Ribocyclophanes A-E, Glycosylated Cyclophanes with Antiproliferative Activity from Two Cultured Terrestrial Cyanobacteria

Article abstract of DOI:10.1021/acs.jnatprod.7b00954

The cell extracts of two cultured freshwater Nostoc spp., UIC 10279 and UIC 10366, both from the suburbs of Chicago, showed antiproliferative activity against MDA-MB-231 and MDA-MB-435 cancer cell lines. Bioassay-guided fractionation led to the isolation of five glycosylated cylindrocyclophanes, named ribocyclophanes A-E (1-5) and cylindrocyclophane D (6). The structure determination was carried out by HRESIMS and 1D and 2D NMR analyses and confirmed by single-crystal X-ray crystallography. The structures of ribocyclophanes A-E (1-5) contain a β-d-ribopyranose glycone in the rare ¹C₄ conformation. Among isolated compounds, ribocyclophane D (4) showed antiproliferative activity against MDA-MB-435 and MDA-MB-231 cancer cells with an IC₅₀ value of less than 1 μM.

Full text of DOI:10.1021/acs.jnatprod.7b00954

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Ribocyclophanes A−E, Glycosylated Cyclophanes with
Antiproliferative Activity from Two Cultured Terrestrial
Cyanobacteria
Daniel S. May,†,∥ Hahk-Soo Kang,†,§,∥ Bernard D. Santarsiero,‡ Aleksej Krunic,† Qi Shen,†
Joanna E. Burdette,† Steven M. Swanson,†,⊥ and Jimmy Orjala*,†
†Department of Medicinal Chemistry and Pharmacognosy, University of Illinois at Chicago, Chicago, Illinois 60612, United States
‡Center for Biomolecular Sciences, University of Illinois at Chicago, Chicago, Illinois 60607, United States
⊥School of Pharmacy, University of Wisconsin at Madison, Madison, Wisconsin 53705, United States
S
* Supporting Information
ABSTRACT: The cell extracts of two cultured freshwater Nostoc
spp., UIC 10279 and UIC 10366, both from the suburbs of Chicago,
showed antiproliferative activity against MDA-MB-231 and MDA-
MB-435 cancer cell lines. Bioassay-guided fractionation led to the
isolation of five glycosylated cylindrocyclophanes, named ribocyclo-
phanes A−E (1−5) and cylindrocyclophane D (6). The structure
determination was carried out by HRESIMS and 1D and 2D NMR
analyses and confirmed by single-crystal X-ray crystallography. The
structures of ribocyclophanes A−E (1−5) contain a β-D-ribopyranose
1
glycone in the rare C4 conformation. Among isolated compounds,
ribocyclophane D (4) showed antiproliferative activity against MDA-
MB-435 and MDA-MB-231 cancer cells with an IC50 value of less
than 1 μM.
olyketides (PKs), one of the largest families of natural
Pproducts, are not commonly found in cyanobacteria
compared to other major natural products families, such as
nonribosomal peptides (NRPs) and PK−NRP hybrids.1−3
However, [7.7]paracyclophanes represent a family of cyano-
bacterial polyketides that has been found in heterocyst-forming
cyanobacteria of the order Nostocales.4−11 The 22-membered
[7.7]paracyclophane core structure, possessing a C2 axis of
symmetry, has been shown to be produced by dimerization of
two halogenated alkyl-resorcinol units made by a type I/type III
polyketide synthase hybrid.12−14 The alkyl-resorcinol units,
termed cylindrofridins, were isolated by Preisitsch et al. and
proposed as biosynthetic intermediates.15 The identification of
the biosynthetic intermediates by Nakamura et al. confirmed
the biosynthetic origin of these alkyl-resorcinol units as well as
their dimerization.13,14,16 The core [7.7]paracyclophane
structure is further modified by tailoring steps such as
methylation, acetylation, carbamate formation, glycosylation,
and/or halogenation, to generate diverse structural variants.17
Cyanobacterial cyclophanes have been shown to possess a
broad range of biological activities, including antibacterial,
antifungal, and cytotoxic activities.6−11
activity and cytotoxicity against several cancer cell lines.
Carbamidocyclophanes F and G were reported from UIC
10274, containing terminal chlorines as well as an uncommon
carbamate moiety.10 In addition, the isolation of merocyclo-
phanes A and B from a Nostoc sp. (UIC 10062) from Grand
Mere State Park in Michigan was reported, as well as
merocyclophanes C and D from the extract of UIC 10110, a
Nostoc sp. from Colorado. Merocyclophanes A−D featured a
new [7.7]paracyclophane carbon skeleton characterized by the
presence of branched methyls at α positions (C-1/14).8,11 The
extract of a cultured Nostoc sp. (UIC 10279), obtained from a
sample collected in the southwest suburbs of Chicago,
displayed antiproliferative activity against the human melanoma
cancer cell line MDA-MB-435. HPLC-based activity profiling
followed by LCMS and 1H NMR-based dereplication suggested
the presence of three glycosylated cyclophanes as active
metabolites. Concurrently, a second Nostoc sp. (UIC 10366),
obtained from a collection in the northwestern suburbs of
Chicago, also displayed antiproliferative activity against MDA-
1
MB-435 cancer cells. LCMS and H NMR-based dereplication
also suggested the presence of glycosylated cyclophanes similar
to UIC 10279, including one unique to UIC 10366. During
Several Nostoc spp. from the University of Illinois at Chicago
(UIC) cyanobacterial library have been identified to produce
[7.7]paracyclophanes. Chlorinated cylindrocyclophanes were
obtained from a Nostoc sp. (UIC 10022A).7 These chlorinated
cyclindrocyclophanes exhibited 20S proteasome-inhibitory
Special Issue: Special Issue in Honor of Susan Horwitz
Received: November 10, 2017
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
isolation, a fifth minor compound was identified from UIC
10366. Herein, we report the isolation, structure determination,
and antiproliferative activities of five glycosylated cyanobacterial
[7.7]paracyclophanes, named ribocyclophanes A−E (1−5).
molecular weights of 848, 716, and 700 Da. Dereplication of
the active peaks of UIC 10366 by LC-MS and H NMR
1
spectroscopy indicated the presence of the previously reported
cylindrocyclophane D as well as two potentially new
glycosylated cyclophanes with molecular weights of 848 and
758 Da.4 Active fractions for each strain were subjected to
reversed-phase HPLC to yield six pure compounds between the
two extracts. During isolation from the extract of UIC 10366, a
low abundance compound, ribocyclophane E (5), was also
obtained. Ribocyclophanes A (1, 0.5%), B (2, 0.1%), and C (3,
0.3%) were isolated from UIC 10279, and ribocyclophanes A
(1, 0.5%), D (4, 0.5%), and E (5, < 0.1%) as well as the known
compound cylindrocyclophane D (6, 0.2%) were isolated from
UIC 10366.
Ribocyclophane A (1), a white amorphous powder obtained
from UIC 10279, displayed a deprotonated molecule at m/z
847.4895 ([M − H]−) in the HRESIMS analysis, correspond-
ing to a molecular formula of C46H71O14. The 1H NMR
spectrum of 1 showed signals characteristic of cylindrocyclo-
phanes with two singlet aromatic protons (H-10/23, δH 6.19;
H-12/25, δH 6.09), benzylic doublet and multiplet methines
(H-1/14, δH 3.65; H-7/20, δH 3.16), shielded diastereotopic
methylenes (H2-3/16 and H2-5/18, δH 0.5−1.0), and a doublet
methyl (H3-35/36, δH1.05). Only half the number of signals
were observed in the 1H and 13C NMR spectra, as indicated by
the molecular formula, suggesting the presence of an element of
symmetry in 1. The 1H NMR spectrum of 1 showed an
additional set of proton signals adjacent to oxygen (δH 3.0−4.0)
and an anomeric proton (δH 4.85) compared to those observed
for the cylindrocyclophanes, suggesting the presence of a
glycoside moiety in 1. The structure and position of a sugar
moiety were determined by 2D NMR analysis. Sequential
COSY correlations from the anomeric proton H-1′/6′ (δH
4.85) to the diastereotopic methylene protons H2-5′/10′ (δH
3.19 and 3.30) via three hydroxylated methine protons, H-2′/7′
(δH 3.73), H-3′/8′ (δH 3.70), and H-4′/9′ (δH 3.59), together
with an HMBC correlation from H2-5′/10′ to C-1′/6′,
indicated the sugar moiety to be a pentopyranose (Figure 1).
The HMBC correlation observed from the anomeric protons
H-1′/6′ to C-1/14 established the linkage between the
pentopyranose and the cylindrocyclophane aglycone.
RESULTS AND DISCUSSION
■
Strain UIC 10279 was obtained from a sample collected
southwest of Chicago near the suburb of Joliet, Illinois, while
strain UIC 10366 was obtained from a sample collected near
Elgin, Illinois, a suburb northwest of Chicago. The strains were
both designated as Nostoc sp. based on morphological and 16S
rRNA gene sequence analysis (Supporting Information, S1 and
S2). Each strain was cultured in Z media,18 and the freeze-dried
cells were extracted with a mixture of CH2Cl2 and MeOH (1:1,
v/v). The resulting extracts displayed antiproliferative activity
against the MDA-MB-435 cancer cell line. Both extracts were
fractionated using Diaion HP-20 resin with an increasing
amount of iPrOH in H2O. Fractions eluting at 50% and 60%
iPrOH showed antiproliferative activity for both strains. To
facilitate the identification of potentially active compounds, the
active fractions were subjected to our HPLC-based activity
profiling that combines HPLC-based fractionation in a 96-well
plate with the cell-based activity assay. Three active compounds
were identified in each strain. Dereplication of the active
Figure 1. Key 2D NMR correlations used for the determination of the
structure of ribocyclophane A (1).
To determine the relative configuration of the sugar moiety,
3JHH analysis and an NOE experiment were initially attempted.
3
The small JHH value (1.8 Hz), observed between H-1′/6′ and
H-2′/7′, indicated that these two protons are in an either ax−
eq or eq−eq relationship. The 3JHH values between H-2′/7′, H-
3′/8′, and H-4′/9′ could not be measured directly from the 1H
1
fractions of UIC 10279 by LC-MS and H NMR spectroscopy
indicated the presence of glycosylated cyclophanes with
B
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
NMR spectrum due to the broad signals observed; however,
the broad line shapes of NMR signals observed for all of the
3
signals indicated small JHH values between these protons. In
addition, no NOE correlation was observed between H-1′/6′
and H-3′/8′ and between H-2′/7′ and H-4′/9′, complicating
the determination of the relative configuration for the sugar
3
moiety. To unambiguously determine the JHH values of the
glycone moiety, spin simulation was performed using Spin-
3
works. The JHH values measured using spin simulation were
3.0 (H-2′/7′ and H-3′/8′), 3.0 (H-3′/8′ and H-4′/9′), 1.2 (H-
4′/9′ and H-5′a/10′a), and 2.4 Hz (H-4′/9′ and H-5′b/10′b),
respectively. These small J values clearly demonstrated that all
the protons are in ax−eq or eq−eq relationships, thus
suggesting β-D-ribopyranose (1C4), β-L-ribopyranose (4C1), β-
D-xylopyranose (1C4), or β-L-xylopyranose (4C1) as a possible
glycone moiety. All of these possible sugar moieties adopt
thermodynamically unfavorable conformations where three or
all four oxygen atoms are in axial positions. To unambiguously
disclose the relative configuration of the sugar moiety as well as
its spatial conformation, single-crystal X-ray crystallographic
analysis was carried out using a crystal obtained from a mixture
of EtOH and H2O (1:1) by slow evaporation. The crystal
structure (Figure 2) of 1 revealed that the relative configuration
Figure 3. β-D-Ribopyranose glycone adopts the less favorable 1C4
conformation to avoid steric interactions with the cylindrocyclophane
aglycone moiety.
spatial volume of the glycone moiety. The selection of the
unfavorable 1C4 conformation is also supported by the
anomeric effect, in which heteroatom substituents preferentially
assume an axial orientation when adjacent to a heteroatom
within a ring. The glycone moiety of 1 obtained from UIC
10366 was found to have the same negative specific rotation
1
and identical H NMR signals to those observed for 1 from
UIC 10279, suggesting that the glycone produced by UIC
10366 is also a β-D-ribopyranose. The crystal structure of 1
allowed the assignment of the stereocenters of the
cylindrocyclophane aglycone as 1R, 2S, 7R, 14R, 15S, 20R,
which are identical to those reported for cylindrocyclophane A.
Ribocyclophane B (2) was obtained as a white amorphous
powder from UIC 10279, and HRESIMS analysis (m/z
715.4477 [M − H]−) suggested the molecular formula of 2
1
to be C41H64O10. The H NMR spectrum of 2 showed an
overall signal distribution pattern similar to that observed for 1.
The appearance of four singlet aromatic (δH 6.07, 6.09, 6.20,
and 6.24) and two doublet methyl (δH 1.05 and 1.06) signals
indicated the structure of 2 to be nonsymmetrical. The
integration of each signal of the glycone moiety corresponded
to that of one aromatic proton signal of the aglycone moiety,
suggesting the presence of only one sugar moiety in 2. The
newly appearing signal of H-14 (δH 3.74) compared to 1,
together with the mass difference of 132 between 2 and 1,
demonstrated that one of the glycoside moieties in 1 was
replaced by a hydroxy group in 2. The ECD spectrum of 2
showed positive Cotton effects at 221 and 275 nm, identical to
those observed for 1, suggesting the same absolute config-
uration for 2.
Figure 2. ORTEP drawing of ribocyclophane A (1).
of the glycone moiety is a β-ribopyranose where oxygen atoms
at C-1′, C-2′, and C-4′ were axially positioned. The absolute
configuration of 1 was determined by measuring the optical
rotation of the glycone moiety and electronic circular dichroism
(ECD) analysis of the aglycone moiety, each liberated by acid
hydrolysis (1 N HCl). The negative rotation ([α]D −9)
observed for the glycone moiety indicated the absolute
Ribocyclophane C (3) was obtained as a white amorphous
powder from UIC 10279. HRESIMS analysis (m/z 699.4533
[M − H]−) suggested the molecular formula of 3 as C41H64O9.
1
configuration to be a β-D-ribopyranose with C4 conformation
(Figure 2). Negative Cotton effects observed at 230 and 280
nm for the aglycone moiety are identical to those observed for
cylindrocyclophane A, indicating the same absolute config-
uration.6 The potential energy of a 1C4 conformation is
1
The H NMR spectrum of 3 closely resembled that of 2, and
four aromatic and two doublet methyl signals were observed,
thus indicating the structure of 2 to be nonsymmetrical due to
the presence of one glycone moiety. The notable difference was
the replacement of an oxygenated methine proton H-14 in 2 by
diastereotopic methylene protons (δH 1.81 and 2.61) in 3. This,
together with the 16 mass difference between 2 and 3, which
corresponded to one oxygen atom, suggested the absence of the
hydroxy group at C-14. This was confirmed by NMR analysis
(Table 2), demonstrating the structure of 3 to be the
deoxygenated analogue of 2. The positive Cotton effects
observed at 222 and 274 nm indicated the same absolute
configuration of 3 as determined for 2.
4
approximately 4 times higher than that of a C1 conformation
for β-D-ribopyranose.19 The adoption of a thermodynamically
unfavorable conformation for the glycone moiety could be
rationalized by a favored spatial arrangement to avoid a steric
interaction between the algycone and glycone moieties in the
overall structure (Figure 3). The β-D-ribopyranose glycone with
4
the preferred C1 conformation occupies a much larger spatial
1
volume than that with a C4 conformation, creating a steric
effect with the aromatic ring and branched methyl of the
cylindrocyclophane aglycone. This steric effect forces the sugar
1
1
moiety into a C4 conformation, and the adoption of a C4
conformation increases the overall stability of 1 by reducing the
Ribocyclophane D (4) was obtained as a white amorphous
powder from UIC 10366. HRESIMS analysis (m/z 757.4489
C
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Ribocyclophane E (5) was obtained as a minor component
from UIC 10366 during the isolation of 4. HRESIMS analysis
showed an isotopic pattern consistent with the presence of one
chlorine atom and suggested a molecular formula of
C43H67ClO11 (m/z 793.9098 [M − H]− and 795.9101 [M −
Table 1. NMR Spectroscopic Data of Ribocyclophane A (1)
in MeOH-d4
b
δH, mult.
a
b
b
no.
δC, type
(J in Hz)
COSY
2/15
HMBC
1/14
90.1, CH 3.65, d
(12.2)
2/15, 3/16, 10/23, 11/24,
12/25, 35/36, 1′/6′
1
H]− in a 3:1 ratio). Comparing the H NMR spectrum to the
spectra for 1−4 indicated 5 to be structurally different. The 1H
c
2/15
41.8, CH 1.65, m
1/14, 3/
16,
NA
1
NMR spectrum of 5 was more similar to the H NMR spectra
of the cylindrofridins published by Preisitsch et al.15 NMR
analysis allowed the assignment of two aliphatic chains as well
as a ribose. The signal at δH 3.83 indicated a proton adjacent to
a halogen, suggesting the presence of the chlorine atom on one
of the aliphatic chains. HSQC and 13C spectra confirmed the
presence of the chlorine, with the signal at δH 3.83 correlating
to C-7 at δC 65.1. COSY correlations from H-7 to H-6 and H-
27 confirmed the placement of the chlorine along the aliphatic
chain. The presence and positions of two resorcinol-like groups
were deduced by HMBC correlations from H-23/25 to C-1 and
likewise from H-10/12 to C-14, H-20 to C-22/26, and H-31/
19 to C-21. No correlations were observed from H-7 to C-9/13
or H-6/27 to C-8, as would be present in other [7.7]-
paracyclophane HMBC spectra. This is consistent with
cylindrofridin-like structures that are linear. HMBC correlation
from H-1 to C-1′ confirmed the position of the ribose, and a
correlation from H-14 to C-37, along with the strong singlet
signal at δH 2.07, suggested the presence of an acetate moiety.
The Cotton effects observed at 222 and 274 nm indicated the
same absolute configuration as the previously described
cylindrofridins and [7.7]paracyclophanes.15 We were unable
to establish the configuration of the stereogenic center at C-7.
Ribocyclophanes A−D (1−4) are glycosylated analogues of
the cyclindrocyclophanes. Ribocyclophane E (5) is a
glycosylated analogue of the cylindrofridins and a likely
biosynthetic intermediate of 4. A notable structural feature of
1−4 is the presence of a β-D-ribopyranose glycone moiety
linked to the benzoyl carbon of the cylindrocyclophane
aglycone. This glycone moiety exclusively adopted the
thermodynamically unfavorable 1C4 conformation in which
three oxygen atoms are in axial positions and one oxygen atom
is in an equatorial position. The crystal structure of 1 showed
that no hydrogen bonding is involved between the glycone and
aglycone moieties to favor the shift of conformational
35/36
d
3/16
4/17
5/18
6/19
34.9, CH2 0.66, m
0.75, m
2/15, 4/17 NA
3/16, 5/18 NA
4/17, 6/19 NA
c
30.0, CH2 0.82, m
1.44, m
d
30.9, CH2 0.71, m
0.96, m
35.5, CH2 1.32, m
5/18, 7/20 4/17, 5/18, 7/20, 27/31
2.03, ddd
(4.2)
7/20
37.0, CH 3.16, m
6/19,
27/31
5/18, 6/19, 8/21, 9/22,
13/26, 27/31, 28/32
8/21
9/22
117.9, C
158.9, C
10/23 105.5, CH 6.19, s
11/24 142.1, C
1/14, 12/25, 8/21, 9/22
1/14, 10/23, 8/21, 13/26
12/25 109.3, CH 6.09, s
13/26 157.1, C
27/31 34.8, CH2 1.47, m
1.95, m
7/20,
28/32
6/19, 7/20, 8/21, 28/32,
29/33
d
28/32 31.9, CH2 1.04, m
1.17, m
27/31,
29/33
NA
d
29/33 24.0, CH2 1.18, m
1.28, m
28/32,
30/34
NA
30/34 14.8, CH3 0.79, t
(7.2)
29/33
28/32, 29/33
35/36 16.7, CH3 1.05, d
(6.0)
2/15
1/14, 2/15, 3/16
1/14, 5′/10′, 2′/7, 3′/8′
1′/6′
2′/7′
3′/8′
4′/9′
104.6, CH 4.85, d
(1.8)
2′/7′
c
72.8, CH 3.73, d
1′/6′, 3′/
8′
NA
(1.8, 3.0)
c
66.9, CH 3.70, t
(3.0)
2′/7′, 4′/
NA
9′
c
71.6, CH 3.59, ddd
(2.4, 3.0)
3′/8′, 5′/
10′
NA
5′/10′ 65.6, CH2 3.19, dd
4′/9′
1′/6′, 3′/8′, 4′/9′
1
(12.6,
2.4)
equilibrium toward the C4 conformation. This suggested that
an avoidance of steric crowding between the β-D-ribopyranose
3.30, dd
(12.6)
glycone and the cylindrocyclophane aglycone is likely to be the
1
major driver of the adoption of the C4 conformation.
a
Assigned from the DEPT-Q spectrum acquired at 226 MHz.
Recorded at 600 MHz. No signals were observed. Could not be
Ribocyclophanes A−E (1−5) and cylindrocyclophane D (6)
were tested for their antiproliferative activity against MDA-MB-
231 and MDA-MB-435 cancer cell lines. Ribocyclophane D (4)
showed the strongest activity among six compounds with IC50
values of less than 1 μM in both cell lines (IC50 0.8 and 0.6 μM,
respectively), whereas moderate activity was observed for
ribocyclophane C (3) (IC50 1.2 and 1.3 μM, respectively).
Ribocyclophane E (5) showed no antiproliferative activity
against either cell line at 25 μM, corroborating the findings of
Preisitsch et al. that an unclosed [7.7]paracyclophane core
structure does not retain antiproliferative activity.15 Ribocyclo-
phanes A and B and cylindrocyclophane D all had modest
activity against the two cell lines, as shown in Supporting
Information S18. Interestingly, the large glycone moieties do
not seem to change the antiproliferative activity, as
ribocyclophane D has similar activity levels to smaller
[7.7]paracyclophanes such as the merocyclophanes.8,11
b
c
d
assigned due to the signal overlap.
[M − H]−) suggested the molecular formula of 4 to be
C43H66O11. The 1H NMR spectrum of 4 was similar to that of 2
and 3, suggesting the compound was nonsymmetrical due to
differing moieties on the cylindrocyclophane scaffold. The
difference in molecular formula between 2 and 4 suggested the
presence of an acetate group. The presence of a singlet at δH 2.0
in the 1H NMR spectrum and signals at δC 21.3 and δC 172.6 in
the 13C NMR spectrum confirmed the presence of an acetate
moiety. HMBC correlations from H-14 to C-37 confirmed the
placement of the acetate moiety α to the resorcinol core. The
positive Cotton effects observed at 222 and 273 nm indicated
the same absolute configuration as all previously described
[7.7]paracyclophanes.
D
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 2. NMR Spectroscopic Data of Ribocyclophanes B (2) and C (3) in MeOH-d4
ribocyclophane B (2)
ribocyclophane C (3)
ribocyclophane B (2)
ribocyclophane C (3)
ab
,
b
ab
,
b
ab
,
b
ab
,
b
no. δC, type δH , mult. (J in Hz) δC, type δH , mult. (J in Hz)
no. δC, type δH , mult. (J in Hz) δC, type δH , mult. (J in Hz)
1
2
3
88.5, CH
40.1, CH
3.65, d (10.2)
1.66, m
88.4, CH
40.1, CH
3.66, d (10.2)
1.66, m
22
23
24
25
26
27
157.7, C
157.2, C
103.7, CH 6.20, s
140.4, C
103.9, CH 6.20, s
140.3, C
33.4, CH2 0.65, m
0.76, m
33.3, CH2 0.66, m
0.79, m
107.8, CH 6.09, s
155.0, C
107.6, CH 6.09, s
155.7, C
4
5
6
28.6, CH2 0.83, m
1.42, m
28.6, CH2 0.81, m
1.43, m
33.3, CH2 1.47, m
1.95, m
33.3, CH2 1.48, m
1.94, m
29.0, CH2 0.72, m
0.95, m
29.3, CH2 0.74, m
0.97, m
28
29
30.2, CH2 1.04, m
1.17, m
30.3, CH2 1.06, m
1.17, m
33.9, CH2 1.32, m
2.04, m
34.0, CH2 1.32, m
2.00, m
22.4, CH2 1.19, m
1.29, m
22.3, CH2 1.20, m
1.29, m
7
35.4, CH
116.1, C
157.7, C
3.16, m
35.4, CH
114.7, C
156.9, C
3.11, m
8
30
31
13.1, CH3 0.79, t (7.2)
33.3, CH2 1.47, m
1.95, m
13.1, CH3 0.82, t (7.2)
33.3, CH2 1.48, m
1.94, m
9
10
11
12
13
14
103.5, CH 6.24, s
142.2, C
106.5, CH 6.02, s
139.4, C
32
33
30.2, CH2 1.04, m
1.17, m
30.3, CH2 1.06, m
1.17, m
107.6, CH 6.07, s
156.0, C
108.5, CH 5.97, s
155.7, C
22.4, CH2 1.19, m
1.29, m
22.3, CH2 1.20, m
1.29, m
80.2, CH
3.74, d (9.6)
44.3, CH2 1.81, t (12.6)
2.61, dd 13.2, 3.6)
1.57, m
34
35
36
1’
13.1, CH3 0.79, t (7.2)
15.4, CH3 1.05, d (7.2)
15.4, CH3 1.06, d (6.6)
103.0, CH 4.85, d (2.4)
13.3, CH3 0.79, t (7.2)
15.4, CH3 1.05, d (6.6)
19.3, CH3 0.95, d (6.6)
103.0, CH 4.85, d (2.4)
15
16
40.7, CH
1.56, m
35.4, CH
33.4, CH2 0.65, m
0.76, m
35.3, CH2 0.62, m
1.01, m
17
18
19
28.6, CH2 0.83, m
1.42, m
28.8, CH2 0.81, m
1.34, m
2’
71.1, CH
65.2, CH
70.1, CH
3.73, brs
71.1, CH
65.3, CH
70.0, CH
3.73, brs
3′
4’
3.70, t (3.0)
3.60, brs
3.70, t (3.0)
3.59, brs
29.0, CH2 0.72, m
0.95, m
29.3, CH2 0.74, m
0.97, m
5′
63.9, CH2 3.20, dd (12.6, 2.4) 64.0, CH2 3.20, dd (12.6, 2.4)
33.9, CH2 1.32, m
2.04, m
34.0, CH2 1.32, m
2.00, m
3.31, overlapped 3.31, overlapped
a
b
Assigned from the HSQC and HMBC spectra. Recorded at 600
MHz.
20
21
35.4, CH
116.1, C
3.16, m
35.4, CH
116.4, C
3.16, m
four 2.8 L Fernbach flasks each containing inorganic Z media (4 × 2
L) with sterile aeration.18 Cultures were illuminated with fluorescent
lamps at 1.03 klx with an 18/6 h light/dark cycle. The temperature of
the culture room was maintained at 22 °C. After 7 weeks, the
biomasses were harvested by centrifugation and then freeze-dried.
Extraction and Isolation. The freeze-dried cells were each
extracted with CH2Cl2−MeOH (1:1 v/v) three times and concen-
trated in vacuo to yield organic extracts (0.5 g of UIC 10279 and 0.65
g of UIC 10366). The resulting extracts were fractionated using Diaion
HP-20 resin with an increasing amount of iPrOH in H2O. The
fractions eluting at 50% and 60% iPrOH were found to be active in the
antiproliferative assay against MDA-MB-435 cells (95% and 97% at 25
μg/mL, respectively). HPLC-based activity profiling of the active
fractions, which associates the HPLC chromatogram with biological
activity, identified three active peaks for each strain. Dereplication of
these peaks by LC-MS and 1H NMR indicated these peaks to be new
cyclophanes with the molecular weights of 848, 758, 716, and 700 Da.
The fractions containing active peaks were combined and subjected to
reversed-phase HPLC (Varian C8 column, 10 mm × 250 mm, 3 mL/
min) eluting with a gradient using aqueous MeOH from 60% to 80%
over 40 min which led to the isolation of ribocyclophanes A (1, 15 mg,
0.5%), B (2, 3 mg, 0.1%), and C (3, 10 mg, 0.3%) from UIC 10279
and ribocyclophanes D (4, 15 mg, 0.5%) and E (5, 1 mg, > 0.1%) and
cylindrocyclophane D (6, 6 mg, 0.2%) from UIC 10366.
EXPERIMENTAL SECTION
■
General Experimental Procedures. Optical rotations were
measured on a PerkinElmer 241 polarimeter. ECD spectra were
recorded on a Jasco J-710 ECD spectrometer. UV and IR spectra were
recorded on a Shimadzu UV2401 UV spectrometer and a Thermo
Nicolet 6700 FT-IR spectrometer, respectively. 1D and 2D NMR
1
spectra including H NMR, COSY, TOCSY, HSQC, HMBC, and T-
ROESY spectra were obtained on a Bruker Avance 600 MHz NMR
spectrometer with a 5 mm CPTXI Z-gradient probe, whereas a Bruker
Avance II 900 MHz NMR spectrometer with a 5 mm ATM CPTCI Z-
gradient probe was used to acquire the DEPT-Q spectra. H and 13C
1
NMR chemical shifts were referenced to the DMSO-d6 solvent signals
(δH 2.49 and δC 39.7, respectively) or MeOH-d4 solvent signals (δH
3.31 and δC 49.15, respectively). A mixing time of 60 ms was set for
the TOCSY experiment and 200 ms for the T-ROESY experiment.
The HMBC spectrum was recorded with the average 3JCH of 8 Hz, and
the HSQC spectrum was measured with the average JCH of 145 Hz.
HRESIMS and LC-MS data were obtained on a Shimadzu IT-TOF
LC-MS spectrometer.
Biological Material. Nostoc sp. (UIC 10279) was isolated from a
sample collected in the southwest suburbs of Chicago near the town of
Joliet, Illinois. The unialgal strain (UIC 10279) was produced through
micropipet isolation techniques.20 Taxonomic identification of the
strain UIC 10279 was carried out using a partial 16S rRNA gene
sequence (GenBank Accession No. JX962720; see SI, S2). Nostoc sp.
(UIC 10366) was isolated from a sample collected in the northwest
suburbs of Chicago near the town of Elgin, Illinois, through micropipet
isolation.20 Taxonomic identification of the strain UIC 10366 was
carried out using a partial 16S rRNA gene sequence (GenBank
Accession No. MF622942; see SI, S2). Both strains were cultured in
1
Ribocyclophane A (1): white, amorphous powder; [α]25 −6 (c
D
0.12, MeOH); UV (MeOH) λmax (log ε) 208 (4.41), 223 (3.88), 276
(3.13) nm; ECD (MeOH 1.0 mM) λmax (Δε) 208 (2.87), 224 (1.27),
282 (1.04) nm; IR (neat) νmax 3365, 2930, 2857, 1628, 1432, 1381,
1
1321 cm−1; H and 13C NMR, COSY, and HMBC data, Table 1;
HRESIMS m/z 847.4894 [M − H] − (calcd for C46H71O14, 847.4844).
E
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Table 3. NMR Spectroscopic Data of Ribocyclophanes D (4) and E (5) in MeOH-d4
ribocyclophane D (4)
ribocyclophane E (5)
ribocyclophane D (4)
ribocyclophane E (5)
a
c
b
c
a
c
b
c
no.
δC, type
δH, mult. (J in Hz)
δC, type
δH, mult. (J in Hz)
no.
δC, type
δH, mult. (J in Hz)
δC, type
δH, mult. (J in Hz)
1
2
3
90.1, CH
41.7, CH
34.9, CH2
3.65, d (10.0)
1.64, m
0.65, m
0.76, m
0.84, m
1.43, m
0.7, m
87.6, CH
40.4, CH
33.9, CH2
4.0, m
24
25
26
27
142.1, C
141.9, C
1.77, m
0.97, m
1.26, m
1.32, m
109.2, CH
157.1, C
6.08, s
107.3, CH
158.3, C
6.22, br
35.0, CH2
1.49, m
39.4, CH2
1.6, m
4
5
6
30.0, CH2
30.8, CH2
35.5, CH2
22.7, CH2
27.6, CH2
39.4, CH2
1.91, m
1.68, m
28
29
31.8, CH2
24.0, CH2
1.05, m
27.6, CH2
23.4, CH2
1.29, m
1.33, m
1.44, m
1.54, m
1.62, m
3.83, m
6.15, s
1.17, m
1.34, m
0.94, m
1.33, m
2.03, m
3.16, m
1.19, m
1.27, m
1.29, m
1.33, m
30
31
14.8, CH3
34.8, CH2
0.79, t (7.3)
1.49, m
14.3, CH3
34.5, CH2
0.92, t (7.3)
1.5, m
7
37.0, CH
118.7, C
159.0, C
109.6, CH
139.3, C
105.3, CH
157.1, C
83.6, CH
40.2, CH
34.6, CH2
65.1, CH
102.7, CH
159.9, C
8
1.95, m
1.94, m
9
32
33
31.9, CH2
24.1, CH2
1.05, m
31.7, CH2
23.8, CH2
1.07, m
10
11
12
13
14
15
16
6.13, s
6.18, s
106.0, CH
144.1, C
6.17, s
6.17, s
1.17, m
1.17, m
1.22, m
1.22, m
106.0, CH
159.9, C
1.29, m
1.30, m
34
35
36
37
38
1’
14.7, CH3
17.0, CH3
16.7, CH3
172.6, C
0.82, t (7.3)
1.05, d (6.4)
0.96, d (6.4)
15.1, CH3
16.5, CH3
14.8, CH3
173.1, C
0.82, t (7.2)
0.99, d (6.5)
0.85, d (6.5)
5.0, d (10.4)
1.77, m
0.71, m
0.78, m
0.84, m
1.43, m
0.7, m
80.8, CH
39.9, CH
30.6, CH2
5.4, m
1.79, m
1.29, m
21.3, CH3
104.5, CH
72.8, CH
67, CH
2.0, s
21.2, CH3
103.8, CH
72.6, CH
66.9, CH
71.3, CH
65.6, CH2
2.07, s
17
18
19
29.8, CH2
30.7, CH2
35.4, CH2
29.8, CH2
29.6, CH2
34.5, CH2
1.05, m
1.34, m
1.09, m
4.85, d (2.0)
3.72, brs
4.89, br
2’
3.76, br
3′
4’
3.7, t (3.2)
3.6, brs
3.78, br
0.94, m
1.33, m
2.03, m
3.16, m
71.6, CH
65.7, CH2
3.67, br
1.47, m
1.92, m
3.19, m
5′
3.2, dd (12.3, 2.5)
3.32, overlapped
3.32, overlapped
3.47, overlapped
a
20
21
22
23
37.0, CH
117.8, C
159.0, C
105.5, CH
36.4, CH
118.2, C
158.3, C
107.3, CH
Assigned from the DEPT-Q spectrum acquired at 226 MHz.
Assigned from HSCQ and HMBC spectra. Recorded at 900 MHz.
b
c
6.2, s
6.22, br
Spectroscopic data of 1 isolated from UIC 10366 were identical to
CCD detector. A total of 240 images of width 1° were collected at a
crystal-to-detector distance of 91 mm and wavelength 0.77 Å, to a
resolution of 0.8 Å. The images were indexed, and the intensities were
integrated and scaled with XDS.21 The unit cell is monoclinic, space
group P21, with two molecules in the asymmetric unit and eight
disordered MeOH molecules. The structure was solved by SHELXS
and refined with SHELXL.22 Crystal data: 2C46H72O14·8C2H6O,
monoclinic, space group P21 (No. 4), a = 10.752(2) Å, b = 34.601(4)
Å, c = 16.762(2) Å, β = 90.480(8)°, V = 6235.7(16) Å3, Z = 4, Dc =
1.10, μ = 0.081, F000 = 2256. Reflections collected/unique = 55 588/
23 169 (Rint = 0.0514), final R1 = 0.0996, wR2 = 0.2478 for reflections
with I > 2σI; R1 = 0.1063, wR2 = 0.2552 for all reflections; goodness-
of-fit = 1.019 for all unique data. Bijvoet analysis: Flack x = −0.4(6),
Parsons z = −0.24(15), probability P2(true) = 1.000. Crystallographic
data (ID entry CCDC 898167) have been deposited in the Cambridge
Crystallographic Data Centre.
those from UIC 10279.
Ribocyclophane B (2): white, amorphous powder; [α]25 −6 (c
D
0.07, MeOH); UV (MeOH) λmax (log ε) 207 (4.10), 221 (3.65), 275
(2.90) nm; ECD (MeOH 0.7 mM) λmax (Δε) 208 (1.62), 222 (0.99),
284 (0.86) nm; IR (neat) νmax 3378, 2930, 2858, 1631, 1469, 1432,
1372 cm−1; 1H and 13C NMR data, Table 2; HRESIMS m/z 715.4471
[M − H]− (calcd for C41H63O10, 715.4421).
Ribocyclophane C (3): white, amorphous powder; [α]25 +6 (c
D
0.16, MeOH); UV (MeOH) λmax (log ε) 209 (4.54), 223 (4.03), 275
(3.23) nm; ECD (MeOH 0.7 mM) λmax (Δε) 211 (2.59), 224 (1.44),
282 (2.00) nm; IR (neat) νmax 3382, 2926, 2856, 1691, 1620, 1592,
1
1431, 1369 cm−1; H and 13C NMR data, Table 2; HRESIMS m/z
699.4533 [M − H]− (calcd for C41H63O9, 699.4472).
Ribocyclophane D (4): white, amorphous powder; [α]25 −3 (c
D
0.14, MeOH); UV (MeOH) λmax (log ε) 208 (4.38), 220 (3.85), 280
(3.05) nm; ECD (MeOH 1.0 mM) λmax (Δε) 211(2.82), 228 (1.21),
284 (0.9); IR (neat) νmax 3380, 2931, 2862, 1625, 1433, 1372, 1256
cm−1; 1H and 13C NMR data, Table 3; HRESIMS m/z 757.4548 [M −
H]− (calcd for C43H65O11, 757.4526).
Acid Hydrolysis of 1. Approximately 5 mg of 1 was hydrolyzed
using 1 N HCl (1 mL) at 100 °C for 1 h. The resulting hydrolysate
was air-dried and extracted with H2O and EtOAc to yield the H2O (1.3
mg) and EtOAc (3.9 mg) fractions. The negative specific rotation
observed for the H2O fraction suggested the sugar moiety to be D-
Ribocyclophane E (5): white, amorphous powder; [α]25 +7 (c
D
0.14, MeOH); UV (MeOH) λmax (log ε) 211 (3.82), 223 (3.61), 278
(2.98) nm; ECD (MeOH 0.7 mM) λmax (Δε) 208 (2.12), 226 (1.12),
277 (−0.8); IR (neat) νmax 3405, 2930, 2858, 1567, 1426, 1387, 1257,
ribopyranose: [α]25 −9 (c 0.13, H2O). Reversed-phase HPLC of the
D
EtOAc fraction (C18, 60−80% aqueous MeOH) yielded the pure
aglycone moiety (0.5 mg; HRESIMS m/z 583.4113 [M − H]−). The
ECD spectrum of the aglycone moiety displayed negative Cotton
effects at 230 and 280 nm, indicating the same absolute configuration
as that reported for cylindrocyclophane A.6
1
830 cm−1; H and 13C NMR data, Table 3; HRESIMS m/z 793.4310
[M − H]− (calcd for C43H66ClO11, 793.4293).
Single-Crystal X-ray Crystallographic Analysis of 1. A small
single crystal, roughly 10 × 10 × 30 μm, was encased in Paratone-N oil
and cooled to 100 K in a nylon loop. Data were collected at the
Advanced Photon Source, LS-CAT, sector 21, on a MAR 300 mm
Antiproliferative Assay. The human melanoma cell line MDA-
MB-435 and human breast cancer cell line MDA-MB-231 were
purchased from the American Type Culture Collection. The cells were
F
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
propagated at 37 °C in 5% CO2 in RPMI 1640 medium supplemented
with fetal bovine serum (10%), penicillin (100 units/mL), and
streptomycin (100 μg/mL). Cells in log phase growth were harvested
by trypsinization. A total of 5000 cells were seeded per well of a 96-
well plate and incubated overnight at 37 °C in 5% CO2. Samples
dissolved in DMSO were then sequentially diluted and added to the
appropriate wells (total volume 100 μL). Each compound was tested
at the following concentrations (μg/mL): 25, 5.0, 1.0, 0.2, and 0.04.
The cells were incubated in the presence of test substance for 96 h at
37 °C and evaluated for viability with a commercial absorbance assay
(CellTiter 96 AQueous One Solution cell proliferation assay, Promega
Corp). Activity was expressed as the percentage of viable cells present
relative to the negative (solvent) control. The positive control was
vinblastine tested at 1 ng/mL, which had 49% viable cells after
treatment.
DEDICATION
■
Dedicated to Susan Band Horwitz, of Albert Einstein College of
Medicine, Bronx, NY, for her pioneering work on bioactive
natural products.
REFERENCES
■
(1) Tan, L. T. Phytochemistry 2007, 68, 954−979.
(2) Van Wagoner, R. M.; Drummond, A. K.; Wright, J. L. C. Adv.
Appl. Microbiol. 2007, 61, 89−217.
(3) Chlipala, G. E.; Mo, S.; Orjala, J. Curr. Drug Targets 2012, 12,
1654−1673.
(4) Moore, B. S.; Chen, J.; Patterson, G. M. L.; Moore, R. E.
Tetrahedron 1992, 48, 3001−3006.
(5) Chen, J. L.; Moore, R. E.; Patterson, G. M. L. J. Org. Chem. 1991,
56, 4360−4364.
(6) Bui, H. T. N.; Jansen, R.; Pham, H. T. L.; Mundt, S. J. Nat. Prod.
2006, 69, 1379−1383.
ASSOCIATED CONTENT
■
S
* Supporting Information
(7) Chlipala, G. E.; Sturdy, M.; Krunic, A.; Lantvit, D. D.; Shen, Q.;
Porter, K.; Swanson, S. M.; Orjala, J. J. Nat. Prod. 2010, 73, 1529−
1537.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.7b00954.
(8) Kang, H. S.; Santarsiero, B. D.; Kim, H.; Krunic, A.; Shen, Q.;
Swanson, S. M.; Chai, H.; Kinghorn, A. D.; Orjala, J. Phytochemistry
2012, 79, 109−115.
Taxonomic identification and phylogenetic tree of UIC
10279 and UIC 10366 strains; 1H NMR, COSY, HSQC,
and HMBC spectra of 1−5; DEPT-Q spectrum of 1 and
4, ECD spectra of 1−5, and antiproliferative data for 1−6
against MDA-MB-435 and MDA-MB-231 cancer cell
lines (PDF)
(9) Preisitsch, M.; Harmrolfs, K.; Pham, H. T.; Heiden, S. E.; Fussel,
̈
A.; Wiesner, C.; Pretsch, A.; Swiatecka-Hagenbruch, M.; Niedermeyer,
T. H.; Muller, R.; Mundt, S. J. Antibiot. 2015, 68, 16510.1038/
̈
ja.2014.118.
(10) Luo, S.; Kang, H.-S.; Krunic, A.; Chlipala, G. E.; Cai, G.; Chen,
W.-L.; Franzblau, S. G.; Swanson, S. M.; Orjala, J. Tetrahedron Lett.
2014, 55, 686−689.
(11) May, D. S.; Chen, W.; Lantvit, D. D.; Zhang, X.; Krunic, A.;
Burdette, J. E.; Eustaquio, A.; Orjala, J. J. Nat. Prod. 2017, 80, 1073−
1080.
Crystallographic data (CIF)
Crystallographic data (CIF)
AUTHOR INFORMATION
(12) Bobzin, S.; Moore, R. Tetrahedron 1993, 49, 7615−7626.
(13) Nakamura, H.; Hamer, H. a.; Sirasani, G.; Balskus, E. P. J. Am.
Chem. Soc. 2012, 134, 18518−18521.
■
Corresponding Author
*Tel: +1-312-996-5583. Fax: +1-312-996-7107. E-mail: orjala@
uic.edu.
(14) Nakamura, H.; Wang, J. X.; Balskus, E. P. Chem. Sci. 2015, 6,
3816−3822.
ORCID
(15) Preisitsch, M.; Niedermeyer, T. H. J.; Heiden, S. E.; Neidhardt,
I.; Kumpfmuller, J.; Wurster, M.; Harmrolfs, K.; Wiesner, C.; Enke, H.;
̈
Hahk-Soo Kang: 0000-0001-9602-6042
Jimmy Orjala: 0000-0002-4420-2048
Muller, R.; Mundt, S. J. Nat. Prod. 2016, 79, 106−115.
̈
(16) Nakamura, H.; Schultz, E. E.; Balskus, E. P. Nat. Chem. Biol.
2017, 13, 916−921.
Present Address
§Department of Biomedical Sciences and Engineering, Konkuk
University, Seoul 05029, Korea.
(17) Preisitsch, M.; Heiden, S. E.; Beerbaum, M.; Niedermeyer, T. H.
J.; Schneefeld, M.; Herrmann, J.; Kumpfmuller, J.; Thurmer, A.;
̈
̈
Neidhardt, I.; Wiesner, C.; Daniel, R.; Muller, R.; Bange, F. C.;
̈
Author Contributions
Schmieder, P.; Schweder, T.; Mundt, S. Mar. Drugs 2016, 14, 21.
(18) Falch, B. S.; Konig, G. M.; Wright, A. D.; Sticher, O.;
Angerhofer, C. K.; Pezzuto, J. M.; Bachmann, H. Planta Med. 1995, 61,
321−328.
∥D. S. May and H.-S. Kang contributed equally.
Notes
The authors declare no competing financial interest.
(19) Vijayalakshmi, K. S.; Rao, V. S. R. Carbohydr. Res. 1972, 22 (2),
413−424.
(20) Chlipala, G.; Mo, S.; Carcache de Blanco, E. J.; Ito, A.; Bazarek,
S.; Orjala, J. Pharm. Biol. 2009, 47, 53−60.
(21) Kabsch, W. J. Appl. Crystallogr. 1993, 26, 795−800.
(22) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
2008, 64, 112−122.
ACKNOWLEDGMENTS
■
This research was supported by PO1 CA125066 from NCI/
NIH and T32AT00753, The Office of the Director, National
Institutes of Health (OD) National Center for Complementary
and Integrative Health (NCCIH) to D.S.M. We thank Dr. B.
Ramirez from Center for Structural Biology at UIC for
assistance with NMR experiments. LC-MS analysis was
performed at UIC Research Resource Center (RRC) Mass
Spectrometry Facility. We acknowledge the use of the
Southeast Regional Collaborative Access Team (SER-CAT),
beamline 22-BM, for data collection of these small crystals. Use
of the Advanced Photon Source at Argonne National
Laboratory was supported by the U.S. Department of Energy,
Office of Science, Office of Basic Energy Sciences, under
Contract No. DE-AC02-06CH11357.
G
DOI: 10.1021/acs.jnatprod.7b00954
J. Nat. Prod. XXXX, XXX, XXX−XXX