Thiol-based probe for electrophilic natural products reveals that most of the ammosamides are artifacts

Article abstract of DOI:10.1021/acs.jnatprod.6b00773

To date, 16 members of the ammosamide family of natural products have been discovered, and except for ammosamide D each of these metabolites is characterized by an unusual chlorinated pyrrolo[4,3,2-de]quinoline skeleton. Several ammosamides have been shown to inhibit quinone reductase 2, a flavoenzyme responsible for quelling toxic oxidative species in cells or for killing cancer cells outright. Treatment of the extract from an ammosamide-producing culture (Streptomyces strain CNR-698) with a thiol-based reagent designed to label electrophilic natural products produced an ammosamide C-thiol adduct. This observation led us to hypothesize, and then demonstrate through experimentation, that all of the other ammosamides are derived from ammosamide C via nonenzymatic processes involving exposure to nucleophiles, air, and light. Like many established electrophilic natural products, reaction with the thiol probe suggests that ammosamide C is itself an electrophilic natural product. Although ammosamide C did not show substantial cytotoxicity against cancer cells, its activity against a marine Bacillus bacterial strain may reflect its ecological role. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.jnatprod.6b00773

Article
pubs.acs.org/jnp
Thiol-Based Probe for Electrophilic Natural Products Reveals That
Most of the Ammosamides Are Artifacts
Daniela Reimer and Chambers C. Hughes*
Center for Marine Biotechnology and Biomedicine, Scripps Institution of Oceanography, University of California, San Diego, La Jolla,
California 92093-0204, United States
S
* Supporting Information
ABSTRACT: To date, 16 members of the ammosamide
family of natural products have been discovered, and except for
ammosamide D each of these metabolites is characterized by
an unusual chlorinated pyrrolo[4,3,2-de]quinoline skeleton.
Several ammosamides have been shown to inhibit quinone
reductase 2, a flavoenzyme responsible for quelling toxic
oxidative species in cells or for killing cancer cells outright.
Treatment of the extract from an ammosamide-producing
culture (Streptomyces strain CNR-698) with a thiol-based
reagent designed to label electrophilic natural products
produced an ammosamide C-thiol adduct. This observation
led us to hypothesize, and then demonstrate through experimentation, that all of the other ammosamides are derived from
ammosamide C via nonenzymatic processes involving exposure to nucleophiles, air, and light. Like many established electrophilic
natural products, reaction with the thiol probe suggests that ammosamide C is itself an electrophilic natural product. Although
ammosamide C did not show substantial cytotoxicity against cancer cells, its activity against a marine Bacillus bacterial strain may
reflect its ecological role.
presence of formaldehyde in the growth medium, is a salient
ne of the many difficulties associated with the study of
Onatural products is the isolation of artifacts, compounds example.5
The ammosamides are a family of biologically active natural
derived from natural products that are formed via non-
enzymatic processes during the cultivation and extraction of
the organism or during the process of purification. Here, we use
the term “artifact” broadly to include metabolites modified by
both biotic and abiotic reagents, provided the transformations
proceed without the need for catalysis. In most cases, an artifact
arises from reaction between an electrophilic site on a natural
product and a nucleophilic metabolite or solvent (Figure 1).
For example, the plant metabolite palmatine (1) has been
shown to react with solvent during isolation to form
pseudobase 2 and artifacts 3−5.1 Trichloromethyl artifact 5
stems from a common extraction procedure for plant alkaloids
involving a mixture of ammonium hydroxide and CHCl3.
Extraction of a Hyrtios sponge with MeOH yielded
puupehenone (6) and dimethoxypuupehenol (7),2 and
aqueous extraction of a Zyzzya sponge gave makaluvamine H
(8) and damirone A (9).3 In both cases, the formation of 7 and
9 is directly observed upon treatment of pure 6 and 8 with the
appropriate conditions, which provides strong evidence that 7
and 9 are indeed artifacts. The oxidation of natural products to
produce artifacts is also common. For instance, cryptolepine
(10) oxidizes to cryptolepinone (11) during isolation and
purification.4 Artifacts arising from a nonenzymatic reaction
between a nucleophile-containing natural product and an
external electrophile are much less common. The formation of
bohemamine dimer 13 from bohemamine (12), due to the
products isolated from marine-derived Streptomyces sp. CNR-
698.6 Although many related pyrroloiminoquinone natural
products have been reported from marine sponges, including
the batzellines, damirones, isobatzellines, makaluvamines,
discorhabdins, prianosins, epidardins, tsitsikammamines, velut-
amine, zyzzyanone, and wakayin,7 the ammosamides and
lymphostin are the only natural products known to possess a
pyrrolo[4,3,2-de]quinoline skeleton.8 In their initial disclosure,
ammosamides A (14) and B (15) were shown to be highly
cytotoxic against the HCT-116 colon cancer cell line.
Ammosamide B was then determined to target the motor
protein myosin using an immunofluorescent probe, and the
activity of the metabolite was thus attributed to inhibition of
crucial myosin-based cellular functions.9 Ammosamide B has
also been shown to inhibit quinone reductase 2.10 In contrast to
the initial reported cytotoxicity of 14 and 15, subsequent
publications have noted their lack of cytotoxicity against several
cancer cell lines.10,11 A later report detailing the first total
synthesis of 15 included the serendipitous discovery of a minor
metabolite, ammosamide C (16).12 Since then, ammosamide D
(17) has been discovered, and ammosamides E−P have been
produced by supplementing the culture medium with a set of
Received: August 23, 2016
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
that the adduct originated from ammosamide C (16).
Curiously, the red-purple labeling reaction mixture of 19, 20,
and triethylamine gradually turned blue due to the formation of
ammosamide A (14) and bromide 21. On the basis of previous
applications of 19 to bacterial extracts, an effective bond-
forming reaction with 16 suggests that the metabolite binds
irreversibly to its cellular target(s) and that the electrophilic
iminium ion is therefore essential for the metabolite’s biological
activity.17 However, to date, both the cellular target and
mechanism of action of ammosamide C have not been
determined.
This observation led us to hypothesize that ammosamide C
(16) is the precursor to all other ammosamides, which are
themselves artifacts formed from uncatalyzed chemical trans-
formations owing to the electrophilicity of the reactive C-2
iminium functionality. In a key experiment, a small quantity of
purified 16 was treated with A1 medium buffered to pH 10 to
mimic the alkalinity associated with the producing culture and
after 5 days yielded ammosamides B (15) and E (18). A1
medium at pH 6.8 effected very little conversion (Figures 2 and
S1). Furthermore, C (16) cleanly converted to B (15) using an
aqueous sodium bicarbonate solution. When treated with
cysteine at pH 10, C (16) converted to B (15), E (18), and A
(14). Although the notion that 14 and 15 are derived from 16
has been previously suggested,12 this transformation has not
been heretofore demonstrated experimentally. To be fair, we do
acknowledge that ammosamides A (14), B (15), and E (18),
and others, may be formed via enzyme-catalyzed processes in
addition to the nonenzymatic conditions described above.
To further investigate the reactivity of ammosamide C (16),
we required larger quantities of material. Originally, this
metabolite was purified as a minor metabolite from cultivations
of strain CNR-698 because its hydrophilicity meant that
extractions with organic solvent and hydrophobic resins were
ineffective. When a 3-day-old culture was directly treated with
MeOH containing 0.1% trifluoroacetic acid, filtered through
Celite, concentrated to remove the organic solvent, fractionated
with C18 silica gel, and purified by HPLC, 100 mg of 16 per
liter of culture was obtained. Acidification of the culture before
concentration preserved the natural product and greatly
reduced the formation of artifacts, while the addition of
organic solvent precipitated cellular material that would
otherwise clog the C18 silica gel column.
With adequate amounts of ammosamide C (16) in hand, we
treated the compound with various nucleophiles in order to
assess the reactivity of the iminium functionality and to obtain a
sufficient quantity of each adduct for full structural character-
ization (Scheme 2). In contrast to a previously reported study
by Pan et al., in which new ammosamide derivatives were
generated by supplementing the culture with amines,11 here
similar derivatives were prepared in vitro using amines, thiols,
and water. Treatment of 16 with amine nucleophiles (n-
propylamine and 4-chloroaniline) furnished 23 and ammosa-
mide G (24), while treatment with thiol nucleophiles (N-
acetylcysteamine, ethanethiol, and 4-bromobenzyl mercaptan)
provided 26−28. Conversion of 16 to ammosamide B (15) was
performed using basic aqueous conditions, as before.
Presumably, these transformations proceed via acetals 22, 25,
and 29, which oxidize when exposed to air. When conducted
under an inert atmosphere, clear color changes accompany both
formation of the acetals and subsequent oxidation. Given the
wide range of possible ammosamide C-nucleophile adducts, the
reactive iminium functionality may represent a strategy in
Figure 1. Isolation artifacts of natural products.
precursors.11,13 Three additional total syntheses of ammosa-
mide B (15) have since been completed.14−16
RESULTS AND DISCUSSION
■
In our ongoing efforts to detect electrophilic natural products
using thiol-based probe 19 and triethylamine,17 we observed
that the addition of 19 to the extract of Streptomyces sp. strain
CNR-698 produced a single adduct (20) (Scheme 1).18 In
accordance with the method, the thiol adduct bore a
conspicuous isotopic pattern by virtue of the probe’s bromine
atom. The structure of 20 was determined using HRMS and
NMR spectrosopy, and retrosynthetic analysis then revealed
B
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Scheme 1. Treatment of the Extract of Strain CNR-698 with Cysteine-Based Probe 19 Gave 20 and Adduct 20 Fragmented to
Ammosamide A (14) and 21
Nature for the deliberate diversification of the ammosamide
scaffold, yielding a whole class of molecules that are capable of
exhibiting a variety of biological functions.
We then noticed that ammosamide B (15) after prolonged
exposure to sunlight and air in MeOH converted to
ammosamide D (17) (Figure 3). This conversion did not
appear to occur in other solvents, such as THF or DMSO,
suggesting that a protic solvent is required (Figure S5d). This
observation is in stark contrast to a previous study describing
the discovery of the natural product from cultures of
Streptomyces sp. strain SNA-020, wherein an enzyme-catalyzed
process yielding 17 was postulated.13 Under standard
Figure 2. UV/vis-HPLC chromatograms (254 nm) showing the
conversion of ammosamide C (16). (A) 16 in A1 medium at pH 6.8
cultivation conditions in the laboratory with artificial light,
after 5 days (green). The conversion of 16 after 5 days into (B)
ammosamide D is formed only in small amounts as a minor
ammosamides B (15) and E (18) in A1 medium at pH 10 (black), (C)
compound.
ammosamide B (15) in an aqueous sodium bicarbonate solution at pH
The cytotoxicity of selected compounds against the HCT-
8.1 (red), and (D) ammosamides B (15), E (18), and A (14) in a PBS
116 colon cancer cell line was assessed in-house using the MTS
solution containing excess cysteine at pH 10 (blue).
a
Scheme 2. Reaction of Ammosamide C (16) with Various Nucleophiles Followed by Oxidation in Air
a
Pyrroloquinolinium compounds 23 and 26−28 were isolated as TFA salts.
C
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
D
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 3. Ammosamide B (15) converts to ammosamide D (17) after prolonged exposure to sunlight and air in MeOH. UV/vis-HPLC
chromatogram (254 nm) at 0 days (black) and after 6 days (blue).
Technologies HPLC system coupled to an ELSD and UV/vis detector
(210, 254, and 360 nm) using a Phenomenex Luna reversed-phase
(RP) C18(2) column (5 μm, 100 Å, 100 × 4.6 mm) with a 10 min
solvent gradient from 10% to 100% or 1% to 60% CH3CN in H2O
containing 0.1% formic acid (FA) and a flow rate of 1.0 mL min−1.
Using the same column and solvent gradients, liquid chromatog-
raphy−high-resolution mass spectrometry (LC-HRMS) was per-
formed on an analytical Agilent 1260 Infinity Series LC system
coupled to a 6530 Series Q-TOF mass spectrometer. Metabolites and
reaction products were purified using RP UV/vis-HPLC (Millipore
Waters 600E) with a semipreparative Phenomenex Luna RP C8(2)
column (5 μm, 100 Å, 250 × 10 mm), semipreparative Phenomenex
Luna RP CN column (5 μm, 100 Å, 250 × 10 mm) or a preparative
Phenomenex Luna RP C18(2) column (5 μm, 100 Å, 250 × 21.2
mm), and a flow rate of 3 and 13 mL min−1, respectively. Thiol 19 was
synthesized as described previously.17 All reagents and solvents were
purchased commercially and were used without further purification.
Extraction and Isolation of Ammosamides A−C (14−16). The
marine Streptomyces sp. strain CNR-698 was cultivated in a 2.8 L
Fernbach flask containing 1 L of a seawater-based A1 medium (10 g
L−1 starch, 4 g L−1 yeast extract, 2 g L−1 peptone, 75% seawater (La
Jolla, CA, Pacific Ocean), 25% deionized (DI) water) at 230 rpm and
27 °C under artificial light. For the isolation of 14 and 15, a starter
culture of 25 mL was grown for 3 days and inoculated into fresh A1
medium (1 L). After 7 days of cultivation, XAD7HP resin was added
to the culture and left to stir for 1 h. The cells and resin were filtered
through cheesecloth, extracted with acetone (200 mL), and
concentrated under reduced pressure to remove the organic solvent.
The aqueous layer was then extracted with EtOAc (500 mL). The
organic layer was dried over sodium sulfate, filtered, and concentrated
to afford 57.7 mg. The crude material was dry-loaded onto C18 silica
gel and fractionated using a solid-phase extraction RP C18 cartridge
(20%, 40%, 60%, 80%, and 100% CH3CN in H2O + 0.1% TFA). The
40% and 60% fractions were combined and evaporated to dryness to
yield 17.1 mg. Amides 14 and 15 were purified by preparative C18
HPLC at 254 nm (25% CH3CN in H2O, 0.1% TFA, for 13 min, then
30% CH3CN in H2O, 0.1% TFA, tR(15) = 12 min, tR(14) = 23 min)
to give 4.0 mg of 14 as blue and 4.0 mg of 15 as purple solids. For the
isolation of 16, a starter culture of 25 mL was grown for 3 days and
inoculated into fresh A1 medium (1 L). After 3 days of cultivation,
MeOH (500 mL) containing 0.1% TFA was added to the broth, and
the resulting heterogeneous solution was filtered through a pad of
Celite. The filtrate was concentrated under reduced pressure to
remove the organic solvent, and the aqueous solution was then loaded
onto a C18 column. The column was flushed thoroughly with H2O +
0.1% TFA, and ammosamide C (16) was then eluted with 20%
CH3CN in H2O + 0.1% TFA. The fraction was concentrated to give
432 mg. A portion of the fraction (43.2 mg, 10%) was dissolved in
DMSO and purified by preparative C18 HPLC at 254 nm (35%
MeOH in H2O, 0.1% TFA, tR = 11 min) to yield 10.0 mg of
ammosamide C (16) as a bright red solid. All HRMS, UV/vis, and
NMR data agreed with reported data. HRMS (ESI-Q-TOF) for 14:
m/z [M + H]+ 308.0372 (calcd for C12H11ClN5OS, 308.0367, Δ0.72
ppm); for 15: m/z [M + H]+ 292.0596 (calcd for C12H11ClN5O2,
assay and against the 60-cell-line panel at the National Cancer
Institute (Figure S2).19 Ammosamide C (16) displayed modest
cytotoxicity, exhibiting the highest potency against the human
colorectal cancer cell line HCT-116 of all tested cancer cell
lines. Importantly, 16 persisted for weeks in the bioassay media
and only slowly converted to ammosamide B (15) with a t1/2
≈
10 days, so the bioactivity can be attributed exclusively to 16.
Ammosamides A (14), B (15), D (17), and G (24) showed
very little cytotoxicity, in agreement with other reports,10,11,13
while propylamine adduct 23 presented activity similar to 16.
Interestingly, ammosamides L and M, carrying the aliphatic
amines hexylamine and isopropylamine, also showed significant
cytotoxicity against the HCC44 cancer cell line (0.50 and 1.1
μM).11 The fact that 23 showed activity was unexpected and
defies our hypothesis that the unsubstituted electrophilic
iminium functionality is tied to its biological activity, assuming
cytotoxicity toward eukaryotic cells is related to the
metabolite’s ecological function.
Table 2. Cytotoxicity of Selected Compounds against
a
Human Colorectal Cancer Cell Line HCT-116
compound
HCT-116 IC50 (μM)
14 (A)
15 (B)
16 (C)
17 (D)
23
NSA
29
5.5
NSA
6.8
24 (G)
26
39
NSA
a
NSA = no significant activity. Growth inhibitions (%) at the highest
test concentration (78.125 μg/mL) were measured as 19% (14), 97%
(17), and 26% (26). Values are averages of three measurements.
Lastly, the antimicrobial activities of ammosamides A−C
(14−16) were determined against Bacillus oceanisediminis
CNY-977 and Escherichia coli MG1655 using a disk diffusion
assay.20 Although little activity was observed against the E. coli
strain, a significant zone of inhibition for 16 alone was apparent
against the Bacillus strain (Figure S3). The antimicrobial
activity of 16 against a Gram-positive marine bacterial strain
may reflect the true ecological role of the natural product.
EXPERIMENTAL SECTION
■
General Experimental Procedures. 1H NMR and 2D NMR
spectra were recorded at 500 MHz in DMSO-d6 (residual solvent
referenced to 2.50 ppm for 1H, 39.52 ppm for 13C) on a Varian Inova
500 MHz NMR spectrometer or a Jeol 500 MHz NMR spectrometer.
Reactions were analyzed with an analytical 1100 Series Agilent
E
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
292.0596, Δ1.79 ppm); for 16: m/z [M]+ 276.0646 (calcd for
showed labeling of 16 (tR = 3.0 min) with probe 19 (tR = 7.1 min) to
give adduct 20 (tR = 3.8 min) (Figure S7). In this instance, a
significant amount of ammosamide A (14) formed via cleavage of 20
during workup and purification.
C12H11ClN5O+, 276.0647, Δ0.27 ppm).
Conversion of Ammosamide C in Cultivation Medium, pH
10. Ammosamide C TFA (16) (1.0 mg, 0.0026 mmol) was dissolved
in 500 μL of seawater-based A1 medium at pH 10.0. As a control
reaction, the same amount of 16 was dissolved in seawater-based A1
medium at pH 6.8. The solutions were kept at room temperature (rt).
HPLC samples were prepared by diluting the reaction mixture (10 μL)
with CH3CN (20 μL), and 5 μL was analyzed for the consumption of
16 (tR = 4.8 min) and the formation of 15 (tR = 6.9 min) and 18 (tR =
5.3 min) by analytical C18 ELSD-UV/vis-HPLC (1% to 60% CH3CN
in H2O, 0.1% formic acid) for 8 days and after 14, 18, 20, 24, and 28
days (Figure 2a, Figures S1a and S1b).
Conversion of Ammosamide C in PBS Buffer, pH 10.
Ammosamide C TFA (16) (2.0 mg, 0.0051 mmol) was dissolved in
1.0 mL of phosphate-buffered saline (PBS) buffer at pH 10 (137 mM
NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4). The
solution was kept at rt, and the conversion of 16 (tR = 4.8 min) to 15
(tR = 6.9 min) and 18 (tR = 5.3 min) was monitored by analytical C18
ELSD-UV/vis-HPLC and LC-HRMS after 1, 2, and 5 days, as
described for the cultivation medium above (Figure S1d).
Conversion of Ammosamide C in Aqueous Sodium
Bicarbonate Solution, pH 8.1. Ammosamide C TFA (16) (2.0
mg, 0.0051 mmol) was dissolved in 1.0 mL of saturated NaHCO3
solution at pH 8.1. The solution was kept at rt, and the conversion of
16 (tR = 4.8 min) to 15 (tR = 6.9 min) was monitored by analytical
C18 ELSD-UV/vis-HPLC and LC-HRMS after 1, 2, and 5 days, as
described for the cultivation medium above (Figure 2b, Figure S1c).
Conversion of Ammosamide C in PBS Buffer, pH 10,
Supplemented with L-Cysteine. Ammosamide C TFA (16) (0.40
mg, 0.0010 mmol, 1 equiv) and L-cysteine (1.3 mg, 0.0072 mmol, 7
equiv) were dissolved in 400 μL of PBS buffer at pH 10. The solution
was kept at rt, and the conversion of 16 (tR = 4.8 min) to 14 (tR = 8.3
min), 15 (tR = 6.9 min), 18 (tR = 5.3 min), and 20 (tR = 4.9 min) was
monitored by analytical C18 ELSD-UV/vis-HPLC and LC-HRMS
after 1, 3, 4, 5, 6, 7, and 10 days as described for the cultivation
medium above (Figure 2c, Figure S1e).
Synthesis and Purification of Ammosamide C Cysteine Thiol
Probe Adduct (20). Ammosamide C TFA (16) (10.0 mg, 0.0257
mmol, 1 equiv) and cysteine thiol probe (19) (17 mg, 0.053 mmol, 2
equiv) were dissolved in dry and sparged DMF (1.0 mL) in a vial
purged with N2. For cleaner conversion to 20 Et3N was not added.
The reaction mixture was kept under N2 for 22 h at rt. The reaction
mixture was concentrated under N2 and high vacuum overnight, and
the product was purified by semipreparative C8 HPLC at 254 nm
(35% CH3CN in H2O, 0.1% TFA, tR = 6.0 min) to afford 6.4 mg
(35%) as a red-purple TFA salt. UV (MeOH) λmax 435 (3.17), 560
(2.67) nm; 1H and 13C NMR, see Table 1; HRMS (ESI-Q-TOF) m/z
[M]+ 591.0230 (calcd for C23H21BrClN6O4S+, 591.0211, Δ3.29 ppm).
Synthesis and Purification of Ammosamide C Propylamine
Adduct (23). Ammosamide C TFA (16) (6.0 mg, 0.015 mmol, 1
equiv) and propylamine (2.7 μL, 0.033 mmol, 2.1 equiv) were each
dissolved in dry and sparged DMF (500 μL) in vials purged with N2.
The propylamine solution was added to the ammosamide C solution,
and the reaction mixture was kept under N2 for 22 h at rt. Analysis of
the mixture by analytical C18 ELSD-UV/vis-HPLC (10% to 100%
CH3CN in H2O, 0.1% FA) showed complete conversion of the
starting material 16 (tR = 3.0 min) to product 23 (tR = 4.0 min)
(Figure S8). The reaction mixture was concentrated under N2 and
high vacuum overnight, and the product was purified by semi-
preparative CN HPLC at 260 nm (19% CH3CN in H2O, 0.1% TFA, tR
= 10.0 min) followed by semipreparative C8 HPLC at 260 nm (25%
CH3CN in H2O, 0.1% TFA, tR = 7.0 min) to give 5.9 mg (82%) as a
blue TFA salt. UV (MeOH) λmax 345 (3.76), 430 (3.34), 575 (3.45)
1
nm; H and 13C NMR, see Table 1; HRMS (ESI-Q-TOF) m/z [M]+
333.1230 (calcd for C15H18ClN6O+, 333.1225, Δ0.27 ppm).
Synthesis and Purification of Ammosamide G (24).
Ammosamide C TFA (16) (10.8 mg, 0.0277 mmol, 1 equiv) and 4-
chloroaniline (7.5 mg, 0.059 mmol, 2.1 equiv) were each dissolved in
dry and sparged DMF (500 μL) in vials purged with N2. The 4-
chloroaniline solution was added to the ammosamide C solution, and
the reaction was kept under N2 for 18 h at rt. Analysis of the mixture
by analytical C18 ELSD-UV/vis-HPLC (10% to 100% CH3CN in
H2O, 0.1% FA) showed a 25−30% conversion of the starting material
16 (tR = 3.0 min) to product 24 (tR = 4.7 min) (Figure S9). The
reaction mixture was concentrated under N2 and high vacuum
overnight, and the product was purified by semipreparative C8 HPLC
at 260 nm (30% CH3CN in H2O, 0.1% TFA, tR = 10.0 min) to give 3.5
mg (32%) as a blue solid. UV (MeOH) λmax 335 (3.38), 575 (2.95)
Conversion of Ammosamide B into Ammosamide D (17)
with Light. For three different small-scale reactions ammosamide B
(15) (1.0 mg, 0.0034 mmol) was dissolved in MeOH (500 μL), THF
(500 μL), or DMSO (500 μL) and exposed to direct sunlight, while
allowing for the exchange of air using a loosened cap for up to 21 days.
The reaction was monitored by analytical C18 ELSD-UV/vis-HPLC
(1% to 60% CH3CN in H2O, 0.1% FA) after 0, 2, 6, 9, 13, and 21 days.
The conversion of 15 (tR = 6.9 min) to 17 (tR = 4.4 min) in MeOH
was indicated by a color change from purple to bright orange (Figure
S5). For the isolation and purification of 17, ammosamide B (15) (6.3
mg, 0.022 mmol) was dissolved in MeOH (1.0 mL) and exposed to
the same conditions. As the reaction is light dependent, cloudy
weather conditions required a longer experiment length (5 weeks) for
good conversion (∼50%). The reaction was concentrated under N2,
and the product was purified by semipreparative C8 HPLC at 260 nm
(18% CH3CN in H2O, 0.1% TFA, tR = 5.5 min) to furnish 3.1 mg
(47%) as a bright orange solid. All HRMS, UV/vis, and NMR data
1
nm; H and 13C NMR, see Table 1; HRMS (ESI-Q-TOF) m/z [M +
H]+ 401.0691 (calcd for C18H15Cl2N6O, 401.0679, Δ1.29 ppm).
Synthesis and Purification of Ammosamide C N-Acetylcyste-
amine Adduct (26). Ammosamide C TFA (16) (6.0 mg, 0.015
mmol, 1 equiv) and N-acetylcysteamine (3.5 μL, 0.033 mmol, 2.1
equiv) were each dissolved in dry and sparged DMF (500 μL) in vials
purged with N2. The N-acetylcysteamine solution was added to the
ammosamide C solution, followed by dry Et3N (6.1 μL, 0.044 mmol,
2.8 equiv), and the reaction mixture was kept under N2 for 22 h at rt.
Analysis of the mixture by analytical C18 ELSD-UV/vis-HPLC (10%
to 100% CH3CN in H2O, 0.1% FA) showed nearly complete
conversion of the starting material (16, tR = 3.0 min) to product 26 (tR
= 3.3 min) (Figure S10). The reaction mixture was concentrated under
N2 and high vacuum overnight, and the product was purified by
semipreparative C8 HPLC at 270 nm (10% CH3CN in H2O, 0.1%
TFA, tR = 12.5 min) followed, again, by semipreparative C8 HPLC at
270 nm (20% CH3CN in H2O, 0.1% TFA, tR = 5.5 min) to give 5.6 mg
(72%) as a red TFA salt. UV (MeOH) λmax 450 (3.37), 545 (3.00)
1
agreed with reported data. H NMR δH 9.36 (1H, s), 9.17 (1H, s),
8.55 (1H, s), 8.18 (1H, q, J = 4.7 Hz), 8.02 (1H, s), 8.00 (1H, s), 2.77
(3H, d, J = 4.7 Hz) ppm; HRMS (ESI-Q-TOF) m/z [M + H]+
309.0389 (calcd for C12H10ClN4O4, 309.0385, Δ1.82 ppm).
Labeling Experiment of Streptomyces sp. CNR-698 with the
Cysteine Thiol Probe. An extract from 9 × 1 L cultures of the
marine Streptomyces sp. CNR-698 was prepared as described
previously.17 The extract was dry-loaded onto C18 silica gel and
fractionated using a solid-phase extraction RP C18 cartridge (20% and
50% CH3CN in H2O, 0.1% TFA). Both fractions were combined and
evaporated to dryness to provide 2.19 g. The crude fraction and
cysteine thiol probe (19) (200 mg, 0.629 mmol) were dissolved in dry
and sparged DMF (20 mL) in a vial purged with N2. Dry Et3N (175
μL, 1.26 mmol) was added, and the reaction mixture was kept
overnight under N2 at rt. Analysis of the reaction mixture by analytical
C18 ELSD-UV/vis-HPLC (10% to 100% CH3CN in H2O, 0.1% FA)
1
nm; H and 13C NMR, see Table 1; HRMS (ESI-Q-TOF) m/z [M]+
393.0888 (calcd for C16H18ClN6O2S+, 393.0895, Δ1.67 ppm).
Synthesis and Purification of Ammosamide C Ethanethiol
Adduct (27). Ammosamide C TFA (16) (3.0 mg, 0.008 mmol, 1
equiv) and ethanethiol (1.2 μL, 0.017 mmol, 2.2 equiv) were each
dissolved in dry and sparged DMF (250 μL) in vials purged with N2.
F
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
The ethanethiol solution was added to the ammosamide C solution,
followed by dry Et3N (3.0 μL, 0.022 mmol, 2.8 equiv), and the
reaction mixture was kept under an atmosphere of air (conversion was
<5% after 48 h under oxygen-free conditions) for 22 h at rt. Analysis of
the mixture by analytical C18 ELSD-UV/vis-HPLC (10% to 100%
CH3CN in H2O, 0.1% FA) showed nearly complete conversion of the
starting material 16 (tR = 3.0 min) to a mixture of products 27 (tR =
3.9 min), 14 (tR = 5.5 min), and 15 (tR = 4.6 min) (Figure S11). The
reaction mixture was concentrated under N2 and high vacuum
overnight, and the product was purified by semipreparative C8 HPLC
at 270 nm (20% CH3CN in H2O, 0.1% TFA, tR = 14.5 min) to give 1.4
at rt or 37 °C for B. oceanisediminis and E. coli, respectively, and
inhibition zone diameters were measured thereafter.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.6b00773.
One-dimensional (1H) and two-dimensional (COSY,
HSQC, HMBC) NMR spectra of 17, 20, 23, 24, 26, and
27 as well as UV/vis and HRMS data of all characterized
compounds (PDF)
1
mg (40%) as a light red TFA salt. H and 13C NMR, see Table 1;
HRMS (ESI-Q-TOF) m/z [M]+ 336.0691 (calcd for C14H15ClN5OS+,
336.0680, Δ0.12 ppm).
Synthesis and Purification of Ammosamide C 4-Bromoben-
zyl Mercaptan Adduct (28). Ammosamide C TFA (16) (6.0 mg,
0.015 mmol, 1 equiv) and 4-bromobenzyl mercaptan (6.6 mg, 0.033
mmol, 2.1 equiv) were each dissolved in dry and sparged DMF (500
μL) in vials purged with N2. The bromobenzyl mercaptan solution was
added to the ammosamide C solution, followed by dry Et3N (6.1 μL,
0.044 mmol, 2.8 equiv), and the reaction was kept under an
atmosphere of air (conversion was <5% after 48 h under oxygen-
free conditions) for 22 h at rt. Analysis of the mixture by ELSD-UV/
vis-HPLC (10% to 100% CH3CN in H2O, 0.1% FA) showed a mixture
of starting material and products 28 (tR = 5.1 min), 14 (tR = 5.5 min),
and 15 (tR = 4.6 min) (Figure S12). The reaction mixture was
concentrated under N2 and high vacuum overnight, and the product
was purified by semipreparative C8 HPLC at 260 nm (30% CH3CN in
H2O, 0.1% TFA, tR = 17.0 min) to give 1.3 mg (14%) as a red TFA
salt. HRMS (ESI-Q-TOF) m/z [M]+ 475.9941 (calcd for
C19H16BrClN5OS+, 475.9942, Δ0.17 ppm). The adduct could not be
characterized by UV and NMR since it readily decomposed to
ammosamide A.
AUTHOR INFORMATION
Corresponding Author
*Tel: (858) 534-7121. E-mail: chughes@ucsd.edu.
ORCID
Chambers C. Hughes: 0000-0002-7000-6438
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS
■
This work was supported by start-up funds from the Scripps
Institution of Oceanography. D.R. thanks the Deutsche
Forschungsgemeinschaft (RE 3704/1-1) for postdoctoral
funding. We acknowledge G. Castro-Falcon and Dr. D. Hahn
́
for providing pure ammosamide C for this study.
Cytotoxicity Assays. The human colorectal cancer cell line HCT-
116 (ATCC CCL-247) was cultured and passaged as described
previously.17 Cytotoxicity in half-maximal inhibitory concentrations
(IC50) of ammosamides A−D (14−17) and the synthetic derivatives
(23, 24, 26) was tested using the colorimetric 3-(4,5-dimethylthiazol-
2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium
(MTS) bioassay.21 Compounds in DMSO were added at a
concentration of 10 mg mL−1, and the bioassay was conducted as
reported previously.17 IC50 determination was performed using a four-
parameter sigmoidal fit. DMSO (ATCC) and the standard anticancer
drug etoposide VP-16 (Sigma-Aldrich), with an IC50 of 0.49−4.9 μM
at a concentration of 4 mg mL−1 in DMSO, were tested as negative
and positive controls, respectively.
NCI60 Cancer Cell Line Screening. Ammosamide C (16) was
submitted to the U.S. National Cancer Institute 60 human tumor cell
line anticancer drug screen including blood cancer (CML, ALL, CLL),
NSCLC, colon cancer, central nervous system cancer, melanoma,
ovarian cancer, renal cancer, prostate cancer, and breast cancer cell
lines.19 16 was tested at a single high dose (10 μM) in the full panel
comprising all 60 cell lines. Values were reported for growth inhibition
(GI50) corresponding to percentage of growth of treated cells relative
to nontreated cells.
Disk Diffusion Assay. Disk diffusion tests for antimicrobial
activity were carried out according to the Kirby−Bauer method.20
Inoculums of the Gram-negative Escherichia coli MG1655 and the
Gram-positive marine Bacillus oceanisediminis CNY-977 were prepared
from 24 h precultures in LB (Luria−Bertani) medium, pH 7.0,
cultivated at rt and 200 rpm on a rotary shaker. LB agar plates were
overlaid with inoculums of the strains, which were streaked over the
agar surface using a sterile swab. Compounds 14−16 (10 μg) were
spotted on sterile paper discs with a 6 mm diameter (Becton
Dickinson, BBL blank paper discs) at a concentration of 10 mg mL−1
in MeOH (10 μL) and dried prior to their application on the
inoculated agar surface. MeOH (10 μL) and ciprofloxacin (5 μg)
(Becton Dickinson, BBL Sensi-Disc) were tested as negative and
positive controls, respectively. All agar plates were incubated for 48 h
REFERENCES
■
(1) Marek, R.; Seckarova, P.; Hulova, D.; Marek, J.; Dostal, J.;
Sklenar, V. J. Nat. Prod. 2003, 66, 481−486.
(2) Pina, I. C.; Sanders, M. L.; Crews, P. J. Nat. Prod. 2003, 66, 2−6.
(3) Utkina, N. K.; Gerasimenko, A. V.; Popov, D. Y. Russ. Chem. Bull.
2003, 52, 258−260.
(4) Fort, D. M.; Litvak, J.; Chen, J. L.; Lu, Q.; Phuan, P.; Cooper, R.;
Bierer, D. E. J. Nat. Prod. 1998, 61, 1528−1530.
(5) Fu, P.; Legako, A.; La, S.; MacMillan, J. B. Chem. - Eur. J. 2016, 22
(10), 3491−3495.
̂
(6) Hughes, C. C.; MacMillan, J. B.; Gaudencio, S. P.; Jensen, P. R.;
Fenical, W. Angew. Chem., Int. Ed. 2009, 48, 725−727.
(7) Antunes, E. M.; Copp, B. R.; Davies-Coleman, M. T.; Samaai, T.
Nat. Prod. Rep. 2005, 22, 62−72.
(8) Aotani, Y.; Nagata, H.; Yoshida, M. J. Antibiot. 1997, 50 (7),
543−545.
̂
(9) Hughes, C. C.; MacMillan, J. B.; Gaudencio, S. P.; Fenical, W.; La
Clair, J. J. Angew. Chem., Int. Ed. 2009, 48, 728−732.
(10) Reddy, P. V. N.; Jensen, K. C.; Mesecar, A. D.; Fanwick, P. E.;
Cushman, M. J. Med. Chem. 2012, 55, 367−377.
(11) Pan, E.; Oswald, N. W.; Legako, A. G.; Life, J. M.; Posner, B. A.;
MacMillan, J. B. Chem. Sci. 2013, 4, 482−488.
(12) Hughes, C. C.; Fenical, W. J. Am. Chem. Soc. 2010, 132, 2528−
2529.
(13) Pan, E.; Jamison, M.; Yousufuddin, M.; MacMillan, J. B. Org.
Lett. 2012, 14, 2390−2393.
(14) Reddy, P. V. N.; Banerjee, B.; Cushman, M. Org. Lett. 2010, 12,
3112−3114.
(15) Wu, Q.; Jiao, X.; Wang, L.; Xiao, Q.; Liu, X.; Xie, P. Tetrahedron
Lett. 2010, 51, 4806−4807.
(16) Takayama, Y.; Yamada, T.; Tatekabe, S.; Nagasawa, K. Chem.
Commun. 2013, 49, 6519−6521.
(17) Castro-Falcon, G.; Hahn, D.; Reimer, D.; Hughes, C. C. ACS
Chem. Biol. 2016, 11, 2328−2336.
(18) Lam, C. F. C.; Grkovic, T.; Pearce, A. N.; Copp, B. R. Org.
Biomol. Chem. 2012, 10, 3092−3097.
G
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
(19) Shoemaker, R. H.; Shoemaker, R. H. Nat. Rev. Cancer 2006, 6
(10), 813−823.
(20) National Committee for Clinical Laboratory Standards. Methods
for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow
Aerobically: Approved Standards, sixth ed.; NCCLS: Wayne, PA, USA,
2003.
(21) Cory, A. H.; Owen, T. C.; Barltrop, J. A.; Cory, J. G. Cancer
Commun. 1991, 3 (7), 207−212.
H
DOI: 10.1021/acs.jnatprod.6b00773
J. Nat. Prod. XXXX, XXX, XXX−XXX