Psammaplin A, a chitinase inhibitor isolated from the Fijian marine sponge Aplysinella rhax

Article abstract of DOI:10.1016/S0968-0896(01)00372-8

Several brominated tyrosine derived compounds, psammaplins A (1), K (2) and (3) as well as bisaprasin (4) were isolated from the Fijian marine sponge Aplysinella rhax during a bioassay guided isolation protocol. Their structures were determined using NMR and MS techniques. Psammaplin A was found to moderately inhibit chitinase B from Serratia marcescens, the mode of inhibition being non-competitive. Crystallographic studies suggest that a disordered psammaplin A molecule is bound near the active site. Interestingly, psammaplin A was found to be a potent antifungal agent. Copyright

Full text of DOI:10.1016/S0968-0896(01)00372-8

Bioorganic & Medicinal Chemistry 10 (2002) 1123–1128
Psammaplin A, a Chitinase Inhibitor Isolated from the Fijian
Marine Sponge Aplysinella Rhax
a
c
b,y
a,
d
J. N. Tabudravu, V. G. H. Eijsink, G. W. Gooday, M. Jaspars, * D. Komander,
e
c
d
M. Legg, B. Synstad and D. M. F. van Aalten
aMarine Natural Products Laboratory, Department of Chemistry, University of Aberdeen, Old Aberdeen AB243UE, Scotland, UK
bDepartment of Molecular and Cell Biology, University of Aberdeen, Foresterhill AB25 2ZD, Scotland, UK
cDepartment of Chemistry and Biotechnology, Agricultural University of Norway, As, Norway
dWellcome Trust Biocentre, School of Life Sciences, University of Dundee, Dundee DD1 5EH, Scotland, UK
eSyngenta Ltd., Jealott’s Hill International Research Centre, Bracknell, Berkshire RG42 6ET, UK
Received 31 July 2001; accepted 22 October 2001
Abstract—Several brominated tyrosine derived compounds, psammaplins A (1), K (2) and L( 3) as well as bisaprasin (4) were iso-
lated from the Fijian marine sponge Aplysinella rhax during a bioassay guided isolation protocol. Their structures were determined
using NMR and MS techniques. Psammaplin A was found to moderately inhibit chitinase B from Serratia marcescens, the mode of
inhibition being non-competitive. Crystallographic studies suggest that a disordered psammaplin A molecule is bound near
the active site. Interestingly, psammaplin A was found to be a potent antifungal agent. # 2002 Elsevier Science Ltd. All rights
reserved.
Introduction
There are few inhibitors of chitinase, the most studied of
which is allosamidin, isolated from Streptomyces sp.,
3
Chitinase is widely distributed among plants, micro-
organisms, marine invertebrates, fish and insects.1 It
plays a variety of important roles in these organisms
ranging from nutrition to defence and control of ecdysis
in insects. All these roles involve the degradation of
chitin [b-(1,4)-linked homopolymer of N-acetyl-d-glu-
cosamine units], which is mainly used as a structural
polymer by these organisms. The degradation of the
chitin molecule is achieved by a variety of chitinolytic
enzymes that contribute to the degradation process in
different ways. Exochitinases release short oligosacchar-
ides (most often chitobiose) from the ends of the chitin
chains. b-N-Acetyl glucosaminidases release N-acetyl-
glucosamine monomers whereas endochitinases cleave
randomly within the chitin chain. The importance of
chitinases in many biological processes makes their
inhibitors important targets for potential antifungal and
insecticidal agents as well as antimalarial agents.2
with an IC50 of <1 mM. Crystal structures of chit-
inase–allosamidin complexes give insights into the mode
of action of this inhibitor, as well as the mechanism of
2
,4
action of the enzymes. The more recently reported
argadin, a modified cyclic peptide isolated from the
fungus Clonostachys sp. FO-7314 has an IC against
5
0
ꢀ
5
Lucilia cuprina chitinase of 3.4 nM at 20 C. Under the
same conditions, allosamidin shows an IC50 of 0.4 nM.
Another fungal strain, Gliocladium sp FTD-0668 pro-
duced the modified cyclic pentapeptide argifin with an
ꢀ
6
IC50 of 100 nM against L. cuprina chitinase at 20 C.
Crystallographic studies of a Serratia marcescens chit-
inase in complex with argadin/argifin has revealed
ordered binding of these cyclic peptides in the active
site. Two less potent inhibitors are also known, Cyclo(l-
Arg-d-Pro), a diketopiperazine isolated from a marine
7
bacterium, Pseudomonas sp. (17% inhibition at 1 mM)
and the styloguanidines isolated from the marine
8
sponge Stylotella aurantium.
*
Corresponding author. Tel.: +44-1224-272895; fax: +44-1224-
This paper reports on the bioassay guided isolation of a
brominated tyrosine derived natural product,
psammaplin A, which moderately inhibits chitinases
and displays anti-fungal activity. In addition, two
2
72921; e-mail: m.jaspers@abdn.ac.uk
y
Graham Gooday passed away whilst this manuscript was being
refereed. His presence in the chitinase research community will be
greatly missed.
0
968-0896/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00372-8

1124
J. N. Tabudravu et al. / Bioorg. Med. Chem. 10 (2002) 1123–1128
psammaplins not previously reported in the literature
are described.
compound had a similar gross structure, but was not
symmetrical. Mass spectrometry indicated that an extra
+
O was present (HRESIMS m/z 700.9385 [M+Na]
7
Á
9
À3.4 mmu from calculated for C H S Br N O ).
2
2
24
2
2
4
7
1
3
0
that C12 was hydroxylated (Table 3). HMBC correla-
0
The C NMR chemical shifts of C8 –C13 indicated
Results and Discussion
Bioassay guided isolation
0
tions from C12 to H13 and between C8 to both H13
0
0
0
0
0
and H9 were observed indicating that the two hydroxyl
groups belong to same benzene ring system.
Small samples of sponges, soft corals (129) and seaweed
(
9) were collected by free diving from reefs in various
locations in the Fiji Islands: Laucala bay, Suva harbour
and Verata (Viti Levu); reefs in Wainunu and Dama
and in the island of Vanua Levu. A small amount of
each sample was extracted with ethanol, and the
extracts were used in the preliminary chitinase inhibi-
tion bioassay. Out of the 138 samples screened, eight
extracts showed between 22 and 50% inhibition versus
the control in the preliminary assay. The organisms
giving the lower inhibition values were not identified
taxonomically. The strongest inhibitory activity was
shown by the sponge Aplysinella rhax (Order Ver-
ongida, Family Druinellidae), giving a 50% inhibition
versus the control. Verongid sponges are known to
produce bromotyrosine derived metabolites such as the
Compound 3 contained only 11 resonances in its 13C
0
13 matching those of the dihydroxylated benzene ring
NMR spectrum with the chemical shifts of C8/8 –C13/
0
0
0
chemical shifts C8 –C13 in 2 (Table 3), suggesting that
3 was a symmetrical version of 2 with two dihy-
droxylated benzene rings, a fact which was confirmed by
+
HRESIMS (m/z 694.9453 [M+H] Á 2.8 mmu calcu-
7
9
lated for C H S Br N O ).
4
2
2
25
2
2
8
Comparison of spectroscopic experimental data of
compound 4, with that found in the literature showed
this compound to be bisaprasin, a dimer of psammaplin
0
16
A coupled at C12 (Tables 2 and 3). This was con-
+
firmed by the LRESIMS at m/z 1345.2 [M+Na] ,
9
10
11
bastadins,
psammaplins, and A. rhax is known to produce
aeroplysinins,
1
psammaplysins,
and
consistent with the compound being dimeric. The iso-
tope pattern also indicated that it contained four bro-
mines.
2
1
3,14
psammaplins.
A larger re-collection of A. rhax showed a 50% inhibi-
tion compared to the control and was subjected to a
bioassay-guided isolation procedure. Solvent partition-
ing of the crude extract followed by size exclusion
chromatography of the dichloromethane fraction gave
an active semi-purified fraction which was further puri-
fied by reversed-phase HPLC, yielding four pure com-
pounds 1–4 (Table 1).
Chitinase inhibition
A chitinase inhibition bioassay using Bacillus sp. chit-
inase of compounds 1–4 showed psammaplin A (1) to
be the main active compound in this sponge. Com-
pounds 2–4 showed no significant inhibition of this
chitinase. The IC50 for 1 of 68 mM was moderate when
compared to allosamidin, the most potent chitinase
inhibitor known. The mode of inhibition was initially
determined using the Bacillus sp. chitinase. The results
showed psammaplin A had a dual mode of inhibition.
At higher concentrations psammaplin A was an
uncompetitive inhibitor whilst at lower concentrations it
inhibited chitinase in a mixed non-competitive way.
Psammaplin A was also assayed using the chitinase
Compound identification
Comparison of spectroscopic experimental data of
compound 1, the predominant compound isolated, with
that found in the literature showed this compound to be
1
2,15
psammaplin A (Tables 2 and 3).
1
7
bioassay described by Gooday. Preliminary results
had shown strong inhibitions of endochitinase enzymes,
particularly against the bacterial enzymes from Strepto-
myces with an IC50 of 50 mM.
The 13C NMR spectrum of 2 was similar except that
nearly every signal was doubled, suggesting that this
Table 1. Structures and tumour cell IC50’s for compounds 1–4
More detailed enzyme kinetics studies were conducted
with Chitinase B from Serratia marcescens. Initial
results showed that psammaplin A had an IC50 of
approximately 100 mM. This is higher than the IC50
values obtained for this inhibitor against chitinases
from Bacillus sp. and Actinomyces sp. The mode of
inhibition was determined by measuring kinetic para-
meters and by plotting the data in Lineweaver–Burk
plots (Fig. 1). The results clearly show that psammaplin
A acts as a non-competitive inhibitor. Calculations
using EnzFitter confirmed that the inhibitor primarily
Compound
R
1
R
2
IC50 (mM)
A2780 ovarian
tumour
K562
leukaemia
1
2
3
4
(Psammaplin A)
(Psammaplin K)
(Psammaplin L)
(Bisaprasin)
H
H
OH
H
H
OH
OH
Dimer
1.2
0.71
>50
>50
>50
affects Vmax and not K . Using the standard reciprocal
m
>50
>50
>50
Michaelis–Menten equation for non-competitive inhibi-
tion, the K for psammaplin A was calculated to be
i
148 mM.

J. N. Tabudravu et al. / Bioorg. Med. Chem. 10 (2002) 1123–1128
1125
1
Table 2. H NMR Data (d/ppm, CD
3
OD, 400 MHz, ppm, multiplicity, J in Hz) for compounds 1–4
H
1
2
3
4
2
3
7
9
2.75 (2H, t, 6.9)
3.49 (2H, t, 6.6)
3.75 (2H, s)
2.77 (2H, t, 6.8)
3.48 (2H, t, 6.8)
3.68 (2H, s)
7.32 (1H, d, 2)
6.72 (1H, d, 8.0)
7.02 (1H, dd, 2.0,8.0)
2.77 (2H, t, 6.8)
3.48 (2H, t, 6.8)
3.76 (2H, s)
2.78 (2H, t, 6.6)
3.50 (2H, t, 6.7)
3.70 (2H, s)
6.81 (1H, d, 2.0)
—
6.68 (1H, d, 2.0)
2.78 (2H, t, 6.7)
3.50 (2H, t, 6.7)
3.70 (2H, s)
6.81 (1H, d, 2.0)
—
6.68 (2H, d, 2.0)
2.75 (2H, m)
3.49 (2H, m)
3.82 (2H, s)
7.35 (1H, d, 2.0)
6.70 (1H, d, 8.5)
7.02 (1H, m)
2.75 (2H, m)
3.49 (2H, m)
3.74 (2H, s)
7.32 (1H, d, 2.0)
—
7.32 (1H, d, 1.8)
6.70 (1H, d, 8.2)
7.02 (1H, dd, 1.8, 8.4)
2.75 (2H, t, 6.9)
3.49 (2H, t, 6.6)
3.75 (2H, s)
7.32 (1H, d, 1.8)
6.70 (1H, d, 8.2)
7.02 (1H, dd, 1.8, 8.4)
1
1
2
3
7
9
1
1
2
3
0
0
0
0
6.81 (1H, d, 2.0)
—
6.68 (2H, d, 1.6)
0
0
2
3
7.00 (1H, m)
Crystallography
up’, towards Glu144 and the N-acetyl group2,4 (Fig. 2).
‘
In the ChiB–psammaplin structure, Asp142 is partially
pointing upwards, suggesting an interaction with a
moiety bound to the À1 site. These results show that
psammaplin A interacts with the active site of ChiB, but
that it does not bind tightly and precisely as does, for
example, allosamidin. Since Psammaplin A acts as a
non-competitive inhibitor, one would not expect the
compound to bind tightly to the active site, at least not
to the subsites occupied by 4MU-(GlcNAc)2.
The structure of ChiB soaked with psammaplin A was
˚
solved and refined with 1.85 A diffraction data, to
R=0.20, Rfree=0.246 (Table 4). In the final maps,
fragmented density for a large molecule bound to the
active site is observed. Although this density is compat-
ible with a model of psammaplin A, repeated attempts
at interpreting it failed, possibly due to the great con-
formational flexibility of the psammaplin A molecule.
Features of the ChiB structure indicate that the inhi-
bitor molecule is bound to the active site. Family 18
chitinases contain a characteristic signature sequence in
their active site, Asp-X-X-Asp-X-Asp-X-Glu, of which
the Glu is the catalytic acid (Glu144 in ChiB). Previous
structural work has shown that the Asp before the Glu
Additional biological activity
Psammaplin A has previously been shown to possess
potent antibiotic activity against Gram-positive bac-
teria, especially methicillin-resistant Staphylococcus
2
2
(Asp142 in ChiB) serves to stabilise a positively charged
oxazolinium ion intermediate formed during enzymatic
aureus (MRSA). The mechanism of inhibition was
believed to be the inhibition of DNA synthesis and
DNA gyrase activity. Psammaplin A has also been
shown to be cytotoxic and to inhibit the human topo-
chitinolysis by attack of the N-acetyl group in the À1 sub-
2
site on the C1 carbon of the same sugar. In all family 18
chitinase apo-structures presently available Asp142 points
2
3
isomerase II enzyme.
,18À21
‘
down’ towards Asp140 to form a hydrogen bond.2
In
all presently known structures of chitinases in complex
with substrates/inhibitors, however, Asp142 is pointing
13
Table 3.
multiplicity) for compounds 1–4
3
C NMR spectral data (d/ppm, CD OD, 100 MHz, ppm,
C
1
2
3
4
2
3
5
6
7
8
9
38.6 (t)
39.2 (t)
165.5 (s)
152.8 (s)
28.3 (t)
130.2 (s)
134.1 (d)
110.1 (s)
153.3 (s)
116.7 (s)
130.0 (d)
38.6 (t)
38.2 (t)
39.2 (t)
165.5 (s)
153.8 (s)
28.3 (t)
130.0 (s)
134.1 (d)
110.1 (s)
153.3 (s)
116.7 (d)
130.2 (d)
38.1 (t)
38.1 (t)
39.2 (t)
165.5 (s)
152.8 (s)
28.5 (t)
130.1 (s)
124.4 (d)
110.1 (s)
142.3 (s)
146.8 (s)
116.1 (d)
38.1 (t)
38.0 (t)
39.8 (t)
166.0 (s)
151.5 (s)
28.9 (t)
130.8 (s)
134.6 (d)
110.6 (s)
153.8 (s)
117.2 (d)
132.7 (d)
38.7 (t)
1
1
1
1
2
3
5
6
7
8
9
1
1
1
1
0
1
2
3
0
0
0
0
0
0
0
39.2 (t)
39.2 (t)
39.2 (t)
39.7 (t)
165.5 (s)
152.8 (s)
28.3 (t)
130.2 (s)
134.1 (d)
110.1 (s)
153.3 (s)
116.7 (d)
130.0 (d)
165.5 (s)
152.8 (s)
28.5 (t)
130.0 (s)
124.4 (d)
110.1 (s)
142.8 (s)
146.8 (s)
116.1 (d)
165.5 (s)
152.8 (s)
28.5 (d)
130.1 (s)
124.4 (d)
110.1 (s)
142.3 (d)
146.8 (s)
116.1 (s)
166.1 (s)
153.3 (s)
28.9 (t)
129.1 (s)
134.0 (d)
110.6 (s)
153.3 (s)
112.9 (s)
130.6 (d)
0
0
0
0
0
1
2
3
Figure 1. Lineweaver–Burk plots. The lines corresponding to mea-
surements with (^) and without (&) psammaplin A have correlation
coefficients of 0.9984 and 0.9988, respectively. Each point represents
the average of three individual measurements, where each measure-
ment was done in duplicate.

1126
J. N. Tabudravu et al. / Bioorg. Med. Chem. 10 (2002) 1123–1128
Table 4. Details of data collection and structure refinementa
against insect pests was shown to be much lower. Only
mild efficacy was shown against one species, the dia-
mondback moth, Plutella xylstella (Table 6).
˚
Cell dimensions (A)
a=55.75
b=103.59
c=186.07
30–1.85 (1.92–1.85)
254155
˚
Resolution range (A)
No. observed reflections
No. unique reflections
Redundancy
I/sI
Completeness (%)
Conclusion
83,126 (6368)
3.9 (3.1)
10.8 (2.4)
89.1 (69.2)
5.5 (44.0)
20.0
This study confirms that marine sponges are a source of
potential inhibitors of chitinase enzymes, and that fur-
ther investigation of the phylum Porifera is warranted.
The X-ray crystal structure of the psammaplin A/ChiB
complex indicates that the psammaplin A interacts with
the active site, but, that due to its conformational flex-
ibility, the interaction is not as tight and precise as more
potent inhibitors such as allosamidin. The enzyme
kinetics studies, showing psammaplin A is a non-com-
petitive inhibitor of ChiB, are consistent with non-spe-
cific binding of psammaplin A to the ChiB active site.
As might be expected for chitinase inhibitors, psamma-
plin A shows anti-fungal activity. As filter feeders,
sponges are prone to microbial infections, and might
have developed these bromotyrosine containing com-
pounds as antifungal agents.
Rmerge (%)
Rcryst (%)
Rfree (%)
24.6
7846
614
0.012
1.6
1.46
No. protein atoms
No. water molecules
RMSd bond ideality (A)
RMSd angle ideality ( )
RMSd B-factors bonds (A )
˚
ꢀ
˚
2
aValues between brackets are for the highest resolution shell. Crystals
were of space group P2 2 2 , and were cryo-cooled to 100 K. All
1 1 1
measured data were included in structure refinement.
The crude extract of A. rhax was tested for inhibition of
cancer cells. Against ovarian cancer cells it showed an
IC50 of 5.37 mg/mLand against leukaemia cells it
showed an IC50 of 4.97 mg/mL. Tests on the pure com-
pounds showed psammaplin A to be the main active
component. The IC ’s (Table 1) were 1.20 and 0.71 mM
5
0
Experimental
for ovarian tumour and leukaemia cells, respectively.
This is not unusual as Verongid sponges have been pre-
4
General experimental procedures
2
viously reported to yield cytotoxic compounds.
HPLC separations were carried out using a Spectra-
physics P100 isocratic pump and monitored using a
Hewlett Packard HP 1050 Series Variable Wavelength
UV Detector and a Waters ODS (10Â250 mm) column.
Low resolution mass spectra were obtained on a Finni-
gan Masslab Navigator (electrospray) while the high
resolution mass data was obtained on a Finnigan Mat-
Because chitinases are common in fungi, and crucial to
the control of ecdysis in insects, it was felt that investi-
gating the agrochemical potential of psammaplin A was
2
justified. Psammaplin A was submitted to Syngenta
Ltd. for antifungal and insecticidal bioassays. The
compound was tested in vivo against commercially
important fungal species and insect pests. Psammaplin
A was found to be particularly effective against vine
downy mildew, Plasmopora viticola (Table 5). Efficacy
1
13
95. H, C and all NMR 2D experiments were recorded
on a Varian Unity INOVA 400 spectrometer in CD OD
solution.
3
Figure 2. Stereo image of the final model with the remaining difference density. Side chains (carbons in black) surrounding the carbohydrate binding
site are shown as sticks, together with relevant stretches of backbone (grey). Some key residues are labeled, the catalytic acid Glu144 and the pre-
ceeding Asp140/Asp142 (the latter of which was refined in two alternate conformations, identified with yellow carbons), and the tryptophans
2
important for chitin binding (Trp97/Trp220). The position a NAG5 chitin oligomer would occupy according to a previously published structure are
drawn in a stick model with green carbons. The unbiased F
o c
–F map (i.e., prior to inclusion of any ligand atom) is contoured at 2.5 s (magenta).
Some hydrogen bonds discussed in the text are drawn as dotted lines. The sugar binding subsites À2 to +3 are labelled in blue.

J. N. Tabudravu et al. / Bioorg. Med. Chem. 10 (2002) 1123–1128
1127
Collection
(100%, MeOH) lmax 290 (2 3438). IR 3377–3185, 2369,
735, 1701, 1655, 1638, 1525, 1430, 1288, 1222, 986.
1
The sample of A. rhax (de Laubenfels 1954, Order Ver-
ongida, Family Druinellidae), collection number
Psammaplin L (3). Light pinkish oil (10 mg); HRESIMS
+
9
of about 5 m by free diving on a wall of Cakaulevu reef,
712SD130, was collected in December 1997 at a depth
m/z 694.9453 [M+H]
7
Á 2.8 mmu calculated for
291.2 (2
9
C H S Br N O . UV (100%, MeOH) l
2
2
25
2
2
4
8
max
ꢀ
4.694 E), and was identified by Dr. John Hooper of
0
ꢀ
Wainunu, Vanua Levu Island, Fiji (17 2.609 S; 178
0
3139). IR 3575–3211, 2360, 1653, 1636, 1541, 1420,
1362, 1287, 1215, 1044.
5
the Queensland Centre for Biodiversity, Queensland
Museum, Australia. A voucher specimen is preserved at
the Regional Herbarium, School of Pure and Applied
Sciences, University of the South Pacific, Fiji (voucher
no. 9712SD130).
Bisaprasin (4). Colourless oil (40 mg); UV (100%,
MeOH) lmax. 325 (e 3230), 289 (e 3652). IR 3561–3286,
2076, 2927, 2340, 1652, 1624, 1418, 1360, 1207, 1043,
979, 878, 800.
Extraction and isolation
Enzyme assays
The frozen sample 1300 g (wet weight) was extracted
The preliminary assay method for chitinase enzyme
inhibition of crude extracts was developed by Dr. Shi-
zuri’s group at the Marine Biotechnology Institute,
Japan. Preliminary assays on crude extracts were per-
formed using Bacillus sp. chitinase with the synthetic
substrate 4-nitrophenyl-b-d-N,N-diacetylchitobioside.
with MeOH (3Â) and CH Cl (3Â), and then the con-
2
2
centrated extracts were combined and shipped to the
home laboratory in Aberdeen. The crude extract (30 g)
was partioned between water and CH Cl . The CH Cl
2
2
2
2
partion fraction was evaporated to give a crude oil
12 g) that was partioned between hexane and 90%
MeOH/H O. The MeOH fraction was phase-adjusted
ꢀ
(
300 mLof each extract was dried under vacuum at 30 C
after which it was properly diluted and assayed. 1-mL
UV cuvettes were filled with 900 mLsubstrate (0.4 mM)
in 50 mM pH 7 phosphate buffer, 50 mLof the extract
solution in 75% EtOH and 50 mLof Bacillus chitinase
2
to 50% MeOH/H O and partitioned with CH Cl . The
2
2
2
CH Cl fraction was then subject to Sephadex LH-20
2
2
size-exclusion chromatography (1:1 CH Cl /MeOH).
2
2
ꢀ
After the pooling together of similar fractions, reverse-
phase ODS HPLC using 70% MeOH/H O as eluent
gave compounds 1–4.
(1 mg/mLin distilled water) at 30 C. In control reac-
tions, 50 mL75% EtOH was added instead of the
extracts. The reaction rate of the enzyme on the syn-
thetic substrate was followed between 20 and 80 s at
lmax 405 nm on a UV spectrophotometer. Inhibition
became apparent by lowered reaction rates. The degree
of inhibition was quantified by expressing the activities
as percentage of the activity of the control samples with
no extract. The same assay was used for determination
of the IC50 of purified psammaplin A, using inhibitor
concentrations between 0.6 mM and 8.0 mM. To get an
impression of the mode of inhibition, assays were con-
ducted in which both substrate concentrations and
psammaplin A concentrations were varied. The results
were analysed by plotting 1/(% remaining activity) ver-
sus 1/[S].
2
Psammaplin A (1). White powder (100 mg); LRESIMS
+
+
m/z 685.8 (M+Na) ; HRESIMS m/z 662.9587 [M+H]
Á À0.5 mmu calculated for C H S Br N O ; UV
79
25 2
2
2
2
4 6
(
2
100%, MeOH) l
920, 2350, 1660, 1538, 1426, 1359, 1285, 1210, 1044, 985.
278 nm (2 3790). IR 3377–3179,
max
Psammaplin K (2). light pinkish oil (40 mg); LRESIMS
+
+
m/z 705.9 [M+Na] ; HRESIMS m/z 700.9385 [M+Na]
9
7
Á À3.4 mmu calculated for C H S Br N O . UV
2
2
24
2
2
4
7
Table 5. In-vivo fungal disease testsa
Fungus
% Control
The effect of psammaplin A on the activity of Chitinase
B from Serratia marcescens was analysed using the syn-
thetic substrate 4-methylumbelliferyl b-d-N,N-dia-
Erysiphe graminis f. sp tritici—Wheat powdery mildew
Phytophthora infestans—Tomato late blight
Plasmopora viticola—Vine downy mildew
Puccinia recondite—Wheat brown rust
0
0
100
67
0
cetylchitobioside [4-MU-(GlcNAc) ]. This substrate
2
Septoria nodorum—Wheat leaf spot
permits accurate determination of kinetic parameters,
despite the fact that one has to use relatively low sub-
a100 ppm of 1 applied to various fungi on test plants. Activity expres-
sed as % disease control compared to controls.
2
5
strate concentrations in the assays. In a standard assay
used for the determination of the IC50 value) the reac-
(
tion mixture (100 mL) contained 1.15 nM enzyme, 12 mM
substrate, various amounts of psammaplin A, 50 mM
citrate–phosphate buffer, pH 6.3 and 0.1 mg/mLBSA.
Table 6. In-vivo insect tests. Insects treated with 500 ppm of 1
Insect
% Mortality
ꢀ
The reaction was incubated for 10 min at 37 C, after
which the reaction was stopped with 1.9 mL0.2 M
Na CO . The amount of 4-MU released was determined
using a DyNA 200 Fluorimeter (Hoefer Pharmacia
Biotech, San Francisco, CA, USA).
Tetranychus urticae—Two spotted spider mite
Myzus persicae—Peach potato aphid
Heliothis virescens—Tobacco budworm
Plutella xylstella—Diamondback moth
Drosophila melanogaster—Fruitfly
0
55*
0
55
0
2
3
Meloidogyne incognita—Root knot nematode
0
Kinetic parameters were determined using substrate
concentrations of 5, 10, 15 and 20 mM. Psammaplin A
Control mortality=55%.

1128
J. N. Tabudravu et al. / Bioorg. Med. Chem. 10 (2002) 1123–1128
was added to a final concentration of 125 mM. To check
for linearity samples were taken after 0, 5, 10 and
Development Research Fellowship. The structure fac-
tors and coordinates have been deposited with the PDB
(entry 1GPF).
1
5 min. The production of 4-MU was linear in all cases,
thus permitting straightforward calculation of enzyme
velocities. The data sets were analysed using Line-
weaver–Burk plots as well as the program EnzFitter
References and Notes
(
Biosoft, Cambridge, UK).
1
2
. Gooday, G. W. Chitin Chitosan Res. 1997, 3, 233.
. van Aalten, D. M. F.; Komander, D.; Synstad, B.;
Antifungal assays
Gaseidnes, S.; Peter, M. G.; Eijsink, V. G. H. Proc. Natl.
Acad. Sci. U.S.A. 2001, in press.
3. Sakuda, S.; Isogai, A.; Masumoto, S.; Suzuki, A. Agric.
Biol. Chem. 1987, 51, 3251.
The sample, dissolved in acetone was diluted (100 ppm)
in sterile de-ionised water was applied on the plants
(
wheat, grape, vine and tomatoes) as a foliar spray using
a hand-held sprayer. The tested compound was also
sprayed on the roots of the plants. The plants were then
inoculated with a calibrated suspension of fungal spores
and maintained under conditions appropriate for dis-
ease development. Assessment was done after 10 days
by visual inspection of the area covered by the disease
and was compared against untreated plants and also
against plants treated with commercial agrochemicals
fungicides. This was expressed as percentage disease
control.
4. Terwisscha van Scheltinga, A. C.; Armand, S.; Kalk, K. H.;
Isogai, A.; Henrissat, B.; Dijkstra, B. W. Biochemistry 1995,
3
4, 15619.
. Arai, N.; Shiomi, K.; Yamaguchi, Y.; Masuma, R.; Iway,
5
Y.; Turberg, A.; Kolbl, H.; Omura, S. Chem. Pharm. Bull.
2
000, 48, 1442.
. Omura, S.; Arai, N.; Yamaguchi, Y.; Masuma, R.; Iway,
6
Y.; Namikoshi, M.; Turberg, A.; Kolbl, H.; Shiomi, K. J.
Antibiot. 2000, 53, 603.
7. Izumida, H.; Imamura, N.; Sano, H. J. Antibiot. 1996, 49,
76.
8
. Kato, K.; Shizuri, Y.; Izumida, H.; Yokoyama, A. Tetra-
hedron Lett. 1995, 36, 2133.
. Jaspars, M.; Rali, T.; Laney, M.; Schatzman, R. C.; Diaz,
M. C.; Schmitz, F. J.; Pordesimo, E. O.; Crews, P. Tetrahedron
Insecticidal assays
9
Adult or larval stages of the insects which were main-
tained on leaf dics were treated with 500 ppm of the
compound. Assessment which was based on% mortality
was made after 3–5 days.
1
1
994, 50, 7367.
0. Cimino, G.; de Rosa, S.; de Stefano, S.; Self, R.; Sodano,
G. Tetrahedron Lett. 1983, 24, 3029.
1. Roll, D. M.; Chang, C. W. J.; Scheuer, P. J.; Gray, P. J.;
Shoolery, J. N.; Matsumoto, G. K.; van Duyne, G. D.;
1
Clardy, J. J. Am. Chem. Soc. 1985, 107, 2196.
12. Quinoa, E.; Crews, P. Tetrahedron Lett. 1987, 28, 3229.
13. Pham, N. B.; Butler, M. S.; Quinn, R. J. J. Nat. Prod.
Crystallography
ChiB was overexpressed and purified as described pre-
2
5
2
000, 63, 393.
4. Shin, J.; Lee, Hui-Seung; Seo, Youngwan; Rho, Jung-Rae;
viously. ChiB was crystallized by vapour diffusion
1
Cho, Ki Woong; Paul Valerie, J. Tetrahedron 2000, 56, 9071.
from solutions containing ammonium sulphate, glycerol
and HEPES pH 7.20 Pure psammaplin A was dissolved
in 1:1 H O/MeOH and used for soaking experiments. A
1
3
1
5. Arabshahi, L.; Schmitz, F. J. J. Org. Chem. 1987, 52,
584.
6. Rodriguez, A. D.; Akee, R. K.; Scheuer, P. J. Tetrahedron
2
5
0-fold molar excess of psammaplin was added to crys-
tallisation drops containing native ChiB crystals. After
2 h of soaking crystals were frozen in a nitrogen
Lett. 1987, 28, 4989.
7. Sampson, M. N.; Gooday, G. W. In Chitin Enzymology;
1
1
stream. Diffraction data (Table 4) were collected at
ESRF, Grenoble, on beamline BM14. The structure was
Muzzarelli, R. A. A., Ed.; Atec Edizioni, 1996; Vol. 2, p 227.
18. Terwisscha van Scheltinga, A. C.; Kalk, K. H.; Beintema,
J. J.; Dijkstra, B. W. Structure 1994, 2, 1181.
2
6
refined with CNS using the published wild type ChiB
2
0
1
9. Perrakis, A. I.; Tews, I.; Dauter, Z.; Oppenheim, A. B.;
structure as a template after simulated annealing runs
at 2500 K to remove bias (Table 4). Model building was
Chet, I.; Wilson, K. S.; Vorgias, C. E. Structure 1994, 2, 1169.
20. van Aalten, D. M. F.; Synstad, B.; Brurberg, M. B.;
Hough, E.; Riise, B. W.; Eijsink, V. G. H.; Wieringa, R. K.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 5842.
performed with O.27
2
1. Hollis, T.; Mozingo, A. F.; Bortone, K.; Ernst, S.; Cox,
R.; Robertus, J. D. Protein Sci. 2000, 9, 544.
2. Kim, D. Y.; Lee, I. S.; Jung, J. H.; Yang, S. I. Arch.
Acknowledgements
2
J.T. wishes to thank Ratu Isoa Bulikiobo and Ratu
Dovibua Tagivetaua for permission to collect samples,
Usaia Tabudravu and Inosa Qativi for sample collec-
tions, Akihiro Yokoyama, Mamoru Endo and Yoshi-
kazu Shizuri of MBI for bioassay procedure and
technical help, the Government of Fiji and the ORS for
a scholarship. M.J. wishes to thank Gary Duncan for
mass spectra and Alan McGown for anticancer bio-
assays. Aberdeen University Chemistry Department is
thanked for the provision of NMR facilities. We thank
the ESRF, Grenoble, for the time at beamline BM14.
D.v.A. is supported by a Wellcome Trust Career
Pharmacol. Res. 1999, 22, 25.
23. Kim, D.; Lee, I. S.; Jung, J. H.; Lee, C. O.; Choi, S. U.
Anticancer Res. 1999, 19, 4085.
2
4. Ciminiello, A.; Carmela, D.; Fattorusso, E.; Magno, S.;
Pansini, M. J. J. Nat. Prod. 1999, 62, 590.
5. Brurberg, M. B.; Nes, I. F.; Eijsink, V. G. H. Micro-
biology 1996, 142, 1581.
6. Brunger, A. T.; Adams, P. D.; Clore, G. M.; Gros, P.;
2
2
Grosse-Kunstleve, R. W.; Jiang, J.-S.; Kuszewski, J.; Nilges,
M.; Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.;
Warren, G. L. Acta Cryst. 1998, D54, 905.
27. Jones, T. A.; Zou, J. Y.; Cowan, S. W.; Kjeldgaard, M.
Acta Cryst. 1991, A47, 110.