An Electrophilic Natural Product Provides a Safe and Robust Odor Neutralization Approach to Counteract Malodorous Organosulfur Metabolites Encountered in Skunk Spray

Article abstract of DOI:10.1021/acs.jnatprod.9b00415

The anal secretions of skunks comprise several types of malodorous organosulfur compounds. The pungent metabolites are used defensively by skunks to repel threats posed by predators, and in many parts of the world, those perceived threats include humans a

Full text of DOI:10.1021/acs.jnatprod.9b00415

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
An Electrophilic Natural Product Provides a Safe and Robust Odor
Neutralization Approach To Counteract Malodorous Organosulfur
Metabolites Encountered in Skunk Spray
Lin Du,† Charissa Munteanu,‡ Jarrod B. King,† Doug E. Frantz,‡ and Robert H. Cichewicz*,†
†Natural Products Discovery Group, Institute for Natural Products Applications and Research Technologies, Department of
Chemistry and Biochemistry, Stephenson Life Science Research Center, University of Oklahoma, 101 Stephenson Parkway,
Norman, Oklahoma 73019, United States
‡Department of Chemistry, The University of Texas at San Antonio, San Antonio, Texas 78249, United States
S
* Supporting Information
ABSTRACT: The anal secretions of skunks comprise several
types of malodorous organosulfur compounds. The pungent
metabolites are used defensively by skunks to repel threats
posed by predators, and in many parts of the world, those
perceived threats include humans and their pets. The
extremely low thresholds for detection of the organosulfur
metabolites make efforts to “de-skunk” people, animals, and
clothing a process fraught with many challenges. The fungal-
derived metabolite pericosine A (4) is a promiscuous yet
stabile electrophilic compound that we propose is used by
some fungi as a novel form of chemical defense. Our
investigations have indicated that pericosine A readily reacts
with skunk-spray secretions to transform them into odorless
products. Mechanistic and computational studies suggested that pericosine A and its synthetic analogues react via SN2′-type
mechanisms with thiols and thioacetates under aqueous conditions to generate stable thioethers. Testing revealed that
pericosine A did not cause skin or eye irritation and was highly effective at deodorizing skunk anal gland secretions when
formulated to include adjunctive cosmetic ingredients.
uman−wildlife conflict is a growing global problem
compounds undergo autohydrolysis or are subjected to
microbial biotransformation into their thiol forms.3 The
pungent odor of skunk thiols is highly persistent (e.g.,
compounds are oily and easily adhere to clothing, hair, and
skin), and they are readily detectable by humans and other
animals at low concentrations (e.g., the human nose can detect
concentrations as low as ∼10 ppb).4 Animals unfortunate
enough to be directly exposed to skunk sprays can develop a
variety of physical symptoms including ocular edema,
conjunctivitis, drooling, and squinting. In some instances, the
thiols in skunk sprays can induce oxidative damage to
hemoglobin and cause Heinz body anemia in affected animals.5
It is rather difficult to fully “de-skunk” (i.e., remove the
detectable scent from) sprayed objects and animals (e.g., pet
dogs). The well-known Krebaum’s formula3 containing
hydrogen peroxide and baking soda is a somewhat effective
home remedy that converts the thiols to odorless sulfonic acids
(Figure 1B). However, this highly oxidative combination of
ingredients can cause dermal and ocular irritation as well as
bleach hair, fur, and textiles. Several commercial products are
H
resulting from adverse interactions between humans and
nondomesticated plants and animals.1,2 The outcomes of these
interactions can range from unpleasant to beastly with humans
and/or wildlife suffering a variety of consequences. One
distasteful type of conflict that occurs with considerable
frequency involves encounters between humans and skunks
(Mephitidae). Skunks possess a unique anal gland system that
has evolved to generate and store cocktails of extremely
odoriferous metabolites. When threatened, skunks expel those
fluids in the form of a spray, which serves to overwhelm
predators and deter further attacks. Exposure to skunk
defensive secretions (also known as skunk spray or musk) is
a common and challenging problem for humans, pets, and
livestock throughout North and South American countries as
well as portions of the Malay Archipelago. Skunk sprays are
generally composed of organic thiols, including (E)-2-butene-
1-thiol and 3-methyl-1-butanethiol (Figure 1A) and their
thioacetate derivatives. The two aforementioned thiols are the
major (>50%) volatile components of skunk sprays and are
largely responsible for their extraordinarily repellent odors.3
Although the thioacetates tend to be less or nonodoriferous,
their presence can contribute to the continual emanation of
odors from affected humans, animals, and objects as the
Received: May 3, 2019
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
members of our research team to the samples treated with
these products confirmed that the thiol remained readily
detectable to the nose). Several of the products did appear to
contain surfactant-like molecules, which implied their intended
functions might be the removal of organosulfur compounds
from affected surfaces. Interestingly, one product (Product A)
was found to readily convert 1 to a water-insoluble, crystalline
product (3) (Figure 1B). Despite the manufacturer’s claim that
the product was “ECO friendly” and contained a “non-toxic
mineral-based deodorizer,” our analyses of the odorless
reaction product [NMR and X-ray diffraction data (Figures
1B and S2)] as well as the product ingredients revealed that
Product A contained chloramine-T, which has been identified
as a potentially toxic and irritating oxidant that is primarily
used as an industrial biocide and disinfectant.6
Considering the problems associated with existing “de-
skunking” strategies, we set out to identify a safer and more
effective skunk-spray-deodorizing solution. One of the lead
substances to emerge from our search was the chemoreactive
fungal metabolite pericosine A (4) (Figure 2), which had been
previously reported by our group to protect fungi from
nucleophilic antibiotics/toxins.7 On the basis of those findings,
we hypothesized that pericosine A and its synthetic analogues
might offer an appealing solution to the neutralization of thiols
and even thioacetates via SN2′ substitution mechanisms under
aqueous conditions. Herein, we describe the unique thiol-
reactive properties of pericosine A and its analogues. In
addition, we report on the development and testing of reaction
conditions using pericosine A that could be employed to “de-
skunk” sensitive surfaces (i.e., skin and eyes) as well as present
a concise summary of preliminary toxicological assay data
aimed at assessing the dermal and ocular safety of pericosine A.
Figure 1. (A) Structures of the major odoriferous thiols in skunk
defensive secretion and the test agents [2-phenylethanethiol (1) and
4-bromo-α-toluene thioacetate (2)] that were used as model
compounds in this study. (B) Current and proposed chemical
approaches to deodorization discussed in this report.
marketed as “de-skunking” agents (refer to the Supporting
Information for examples), which are purported to break down
or eliminate skunk spray components. However, in our testing
of these products by LCMS monitoring using the model
compound 2-phenylethanethiol (1) (Figure 1A), five of the six
formulas that we tested failed to provide evidence that the
target thiol had been chemically altered (passive exposure of
RESULTS AND DISCUSSION
■
Pericosine A is composed of a methyl shikimate skeleton with a
chlorine atom attached to the C-6 position, which gives this
Figure 2. Reactivity of pericosine A (4) with selected sulfur-containing compounds (1, 2, 5, and 6) in 50% MeOH with Et3N as the catalyst.
Reactant/reagent ratios for all reactivity tests, pericosine A (1 equiv)/sulfur-containing compound (1 equiv)/Et3N (2 equiv). Products 7 (scalemic)
and 8 (racemic) were generated in large scale under modified reaction conditions [pericosine A (1 equiv)/1 (1.5 equiv)/Et3N (2 equiv) for making
7 and pericosine A (1 equiv)/2 (1.5 equiv)/Et3N (10 equiv) for making 8].
B
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
molecule a unique set of electrophilic properties and makes it
an excellent substrate for reactions with a wide range of
nucleophiles (Figures 2 and S3).7 Our studies began by testing
the reactivity of pericosine A against four classes of sulfur-
containing model compounds: thiol (2-phenylethanethiol, 1),
thioester (4-bromo-α-toluene thioacetate, 2), thioether
(dibenzyl sulfide, 5), and disulfide (dibenzyl disulfide, 6)
(Figure 2). Pericosine A reacted with thiol 1 under both
neutral and basic conditions to generate product 7 (Figure
S4A,B). The catalytic base (Et3N in this case) significantly
improved the rate of the reaction. Notably, compound 7 and
similar thioether addition products did not produce noticeable
odors; when reaction vessels were removed from the hood, no
significant foul smells were encountered. Similarly, thioester 2
readily reacted with pericosine A under basic conditions to
form 8. Neither thioether 5 nor disulfide 6 reacted with
pericosine A under the experimental conditions we inves-
tigated (Figures 2 and S4).
To understand (i) the spectrum of reactivity for pericosine A
with organosulfur compounds under different reaction
conditions, (ii) opportunities that existed to improve its
thiol-neutralizing capabilities, and (iii) possible approaches for
using pericosine A in real-world situations to neutralize
odoriferous skunk metabolites, we set about determining
how the natural product mechanistically functioned with
different sulfur-containing substrates as well as explored
alternative reagents that would be appropriate for topical use.
As an entry point to these studies, we began by examining the
structures of products 7 and 8, which were solved through
analyses of their 1D (1H and 13C) and 2D [1H−1H COSY
(correlation spectroscopy), HSQC (heteronuclear single
quantum coherence), and HMBC (heteronuclear mulitple
bond correlation)] NMR data (Tables 1 and 2; Figures S25−
S31) as well as high-resolution ESIMS (electrospray ionization
mass spectrometry) data. Further chemical studies examining
the absolute configurations of 7 and 8 and the reaction
mechanisms leading to their formation revealed that the
product distributions for these reactions were highly depend-
ent on both the reaction conditions used and stoichiometric
ratios of the substrates and reagents employed. During our
initial synthesis of 7, an excess of 1 (1.5 equiv) and the
catalytic base Et3N (2 equiv) was used to maximize product
yield. Under these conditions, product 7 was identified as a
scalemic mixture by chiral HPLC (high-performance liquid
chromatography) (Figures 3F and S5E). The relative
configurations of C-5 and C-6 were determined as 5R*,6R*
via comparisons of calculated 13C NMR data for the C-6
epimers 7a (5R,6R) and 7b (5R,6S) with experimentally
derived data (Figure S5A,B). Interestingly, when the reaction
was performed using an equal molar amount of 1 and
pericosine A in the absence of Et3N, it resulted in the
formation of pure stereoisomer 7a (Figures 3E and S5E). The
absolute configuration of 7a was confirmed as 3R,4R,5R,6R
through an analysis of its experimental and calculated ECD
(electronic circular dichroism) spectra and specific rotation
values (Figures S5A,G). The reaction between pericosine A (1
equiv) and 2 (1.5 equiv) was also promoted when excess Et3N
(10 equiv) was used to form racemic 8 (Figures S5C,F).
Enantiomers 8a (3R,4R,5R,6R) and ent-8a (3S,4S,5S,6S) were
separated by chiral HPLC, and their absolute configurations
were determined by comparing their ECD spectra with data
obtained for 7a (Figure S5G).
C
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. 13C NMR Data of 7, 8, and 23−28
a
a
b
b
b
b
b
b
no.
7
8
23
24
25
26
27
28
1
2
3
130.9
136.4
65.7
130.6
136.9
65.7
130.3
141.1
70.1
134.3
137.1
69.8
134.3
136.8
66.8
130.0
141.4
70.1
134.0
137.7
69.7
133.6
137.6
66.8
4
66.5
66.6
72.2
75.4
72.1
72.3
75.4
71.8
5
74.5
74.2
75.9
69.4
73.9
75.4
69.9
74.2
6
46.2
45.3
47.7
48.0
48.1
46.8
46.8
46.9
7
8
166.6
52.5
166.5
52.5
167.9
52.5
168.7
52.5
168.7
52.4
167.8
52.4
168.7
52.5
168.6
52.4
1′
2′
3′
4′
5′
6′
7′
8′
35.3
36.5
37.5
36.8
37.4
37.4
37.7
37.4
35.8
38.5
39.2
38.0
136.9
130.9
131.9
121.4
131.9
130.9
139.1
132.1
132.6
121.9
132.6
132.1
139.4
132.3
132.5
121.7
132.5
132.3
139.4
132.2
132.5
121.6
132.5
132.2
140.2
128.7
128.7
126.6
128.7
128.7
141.8
129.6
129.4
127.3
129.4
129.6
142.1
129.6
129.4
127.2
129.4
129.6
142.1
129.6
129.4
127.2
129.4
129.6
a
b
Spectra measured in CDCl3. Spectra measured in methanol-d4.
While optimizing the reaction conditions to prepare 7,
fully rule out the potential contribution that a Michael-
addition-like pathway may play in this process.
varying ratios of 1, pericosine A, and Et3N were tested. It was
observed that, in the presence of excess Et3N, pericosine A was
converted to an epoxide, pericoxide (9),7 as well as the
methoxy analogue (−)-pericosine C (10)8 (Figures 3A−D and
S6A−D). Considering the conversion of pericosine A to
epoxide 9 under basic conditions, the question arose whether
pericosine A reacted directly with 1 or, instead, was being
converted first to reactive species 9. To address this problem,
the hydroxy groups in pericosine A were protected, yielding
triacetylated derivative 11. This compound proved readily able
to react with 1 in the presence of Et3N to form 12 (Figure 3I).
Notably, 12 was determined to bear the same absolute
configuration as 7a (Figure S5D,G). These results indicated
that pericosine A could serve as a direct precursor to 7a (i.e.,
pericosine A did not require activation to its epoxide form for
the reaction to occur). Additionally, we tested whether 9 might
be involved in the formation of ent-7a, but that test revealed 7a
was only produced when 9 and 1 were reacted in the absence
of Et3N (Figure 3G). Furthermore, we noted that, in reactions
carried out with either excess pericosine A or 9, no ent-7a was
detected. It was also determined that incubation of 7 with Et3N
did not contribute to its racemization (Figure S6E). Taken in
combination, the accumulated evidence suggested that ent-7a
was not produced through a reaction with 9, but instead, its
production was based on the amount of 1 used in the
reactions. Subsequently, we observed that racemization of 7a
occurred when it was exposed to 1 and that no basic catalyst
was required for this process to proceed (Figure 3H). Thus,
the overall reaction mechanism was proposed to occur as
depicted in Figure 3J. To summarize, under basic conditions,
pericosine A was partially converted to epoxide analogue 9,
and both pericosine A and 9 are believed to have competed to
react with 1 (syn-SN2′ for pericosine A and anti-SN2′ for 9).
This was followed by racemization of product 7a when excess
1 was present. A similar mechanistic process was determined to
occur when pericosine A was reacted with thioacetate 2
(Figure S7). These observations agree with our previous
computational studies,7 which supported the stereoselective
SN2′ substitution-type reaction mechanisms for pericosine A
and 9. Although our accumulated chemical and spectroscopic
evidence support this general pathway, at this time, we cannot
To further investigate and optimize the reactivity of
pericosines with thiols and thioesters, two chlorinated
pericosine analogues, 19 and 20, were synthesized from
(−)-shikimic acid as described8−13 with some procedural
modifications (Scheme 1). Both 19 and 20 readily reacted with
1 and 2 in 50% MeOH in the presence of Et3N (Figure S8B).
Although the reaction rates of thiols with 19 and 20 were
comparable to what had been observed for pericosine A, 19
and 20 appeared to be more water-soluble than pericosine A
(i.e., solutions containing pericosine A appeared slightly
cloudy, whereas solutions of 19 and 20 were clear). This
proposition was also supported by the observed increased
relative polarity of 19 and 20 as compared to pericosine A as
estimated by LCMS analysis (Figure S8A). Since 19 and 20
showed similar substrate reactivities (Figure S8B), the more
hydrophilic analogue 19 (Figure S8A) was selected for
reactivity studies. Like pericosine A, 19 was converted to
epoxide analogue 21 and the methoxy analogue 22 in 50%
MeOH in the presence of Et3N (Scheme 1 and Figures S12A−
C). The reaction of 19 with 1 yielded three thiol-substituted
products, 23−25, and the ratios of those compounds varied on
the basis of the amount of Et3N (2 to 10 equiv) used in each
reaction (Scheme 1 and Figure S8B). A similar reactivity
profile was observed for the reaction between 19 and 2
resulting in the formation of three substituted products 26−28
(Scheme 1 and Figure S8B). The absolute configurations of
3
21−28 were determined by comparing their JH,H coupling
constants and 13C NMR data with those of selected model
compounds (Figures S9−S11). The results of those tests
(Figure S12D) suggested that both pericosines reacted with
thiol 1 via the same SN2′ mechanisms (syn-SN2′ for 19 and
anti-SN2′ for 21) that had been identified for pericosine A and
9 (Figure S6E). Despite the similarities, two minor thiol-
substituted products, 23 and 25, were consistently detected
under all tested conditions even when excess 19 or 20 was
reacted with 1 (Scheme 1 and Figure S12D). We speculate that
these products were formed via dual mechanisms: isomer-
ization of the major product 24 due to excess 1 as well as
through direct substitution of 19 and 21 by 1 through
alternative SN2′ and/or SN2 mechanisms (Figure S12).
D
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
S1), which facilitated the complete consumption of 2 and
pericosine A in H2O (Figure S14B). This compound is a
ubiquitous component of eukaryotic cells and has been
reported to be incorporated into some cosmetics as an
antioxidant agent.14,15 While experiments with 29 were
underway, parallel efforts to uncover additional alternative
reaction conditions that were even more favorable and safer
revealed a noteworthy surprise; propylene glycol (PG) was
determined to be a highly effective cosolvent (20−50% PG)
that facilitated the rapid solubilization of the hydrophobic
thiols and thioacetates but did not impede the catalytic
properties of 29 (Figure S14C). Importantly, PG is widely
used in skin care products as a humectant, making it a useful
addition to the suite of optimized reaction conditions and
reagents.
While ascertaining the mechanistic role that 29 played in
substitution reactions between pericosines and thiols/thio-
esters, an analysis of the reaction products revealed that some
significant changes had occurred. For example, it was observed
that, in reactions between pericosine A and 2 in the presence
of 29, 29 itself was rapidly reacting with pericosine A and
epoxide 9 to form the new substitution product 30 (Figures 4,
S1, and S14E−H). The reaction between 29 (1 equiv) and
[13C]-labeled pericosine A (1 equiv) revealed 30 was formed
via an SN2′ mechanism (Figure 4A), and the absolute
configuration of C-6 was determined to be R on the basis of
3
1
the large JH5,H6 value (9.4 Hz) observed in the H NMR
spectrum of the TFA salt of 30 (TFA-30) (Figure S15).
Interestingly, unlike the reactions of 2 with pericosine A that
were catalyzed by Et3N (Figure S7E), catalysis involving 29
uniformly led to 1:1 racemates of 8 regardless of the reaction
conditions employed (Figure S16E−H). Thus, we hypothe-
sized that racemization of 8 was not only induced by excess 2′
(Figure S7F) but also facilitated by the presence of 29. The
reactions between 30 and 2 (Figures S1 and S16A,B) and
between 8 and 29 (Figures S1 and S16C,D) revealed that 8
and 30/ent-30 were likely to undergo a virtually continuous
cycle of interconversion via a syn-SN2′ pathway resulting in the
racemization of both structures (Figure 4B). Furthermore,
catalysis by 29 facilitated the 1,4-addition of 2′ (Figure S1) to
the alkene group of 8 forming 31 (Figures 4B, S14, and S16),
which had previously only appeared as a minor product in the
reaction of excess 2 with pericosine A in the presence of Et3N
(Figure S7D). However, the formation of 31 was completely
suppressed when excess pericosine A was used in the reaction
(Figure S14D), indicating the 1,4-addition of 2′ is a much
slower process. Finally, catalysis by 29 led to the virtually
instantaneous (immediately after the mixing of reagents) and
complete conversion of thiol 1 (Figure 4C) to racemic 7
(Figure 4D). Thus, reaction conditions employing spermine in
50% aqueous PG exhibited enhanced rapidity for the
consumption of thiols compared to the system using Et3N
and 50% aqueous MeOH (Figures 4C and S6A).
Figure 3. Investigating the mechanism of Et3N-catalyzed substitution
involving 1 and pericosine A (4). (A−D) Time-dependent conversion
of pericosine A. Pericosine A (1 equiv) was mixed with Et3N (2
equiv) in 50% MeOH, and the reactant mixture was monitored by
LCMS at several time points [(A) 1 min; (B) 15 min; (C) 30 min;
(D) 100 min]. (E−H) Chiral-HPLC analysis of 7 derived from
different conditions. The reactants/reagents [initial ratios: (E)
pericosine A (1 equiv)/1 (1 equiv); (F) pericosine A (1 equiv)/1
(1.5 equiv)/Et3N (2 equiv); (G) 9 (1 equiv)/1 (1 equiv); (H) 7a (1
equiv)/1 (1 equiv)] were incubated in 50% MeOH overnight prior to
HPLC analysis. HPLC conditions: Lux cellulose-2 analytical column,
30% MeCN elution, monitored at 200 nm. (I) Synthesis of the fully
acetylated pericosine analogue (11) and its reaction with 1. (J)
Proposed mechanism for the Et3N-catalyzed substitution involving
pericosine A with 1 or MeOH.
Encouraged by the robust reactivity of pericosine A and
analogues 19 and 20, we set about identifying reagents that
would be potentially safer for topical applications on humans
and other animals. First, we sought to replace Et3N with a
more suitable base because its presence was essential for the
successful conversion of thioacetates into reactive thiols. To
begin, 21 inorganic and organic bases (2 equiv) were tested for
their catalytic capabilities in reactions between pericosine A (1
equiv) and 2 (1 equiv) under 100% aqueous conditions. Such
reaction conditions had been noted by us as unfavorable for
supporting Et3N-facilitated catalysis (Figure S14A). From that
initial set, 6 bases including K2CO3 and 5 types of polyamines
(Figure S13) were found to have varying degrees of catalytic
capabilities. One of the most active of those agents was the
naturally occurring polyamine spermine (29) (Figures 4A and
On the basis of our success identifying an efficacious set of
reaction conditions and reagents expected to be safe for dermal
use, we proceeded to test our processes on the complex
cocktail of odoriferous sulfur-containing metabolites directly
harvested from skunk anal glands. Using a prior inventory of
the thiols and thioesters found in skunk glands,3 we tested
pericosine A and analogues 19 and 20 in both the Et3N/
MeOH and aqueous spermine/PG systems. The results of
those reactions were analyzed by LCMS, which revealed a
multitude of thioether substitution products had formed [for
E
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Scheme 1. Synthesis of the Pericosine A Analogues 19 and 20 and Base-Catalyzed Nucleophilic Substitutions (Thiol 1 and/or
a
Thioacetate 2) of 19
a
(a) CSA, MeOH; (b) cyclohexanone, CSA; (c) Tf2O, pyr., DMAP (cat.); (d) CsOAc, DMF; (e) oxone, NaHCO3, 1,1-trifluoroacetone/H2O; (f)
HCl in Et2O; (g) MeOH, acetyl chloride (cat.), (h) incomplete deprotection by MeOH/acetyl chloride (cat.) (thus, reprotected by CSA in
acetone and then complete deprotection by MeOH/acetyl chloride (cat.)); (i) Et3N, 50% MeOH; (j) 1, Et3N, 50% MeOH; (k) 2, Et3N, 50%
MeOH.
example, reaction products 32−36 (Figure S1) were detected
as having been formed from pericosine A, Figures S17A−G].
After overnight reactions, the spermine/PG-based reaction
system exhibited comparable levels of consumption of the
skunk-gland-derived sulfur-containing metabolites as had been
observed for the Et3N/MeOH system (Figures S17B,C). While
formal odor identification tests were not performed, the
effectiveness of the reactions at neutralizing the odor of skunk-
gland secretions was readily apparent to the research team:
whereas even single-digit microliter volumes of the skunk scent
were capable of creating an overwhelmingly foul aroma (even
when briefly removed from the chemical fume hood), samples
of skunk secretions that had been properly treated with
pericosine A, 19, and 20 could be handled on the benchtop for
extended periods without emitting any trace of unpleasant
odor.
0.45 μM).16 However, the same group also reported that
pericosine A only showed weak antiproliferative activities
against 20 human cancer cell lines with an average GI50 value
of 15.1 μM. Moreover, our in vitro antiproliferative/
cytotoxicity tests also showed that pericosine A only weakly
inhibited the growth of the Ect 1 human normal cervical cell
line (IC50 value of 18.3 μM) and the NIH/3T3 normal mouse
fibroblast cell line (IC50 value of 8.4 μM). Thus, we
determined it was prudent to test pericosine A for its potential
to cause skin and eye damage or irritation. To test the dermal
and ocular toxicity of pericosines, we evaluated the test
substances using the EpiDerm skin irritation and EpiOcular
eye irritation models, respectively (Figure 5).18,19 Pericosine A
did not exhibit toxic or irritating effects at concentrations of 0.5
and 1 mM (Figure 5A), nor did it induce the expression of the
cytokine interleukin-1α (IL-1α) at 0.5 mM (Figure 5B).
Furthermore, pericosine A (0.5 and 1 mM) was determined to
be nonirritating in the EpiOcular eye irritation assay, and it did
not reduce the viability of the EpiOcular tissues as compared
to the recommended control (Figure 5C). When the
EpiOcular tissues were exposed to pericosine A (1 mM) for
Tests were performed to assess the safety of pericosine A
alone and under the optimized reaction conditions we
developed. We were initially taken aback by a report suggesting
pericosine A exhibited notable in vitro cytotoxicity against the
murine P388 lymphocytic leukemia cell line (ED50 value of
F
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 4. (A) The conjugation of [13C]-labeled pericosine A (4) (accomplished by feeding [U−13C6]-D-glucose to the producing fungus) with
spermine (29). (B) The proposed mechanism of the reaction between 2 and pericosine A with 29 as the catalyst. (C) LCMS analysis (UV 210 nm)
of the reaction between pericosine A (1 equiv) and 1 (1 equiv) in 50% propylene glycol (PG) with spermine (2 equiv) as the catalyst. The reaction
mixture was injected into LCMS immediately upon mixing of the reagents. Product 7 was analyzed by chiral HPLC (UV 210 nm) (D).
an extended period (24 h), it was found to meet the
classification of an ultramild chemical agent with an ET-50
value of >24 h (Figure 5D). Additionally, the safety of
pericosine A remained unchanged in both the EpiDerm and
EpiOcular tests when 50% PG was used as the test vehicle as
well as when mixtures of 29 (0.5 mM) and pericosine A (0.5
mM) were used as the test sample (Figure 5A−D). Further
testing was performed to assess the potential mutagenic
properties of pericosine A. Pericosine A showed no mutagenic
properties in the Ames II mutagenicity assay,17 even at a
concentration as high has 1 mg/mL (Table S3). Taken in
combination, these data suggested that pericosine-based odor
neutralization strategies might be safe and effective for dermal
uses on humans and other animals. Moreover, pericosine A
appeared to have no immediately detectable problematic
properties as it would relate to its accidental introduction to
the eye.
In summary, our studies provide a scientific rationale for
using pericosines as a new natural-product-based odor
neutralization strategy. Studies with pericosine A and its
analogues (19 and 20) have revealed an unusual suite of
electrophilic properties incorporated into this family of fungal
metabolites, which render them safe yet highly reactive (even
under ultramild aqueous conditions) toward a variety of
noxious sulfur-containing chemicals. Our team is continuing to
probe the potential applications of pericosines as well as pursue
studies aimed at improving routes toward the production of
these shikimate-based metabolites. For example, in addition to
the synthetic efforts reported in the manuscript, we have
engaged in the process of subjecting the source fungus
(Tolypocladium sp. MEA-2) to chemical mutagenesis and
G
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
column (110 Å, 250 × 10.0 mm) with a flow rate of 4 mL/min. Chiral
analytical HPLC analyses were performed on a Waters 1525 system
using Lux 5 μm cellulose-2 and Lux 5 μm cellulose-3 columns (110 Å,
250 × 4.6 mm) with a flow rate of 1 mL/min. All solvents were of
ACS grade or better.
Reactivity Tests of the Commercial “De-Skunking” Prod-
ucts. Six “de-skunking” products were purchased from Amazon.com,
including Skunk-Off Liquid Soaker, Mold Monster Stink Slayer, Clean
+Green Deskunk Coat Cleaner, Nature’s Miracle Skunk Odor
Remover, SynergyLabs De-Skunk Odor Destroyer, and Skout’s
Honor Skunk Odor Eliminator (referred to as Product A). For each
product, 50 μL was mixed with 2-phenylethanethiol (1, 3 μL from
100 mM DMSO stock solution). The reactant mixtures were kept
static overnight (∼16 h) at room temperature. Reaction products
were mixed with MeOH (50 μL) and analyzed by LCMS.
Conversion of 2-Phenylethanethiol (1) by Commercial
Product A. An aliquot of 1 mL of Product A was added to 20 μL
of 2-phenylethanethiol (1). The mixture was stirred for 3 h at room
temperature. The solvent was evaporated in vacuo, and the residue
was subjected to semipreparative HPLC purification using a Kinetex 5
μm F5 column (isocratic, 40% MeCN in 10% TFA) to afford
compound 3 (38 mg).
1
Compound 3: white solid; H NMR (400 MHz, methanol-d4): δ
7.62 (d, J = 8.3 Hz, 4H), 7.26 (d, J = 8.3 Hz, 4H), 7.23 (m, 3H), 7.09
(d, J = 8.3 Hz, 2H), 3.52 (t, J = 7.4 Hz, 2H), 2.88 (t, J = 7.4 Hz, 2H),
2.40 (s, 6H); 13C NMR (100 MHz, methanol-d4): δ 144.8, 140.5,
138.5, 130.6 (4C), 129.9 (2C), 129.6 (2C), 128.1, 127.8 (4C), 57.0,
30.5, 21.5 (2C); HRESIMS m/z 499.0800, [M + Na]+ (calcd for
C22H24N2O4S3Na, 499.0790).
General Methods for Reactivity Tests of Pericosine A (4)
and Analogues. Reactions were performed in Eppendorf tubes with
50 μL of the test solvent [e.g., H2O, 50% MeOH, and 50% propylene
glycol (PG)]. Reagents were then added to the test solvent:
pericosine A or analogues (1 equiv, 3 μL from 100 mM DMSO
stock solution), sulfur-containing compounds [e.g., 2-phenylethane-
thiol (1), 4-bromo-α-toluene thioacetate (2), dibenzyl sulfide (5), and
dibenzyl disulfide (6)] (1 equiv, 3 μL from 100 mM DMSO stock
solution), and/or bases (e.g., Et3N, spermine) (2 equiv, 6 μL from
100 mM DMSO stock solution). The reactant mixtures were then
kept static overnight (∼16 h, default incubation time unless otherwise
specified) at room temperature. Test mixtures were combined with
MeOH (50 μL) and analyzed by LCMS to measure reagent
consumption as well as product formation.
Figure 5. Safety tests of pericosine A (4) (A, B) using the in vitro
EpiDerm skin irritation test. The EpiDerm tissues were exposed to the
test chemicals for 60 min and then incubated in fresh medium for 24
h. The medium was collected for analysis of the cytokine interleukin-
1α (IL-1α) (B), and the viability of the tissues was evaluated using the
MTT assay. (C) For the EpiOcular eye irritation test, tissues were
incubated with the test chemicals for 30 min, and cell viability was
evaluated using the MTT assay. Viability: ≥60% is considered
nonirritating while viability levels <60% are interpreted as irritating.
(D) EpiOcular sub-draize ultra-mildness test. The ET-50 is defined as
the time of exposure needed for a test substance to reduce the tissue
viability by 50% relative to control tissues. For all tests: n = 3; ***, p <
0.001; 50% PG; 50% propylene glycol aqueous solution.
culture medium optimization, which has already led to a 15-
fold increase in the yield of pericosine A (156 vs 10 mg/L, data
not shown). The encouraging results obtained for pericosine A
and its analogues have revealed a promising suite of
commercial applications utilizing this robust odor neutraliza-
tion technology. Further studies examining the activities of
additional synthetic pericosines are anticipated to provide
inspiration for the creation of a new generation of nontoxic
materials engineered to counteract a range of noxious
nucleophilic substances.
Reaction of Pericosine A (4) with 2-Phenylethanethiol (1)
Catalyzed by Et3N. Pericosine A (4, 10 mg, 1 equiv) was mixed with
2-phenylethanethiol (1, 6 μL, 1.5 equiv) and Et3N (12.5 μL, 2 equiv)
in 50% MeOH (2 mL). The reactant mixture was stirred at room
temperature for 3 h, and the solvent was removed in vacuo. The
residue was purified by semipreparative HPLC using a Gemini 5 μm
C18 column (isocratic, 50% MeCN) to yield compound 7 (10 mg,
scalemic mixture).
EXPERIMENTAL SECTION
Scalemic-(3R*,4R*,5R*,6R*)-methyl 3,4,5-Trihydroxy-6-
■
(phenethylthio)cyclohex-1-enecarboxylate (Scalemic-7): white
General Experimental Procedures. Optical rotation data were
obtained on a Rudolph Research Autopol III automatic polarimeter.
NMR data were obtained on Varian VNMR spectrometers (400 and
1
solid; [α]20 −93 (c 0.15, MeOH); for H and 13C NMR data, see
D
Tables 1 and 2; HRESIMS m/z 347.0932, [M + Na]+ (calcd for
C16H20O5SNa, 347.0924).
1
500 MHz for H, 100 and 125 MHz for 13C) with broad band and
triple resonance probes at 25 0.5 °C. HRESIMS (high-resolution
electrospray ionization mass spectrometry) data were collected on an
Agilent 6538 high-mass-resolution QTOF mass spectrometer.
LCESIMS (liquid chromatography-electrospray ionization mass
spectrometry) data were obtained on a Shimadzu LC−MS 2020
system (ESI quadrupole) coupled to a photodiode array detector with
a Phenomenex Kintex column (2.6 μm C18 column, 100 Å, 75 × 3.0
mm). Preparative HPLC separations were performed on a Shimadzu
system using a SCL-10A VP controller and a Gemini 5 μm C18
column (110 Å, 250 × 21.2 mm) with a flow rate of 10 mL/min.
Semipreparative HPLC separations were performed on a Waters 1525
system using a 2998 PDA detector and HPLC columns including a
Gemini 5 μm C18 column (110 Å, 250 × 10.0 mm), a Kinetex 5 μm
biphenyl column (110 Å, 250 × 10.0 mm), and a Kinetex 5 μm F5
Pericosine A (4, 10 mg, 1 equiv) was mixed with 2-phenyl-
ethanethiol (1, 4 μL, 1 equiv) without Et3N. After stirring for 16 h,
the standard purification procedures were applied to generate the pure
3R,4R,5R,6R isomer of 7 (5.2 mg).
(3R,4R,5R,6R)-Methyl 3,4,5-Trihydroxy-6-(phenethylthio)-
cyclohex-1-enecarboxylate (7a): [α]20 −145 (c 0.1, MeOH).
D
Reaction of Pericosine A (4) with 4-Bromo-α-toluene
Thioacetate (2) Catalyzed by Et3N. Pericosine A (4, 10 mg, 1
equiv) was mixed with 4-bromo-α-toluene thioacetate (2, 11.3 μL, 1.5
equiv) and Et3N (62.5 μL, 10 equiv) in 50% MeOH (2 mL). The
reactant mixture was stirred at room temperature overnight and
before the solvent was removed in vacuo. The residue was purified by
semipreparative HPLC using a Gemini 5 μm C18 column (isocratic,
50% MeCN) to yield compound 8 (3.7 mg, racemic mixture).
H
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
a
1
Table 3. H and 13C NMR Data of 19−22
19
20
21
22
no.
13C
1H (J)
13C
1H (J)
13C
1H (J)
13C
1H (J)
1
2
3
4
5
6
7
8
9
133.0
139.5
66.7
71.7
74.5
58.5
167.2
52.6
129.8
144.7
69.9
71.1
75.5
55.8
166.8
52.7
128.6
146.2
67.5
66.4
53.7
47.3
166.8
52.5
132.1
141.0
70.1
75.6
69.8
77.9
168.6
52.4
58.7
6.78, dd (4.3, 1.2)
4.35, dd (4.3, 4.1)
3.62, dd (8.2, 4.1)
4.13, dd (8.2, 5.4)
4.58, ddd (5.4, 1.1, 1.0)
6.87, d (2.5)
4.84, ddd (2.0, 1.9, 1.8)
4.19, dd (5.4, 1.9)
4.37, m
3.56, dd (4.0, 3.0)
3.91, dd (4.0, 2.0)
6.59, dd (2.6, 1.6)
4.40, ddd (7.1, 2.6, 2.6)
3.52, dd (7.1, 2.1)
4.23, dd (4.0, 2.1)
4.20, m
4.38, dd (8.5, 2.5)
3.97, dd (8.5, 2.6)
4.14, dd (3.0, 2.6)
4.76, dd (3.0, 0.9)
3.78, s
3.78, s
3.81, s
3.50, s
3.76, s
a
In methanol-d4.
Rac-(3R*,4R*,5R*,6R*)-methyl 6-((4-Bromobenzyl)thio)-3,4,5-tri-
(1R,4R,5R,6S)-Methyl 4,5-dihydroxy-7-oxabicyclo[4.1.0]hept-2-
hydroxycyclohex-1-enecarboxylate (Racemic-8): white solid; [α]20
ene-2-carboxylate (21): colorless solid; [α]20 −54 (c 0.3,
D
D
1
MeOH); for H and 13C NMR data, see Table 3; HRESIMS m/z
1
0 (c 0.35, MeOH); for H and 13C NMR data, see Tables 1 and 2;
185.0459, [M − H]¯ (calcd for C8H9O5, 185.0455).
HRESIMS m/z 410.9881, [M + Na]+ (calcd for C15H17O5SBrNa,
410.9782).
(3R,4S,5S,6R)-Methyl 3,4,5-Trihydroxy-6-methoxycyclohex-1-
enecarboxylate (22): colorless solid; [α]20 −58 (c 0.25, MeOH);
Transformation of Pericosine A (4) in 50% MeOH Catalyzed
by Et3N. Pericosine A (4, 15 mg, 1 equiv) was mixed with Et3N (18.8
μL, 2 equiv) in 50% MeOH (12 mL). The reactant mixture was
stirred at room temperature for 1 h before the solvent was removed in
vacuo. The residue was purified by semipreparative HPLC using a
Gemini 5 μm C18 column (isocratic, 10% MeCN) to yield
compounds 9 (3.5 mg)7 and 10 (9.4 mg).8
D
for 1H and 13C NMR data, see Table 3; HRESIMS m/z 241.0694, [M
+ H]+ (calcd for C9H14O6Na, 241.0683).
Reaction of Compound 19 with 2-Phenylethanethiol (1)
Catalyzed by Et3N. Compound 19 (20 mg, 1 equiv) was mixed with
2-phenylethanethiol (1, 8 μL, 1 equiv) and Et3N (125 μL, 10 equiv)
in 50% MeOH (8 mL). The reactant mixture was stirred overnight at
room temperature, and the solvent was removed in vacuo. The
products were purified from the residue by semipreparative HPLC
using a Gemini 5 μm C18 column (isocratic, 30% MeCN) to yield
compounds 23 (3.4 mg), 24 (9.5 mg), and 25 (9.0 mg).
Acetylation of Pericosine A (4). Pericosine A (4, 10 mg) was
mixed with Ac2O (1 mL) in DCM (1 mL) followed by the addition of
4 Å molecular sieves (100 mg). The reactant mixture was stirred at 40
°C overnight, and the solvent was removed in vacuo. The residue was
purified by semipreparative HPLC using a Gemini 5 μm C18 column
(isocratic, 40% MeCN) to yield compound 11 (8.1 mg).
(1S,2S,3S,6S)-6-Chloro-5-(methoxycarbonyl)cyclohex-4-ene-
(3R,4S,5S,6S)-Methyl 3,4,5-Trihydroxy-6-(phenethylthio)-
cyclohex-1-enecarboxylate (23): white solid; [α]20 17 (c 0.23,
D
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
m/z 347.0928, [M + Na]+ (calcd for C16H20O5SNa, 347.0924).
1,2,3-triyl Triacetate (11): white solid; [α]20 115 (c 0.29, MeOH);
D
1H NMR (500 MHz, methanol-d4): δ 6.80 (d, J = 3.4 Hz, 1H), 5.84
(3R,4S,5S,6R)-Methyl 3,4,5-Trihydroxy-6-(phenethylthio)-
cyclohex-1-enecarboxylate (24): white solid; [α]20 −137 (c 0.63,
D
(m, 1H), 5.60 (dd, J = 2.3, 4.5 Hz, 1H), 5.43 (dd, J = 2.3, 5.9 Hz,
1H), 4.98 (d, J = 5.9 Hz, 1H), 3.82 (s, 3H), 2.07 (s, 3H), 2.05 (s,
3H), 2.04 (s, 3H); 13C NMR (125 MHz, methanol-d4): δ 171.5,
171.3, 171.2, 165.9, 137.4, 133.3, 73.7, 68.5, 66.7, 54.0, 53.0, 20.5
(2C), 20.4; HRESIMS m/z 371.0504, [M + Na]+ (calcd for
C14H17O8ClNa, 371.0504).
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
m/z 347.0930, [M + Na]+ (calcd for C16H20O5SNa, 347.0924).
(3R,4R,5S,6S)-Methyl 3,4,5-Trihydroxy-6-(phenethylthio)-
cyclohex-1-enecarboxylate (25): white solid; [α]20 24 (c 0.60,
D
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
m/z 347.0928, [M + Na]+ (calcd for C16H20O5SNa, 347.0924).
Reaction of 19 with 4-Bromo-α-toluene Thioacetate (2)
Catalyzed by Et3N. Compound 19 (20 mg, 1 equiv) was mixed with
4-bromo-α-toluene thioacetate (2, 15 μL, 1 equiv) and Et3N (125 μL,
10 equiv) in 50% MeOH (8 mL). The reactant mixture was stirred at
room temperature overnight, and the solvent was removed in vacuo.
The compounds were purified from the residue by semipreparative
HPLC using a Gemini 5 μm C18 column (isocratic, 35% MeCN) to
yield compounds 26 (2.8 mg), 27 (6.8 mg), and 28 (7.0 mg).
(3R,4S,5S,6S)-Methyl 6-((4-Bromobenzyl)thio)-3,4,5-trihydroxy-
Reaction of Compound 11 with 2-Phenylethanethiol (1)
Catalyzed by Et3N. Compound 11 (16 mg, 1 equiv) was mixed with
2-phenylethanethiol (1, 6 μL, 1.5 equiv) and Et3N (12.8 μL, 2 equiv)
in 50% MeOH (2 mL). The reactant mixture was stirred at room
temperature overnight before the solvent was removed in vacuo. The
residue was purified by semipreparative HPLC using a Gemini 5 μm
C18 column (isocratic, 70% MeCN) to yield compound 12 (13.5 mg).
(1R,2R,3R,6R)-5-(Methoxycarbonyl)-6-(phenethylthio)cyclohex-
4-ene-1,2,3-triyl Triacetate (12): white solid; [α]20 −98 (c 0.68,
D
1
cyclohex-1-enecarboxylate (26): white solid; [α]20 38 (c 0.19,
MeOH); H NMR (500 MHz, methanol-d4): δ 7.21−7.28 (m, 4H),
D
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
7.17 (m, 1H), 6.66 (H-2, dd, J = 1.0, 4.2 Hz, 1H), 5.71 (H-3, t, J = 5.0
Hz, 1H), 5.57 (H-4, dd, J = 2.4, 5.0 Hz, 1H), 5.34 (H-5, dd, J = 2.4,
4.2 Hz, 1H), 3.96 (H-6, d, J = 4.2 Hz, 1H), 3.79 (CH3-8, s, 3H),
2.86−3.06 (m, 4H), 2.06 (s, 3H), 2.05 (s, 3H), 2.02 (s, 3H); 13C
NMR (125 MHz, methanol-d4): δ 172.0, 171.7, 171.5, 167.1, 141.6
(C-2), 134.2, 134.1, 129.7 (2C), 129.5 (2C), 127.4, 72.5 (C-5), 67.9
(C-4), 65.3 (C-3), 52.9 (CH3-8), 44.3 (C-6), 37.2, 35.5, 20.8, 20.6
(2C); HRESIMS m/z 473.1255, [M + Na]+ (calcd for C22H26O8SNa,
473.1241).
m/z 410.9878, [M + Na]+ (calcd for C15H17BrO5SNa, 410.9872).
(3R,4S,5S,6R)-Methyl 6-((4-bromobenzyl)thio)-3,4,5-trihydroxy-
cyclohex-1-enecarboxylate (27): white solid; [α]20 −181 (c 0.45,
D
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
m/z 410.9885, [M + Na]+ (calcd for C15H17BrO5SNa, 410.9872).
(3R,4R,5S,6S)-Methyl 6-((4-bromobenzyl)thio)-3,4,5-trihydroxy-
cyclohex-1-enecarboxylate (28): white solid; [α]20 80 (c 0.47,
D
1
MeOH); for H and 13C NMR data, see Tables 1 and 2; HRESIMS
m/z 410.9885, [M + Na]+ (calcd for C15H17BrO5SNa, 410.9872).
Reaction of Pericosine A (4) with Spermine (29). Pericosine A
(4, 4.4 mg, 1 equiv) was mixed with spermine (29, 4 μL, 1 equiv) in
50% MeOH (1 mL). The reactant mixture was stirred for 1 h at room
temperature before the solvent was removed in vacuo to yield
compound 30 (8.2 mg). Compound 30 (8.2 mg) was dissolved in
50% MeOH (1 mL) followed by the addition of TFA (7.6 μL), and
Transformation of Compound 19 in 50% MeOH Catalyzed
by Et3N. Compound 19 (15 mg, 1 equiv) was mixed with Et3N (18.8
μL, 2 equiv) in 50% MeOH (12 mL). The reactant mixture was
stirred at room temperature for 1 h, and the solvent was removed in
vacuo. The reaction products were purified from the residue by
semipreparative HPLC using a Gemini 5 μm C18 column (isocratic,
10% MeCN) to yield compounds 21 (4.5 mg) and 22 (3.8 mg).
I
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
1
Table 4. H and 13C NMR Data of 30 and 31
a
a
a
b
c
30
TFA-30
[13C]-30
31
31
no.
13C
1H (J)
13C
1H (J)
JC,C (Hz)
13C
1H (J)
1H (J)
1
2
3
4
5
6
7
8
132.0
142.7
68.6
72.3
72.2
58.6
168.6
52.6
124.1
149.8
69.0
73.9
70.6
60.0
167.3
53.3
44, 73
43
43, 38
37, 38
37
50.7
46.1
69.1
71.5
69.1
46.1
171.5
51.9
36.1
2.90, br s
3.15, t (4.6)
2.94, dd (4.6, 11.1)
3.80, m
3.80, m
3.80, m
6.77, br s
4.31, br s
4.04, br s
3.87, br d (6.8)
3.70, br d (6.8)
7.04, ddd (1.8, 1.8, 1.9)
4.45, m
4.10, m
4.05, dd (2.0, 9.4)
4.18, ddd (2.6, 2.6, 9.4)
2.89, br d (11.0)
4.02, br dd (2.9, 11.0)
4.28, br t (2.9)
4.02, br dd (2.9, 11.0)
2.89, br d (11.0)
44
73
2.94, dd (4.6, 11.1)
3.77, s
3.85, s
3.72, s
3.69, d (13.4)
3.74, d (13.4)
3.60, s
3.76, d (13.0)
3.82, d (13.0)
1′/1″
2′/2″
45.5
2.74, m
45.1
3.41, m
3.25, m
2.20, m
3.15, m
136.7
3′/3″
4′/4″
5′/5″
6′/6″
7′/7″
8′
9′
11′
12′
13′
28.3
49.2
1.74, m
2.90, m
24.4
45.9
130.8
131.9
121.5
131.9
130.8
7.16, d (8.0)
7.46, d (8.0)
7.24, d (8.4)
7.50, d (8.4)
49.2
26.8
27.6
49.8
47.9
30.7
40.3
2.81, m
1.68, m
1.61, m
2.68, m
2.74, m
1.76, m
48.6
24.2
24.2
48.6
45.8
25.4
37.8
3.08, m
1.82, m
1.82, m
3.08, m
3.12, m
2.09, m
3.06, m
7.46, d (8.0)
7.16, d (8.0)
7.50, d (8.4)
7.24, d (8.4)
2.82, m
a
b
c
Measured in methanol-d4. Measured in CDCl3. Measured in DMSO-d6.
the mixture was stirred for 5 min. The solvents and unreacted TFA
were removed in vacuo to yield the TFA salt of 30, TFA-30 (15.2
mg).
(3R,4R,5S,6R)-Methyl 6-((3-((4-((3-Aminopropyl)amino)butyl)-
amino)propyl)amino)-3,4,5-trihydroxycyclohex-1-enecarboxylate
values of these conformers were summed after a Boltzmann statistical
weighting. Single ECD spectra of the calculated conformers were
determined by SpecDis 1.60 using a sigma value of 0.2−0.3 eV. After
applying a proper scaling factor and a UV-shift correction, the
Boltzmann-summed computed ECD spectra were compared with the
experimentally obtained ECD data.
Biological Tests. The in vitro antiproliferative/cytotoxic proper-
ties of pericosine A were tested in the MTT (3-(4,5-dimethylthiazol-
2-yl)-2,5-diphenyltetrazolium bromide) assay against the Ect 1 human
normal cervical cell line and the NIH/3T3 normal mouse fibroblast
cell line.20 The potential mutagenic activity of pericosine A was
evaluated in the Ames II mutagenicity assay17 performed by
BioReliance Corporation. The dermal and ocular toxicities of
pericosine A were evaluated using the EpiDerm skin irritation and
EpiOcular eye irritation models18,19 following the standard protocols
provided by the manufacturer (MatTek Corporation).
1
(30): colorless solid; [α]20 −14 (c 1.0, MeOH); for H and 13C
D
NMR data, see Table 4; HRESIMS m/z 389.2762, [M + H]+ (calcd
for C18H37N4O5, 389.2758).
TFA-30. [α]20D −24 (c 1.0, MeOH); for 1H and 13C NMR data, see
Table 4.
Production of [13C]-Labeled 30. To produce the [13C]-labeled
30, [13C]-labeled pericosine A (3.0 mg, 1 equiv, accomplished by
feeding [U-13C6]-D-glucose to the source fungus)7 was reacted with
spermine (30, 3.0 μL, 1 equiv) under the same conditions as
previously described. The JC,C coupling constants of the [13C]-labeled
30 (6.6 mg) are listed in Table 4.
Reaction of Pericosine A (4) with 4-Bromo-α-toluene
Thioacetate (2) Catalyzed by Spermine (29). Pericosine A (4,
20 mg, 1 equiv) was mixed with 4-bromo-α-toluene thioacetate (2, 15
μL, 1 equiv) and spermine (20 μL, 2 equiv) in 50% propylene glycol
(PG) (2 mL). The reactant mixture was stirred at room temperature
overnight, and the solvent was removed in vacuo. Compounds were
obtained from the organic residue by semipreparative HPLC using a
Gemini 5 μm C18 column (isocratic, 65% MeCN) to yield
compounds 8 (8.7 mg, racemic mixture) and 31 (12.5 mg).
(1R,2R,3R,4R,5S,6S)-Methyl 2,6-bis((4-Bromobenzyl)thio)-3,4,5-
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.9b00415.
■
S
Synthesis of the pericosine A analogs 19 and 20; crystal
data and structure refinement for compound 3;
calculated carbon chemical shifts; results of the AMES
II assay; structures of numbered compounds 1−36,
including examples of atom-numbering schemes for the
different chemical classes (as a convenience to the
reader); NMR spectra; LCMS analysis; absolute
configuration and mechanism of formation; principal
trihydroxycyclohexanecarboxylate (31): white solid; [α]20 −4 (c
D
0.4, MeOH); for 1H and 13C NMR data, see Table 4; HRESIMS m/z
612.9323, [M + Na]+ (calcd for C22H24O5S2Br2Na, 612.9324).
Computational Details. Conformational analyses were carried
out using ComputeVOA v1.1. Geometry, frequency, 13C NMR, ECD,
and specific rotation calculations were applied at the DFT (density
functional theory) and TD-DFT (time-dependent density functional
theory) levels [B3LYP functional/6-31G(d,p) or 6-311+G(2d,p)
basis set] with Gaussian ’09 carried out in the gas phase. For each
substance, subsets of the lowest energy conformers in the gas phase
were obtained by selecting only those conformers with energies
predicted to be within 2.0 kcal/mol of their respective lowest-energy
conformers. The 13C NMR data, ECD spectra, and specific rotation
component analysis; 13C chemical shifts; JH,H coupling
3
constants; catalytic effects; catalyzed reactions; gener-
ation of the TFA salt of 30; optimized geometries,
relative energies, and Boltzmann populations; optimized
1
1
conformers; H and 13C NMR data; H−1H COSY,
HSQC, and HMBC data (PDF)
J
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
AUTHOR INFORMATION
Corresponding Author
*Tel: (405) 325-6969. E-mail: rhcichewicz@ou.edu (R.H.C.).
■
ORCID
Lin Du: 0000-0001-8751-4482
Doug E. Frantz: 0000-0002-4509-4579
Robert H. Cichewicz: 0000-0003-0744-4117
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Financial support from the University of Oklahoma, Office of
Technology Development, Growth Fund, is appreciated. We
are grateful for helpful recommendations provided by the
University of Oklahoma, Center for the Creation of Economic
Wealth, Technology Commercialization Program.
REFERENCES
■
161.
(1) Anand, S.; Radhakrishna, S. J. Asia Pac. Biodivers. 2017, 10, 154−
(2) Nyhus, P. J. Annu. Rev. Environ. Resour. 2016, 41, 143−171.
(3) Wood, W. F. Chem. Educ. 1999, 4, 44−50.
(4) Aldrich, T. B. J. Exp. Med. 1896, 1, 323−340.
(5) Zaks, K. L.; Tan, E. O.; Thrall, M. A. J. Am. Vet. Med. Assoc.
2005, 226, 1516−1518.
(6) Shim, I.; Seo, G. B.; Oh, E.; Lee, M.; Kwon, J. T.; Sul, D.; Lee, B.
W.; Yoon, B. I.; Kim, P.; Choi, K.; Kim, H. M. J. Toxicol. Sci. 2013, 38,
937−946.
(7) Du, L.; You, J.; Nicholas, K. M.; Cichewicz, R. H. Angew. Chem.,
Int. Ed. 2016, 55, 4220−4225.
(8) Usami, Y.; Ohsugi, M.; Mizuki, K.; Ichikawa, H.; Arimoto, M.
Org. Lett. 2009, 11, 2699−2701.
(9) Zhang, E. S.; Xu, T. L.; Wang, D. J.; Huang, T. K.; Yuan, M.; Li,
J.; Zou, Y. RSC Adv. 2014, 4, 10022−10027.
(10) Valsecchi, E.; Tacchi, A.; Prosperi, D.; Compostella, F.; Panza,
L. Synlett 2004, 2004, 2529−2532.
(11) Mizuki, K.; Iwahashi, K.; Murata, N.; Ikeda, M.; Nakai, Y.;
Yoneyama, H.; Harusawa, S.; Usami, Y. Org. Lett. 2014, 16, 3760−
3763.
(12) Usami, Y.; Mizuki, K.; Ichikawa, H.; Arimoto, M. Tetrahedron:
Asymmetry 2008, 19, 1461−1464.
(13) Usami, Y.; Mizuki, K. J. Nat. Prod. 2011, 74, 877−881.
(14) Lovaas, E. Adv. Pharmacol. 1996, 38, 119−149.
(15) Lupo, M. P. Clin. Dermatol. 2001, 19, 467−473.
(16) Yamada, T.; Iritani, M.; Ohishi, H.; Tanaka, K.; Minoura, K.;
Doi, M.; Numata, A. Org. Biomol. Chem. 2007, 5, 3979−3986.
(17) Gee, P.; Maron, D. M.; Ames, B. N. Proc. Natl. Acad. Sci. U. S.
A. 1994, 91, 11606−11610.
(18) Kandarova, H.; Hayden, P.; Klausner, M.; Kubilus, J.;
Sheasgreen, J. J. Vis. Exp. 2009, 29, 1366.
(19) Kaluzhny, Y.; Kandarova, H.; Handa, Y.; DeLuca, J.; Truong,
T.; Hunter, A.; Kearney, P.; d’Argembeau-Thornton, L.; Klausner, M.
ATLA, Altern. Lab. Anim. 2015, 43, 101−127.
(20) Mosmann, T. J. Immunol. Methods 1983, 65, 55−63.
K
DOI: 10.1021/acs.jnatprod.9b00415
J. Nat. Prod. XXXX, XXX, XXX−XXX