LCMS in
negative ion mode and this time found to contain the derivative of
L-homotyrosine.
The absolute configuration of the 2-methoxyglyceric acid was
established by analytical HPLC of the p-bromophenacyl ester
derivative on a chiral stationary phase (Chiracel OJ) and comparison
with authentic standards. These standards were prepared from D-
and L-serine by diazotization with isoamyl nitrite in methanol at
elevated temperature under pressure and purified by ion exchange
chromatography. Although not completely stereoselective, this
method is considerably simpler than the published procedure.15 With
standards in hand, a second portion of the nonoxidized hydrolysate
was derivatized as the p-bromophenacyl ester. The ester of
2-methoxyglyceric acid was isolated by silica chromatography and
The absolute configuration of the constituent amino acids of 1
was determined after acidic hydrolysis by HPLCMS analysis of
the Marfey derivatives and comparison of retention times with those
obtained from authentic standards.10 The required N-methyl-D-

174 Journal of Natural Products, 2009, Vol. 72, No. 1
Notes
analyzed by HPLC on chiral stationary phase. Comparison of the
retention time of the derivative with that of authentic standards
suggested the configuration for the 2-methoxyglyceric acid in 1 to
be R.
Table 1. 500 MHz (1H) and 125 MHz (13C) NMR Data for 3 in
CD3OH
unit
C/H no.
δC
δH
(J in Hz)
HMBC
Ile
1
2
3
4
174.2
1H NMR-guided screening for other peptides containing para-
disubstituted phenols that might be part of a tyrosine or homoty-
rosine revealed the presence of another peptide in extracts of P.
agardhii CYA 126/8. Anabaenopeptin 908 (3)7 yielded a [M +
H]+ pseudomolecular ion at m/z 909.5198 in HRESITOFMS
analysis, consistent with a molecular formula for 3 of C45H69N10O10.
The planar structure of 3 was determined on the basis of gCOSY,
gHSQC, gHMBC, and 1D-TOCSY data, which are summarized in
Table 1. Particularly significant was the observation of two gHMBC
correlations of signals for NH protons resonating at δH 6.47 and
6.62 ppm, respectively, to one carbonyl resonance at δC 160.1,
indicating the presence of a ureido function, a hallmark of the
anabaenopeptin class of compounds.16
60.0 4.32,
37.7 2.05,
26.1a 1.36,
1.03,
m
1, 3, 4, 3′, 1N-Me-Hty
m
m
m
2, 1N-Me-Hty
5
11.8 0.82, dd (7.6, 7.3)
4
3′
NH
1
16.6 0.86, d (6.9)
3, 4
8.44
N-Me-
171.7
Hty
2
3
61.5 4.53, dd (6.2, 8.2) 1, N-Me, 1Hty
32.1b 2.12,
1.78,
m
m
4
1′
33.0 2.28
133.2
1′
2′/6′
3′/5′
4′
130.2 6.98, d (8.5)
116.5 6.71, d (8.5)
157.0
1′, 4, 3′/5′, 4′
2′/6′, 4′
Analysis of 1D TOCSY and gCOSY data suggested the presence
of one valine, one isoleucine, and two γ-substituted aminobutyrate
residues as well as one lysine residue. The amino acid attached via
the ureido function was identified as an arginine residue. Analysis
of gHMBC spectra next allowed the attachment of two phenol rings
to the aminobutyrate moieties to yield the homotyrosines, one of
which was N-methylated as suggested by gHMBC data. A
combination of gCOSY and gHMBC data also allowed the sequence
of amino acids to be determined: the isoleucine is attached via its
carboxyl group to the ε amino group of the lysine and at the
N-terminus to the carboxyl of N-methyl homotyrosine. The latter
is attached to a second homotyrosine residue, which is linked to a
valine residue. The amino group of valine is connected to the
carboxyl of the lysine residue, thereby closing a 19-membered ring.
The R amino group of the lysine bears the urea carbonyl, to which
the arginine side chain is connected. The most important correlations
are depicted in Figure 1.
N-Me
1
2
29.8 2.71,
175.2
s
1Hty
1
Hty
49.5c 4.67,
34.9 2.12,
1.78,
m
m
m
m
m
3
4
32.1b 2.78,
2.65,
132.8
131.1 7.07, d (8.5)
116.5 6.69, d (8.5)
157.2
1′, 2, 2′/6′
1′
2′/6′
2′/6′
3′/5′
4′
NH
1
2
3
4
4′
1′, 3′/5′
4′, 2′/6′
3′/5′
8.99, d (5.2)
1Val
Val
Lys
175.0
61.0 3.95, dd (5.8, 8.1)
31.5 1.99,
19.9 0.98, d (6.6)
19.8 0.97, d (6.5)
7.25, d (5.8)
1
m
4, 4′
2, 3, 4′
2, 3, 4
1Lys
NH
1
2
3
175.8
With this structure in hand we continued screening for other
members of the anabaenopeptin family and tentatively identified a
compound of molecular mass 915, which also contained para-
disubstituted phenols.
56.8 4.09,
32.0b 1.91,
1.78,
22.0 1.58,
1.36,
29.3 1.58,
40.3 3.57,
3.00,
m
m
m
m
m
m
m
m
1, 3, 4, CdOureido
2
4
Anabaenopeptin 915 (4) was isolated as a colorless optically
active solid. Compound 4 yielded a [M + H]+ pseudomolecular
ion of m/z 916.4828 upon HRESITOFMS analysis, consistent with
a molecular formula for 4 of C48H65N7O11. Analysis of 1D-TOCSY
data revealed that the amino acids constituting the ring in 4 were
identical with those in 3. The most significant difference between
3 and 4 was evident in the 1H NMR spectrum, which indicated the
presence of three para-disubstituted phenols instead of two as in
3. Analysis of gHMBC data in conjunction with gCOSY and
gHSQC data then suggested that the side chain in compound 4 was
comprised of a tyrosine residue instead of the arginine present in
3. The absolute configuration of the amino acids in 3 and 4 was
determined by Marfey analysis after acid hydrolysis.
5
6
4, 5
NH
NHε
6.62, d (5.1)
7.35, dd (6.9, 3.2) 1Ile, 6
2, CdOureido
Arg
1
2
3
176.2
53.6 4.32,
30.8 1.91,
1.68,
m
m
m
m
m
1, 3
4
5
NH
NH′
6
26.3a 1.68,
42.0 3.21,
3, 4
6
6
6.47, d (8.5)
7.45, t (5.3)
158.9
160.1
With these structures in hand, renewed literature search for
cyanopeptolins and anabaenopeptins revealed that, in fact, homo-
tyrosine-containing anabaenopeptin-type compounds had been
previously discovered from Oscillatoria agardhii.17a,b The peptide
in the first paper is structurally quite different,17a bearing an
N-methyl alanine residue in place of the N-methyl homotyrosine.
However, the anabaenopeptins from strain NIES 595 differ only
in the amino acid N-terminal to the Hty residue, isoleucine in place
of valine in 3 and 4.17b It should be noted that the structures of the
two anabaenopeptins reported in this study are closely related to
anabaenopeptin T isolated from field-collected material of unre-
ported taxonomy, the only difference being the structure of the
ureido-bound exocyclic amino acid, which is isoleucine in the case
of anabaenopeptin T.16e The failure of the keyword-based search
to uncover the references for the Oscillatoria metabolites is puzzling
Ureido CdO
a
b
13C NMR signals are interchangeable. 13C NMR signals are
interchangeable. c Overlapped with solvent signal.
and may serve as a cautionary tale; a structure-based search had
no such difficulties, however.
In conclusion, we note that all of the four peptides described in
this paper contain homotyrosine. In order to establish whether these
compounds are being produced only in a precursor-directed
biosynthesis experiment or whether they are part of the metabolite
spectrum in unsupplemented medium, we also analyzed cell mass
from cultures to which exogenous homotyrosine had not been
supplied. We found that three of the four compounds described in
this paper are also biosynthesized by strain CYA126/8 of P.
agardhii in the absence of added homotyrosine. The nonsulfated

Notes
Journal of Natural Products, 2009, Vol. 72, No. 1 175
Cyanopeptolin 880 (1): white powder; [R]D -38 (c 0.07, MeOH);
UV (MeOH) λmax (log ε) 279 (2.95) nm; IR (film) νmax 3435, 1645,
1456, 1384 cm-1; 1H NMR (500 MHz, MeOH-d3) δ Ahp: 7.69 (1H, d,
J ) 9.0 Hz, N-H), 6.34 (1H, brd, O-H), 5.04 (1H, d, J ) 1.7 Hz, H-5),
4.56 (1H, m, H-2), 2.78 (1H, m, H-3a), 1.88 (3H, m, H-3b, H-4); Hty:
8.41 (1H, d, J ) 7.5 Hz, N-H), 6.97 (2H, d, J ) 8.5 Hz, H-2′, H-6′),
6.66 (2H, d, J ) 8.5 Hz, H-3′, H-5′), 4.32 (1H, m, H-2), 2.66 (1H, m,
H-4a), 2.51 (1H, m, H-4b), 2.51 (1H, m, H-3a), 1.88 (1H, m, H-3b);
Ile-1: 4.50 (1H, d, J ) 10.7 Hz, H-2), 1.88 (1H, m, H-3), 1.12 (1H, m,
H-4a), 0.70 (1H, m, H-4b), 0.66 (3H, dd, J ) 6.8, 5.0 Hz, H-5), -0.13
(3H, d, J ) 6.5 Hz, H-3′); Ile-2: 8.13 (1H, d, J ) 9.0 Hz, N-H), 4.42
(1H, t, J ) 8.6 Hz, H-2), 1.88 (1H, m, H-3), 1.45 (1H, m, H-4a), 1.21
(1H, m, H-4b), 0.96 (3H, d, J ) 6.9 Hz, H-3′), 0.88 (3H, t, J ) 7.4
Hz, H-5); N-Me-Phe: 7.29 (2H, d, J ) 7.4 Hz, H-2, H-6), 7.24 (2H, t,
J ) 7.4 Hz, H-3, H-5), 7.16 (1H, t, J ) 7.4 Hz, H-4), 5.24 (1H, dd, J
) 11.6, 2.8 Hz, H-2), 3.52 (1H, dd, J ) -14.5, 2.8 Hz, H-3a), 2.83
(3H, s, N-Me), 2.80 (1H, dd, J) -14.5, 11.6 Hz, H-3b); Thr: 8.08
(1H, d, J ) 9.0 Hz, N-H), 5.60 (1H, q, J ) 6.6 Hz, H-3), 4.83 (1H, d,
J ) 9.0 Hz, H-2), 1.39 (3H, d, J ) 6.6 Hz, H-4); 2-OMe-GA: 3.96
(1H, dd, J ) 6.0, 3.7 Hz, H-2), 3.85 (1H, dd, J ) -11.7, 3.7 Hz,
H-3a), 3.77 (1H, dd, J ) -11.7, 6.0 Hz, H-3b), 3.54 (3H, s, O-Me);
13C NMR (125 MHz, MeOH-d3) δ Ahp: 171.1, 76.2, 51.2, 31.4, 22.3;
Hty: 174.3, 156.8, 132.8, 130.6, 116.3, 54.8, 33.7, 32.1; Ile-1: 172.8,
56.6, 34.6, 25.1, 15.0, 10.8; Ile-2: 175.5, 58.2, 38.0, 26.6, 16.2, 10.2;
N-Me-Phe: 172.0, 139.0, 130.9, 130.0, 128.0, 63.0, 35.4, 31.5; Thr:
171.3, 73.4, 56.8, 18.9; 2-OMe-GA: 173.9, 84.5, 63.9, 58.9; HRES-
ITOFMS m/z [M + H]+ 881.4645 (calcd for C45H65N6O12, 881.4660,
1.5 ppm error).
cyanopeptolin 880 (1) was not observed when the strain was grown
in nonsupplemented medium. This observation may mean that
sulfation is rate limiting and that supplementation with homotyrosine
results in a higher flux through the early part of the pathway.
However, this observation was not pursued further.
Members of the cyanopeptolin8 and anabaenopeptin16 classes
of compounds are known protease inhibitors. We did not observe
trypsin/chymotrypsin inhibition by either 3 or 4 at concentrations
up to 100 µg/mL. However, in a standard assay 4 inhibited
carboxypeptidase A with an IC50 of 0.12 µg/mL, while 3 was
inactive at concentrations up to 20 µg/mL, the highest concentration
tested.
The present investigation was concerned with the peptide
metabolite spectrum of strain CYA126/8 as a foundation on which
a better understanding of peptide biosynthesis in cyanobacteria may
be built. Last, it should be noted that 3 and 4 are obviously products
of the NRPS pathway. Our current understanding of NRPS
biochemistry cannot readily explain the co-occurrence of two closely
related metabolites bearing two quite different amino acids, such
as arginine and tyrosine in equivalent positions.
Experimental Section
General Experimental Procedures. Optical rotation data were
recorded on a JASCO DIP-700 polarimeter. IR and UV spectra were
recorded using a Thermo Nicolet Avatar 380 and a Beckman DU-7000
spectrometer, respectively. Nuclear magnetic resonance (NMR) spectra
were recorded using Shigemi tubes on a Varian UNITY INOVA 500
instrument equipped with a 3 mm microprobe. Chemical shifts are
referenced to residual protiated solvent. High-performance liquid
chromatography (HPLC) was performed on a Shimadzu AS-10VP
gradient system using either a 5 µm Lichrospher ODS column (4.6 ×
250 mm) or a 5 µm Phenomenex Luna C-18(2) column (4.6 × 250
mm or 10 × 250 mm). HRESITOFMS spectra were recorded in the
flow injection mode on an Agilent 6100 LC-MSDTOF system equipped
with an Agilent 1100 LC module. HRFABMS data were recorded on
a VG ZAB-70 instrument using a glycerol matrix.
Cyanopeptolin 960 (2): white powder; [R]D -39 (c 0.05, MeOH);
UV (MeOH) λmax (log ε) 279 (2.94) nm; IR (film) νmax 3440, 1645,
1
1455, 1361, 1206 cm-1; H NMR (500 MHz, MeOH-d3) δ Ahp: 7.69
(1H, d, J ) 9.0 Hz, N-H), 6.34 (1H, brd, O-H), 5.04 (1H, d, J ) 5.4
Hz, H-5), 4.56 (1H, m, H-2), 2.78 (1H, m, H-3a), 1.88 (3H, m, H-3b,
H-4); Hty: 8.41 (1H, d, J ) 7.5 Hz, N-H), 6.97 (2H, d, J ) 8.5 Hz,
H-2′, H-6′), 6.66 (2H, d, J ) 8.5 Hz, H-3′, H-5′), 4.32 (1H, m, H-2),
2.66 (1H, m, H-4a), 2.51 (1H, m, H-4b), 2.38 (1H, m, H-2a), 1.88
(1H, m, H-2b); Ile-1: 4.50 (1H, d, J ) 10.7 Hz, H-2), 1.88 (1H, m,
H-3), 1.12 (1H, m, H-4), 0.70 (1H, m, H-4), 0.66 (3H, dd, J ) 6.8, 5.0
Hz, H-5), -0.13 (3H, d, J ) 6.5 Hz, H-3′); Ile-2: 8.13 (1H, d, J ) 9.0
Hz, N-H), 4.42 (1H, dd, J ) 8.6, 8.6 Hz, H-2), 1.88 (1H, m, H-3),
1.45(1H, m, H-4a), 1.21 (1H, m, H-4b), 0.96 (3H, d, J ) 6.9 Hz, H-3′),
0.88 (3H, t, J ) 7.4 Hz, H-5); N-Me-Phe: 7.29 (2H, m, H-2′, H-6′),
7.24 (2H, m, H-3′, H-5′), 7.16 (1H, m, H-4′), 5.24 (1H, dd, J ) 11.6,
2.8 Hz, H-2), 3.52 (1H, dd, J ) -14.5, 2.8 Hz, H-3a), 2.80 (1H, dd,
J) -14.5, 11.6 Hz, H-3b); 2-OMe-GA sulfate: 4.49 (1H, dd, J ) 4.7,
2.7 Hz, H-2), 4.35 (1H, dd, J) -10.4, 2.7 Hz, H-3a), 4.28 (1H, dd, J
) -10.4, 4.7 Hz, H-3b), 3.54 (3H, s, O-Me); Thr: 8.08 (1H, d, J )
7.5 Hz, N-H), 5.60 (1H, q, J ) 6.6 Hz, H-3), 1.39 (3H, d, J ) 6.6 Hz,
H-4); 13C NMR (125 MHz, MeOH-d3) δ Ahp: 171.1, 76.2, 51.2, 31.4,
22.3; Hty: 174.3, 156.8, 132.8, 130.6, 116.3, 54.8, 33.7, 32.1; Ile-1:
172.8, 56.6, 34.6, 25.1, 15.0, 10.8; Ile-2: 175.5, 58.2, 38.0, 26.6, 16.2,
10.6; N-Me-Phe: 172.0, 139.0, 130.9, 130.0, 128.0, 63.0, 35.4, 31.5;
2-OMe-GA sulfate: 173.9, 84.5, 63.9, 58.9; Thr: 171.3, 73.4, 56.8, 18.9;
HRESITOFMS m/z [M + H]+ 961.4223 (calcd for C45H65N6O15S,
961.4229, 0.6 ppm error).
Biological Material. Planktothrix agardhii. CYA126/8 was a kind
gift of Kaarina Sivonen (University of Helsinki). This strain is deposited
in the Norwegian Institute of Water Research culture collection and is
maintained in the culture collection in the Chemistry Department at
the University of Hawaii. Cultures were grown in 20 L carboys in Z+G
medium.18 Continuous illumination was by soft white fluorescent lights
at 100 µE s-1
.
Isolation of Peptides. Cultures were concentrated by low-pressure
filtration over ultrafiltration membranes. The concentrated cell suspen-
sion was pelleted by centrifugation to yield 7.4 g of cell paste. The
pellet was freeze-dried (1.6 g dry wt), suspended in 50% MeOH(aq)
(10 mL/g dry weight), sonicated, and extracted twice with vigorous
shaking for 30 min. The combined extracts were evaporated and applied
to a YMC-ODS-A flash column (4.0 g) equilibrated in 30% aqueous
MeOH. The column was washed with additional 30% MeOH(aq) and
then eluted with a step gradient of MeOH in H2O (30 mL/step).
Compounds 1, 2, and 3 eluted in the 70% fraction, whereas compound
4 eluted in the 50% fraction. Further purification was achieved by
reversed-phase HPLC (Luna C-18(2), 10 × 250 mm, 3 mL/min)
employing a gradient of CH3CN(aq) containing 0.05% TFA (30%
CH3CN(aq) from 0-10 min, to 50% CH3CN(aq) over 15 min, 50%
CH3CN(aq) for 5 min, and to 100% CH3CN in 5 min). Compound 3
elutes after 8.6 min (4.9 mg), compound 4 after 21.5 min (3.9 mg),
and compounds 1 and 2 as a mixture of anomers after 24.4 min. The
cyanopeptolins are resolved by HPLC on a Lichrosphere C-18 column
(4.6 × 250 mm, 1 mL/min) with a gradient starting at 45% CH3CN(aq)
ramping to 50% CH3CN(aq) in 20 min and holding for 5 min.
Cyanopeptolin (1) (1.5 mg) elutes after 9.5 min (anomer (1.4 mg) at
15.1 min), cyanopeptolin (2) (0.5 mg) after 3.4 min, and its anomer
(0.8 mg) after 4.5 min, respectively.
Anabaenopeptin 908 (3): [R]D -29 (c 0.05, MeOH); UV (MeOH)
1
λmax (log ε) 280 (3.39) nm; IR (film) νmax 1652 cm-1; H NMR and
13C NMR, see Table 1; HRESITOFMS m/z [M + H]+ 909.5198 (calcd
for C45H69N10O10, 909.5204, 0.7 ppm error).
Anabaenopeptin 915 (4): white powder; [R]D -21 (c 0.05, MeOH);
UV (MeOH) λmax (log ε) 280 (3.46) nm; IR (film) νmax 1652 cm-1; 1H
NMR (500 MHz, MeOH-d3) δ Ile: 8.42 (1H, d, J ) 9.6 Hz, N-H),
4.30 (1H, dd, J ) 9.6, 6.2 Hz, H-2), 2.03 (1H, m, H-3), 1.38 (1H, m,
H-4a), 1.10 (1H, m, H-4b), 0.85 (3H, d, J ) 6.9 Hz, H-3′), 0.79 (3H,
t, J ) 7.4 Hz, H-5); N-Me-Hty: 6.97 (2H, d, J ) 8.5 Hz, H-2′, H-6′),
6.68 (2H, d, J ) 8.5 Hz, H-3′, H-5′), 4.52 (1H, dd, J ) 6.5, 7.2 Hz,
H-2), 2.70 (3H, s, N-Me), 2.32 (1H, dt, J ) -12.4, 5.0 Hz, H-4a),
2.26 (1H, dt, J ) -12.4, 4.6 Hz, H-4b), 2.12 (1H, m, H-3a), 1.80 (1H,
m, H-3b); Hty: 8.96 (1H, d, J ) 4.7 Hz, N-H), 7.06 (2H, d, J ) 8.3
Hz, H-2′, H-6′), 6.68 (2H, d, J ) 8.3 Hz, H-3′, H-5′), 4.66 (1H, m,
H-2), 2.78 (1H, m, H-4a), 2.64 (1H, m, H-4b), 2.12 (1H, m, H-3a),
1.80 (1H, m, H-3b); Val: 7.28 (1H, d, J ) 4.7 Hz, N-H), 3.92 (1H, dd,
J ) 8.0, 5.9 Hz, H-2), 1.97 (1H, m, H-3), 0.98 (6H, d, J ) 6.8 Hz,
H-4, H-4′); Lys: 7.31 (1H, dd, J ) 7.0, 3.0 Hz, N-Hε), 6.64 (1H, d, J
) 4.7 Hz, N-H), 4.06 (1H, m, H-2), 3.54 (2H, m, H-6), 1.91 (1H, m,
Feeding Experiment. DL-Homotyrosine (100 mg) was suspended
in 14 mL of distilled water, and the minimum amount of diluted HCl
was added to effect dissolution. The resulting solution was dispensed
through a sterile filter in equal 1 mL portions every day over two weeks
starting 10 days after inoculation. Fermentation was continued for
another 10 days after the last addition of compound.

176 Journal of Natural Products, 2009, Vol. 72, No. 1
Notes
(3) Stachelhaus, T.; Mootz, H. D.; Marahiel, M. A. Chem. Biol. 1999, 6,
493–505.
(4) Laub, J.; Henriksen, P.; Brittain, S. M.; Wang, J.; Carmichael, W. W.;
Rinehart, K. L.; Moestrup, O. EnViron. Toxicol. 2002, 17, 351–357.
(5) Sano, T.; Srivashava, V. C.; Kaya, K. Tennen Yuki Kagobutsu Toronkai
Koen Yoshishu 1996, 38, 433–438.
H-3a), 1.76 (1H, m, H-3b), 1.56 (1H, m position, H-5a), 1.53 (1H, m,
H-4a), 1.31 (2H, m, H-4b, H-5b); Tyr: 7.02 (2H, d, J ) 8.3 Hz, H-2′,
H-6′), 6.68 (2H, d, J ) 8.3 Hz, H-3′, H-5′), 6.06 (1H, d, J ) 8.0 Hz,
N-H), 4.45 (1H, m, H-2), 3.03 (1H, m, H-3a), 2.91 (1H, m, H-3b);
13C NMR (125 MHz, MeOH-d3) δ Ile: 174.2, 60.0, 37.7, 26.1, 16.6,
11.7; N-Me-Hty: 171.7, 156.9, 133.1, 130.2, 116.3, 61.4, 34.9, 33.0,
29.8; Hty: 175.2, 157.5, 132.8, 131.1, 116.5, 51.3, 34.9, 31.9; Val: 175.0,
61.2, 31.5, 20.1, 19.8; Lys: 175.9, 56.5, 40.4, 32.0, 29.2, 21.9; Tyr:
177.0, 157.2, 131.5, 129.1, 116.5, 56.6, 38.4; ureido: 159.8; HRES-
ITOFMS m/z [M + H]+ 916.4828 (calcd for C48H65N7O11, 916.4821,
0.7 ppm error).
Carboxypeptidase A Assay. In an Eppendorf tube, 100 µL of 25
mM Tris, 0.5 M NaCl, pH 7.5, 10 µL of carboxypeptidase A [6 U/mL,
Sigma Aldrich], and 10 µL of anabaenopeptin solution of varying
concentrations (dissolved in 50% MeOH(aq)) were combined and
incubated for 30 min at 37 °C. The mixture was cooled to room
temperature, and the reaction was initiated by addition of 380 µL of
hippuryl-L-phenylalanine (1 mM in 25 mM Tris, 0.5 M NaCl, pH 7.5).
After 5 min a 100 µL aliquot of the reaction was quenched with 1 µL
of 50% trichloroacetic acid(aq). The amount of phenylalanine liberated
was determined by HPLC using the OPA derivatization reaction.19
Briefly, the OPA derivatization buffer was prepared as follows, 1
mL of o-phthalaldehyde [0.25 mg per 50 mL of MeOH], 4 mL of borate
buffer [0.1 M boric acid, 0.1 M NaOH, pH 9.3], and 162 µL of
3-mercaptopropionic acid were added in that order and mixed thor-
oughly, and the pH was adjusted to 9.3 with NaOH. The derivatization
buffer was stored at 4 °C until use (buffer was prepared at least 90
min before use and used within three days).
(6) Melillo, D. G.; Larsen, R. D.; Mathre, D. J.; Shukis, W. F.; Wood,
A. W.; Colleluori, J. R. J. Org. Chem. 1987, 52, 5143–5150.
(7) Cyanopeptolin-and anabaenopeptin-type compounds have been pub-
lished under a variety of common names with differing extensions,
such as letters of the alphabet or strain numbers. For the present work
we have chosen an extension to the base term “anabaenopeptin” or
“cyanopeptolin”, respectively, that is based on molecular mass. This
naming system therefore contains some structural information and
should facilitate dereplication as well as communication.
(8) Cyanopeptolin-type compounds: (a) Bister, B.; Keller, S.; Baumann,
H. I.; Nicholson, G.; Weist, S.; Jung, G.; Su¨ssmuth, R. D.; Ju¨ttner, F.
J. Nat. Prod. 2004, 67, 1755–1757. (b) Matern, U.; Oberer, L.; Erhard,
M.; Herdman, M.; Weckesser, J. Phytochemistry 2003, 64, 1061–1067.
(c) Matern, U.; Oberer, L.; Falchetto, R. A.; Erhard, M.; Ko¨nig, W. A.;
Herdman, M.; Weckesser, J. Phytochemistry 2001, 58, 1087–1095.
(d) von Elert, E.; Oberer, L.; Merkel, P.; Huhn, T.; Blom, J. F. J. Nat.
Prod. 2005, 68, 1324–1327. (e) Jakobi, C.; Oberer, L.; Quiquerez,
C.; Ko¨nig, W. A.; Weckesser, J. FEMS Microbiol. Lett. 1995, 129,
129–134. (f) Shin, H. J.; Murakami, M.; Matsuda, H.; Ishida, K.;
Yamaguchi, K. Tetrahedron Lett. 1995, 36, 5235–5238. (g) Ishida,
K.; Matsuda, H.; Murakami, M.; Yamaguchi, K. J. Nat. Prod. 1997,
60, 184–187. (h) Banker, R.; Carmeli, S. Tetrahedron 1999, 55, 10835–
10844. (i) Fujii, K.; Sivonen, K.; Naganawa, E.; Harada, K. Tetra-
hedron 2000, 56, 725–733. (j) Yamaki, H.; Sitachitta, N.; Sano, T.;
Kaya, K. J. Nat. Prod. 2005, 68, 14–18. (k) Ploutno, A.; Shoshan,
M.; Carmeli, S. J. Nat Prod. 2002, 65, 973–978. (l) Ishida, K.;
Matsuda, H.; Murakami, M.; Yamaguchi, K. J. Nat. Prod. 1997, 60,
184–187.
To analyze the amount of phenylalanine liberated, 25 µL of the
quenched reaction mixture was added to 50 µL of OPA derivatization
buffer. The mixture was allowed to incubate at room temperature for
20 min, at which time it was subjected to HPLC analysis.
(9) Beresovsky, D.; Hadas, O.; Livne, A.; Sukenik, A.; Kaplan, A.;
Carmeli, S. Isr. J. Chem. 2006, 46, 79–87.
HPLC was performed on a 5 µm LUNA C18(2) column (4.6 × 250
mm) pre-equilibrated with an isocratic solvent system of 48% 40 mM
sodium phosphate buffer, pH 7.8/52% CH3CN/MeOH/H2O (45:45:10).
The peaks were observed at 338 nm and recorded on an HP 3995
integrator. At a flow rate of 1 mL/min the phenylalanine derivative
eluted at 5.6 min.
(10) Marfey, P. Carlsberg Res. Commun. 1984, 49, 591–596.
(11) White, K. N.; Konopelski, J. P. Org. Lett. 2005, 7, 4111–4112.
(12) Albericio, F.; Bryman, L. M.; Garcia, J.; Michelotti, E. L.; Nicolas,
E.; Tice, C. M. J. Comb. Chem. 2001, 3, 290–300.
(13) Biron, E.; Kessler, H. J. Org. Chem. 2005, 70, 5183–5189.
(14) Bischofberger, N.; Waldmann, H.; Saito, T.; Simon, E. S.; Lees, W.;
Bednarski, M. D.; Whitesides, G. M. J. Org. Chem. 1988, 53, 3457–
3465.
Acknowledgment. We thank W. Yoshida for NMR support.
Acquisition of instruments used in this work was made possible by
support from the Army Research Office, the Air Force Office of
Scientific Research, the Pardee Foundation, and the NSF. This work
was supported by the NSF (OCE04-32479), the NIH (P50ES012740),
and NOAA OHHI award NA04-32479.
(15) Itou, Y.; Ishida, K.; Shin, H. J.; Murakami, M. Tetrahedron 1999, 55,
6871–6882.
(16) Other anabaenopeptin-type compounds: (a) Reshef, V.; Carmeli, S.
J. Nat. Prod. 2002, 65, 1187–1189. (b) Grach-Pogrebinsky, O.;
Sedmak, B.; Carmeli, S. Tetrahedron 2003, 59, 8329–8336. (c) Sano,
T.; Usui, T.; Ueda, K.; Osada, H.; Kaya, K. J. Nat. Prod. 2001, 64,
1052–1055. (d) Murakami, M.; Suzuki, S.; Itou, Y.; Kodani, S.; Ishida,
K. J. Nat. Prod. 2000, 63, 1280–1282. (e) Kodani, S.; Suzuki, S.;
Ishida, K.; Murakami, M. FEMS Microbiol. Lett. 1999, 178, 343–
348. (f) Shin, H. J.; Matsuda, H.; Murakami, M.; Yamaguchi, K. J.
Nat. Prod. 1997, 60, 139–141. (g) Harada, K.; Fujii, K.; Shimada, T.;
Suzuki, M.; Sano, H.; Adachi, K.; Carmichael, W. W. Tetrahedron
Lett. 1995, 36, 1511–1514.
Supporting Information Available: Procedures and analytical data
1
for the synthesis of standards, H NMR spectra for compounds 1-4,
and 13C NMR spectra for 1, 3, and 4. This material is available free of
charge via the Internet at http://pubs.acs.org.
(17) (a) Murakami, M.; Shin, H. J.; Matsuda, H.; Ishida, K.; Yamaguchi,
K. Phytochemistry 1997, 44, 449–452. (b) Itou, Y.; Suzuki, S.; Ishida,
K.; Murakami, M. Bioorg. Med. Chem. Lett. 1999, 9, 1243–1246.
(18) Zehnder, A.; Gorham, P. R. Can. J. Microbiol. 1960, 6, 645–660.
(19) Kutla´n, D.; Molna´r-Perl, I. J. Chromatogr. A 2003, 987, 311–322.
References and Notes
(1) Christiansen, G.; Yoshida, W.; Blom, J.; Portmann, C.; Gademann,
K.; Hemscheidt, T.; Kurmayer, R. J. Nat. Prod. 2008, 71, 1881–1886.
(2) Challis, G. L.; Ravel, J.; Townsend, C. M. Chem. Biol. 2000, 7, 211–
224.
NP800557M