Viridicatumtoxins: Expanding on a Rare Tetracycline Antibiotic Scaffold

Article abstract of DOI:10.1021/acs.joc.5b02367

Viridicatumtoxins, which belong to a rare class of fungal tetracycline-like mycotoxins, were subjected to comprehensive spectroscopic and chemical analysis, leading to reassignment/assignment of absolute configurations and characterization of a remarkably

Full text of DOI:10.1021/acs.joc.5b02367

Subscriber access provided by The Chinese University of Hong Kong
Article
Viridicatumtoxins: Expanding on a rare tetracycline antibiotic scaffold
Zhuo Shang, Angela Aguslyarti Salim, Zeinab Khalil, Michelle
Quezada, Paul Vincent Bernhardt, and Robert John Capon
J. Org. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.joc.5b02367 • Publication Date (Web): 25 Nov 2015
Downloaded from http://pubs.acs.org on November 26, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Organic Chemistry is published by the American Chemical Society.
1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
Viridicatumtoxins: Expanding on a rare tetracycline
antibiotic scaffold
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Zhuo Shang,a Angela A. Salim,a Zeinab Khalil,a Michelle Quezada,a Paul V. Bernhardtb and
Robert J. Capon*a
aInstitute for Molecular Bioscience, The University of Queensland, St. Lucia, QLD 4072,
Australia
bSchool of Chemistry and Molecular Biosciences, The University of Queensland, St. Lucia, QLD
4072, Australia
*r.capon@uq.edu.au
KEYWORDS (viridicatumtoxins, tetracycline antibiotics, marineꢀderived fungus)
ACS Paragon Plus Environment

The Journal of Organic Chemistry
Page 2 of 26
1
2
3
4
5
Abstract
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Viridicatumtoxins, which belong to a rare class of fungal tetracyclineꢀlike mycotoxins, were
subjected to comprehensive spectroscopic and chemical analysis, leading to
reassignment/assignment of absolute configurations and characterisation of a remarkably acidꢀ
stable antibiotic scaffold. Structure activity relationship studies revealed exceptional growth
inhibitory activity against vancomycinꢀresistant Enterococci (IC50 40 nM), >270ꢀfold more
potent than the commercial antibiotic oxytetracycline.
ACS Paragon Plus Environment
2

Page 3 of 26
The Journal of Organic Chemistry
1
2
3
4
Introduction
5
6
7
8
Multidrug resistant (MDR) infectious diseases represent an extremely serious threat to modern
healthcare, impacting the lives of people and communities across the world. As current and
emerging mechanisms for MDR render existing antibiotics less effective, the need to develop
new and improved antibiotics is becoming ever more urgent. One promising approach is to
explore rare classes of microbial metabolites, first recognized but not developed as commercial
antibiotics mid to late last century. Our investigations into the secondary metabolism of
Australian marineꢀderived microbes provided an ideal platform from which to embark on such
an exploration, delivering access to a wealth of bacterial and fungal strains rich in rare and novel
metabolites, many with promising antibiotic properties.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In this report we describe a comprehensive exploration of the viridicatumtoxin scaffold, a rare
class of fungal polyketides closely related to the tetracycline antibiotics. Tetracyclines have been
at the forefront in treating infectious diseases for over 60 years, with first generation tetracycline
antibiotics (e.g. chlortetracycline, tetracycline and oxytetracycline) inspiring second (e.g.
minocycline and doxycycline) and third (e.g. tigecycline and eravacycline) generation variants,
many in use today. Notwithstanding the considerable academic and industry investment in the
tetracyclines over many decades, and the role they continue to play in modern healthcare, even
this remarkably successful antibiotic pharmacophore is in need of renewal. We hypothesized that
a better understanding of the chemical and antibiotic properties of their fungalꢀmetabolite
cousins, the viridicatumtoxins, would inform the future development of new and improved
tetracycline antibiotics.
Results and Discussion
ACS Paragon Plus Environment
3

The Journal of Organic Chemistry
Page 4 of 26
1
2
3
4
5
6
7
8
9
An agar plate (PYGA) cultivation of a Paecilomyces sp. (CMBꢀMF010), isolated from the inner
tissues of an intertidal pulmonate mollusk (Siphonaria sp.) collected near Shorncliffe,
Queensland, Australia, exhibited promising Gramꢀpositive antibacterial properties. Fractionation
of a scaled up cultivation yielded viridicatumtoxins A (1)1 and B (2),2,3,4 and the new
viridicatumtoxins D–F (4–6). Subsequent fractionation of a rice solid phase cultivation added to
this diversity, yielding all other known members of this structure class, spirohexaline (7)5 and
previridicatumtoxin (8),6,7 together with the new viridicatumtoxin C (3). In this report we present
a comprehensive study of 1–8, inclusive of detailed spectroscopic analysis leading to structure
elucidation, assignment/reꢀassignment of absolute configurations, inclusive of commentary on
biosynthetic relationships and chemical stability, as well as antibiotic and cytotoxic properties.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Paecilomyces sp. (CMBꢀMF010) viridicatumtoxins 1–8
ACS Paragon Plus Environment
4

Page 5 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
HRESI(+)MS analysis of 1 and 2 revealed pseudoꢀmolecular ions consistent with the
molecular formula C30H31NO10 (ꢁmmu –0.4) and C30H29NO10 (ꢁmmu –0.2), which detailed
analysis of the 1D and 2D NMR (CDCl3) data attributed to the rare fungal polyketides
viridicatumtoxin1 and viridicatumtoxin B.2,3,4 Viridicatumtoxin was first described in 1973 by
Hutchison et al. as a mycotoxin produced by maize meal cultures of Penicillium viridicatum.1
Subsequent Xꢀray analyses permitted assignment of firstly its relative stereochemistry8 then
subsequently its absolute structure.9 Further studies explored its biosynthesis.6ꢀ7,10ꢀ12
Viridicatumtoxin B was reported in 2008 by Kim et al. as a coꢀmetabolite with viridicatumtoxin
from a Korean soil Penicillium sp. (FR11), with both metabolites exhibiting antiꢀMRSA
activity.2 A 2013 synthesis by Nicolaou et al. revised the structure for viridicatumtoxin B
(relative configuration only) including the crystal structure of its synthetic racemate, and saw
viridicatumtoxin renamed viridicatumtoxin A.3,4 In 2013 Tomoda et al. described spirohexaline
(7) as a coꢀmetabolite with a purported (+)ꢀenantiomer of viridicatumtoxin A from a riceꢀbased
cultivation of Penicillium brasilianum (FKIꢀ3368).5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Although (+)ꢀ1 isolated from Paecilomyces sp. (CMBꢀMF010) correlated with the (+)ꢀ
enantiomer encountered by Tomoda et al.,5 this enantiomeric configuration was contradicted by
our independent Xꢀray analysis which is essentially the same as that reported by Silverton et al.9
The crystal structure of 1 (isomorphous with the previously published structures of the diꢀ
methanol solvate8,9) was found to be the diꢀmethanol hemihydrate. The absolute structure was
confirmed by statistical analysis of anomalous dispersion effects of 2045 Bijvoet pairs collected
from an entire sphere of data with CuꢀKα radiation (see ESI). The absolute configurations at all
four chirotopic Cꢀatoms (Cꢀ4a, Cꢀ5, Cꢀ12a and Cꢀ15) are S. To explain this contradiction we
speculated that viridicatumtoxin A can exist as (+)ꢀacid and (–)ꢀsalt forms with [α]D
ACS Paragon Plus Environment
5

The Journal of Organic Chemistry
Page 6 of 26
1
2
3
4
measurements of opposite sign and comparable magnitude. Consistent with this hypothesis, the
5
6
7
(+)ꢀacid form of 1 ([α]D +31.8, EtOH) isolated from CMBꢀMF010 in the presence of TFA, was
8
9
treated with MgSO4 to yield a Mg (–)ꢀsalt form ([α]D –10.0, EtOH). As predicted, on exposure to
HCl the (–)ꢀsalt reverted to the (+)ꢀacid. Significantly, a commercial sample of viridicatumtoxin
A was determined to be a (–)ꢀsalt, with HCl treatment leading to conversion to the (+)ꢀacid.
Comparable changes in the [α]D of 1 and oxytetracycline (9) were also observed in a range of
solvents, suggesting that this differential acid/salt optical rotation phenomenon is a characteristic
of the tetracycline scaffold. Based on these observations, and on consideration of published
isolation protocols, we propose that the (+)ꢀenantiomers of viridicatumtoxins A and B reported
by Tomoda et al.5 and Kim et al.2 respectively, are in fact the free acids (+)ꢀ1 and (+)ꢀ2
respectively, as isolated and characterized herein.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Furthermore, with detailed spectroscopic analysis identifying 7 (C31H32O10, ꢁmmu +0.1) as
spirohexaline,5 on biosynthetic grounds we take this opportunity to correct its absolute
configuration, aligning it with the coꢀmetabolite 1 (Figure 1). Likewise, spectroscopic analysis
identified 8 (C30H33NO10, ꢁmmu –0.3) as previridicatumtoxin, first reported in 2012 as a lateꢀ
stage biosynthetic precursor to 1 accumulated in the gene deletion strain Penicillium aethiopicum
ꢁgsfA,6,7 we now acknowledge 8 as a natural product with an absolute configuration in common
with its coꢀmetabolite 1. Having fully characterized and assigned/corrected absolute
configurations for all four known viridicatumtoxins (1–2, 7–8), we now describe the new
viridicatumtoxins C–F (3–6).
HRESI(+)MS analysis of 3 returned a molecular formula (C30H29NO11, ꢀmmu –0.1) consistent
1
with an oxygenated homologue of 2. Comparison of the H NMR (CDCl3) data for 3 (Table 1)
with those of 2 revealed the only significant differences as replacement of Hꢀ9 (δH 6.80) in 2 with
ACS Paragon Plus Environment
6

Page 7 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
a phenolic hydroxyl (δH 5.97) in 3, and a deshielding of 8ꢀOMe (δH 3.90 to δH 4.05). Analysis of
the 13C NMR data (Table 2) revealed substitution of the Cꢀ9 methine in 2 (δC 102.6) with a
7
8
9
phenolic moiety in 3 (δC 140.3), the latter featuring an HMBC correlation to 10ꢀOH (δH 8.44).
These observations, together with diagnostic 2D NMR correlations (Figure 2), permitted
assignment of the structure for viridicatumtoxin C (3) as indicated, with the absolute
configuration assigned on the basis of biosynthetic links to 1. Supportive of this assignment,
analytical scale treatment of 3 with acid or heat resulted in oxidative conversion to an orthoꢀ
quinone (3a), as evidenced by conversion to an orthoꢀphenylenediamine adduct (3b) (Figures 2
and S3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2. Diagnostic 2D NMR correlations for 3 and derivatives 3a–b
HRESI(+)MS analysis of 4 returned a molecular formula (C30H31NO10, ꢁmmu +0.2) isomeric
with 1. Comparison of the 1D NMR (CDCl3:MeOHꢀd4 [4:1]) data for 4 (Tables 1–2) with 1,
together with diagnostic 2D NMR correlations (Figure 3), confirmed common ring AꢀD and ring
F (Cꢀ1 to Cꢀ4) substructures. Further comparisons revealed the oxymethine resonances for Hꢀ
5/Cꢀ5 in 4 (δH 5.65; δC 80.2) were significantly deshielded compared to 1 (δH 4.50; δC 71.8), with
1 and 4 exhibiting common HMBC correlations from Hꢀ5 to Cꢀ4, Cꢀ4a, Cꢀ5a, Cꢀ6, Cꢀ11a and Cꢀ
12a, and with the quaternary Cꢀ12a (δC 80.3) and Cꢀ12 (δC 195.3) in 1 replaced by oxymethine
(δH 4.53; δC 74.3) and ester/lactone (δC 155.0) resonances in 4. As further evidence of a Cꢀ12a
ACS Paragon Plus Environment
7

The Journal of Organic Chemistry
Page 8 of 26
1
2
3
4
5
6
secondary OH moiety, a sample of 4 was converted to the tetraacetate 4a (Figure S28) to reveal
the predicted deshielding of the Hꢀ12a methine (4 δH 4.53; 4a δH 5.74).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 3. Diagnostic 2D NMR correlations for 4, 5 and 6
Based on these observations we propose that 4 is a secoꢀring E analogue of 1, formed by
nucleophilic attack at Cꢀ12, leading to cleavage of the Cꢀ12 to Cꢀ12a bond and subsequent
intramolecular lactonisation from 5ꢀOH to Cꢀ12 (Figure 4). We anticipate that this sequence
proceeds with retention of configuration about Cꢀ15, Cꢀ4a and Cꢀ5, and racemisation about Cꢀ
12a. Consistent with this hypothesis, viridicatumtoxin E (5) was identified as the Cꢀ12a epimer
of 4, with individual Cꢀ12a configurations inferred from ROESY correlations (Table S11).
Diagnostic ROESY correlations (Figures 5 and S41–S42) established syn relationships between
H4α/H12a in 4, and H4β/H12a in 5, with both 4 and 5 possessing a common H5 to H14α configuration.
Significantly, a key ROESY correlation positioned H12a and H14β in close proximity in 5 (C12a S),
necessitating an epimeric configuration for 4 (C12a R).
ACS Paragon Plus Environment
8

Page 9 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Plausible biosynthetic pathway linking 1 with 2–6 and 8
HRESI(+)MS analysis of 6 returned a molecular formula (C30H29NO11, ꢀmmu +0.2)
suggestive of an oxidized analogue of 4 (and 5). Comparison of 1D and 2D NMR
(CDCl3:MeOHꢀd4 [4:1]) (Tables 1–2) data for 6 with 4 revealed common rings AꢀE, and Cꢀ2 to
Cꢀ4 ring F substructures (Figure 1), with 1D NMR differences focused around replacement of
the Cꢀ12a/Hꢀ12a oxymethine (δC 74.3; δH 4.53) and Cꢀ1 ketone (δC 194.5) in 4 with a Cꢀ12a
carboxylic acid (δC 170.3) and Cꢀ1 ester/lactone (δC 166.2) in 6, and a deshielding of H2ꢀ4 (ꢀδH
0.71 and 0.96) and Cꢀ4a (ꢀδC 5.3) in 6 compared to 4. Diagnostic 2D HMBC correlations (Figure
3) positioned the carboxylic acid moiety at Cꢀ4a, necessitating closure of a Cꢀ1 to Cꢀ4a lactone,
and permitting assignment of the planar structure as indicated. A plausible biosynthetic pathway
(Figure 4) provides an alternate oxidative Cꢀ1 to Cꢀ12a cleavage of the precursor linking 1 with 4
ACS Paragon Plus Environment
9

The Journal of Organic Chemistry
Page 10 of 26
1
2
3
4
and 5, to deliver viridicatumtoxin F (6), with conservation of configuration about all chiral
5
6
centers.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Diagnostic ROESY correlations for 4 and 5
Viridicatumtoxins D (4) and E (5) bear a striking structural similarity to αꢀapoꢀoxytetracycline
(11) and βꢀapoꢀoxytetracycline (12), wellꢀknown but poorly characterized acid degradation
products and impurities detected in the commercial antibiotic oxytetracycline (9).13 To support
structure assignments for 4 and 5, we subjected 9 to two acid degradation protocols, (i) 1 h at
75°C in 0.1 M HClaq (pH 1), and (ii) 4 h at 37°C in 0.01 M HClaq (pH 2), to yield authentic
standards of 11 and 12 together with the associated artifacts 10 and 13 (Figures 6 and S7–S8).
Structures for 11–13 were confirmed by detailed spectroscopic analysis (Tables S17–S19). By
contrast, extended treatment of 1 under these conditions (>10 h) resulted in no degradation
(Figure S6), with the more forcing conditions of 20 h at 100 °C in 5% pꢀtoluene sulfonic
acid/toluene delivering very minor yields (~1.5%) of the quinoneꢀmethides 14 (C28H28O7, ꢀmmu
–0.3) and 15 (C27H24O6, ꢀmmu –0.5) (Figures 7 and S9). Our studies reveal for the first time that
viridicatumtoxin A is remarkably acid stable, in stark contrast to the highly acid labile
tetracycline framework, confirming 4–6 as natural products not handling artifacts.
ACS Paragon Plus Environment
10

Page 11 of 26
The Journal of Organic Chemistry
1
2
Table 1. 1H NMR (600 MHz) data for viridicatumtoxins C–F (3–6)
3
4
3a
4b
5b
6b
5
6
7
8
Pos.
4
α 2.83 (d, 18.6)
β 3.08 (d, 18.6)
−
−
−
α 4.00 (d, 19.8)
β 3.14 (d, 19.8)
5.58 (br s)
α 2.13 (m)
β 2.23 (m)
1.71 (2H, m)
−
1.45 (s)
0.88 (s)
0.63 (s)
4.05 (s)
17.91 (br s)
4.23 (br s)
5.97 (br s)c
8.44 (br s)
14.14 (br s)
4.89 (br s)
a 9.21 (br s)
b 5.97 (br s)c
α 2.44 (d, 17.6)
β 2.02 (d, 17.6)
5.65 (s)
6.61 (s)
4.53 (br s)
α 3.19 (d, 18.1)
β 3.09 (dd, 18.1, 1.6)
5.37 (br s)
α 1.92 (br d, 18.0)
β 2.14 (m)
α 1.74 (m)
β 1.25 (dd, 13.0, 5.9)
1.47 (s)
0.85 (s)
0.34 (s)
α 2.39 (d, 18.3)
β 2.66 (d, 18.3)
5.72 (s)
6.60 (s)
4.29 (br s)
α 3.25 (d, 17.6)
β 3.09 (d, 17.6)
5.42 (br s)
α 1.95 (br d, 18.4)
β 2.17 (m)
α 1.78 (m)
β 1.27 (dd, 13.3, 5.7)
1.51 (s)
0.85 (s)
0.33 (s)
α 3.15 (d, 17.2)
β 2.98 (d, 17.2)
5.85 (s)
6.60 (s)
5
9
12a
14
9
−
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
α 3.20 (d, 17.5)
β 2.92 (d, 17.5)
5.39 (br s)
α 1.91 (br d, 18.6)
β 2.15 (m)
α 1.76 (m)
β 1.25 (dd, 13.1, 5.8)
1.46 (s)
0.81 (s)
0.30 (s)
17
18
19
21
22
23
8ꢀOCH3
3ꢀOH
4aꢀOH
9ꢀOH
10ꢀOH
11ꢀOH
12aꢀOH
13ꢀNH2
3.80 (s)
3.80 (s)
3.78 (s)
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
−
a 1H NMR spectra were acquired in CDCl3; b 1H NMR spectra were acquired in CDCl3/CD3OD (4:1); c Overlapping signals
Table 2. 13C NMR (150 MHz) data for viridicatumtoxins A–F (1–6), spirohexaline (7) and
previridicatumtoxin (8)
Pos.
1
2
3
4
4a
5
5a
6
6a
7
8
9
10
10a
11
11a
12
12a
13
1a
190.6
99.7
192.9
40.5
71.6
2a
194.3
99.7
192.8
42.2
77.8
188.8
146.4c
124.9c
144.8c
127.3
161.3
102.6
158.4
106.9
165.4
107.3
195.2
80.7
173.0
44.6
60.6
135.8
122.0
23.0
34.2
38.5
20.9
24.5
3a
4b
194.5
99.2
192.7
36.8
76.0
5b
192.8
99.4
191.3
39.0
75.9
6b
166.2
99.1
189.7
35.9
81.3
7a
8a
190.1
99.6
193.0
40.4
71.3
193.6
99.7
192.9
42.4
190.3
110.6
194.5
40.8
71.3
71.7
137.3
123.8
147.3
122.8
160.9
100.0
158.1
105.6
166.1
105.2
195.4
80.8
202.4
41.3
60.2
136.7
121.5
23.0
34.0
38.7
21.1
24.1
77.3
71.8
188.7
145.6c
133.0c
137.5c
114.9
150.4
140.3
141.6
108.5
162.3
108.6
196.6
80.8
80.2
81.2
80.5
71.3
137.2
123.9
147.3
122.8
160.9
100.0
158.1
105.6
166.1
105.2
195.3
80.3
172.8
41.3
60.2
136.7
121.5
23.0
130.5
130.0
148.7
121.9
157.8
98.9
155.8
107.2
171.3
103.6
155.0
74.3
172.8
43.8
60.0
137.5
120.7
23.0
130.2
129.4
148.1
121.3
157.6
99.0
155.3
106.2
170.7
102.4
153.5
77.8
172.0
43.0
59.5
137.0
120.4
22.5
129.5
128.7
148.5
121.7
157.8
99.1
155.9
107.3
170.5
103.4
154.8
170.3
172.7
42.0
130.5
130.8
140.9
99.9
164.8
102.2
161.1
108.7
167.2
104.2
195.7
79.4
172.8
27.3
122.3
137.0
39.7
173.0
43.4
14
15
16
17
18
19
20
21
61.4
60.2
135.1
122.8
23.0
137.3
121.0
22.9
33.8
38.3
21.0
23.7
25.5
26.7
34.0
38.7
21.1
24.1
34.3
37.9
20.8
25.2
25.7
−
60.7
34.0
38.3
21.3
23.9
25.6
−
55.5
33.5
37.9
20.8
23.4
25.1
−
55.1
123.9
131.9
16.7
25.8
17.9
22
23
25.6
25.6
−
55.8
25.6
27.9
55.7
2ꢀCOCH3
8ꢀOCH3
−
−
55.4
−
55.7
55.7
a 13C NMR spectra were acquired in CDCl3; 13C NMR spectra were acquired in CDCl3/CD3OD (4:1); The chemical shifts of carbons were
b
c
tentatively assigned by ACD simulation
ACS Paragon Plus Environment
11

The Journal of Organic Chemistry
Page 12 of 26
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 6. Oxytetracycline (9) and degradation products 10–13
Figure 7. Viridicatumtoxin A (1) minor degradation products 14–15
Although originally described as a mycotoxin,1 oral dosage of 1 has been judged non–toxic to
mice (up to 350 mg/kg) and rats (up to 150 mg/kg).14 Armed with a comprehensive suite of all
known and new viridicatumtoxins (1–8), we compared their growth inhibitory properties against
a panel of drugꢀresistant Gram positive and negative bacteria, with those of the tetracycline
antibiotic 9 and its acid degradation products 11–13 (Table S4). While 1–5 and 7–8 exhibited
antiꢀMRSA properties comparable with 9, viridicatumtoxin B (2) was especially effective against
VRE, with an IC50 >270ꢀfold more potent than 9, albeit with moderate levels of cytotoxicity
against three human cancer cell lines (Table 3).
ACS Paragon Plus Environment
12

Page 13 of 26
The Journal of Organic Chemistry
1
2
3
4
Table 3. Antibacterial and cytotoxicity properties (IC50 ꢁM) of 1–9 and 11–13
5
6
7
8
MRSAꢀ2A
1.2
VREB
1.1
0.04
1.5
3.4
1.5
>30
3.1
4.8
11.0
>30
>30
>30
NCIꢀH460C
1.0
KB3ꢀ1D
2.5
SW620E
1.0
1
2
3
4
5
6
7
8
9
0.15
1.7
9.7
3.3
>30
2.3
4.4
0.5
>30
>30
>30
0.6
9.2
24.2
>30
>30
11.3
5.3
15.9
>30
>30
>30
1.6
0.6
9
11.0
>30
>30
>30
6.3
17.4
20.7
>30
>30
16.7
6.0
>30
>30
>30
>30
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4.1
21.3
>30
>30
11
12
13
>30
A AUSꢀRBWHꢀMRSAꢀ02 (methicillinꢀresistant Staphylococcus aureus); AUSꢀRBWHꢀVREꢀ01
B
C
(vancomycinꢀresistant Enterococcus faecalis), NCIꢀH460 is a human lung carcinoma cell line,
D KB3ꢀ1 is a human cervix carcinoma cell line, E SW620 is a human colon carcinoma cell line
Figure 8. Viridicatumtoxin inspired modifications to tetracyclines
Knowledge of the acidꢀstable viridicatumtoxin scaffold informs our ability to increase the acid
stability of the tetracycline antibiotics, which are easily degraded into the nonꢀactive analogues
when taken orally. Possible modifications to the tetracycline framework includes deamination at
Cꢀ4, oxidation at Cꢀ4a and Cꢀ5, dehydration across Cꢀ5a and Cꢀ6, alkylation at Cꢀ6 and Cꢀ7, and
oxidation at Cꢀ8 (Figure 8).
ACS Paragon Plus Environment
13

The Journal of Organic Chemistry
Page 14 of 26
1
2
3
4
5
6
7
8
9
In conclusion, our investigations into the chemistry of the marine molluskꢀderived
Paecilomyces sp. (CMBꢀMF010) have greatly expanded knowledge of the viridicatumtoxins, a
rare family of mycotoxins closely related to the tetracycline class of antibiotics. In the course of
our studies we documented the exceptional acid stability of the viridicatumtoxins, with coꢀ
metabolite inspired structure activity relationship investigations revealing the 5ꢀoxo analogue,
viridicatumtoxin B (2), as a potent vancomycinꢀresistant Enterococci (VRE) antibiotic.
Knowledge acquired during our investigations has the potential to inform the redesign of the
tetracycline framework, enhancing acid stability and oral bioavailability, while simultaneously
optimizing for and extending antibiotic efficacy to include therapeutically challenging pathogens
such as VRE.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental Section
General Experimental Details. Specific optical rotations ([α]D) were acquired on a
polarimeter in a 100 × 2 mm cell at room temperature. UVꢀvisible spectra were obtained on a
UVꢀvisible spectrophotometer with 1 cm quartz cells. Circular dichroism (CD) spectra were
recorded on a spectropolarimeter in 1 mm quartz cells at room temperature with scanning speed
of 50 nm/min and 0.5 nm step scan over the wavelength from 200–600 nm. Nuclear magnetic
resonance (NMR) spectra were acquired on a 600 MHz spectrometer with either a 5 mm PASEL
1H/Dꢀ13C ZꢀGradient probe or 5 mm CPTCI 1H/19Fꢀ13C/15N/DZꢀGradient cryoprobe. In all
cases spectra were acquired at 25 °C (unless otherwise specified) in solvents as specified in the
1
text, with referencing to residual H or 13C signals in the deuterated solvents. Electrospray
ionization mass spectrometry (ESIMS) experiments were carried out on a LC/MSD (quadrupole)
instrument in both positive and negative modes. Highꢀresolution ESIMS spectra were obtained
on a micrOTOF mass spectrometer either by direct injection in MeCN at 3 ꢁL/min using sodium
ACS Paragon Plus Environment
14

Page 15 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
formate clusters as an internal calibrant or by hyphenating with LC system comprising of
corresponding pump, column compartment, autosampler, photodiode array (PDA) detector.
7
8
9
Collection and Isolation of Paecilomyces sp. (CMB-MF010). The fungus CMBꢀMF010 was
isolated in 2012 from the inner tissue of a marine pulmonate false limpet Siphonaria sp.
collected at the rocky intertidal zone of Moora Park, Shorncliffe, Queensland. The fresh
Siphonaria sample was transported in a sterile tube (50 mL) on ice to the laboratory, where it
was rinsed in sterile natural seawater for 1 min and subjected to surface sterilization in 70%
ethanol (v/v) for 30 s, after which it was washed with sterile seawater to remove traces of EtOH.
Subsequently, the sample was dissected under aseptic conditions and the inner tissue placed on
PYG agar plates (comprising 2% glucose, 1% peptone, 0.5% yeast extract, 0.02%
chloramphenicol and 1.5% agar in 50% artificial seawater (1.7% salinity)). The plates were
wrapped in parafilm and incubated at 26.5 °C for 3–4 weeks. A pure culture of fungus CMBꢀ
MF010 was obtained by singleꢀcolony serial transfer on agar plates and then cryopreserved at –
80 °C in 15% aqueous glycerol. Taxonomic analysis identified CMBꢀMF010 as a Paecilomyces
sp. (see ESI).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Analytical Cultivation and Chemical Profiling of Paecilomyces sp. (CMB-MF010).
Paecilomyces sp. (CMBꢀMF010) was screened in PSB broth (1% sucrose, 1% mannitol, 0.5%
peptone, 0.3% yeast extract and 20% potato juice in water; 200 rpm at 26.5 °C for 7 d in microꢀ
bioreactor), on PYG agar plates (2% glucose, 1% peptone, 0.5% yeast extract and 1.5% agar;
26.5 °C for 25 d) with and without natural sea salts and on rice solid medium (70 g rice, 0.3%
peptone, 0.3% yeast extract, 0.1% monosodium glutamate and 1.7 g natural sea salt prepared in
100 mL distilled water; 26 °C for 25 d). After cultivation, the broth and solid matrix were
ACS Paragon Plus Environment
15

The Journal of Organic Chemistry
Page 16 of 26
1
2
3
4
5
6
7
8
9
extracted exhaustedly with EtOAc and the organic phase was dried in vacuo to yield crude
extracts. Analytes of crude extracts (5 mg/mL) were prepared in MeOH for HPLCꢀDADꢀ
ESI(±)MS analysis by standard gradient elution (Zorbax SBꢀC8 column, 150 × 4.6 mm , 5 ꢁm, 1
mL/min gradient elution from 90% H2O/MeCN to 100% MeCN with 0.05% formic acid in
MeCN as modifier over 15 min). The fungus CMBꢀMF010 produced different secondary
metabolites when grown on PYG agar and rice solid media (Figure S1).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Preparative Cultivation and Fractionation of Paecilomyces sp. (CMB-MF010). A single
colony of Paecilomyces sp. (CMBꢀMF010) was subꢀsampled on PYG agar plates (×100) (2%
glucose, 1% peptone and 0.5% yeast extract, prepared in distilled water) and incubated at 26.5°C
for 25 days. The resulting agar plates were harvested, diced (~1 cm3) and extracted with EtOAc
until the organic phase was almost colorless. The combined organic phase was concentrated in
vacuo to yield a crude extract (1173 mg) which was sequentially triturated to afford hexane
(153.9 mg), CH2Cl2 (700.0 mg) and MeOH (144.0 mg) soluble partitions. The CH2Cl2 solubles
were subjected to SPE chromatography (GracePureTM C18ꢀMax with a 10% stepwise gradient
elution from 70% H2O/MeOH to MeOH) and the peaks of interest (HPLCꢀDADꢀMS)
concentrated in two fractions. The first fraction (38.7 mg) was fractionated with HPLC (Zorbax
SBꢀC8 column, 250 × 9.4 mm, 5 ꢂm, 3 mL/min gradient elution from 65% to 48% H2O/MeCN
over 5 min, followed by 48% to 43% H2O/MeCN over 10 min, with a constant 0.01%
TFA/MeCN modifier) to yield viridicatumtoxin D (4) (tR = 12.4 min; 8.2 mg, 0.7%),
viridicatumtoxin E (5) (tR = 11.8 min; 7.5 mg, 0.6%) and viridicatumtoxin F (6) (tR = 11.2 min;
2.6 mg, 0.2%). The second fraction (96.3 mg) was fractionated by HPLC (Zorbax SBꢀC8 column,
250 × 9.4 mm, 5 ꢂm, 3 mL/min gradient elution from 65% to 48% H2O/MeCN over 5 min,
followed by 48% to 43% H2O/MeCN over 10 min, and from 43% to 30% H2O/MeCN over 5
ACS Paragon Plus Environment
16

Page 17 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
min, with a constant 0.01% TFA/MeCN modifier) to afford viridicatumtoxin A (1) (tR = 18.9
min; 14.1 mg, 1.2%) and viridicatumtoxin B (2) (tR = 20.3 min; 0.6 mg, 0.05%).
7
8
9
For cultivation on rice solid media, a small sample (~1 cm3) of CMBꢀMF010 colony on agar
was used to inoculate a 1 L Erlenmeyer flask containing sterile rice medium, and the mixture
was incubated at 26 °C for 25 days. The fungal mycelia as well as the rice media were
exhaustively extracted with EtOAc and concentrated in vacuo to obtain a combined EtOAc
extract (673.0 mg), which was sequentially triturated to yield hexane (375.9 mg), CH2Cl2 (161.1
mg) and MeOH (22.1 mg) solubles, respectively. The CH2Cl2 solubles (161.1 mg) were
subjected to C18 SPE fractionation (90% H2O/MeOH to MeOH) to yield 10 fractions. After
HPLCꢀDADꢀMS analysis the 50%, 40% and 30% H2O/MeOH fractions were combined (18.4
mg) and further fractionated by HPLC (Zorbax SBꢀC8 column, 250 × 9.4 mm, 5 ꢂm, 3 mL/min
isocratic elution at 45% H2O/MeCN over 20 min with a constant 0.01% TFA/MeCN modifier) to
afford viridicatumtoxin A (1) (tR = 13.9 min; 7.4 mg, 1.1%) and viridicatumtoxin C (3) (tR = 10.5
min; 4.0 mg, 0.6%). After freezeꢀdrying the slightly impure viridicatumtoxin C (3) sample was
reꢀpurified by HPLC (same elution condition as above) to yield pure viridicatumtoxin C (3) (tR =
10.5 min; 2.9 mg, 0.4%) and pure orthoꢀquinone viridicatumtoxin C (3a) (tR = 9.8 min; 0.6 mg).
The 10% H2O/MeOH fraction (53.0 mg) was further fractionated by HPLC (Zorbax SBꢀC3
column, 250 × 9.4 mm, 5 ꢂm, 3 mL/min gradient elution from 50% to 35% H2O/MeCN over 15
min followed by a 5 min hold at 100% MeCN, with a constant 0.01% TFA/MeCN modifier) to
yield spirohexaline (7) (tR = 15.4 min; 0.3 mg, 0.04%) and previridicatumtoxin (8) (tR = 15.9
min; 0.4 mg, 0.06%).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Note: % yields for compounds 1–8 are calculated as weightꢀtoꢀweight estimate against the crude
extract (1172.7 mg for agar plates and 673.0 mg for rice solid media).
ACS Paragon Plus Environment
17

The Journal of Organic Chemistry
Page 18 of 26
1
2
3
4
Characterization of Paecilomyces sp. (CMB-MF010) Metabolites. Viridicatumtoxin A (1).
5
6
7
22
Bright yellow powder; [α]D +31.8 (c 0.2, EtOH); UV (MeOH) λmax (log ε) 236 (4.44), 282
8
9
(4.53), 424 (3.94) nm; CD (MeOH) λmax (ꢀε) 237 (–14.4), 287 (+29.4), 346.5 (–3.0), 361.5 (–
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0.1), 422.5 (–2.7) nm; NMR (600 MHz, CDCl3) see Tables 2 and S6; ESI(+)MS m/z 548 [M –
H2O + H]+, 588 [M + Na]+, ESI(−)MS m/z 564 [M − H]–; HRESI(+)MS m/z 588.1844 [M + Na]+
(calcd for C30H31NO10Na+, 588.1840).
22
Viridicatumtoxin B (2). Bright yellow powder; [α]D +43.3 (c 0.05, EtOH); UV (MeOH) λmax
(log ε) 255 (4.40), 285 (4.35, sh), 431 (3.94), 455 (3.99) nm; CD (MeOH) λmax (ꢀε) 206.5 (–5.4),
227.5 (+4.7), 256 (–1.8), 276.5 (+17.9), 328 (–1.4), 351 (–0.3), 385 (–1.5), 421 (+0.9), 434 (–
0.7), 444.5 (–0.1), 459.5 (–4.2) nm; NMR (600 MHz, CDCl3) see Tables 2 and S7; ESI(+)MS
m/z 564 [M + H]+, 586 [M + Na]+, ESI(−)MS m/z 562 [M − H]–; HRESI(–)MS m/z 562.1721 [M
– H]– (calcd for C30H28NO10, 562.1719).
Viridicatumtoxin C (3). Yellow powder; [α]D22 +41.1 (c 0.08, EtOH); UV (MeOH) λmax (log ε)
257 (4.42), 303 (4.40), 448 (4.04) nm; CD (MeOH) λmax (ꢀε) 226 (+6.1), 251 (–6.3), 271 (+7.9),
298 (+0.2), 321.5 (+2.5) nm; NMR (600 MHz, CDCl3) see Tables 1–2 and S8; ESI(+)MS m/z
580 [M + H]+, 602 [M + Na]+, ESI(−)MS m/z 578 [M − H]–; HRESI(+)MS m/z 602.1634 [M +
Na]+ (calcd for C30H29NO11Na+, 602.1633).
Viridicatumtoxin D (4). Brown amorphous powder; [α]D22 +21.7 (c 0.06, EtOH); UV (MeOH)
λmax (log ε) 263 (4.64), 344 (3.89), 396 (3.98) nm; CD (MeOH) λmax (ꢀε) 206 (–2.0), 225.5 (–
10.7), 260 (+38.4), 279 (–6.7), 308 (–0.9), 378.5 (–2.7) nm; NMR (600 MHz, CDCl3/CD3OD =
4:1) see Tables 1–2 and S9; ESI(+)MS m/z 566 [M + H]+, ESI(−)MS m/z 564 [M − H]–;
ACS Paragon Plus Environment
18

Page 19 of 26
The Journal of Organic Chemistry
1
2
HRESI(+)MS m/z 588.1838 [M + Na]+ (calcd for C30H31NO10Na+, 588.1840).
3
4
5
6
7
8
9
Viridicatumtoxin D tetratacetate (4a). A solution of 4 (2 mg, 3.54 ꢂmol) in pyridine (250 ꢂL)
and acetic anhydride (200 ꢂL) was stirred at room temperature for 30 h, after which it was dried
under N2 and the crude product subjected to HPLC fractionation (Zorbax SBꢀC8 column, 250 ×
9.4 mm, 5 ꢂm, 3 mL/min gradient elution from 35% to 20% H2O/MeCN over 15 min, with an
isocratic 0.01% TFA/MeCN modifier) to yield the tetraacetate 4a (tR = 9.92 min; 1.5 mg, 75%)
as light brown powder; UV (MeOH) λmax (log ε) 268 (4.75), 347 (3.91) nm; 1H NMR (600 MHz,
CDCl3/CD3OD = 4:1) δH 7.07 (1H, s, Hꢀ9), 5.74 (1H, s, Hꢀ12a), 5.58 (1H, s, Hꢀ5), 5.46 (1H, br s,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hꢀ17), 3.86 (3H, s, 8ꢀOCH3), 3.43 (1H, d, J = 18.6 Hz, Hꢀ14α), 3.25 (1H, d, J = 18.6 Hz, Hꢀ
14β), 2.58 (1H, d, J = 17.0 Hz, Hꢀ4α), 2.45 (3H, s, COCH3), 2.41 (3H, s, COCH3), 2.33 (3H, s,
COCH3), 2.30 (3H, s, COCH3), 2.19 (1H, m, Hꢀ18β), 2.05 (1H, d, J = 17.0 Hz, Hꢀ4β), 1.97 (1H,
br d, J = 18.2 Hz, Hꢀ18α), 1.80 (1H, ddd, J = 18.4, 12.6, 6.2 Hz, Hꢀ19α), 1.52 (3H, s, Hꢀ21),
1.31 (1H, dd, J = 13.4, 6.2 Hz, Hꢀ19β), 0.89 (3H, s, Hꢀ22), 0.33 (3H, s, Hꢀ23); ESI(+)MS m/z
734 [M + H]+, 756 [M + Na]+, ESI(–)MS m/z 732 [M – H]–; HRESI(+)MS m/z 756.2263 [M +
Na]+ (calcd for C38H39NO14Na+, 756.2263).
Viridicatumtoxin E (5). Brown amorphous powder; [α]D22 +272.2 (c 0.05, EtOH); UV (MeOH)
λmax (log ε) 263 (4.64), 344 (3.91), 383 (3.99) nm; CD (MeOH) λmax (ꢀε) 211.5 (+0.9), 225.5 (–
7.5), 266 (+40.6), 326 (–2.7), 360.5 (+2.1) nm; NMR (600 MHz, CDCl3/CD3OD = 4:1) see
Tables 1–2 and S10; ESI(+)MS m/z 566 [M + H]+, 588 [M + Na]+, ESI(−)MS m/z 564 [M − H]–;
HRESI(+)MS m/z 588.1840 [M + Na]+ (calcd for C30H31NO10Na+, 588.1840).
Viridicatumtoxin F (6). Brown amorphous powder; [α]D22 +146.2 (c 0.06, EtOH); UV (MeOH)
ACS Paragon Plus Environment
19

The Journal of Organic Chemistry
Page 20 of 26
1
2
3
4
λmax (log ε) 263 (4.55), 347 (3.85), 376 (3.90) nm; CD (MeOH) λmax (ꢀε) 213.5 (+5.9), 247 (–
5
5.4), 265 (+43.9), 328.5 (–1.3), 359 (+0.7) nm; NMR (600 MHz, CDCl3/CD3OD = 4:1) see
Tables 1–2 and S12; ESI(+)MS m/z 580 [M + H]+, ESI(−)MS m/z 578 [M − H]–, 534 [M − CO2 −
H]+; HRESI(+)MS m/z 602.1631 [M + Na]+ (calcd for C30H29NO11Na+, 602.1633).
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
Spirohexaline (7). Bright yellow powder; [α]D +15.5 (c 0.03, EtOH); UV (MeOH) λmax (log
ε) 236 (4.15), 283 (4.34), 426 (3.69) nm; NMR (600 MHz, CDCl3) see Tables 2 and S13;
ESI(+)MS m/z 547 [M − H2O + H]+, 587 [M + Na]+, ESI(−)MS m/z 563 [M − H]–; HRESI(−)MS
m/z 563.1922 [M − H]– (calcd for C31H31O10, 563.1923).
Previridicatumtoxin (8). Bright yellow powder; [α]D22 +98.8 (c 0.04, EtOH); UV (MeOH) λmax
(log ε) 241 (4.50), 273 (4.65), 415 (4.15) nm; NMR (600 MHz, CDCl3) see Tables 2 and S14;
ESI(+)MS m/z 550 [M − H2O + H]+, 568 [M + H]+, 590 [M + Na]+, ESI(−)MS m/z 566 [M − H]–;
HRESI(−)MS m/z 566.2035 [M − H]– (calcd for C30H32NO10, 566.2032).
Oxidation and Derivatization of Viridicatumtoxin C (3). A mixture sample of 3 (65%, w/w)
and 3a (35%, w/w) in MeOH (40 ꢂL) was heated at 40 °C overnight, and the product dried under
N2, reꢀdissolved in MeOH (20 ꢂL) and subjected to HPLCꢀDADꢀMS analysis (Zorbax SBꢀC8
column, 150 × 4.6 mm, 5 ꢁm, 1 mL/min gradient elution from 90% H2O/MeCN to 100% MeCN
with 0.05% formic acid as modifier over 15 min) to detect the significant transformation of
viridicatumtoxin C (3) to orthoꢀquinone viridicatumtoxin C (3a) (90%, w/w); UV (MeOH) λmax
(log ε) 265 (4.48), 468 (3.65), 591 (3.72) nm; ESI(+)MS m/z 578 [M + H]+, ESI(−)MS m/z 576
[M − H]–; HRESI(+)MS m/z 600.1481 [M + Na]+ (calcd for C30H27NO11Na+, 600.1476). To
confirm the structure, 3a was treated with 1,2ꢀphenylenediamine (PLDA) and analyzed by
ACS Paragon Plus Environment
20

Page 21 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
HPLCꢀDADꢀHRMS (GeminiꢀNX C18 column, 150 × 2.0 mm, 5 µm, 250 ꢂL/min gradient elution
from 90% H2O/MeCN to 100% MeCN (with isocratic 0.05% formic acid as modifier) over 10
min at 35°C, then hold for 5 min and equilibrate for 4 min; 100 ꢁL/h injection of sodium formate
clusters as an internal calibrant) to detect 95% conversion to the adduct 3b; C36H31N3O9 m/z
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
+
650.2098 [M + H]+ (calcd for C36H32N3O9 , 650.2133) (Figure S3).
Acid Stability of Viridicatumtoxin A (1) and Oxytetracycline (9). Experiment 1: Individual
aliquots of 1 (100 ꢂg) and 9 (100 ꢂg) were treated with 0.1 M HCl (200 ꢂL, pH = 1) at 75 °C,
with sampling (6 ꢂL) at 1, 2, 3, 4, 7 and 10 h intervals. Individual analyte samples were analysed
by HPLCꢀDADꢀESI(±)MS (Zorbax SBꢀC8 column, 150 × 4.6 mm, 5 ꢁm, 1 mL/min gradient
elution from 90% H2O/MeCN to 100% MeCN with 0.05% formic acid as modifier over 15 min)
(Figures S6 and S7). Experiment 2: For testing the stability of oxytetracycline (9) in simulated
gastric acid condition an aliquot of 9 (100 ꢂg) was treated with 0.01 M HCl (200 ꢂL, pH = 2) at
37 °C, with sampling (6 ꢂL) at 1, 2, 4, 6, 10 and 20 h intervals. Individual analyte samples were
analysed by HPLCꢀDADꢀESI(±)MS (Zorbax SBꢀC8 column, 150 × 4.6 mm, 5 ꢁm, 1 mL/min
gradient elution from 90% H2O/MeCN to 100% MeCN with 0.05% formic acid as modifier over
15 min) (Figure S8). Experiment 3: For testing the stability of viridicatumtoxin A (1) under
harsher acidic conditions an aliquot of 1 (0.8 mg) dissolved in 5% (w/v) pꢀTsOH toluene (200
ꢂL) was treated with molecular sieves to remove residual water. The filtered solution was then
heated to 100 °C with sampling (5 ꢂL) at 1, 2, 3, 4, 6, 8, 10, 12, 22, 27, 32 and 48 h intervals.
Individual analyte samples were dried, redissolved in MeOH and subjected to HPLCꢀDADꢀ
ESI(±)MS analysis (Zorbax SBꢀC8 column, 150 × 4.6 mm column, 5 ꢂm, 1 mL/min gradient
elution from 90% H2O/MeCN to 100% MeCN over 15 min, with constant 0.05% formic acid
modifier) (Figure S9).
ACS Paragon Plus Environment
21

The Journal of Organic Chemistry
Page 22 of 26
1
2
3
4
5
6
7
8
9
Purification of Oxytetracycline and Viridicatumtoxin A Acid Degradation Products (11–
15). A sample of 9 (150 mg) was dissolved in 0.1 M HCl (1.5 mL) and stirred at 75 °C for 2 h,
after which the solution was dried in vacuo to yield a crude product (157.2 mg) that was
fractionated by preparative HPLC (Luna C18 column, 250 × 21.2 mm, 10 ꢂm, 20 mL/min
gradient elution from 90% to 15% H2O/MeCN over 20 min, with an isocratic 0.01% TFA
modifier) to afford anhydroꢀoxytetracycline (10) (tR = 9.9 min; 8.5 mg, 5.4%), αꢀapoꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
oxytetracycline (11) (tR = 7.5 min; 36.4 mg, 23.2%), βꢀapoꢀoxytetracycline (12) (tR = 10.9 min;
12.1 mg, 7.7%) and terrinolide (13) (tR = 12.8 min; 52.1 mg, 33.1%). Note: anhydroꢀ
oxytetracycline (10) could not be obtained in pure form as it was unstable to the HPLC acidified
mobile phase (0.01% TFA), undergoing facile conversion to 11–13.
A sample of 1 (45 mg) dissolved in 5% (w/v) pꢀTsOH toluene (3 mL) was treated with
molecular sieves to remove residual water, and the filtered reaction mixture was heated at 100 °C
for 20 h, after which it was dried in vacuo and fractionated by preparative HPLC (Luna C18
column, 250 × 21.2 mm, 10 ꢂm, 20 mL/min gradient elution from 35% H2O/MeCN to 100%
MeCN over 20 min, with an isocratic 0.01% TFA modifier) to afford 14 (tR = 16.7 min; 0.8 mg,
1.8%) and 15 (tR = 17.7 min; 0.6 mg, 1.3%).
Characterization of Oxytetracycline and Viridicatumtoxin A Acid Degradation Products
22
(11–15). Oxytetracycline (9). Pale yellow powder; [α]D –119.5 (c 0.2, MeOH); UV (MeOH)
λmax (log ε) 266 (3.99), 360 (3.86) nm; NMR (600 MHz, DMSOꢀd6) see Table S16; ESI(+)MS
m/z 461 [M + H]+, ESI(−)MS m/z 459 [M − H]–; HRESI(+)MS m/z 461.1560 [M + H]+ (calcd for
+
C22H25N2O9 , 461.1555).
22
α
ꢀApoꢀoxytetracycline (11). Pale yellow powder; [α]D +61.6 (c 0.2, MeOH); UV (MeOH)
ACS Paragon Plus Environment
22

Page 23 of 26
The Journal of Organic Chemistry
1
2
3
4
5
6
7
8
9
λmax (log ε) 250 (4.53), 332 (3.78), 381 (3.92), 397 (3.93) nm; NMR (600 MHz, DMSOꢀd6) see
Table S17; ESI(+)MS m/z 443 [M + H]+, ESI(−)MS m/z 441 [M − H]–; HRESI(+)MS m/z
+
443.1454 [M + H]+ (calcd for C22H23N2O8 , 443.1449).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
22
β
ꢀApoꢀoxytetracycline (12). Pale yellow powder; [α]D –70.0 (c 0.2, MeOH); UV (MeOH)
λmax (log ε) 244 (4.46), 332 (3.74), 394 (3.85) nm; NMR (600 MHz, DMSOꢀd6) see Table S18;
ESI(+)MS m/z 443 [M + H]+, ESI(−)MS m/z 441 [M − H]–; HRESI(+)MS m/z 443.1454 [M +
+
H]+ (calcd for C22H23N2O8 , 443.1449).
22
Terrinolide (13). Pale yellow powder; [α]D –29.3 (c 0.2, MeOH); UV (MeOH) λmax (log ε)
249 (4.53), 332 (3.91), 380 (3.98), 397 (3.98) nm; NMR (600 MHz, DMSOꢀd6) see Table S19;
ESI(+)MS m/z 398 [M + H]+, ESI(−)MS m/z 396 [M − H]–; HRESI(−)MS m/z 396.0721 [M −
H]– (calcd for C20H14NO8 , 396.0719).
–
22
14. Yellow amorphous powder; [α]D +108.0 (c 0.04, EtOH); UV (MeOH) λmax (log ε) 251
1
(4.69), 291 (4.16), 303 (4.22), 394 (4.12) nm; H NMR (600 MHz, CDCl3) δH 13.03 (1H, s, 12ꢀ
OH), 10.96 (1H, s, 10ꢀOH), 7.33 (1H, s, Hꢀ5), 7.05 (1H, s, Hꢀ14), 6.38 (1H, s, Hꢀ9), 6.05 (1H, br
s, 12aꢀOH), 5.73 (1H, m, Hꢀ17), 3.87 (3H, s, 8ꢀOCH3), 3.80 (2H, d, J = 3.5, Hꢀ4), 3.75 (3H, s, 4ꢀ
COOCH3), 2.26 (2H, m, Hꢀ18), 1.91 (1H, dt, J = 13.2, 6.2 Hz, Hꢀ19α), 1.63 (1H, dt, J = 13.2, 6.8
13
Hz, Hꢀ19β), 1.11 (3H, d, J =1.5 Hz, Hꢀ21), 0.87 (3H, s, Hꢀ22), 0.85 (3H, s, Hꢀ23); C NMR
(150 MHz, CDCl3) δC 191.1 (Cꢀ11), 171.5 (Cꢀ3), 163.0 (Cꢀ10), 162.0 (Cꢀ8), 149.4 (Cꢀ12), 149.3
(Cꢀ6), 144.3 (Cꢀ14), 143.4 (Cꢀ12a), 132.8 (Cꢀ5a), 131.0 (Cꢀ16), 126.2 (Cꢀ4a), 124.6 (Cꢀ7), 124.5
(Cꢀ17), 124.4 (Cꢀ6a), 117.5 (Cꢀ5), 114.7 (Cꢀ11a), 106.2 (Cꢀ10a), 97.2 (Cꢀ9), 68.3 (Cꢀ15), 55.8
(8ꢀOCH3), 52.5 (4ꢀCOOCH3), 37.0 (Cꢀ20), 35.9 (Cꢀ19), 35.9 (Cꢀ4), 27.0 (Cꢀ23), 26.9 (Cꢀ22),
23.2 (Cꢀ18), 19.7 (Cꢀ21); ESI(+)MS m/z 477 [M + H]+, ESI(−)MS m/z 475 [M − H]–;
ACS Paragon Plus Environment
23

The Journal of Organic Chemistry
Page 24 of 26
1
2
+
HRESI(+)MS m/z 477.1911 [M + H]+ (calcd for C28H29O7 , 477.1908).
3
4
5
6
7
8
22
15. Yellow amorphous powder; [α]D +125.3 (c 0.05, EtOH); UV (MeOH) λmax (log ε) 251
1
9
(4.74), 292 (4.13), 302 (4.16), 387 (4.15) nm; H NMR (600 MHz, CDCl3) δH 13.11 (1H, s, 12ꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
OH), 10.92 (1H, s, 10ꢀOH), 7.42 (1H, s, Hꢀ5), 7.16 (1H, s, Hꢀ14), 6.41 (1H, s, Hꢀ9), 5.76 (1H, m,
Hꢀ17), 3.89 (3H, s, 8ꢀOCH3), 3.86 (2H, br s, Hꢀ4), 2.27 (2H, m, Hꢀ18), 1.93 (1H, dt, J = 13.3, 6.2
Hz, Hꢀ19α), 1.62 (1H, dt, J = 13.3, 6.7, Hꢀ19β), 1.12 (3H, d, J = 1.5 Hz, Hꢀ21), 0.88 (3H, s, Hꢀ
13
22), 0.86 (3H, s, Hꢀ23); C NMR (150 MHz, CDCl3) δC 191.2 (Cꢀ11), 172.5 (Cꢀ3), 163.2 (Cꢀ
10), 162.2 (Cꢀ8), 148.5 (Cꢀ6), 147.5 (Cꢀ12), 146.5 (Cꢀ14), 142.0 (Cꢀ12a), 132.7 (Cꢀ5a), 130.5 (Cꢀ
16), 130.2 (Cꢀ4a), 129.6 (Cꢀ6a), 124.9 (Cꢀ17), 124.5 (Cꢀ7), 116.6 (Cꢀ11a), 111.0 (Cꢀ5), 106.1 (Cꢀ
10a), 97.5 (Cꢀ9), 68.5 (Cꢀ15), 55.8 (8ꢀOCH3), 37.1 (Cꢀ20), 35.9 (Cꢀ19), 33.9 (Cꢀ4), 27.0 (Cꢀ23),
26.9 (Cꢀ22), 23.1 (Cꢀ18), 19.7 (Cꢀ21); ESI(+)MS m/z 445 [M + H]+, ESI(−)MS m/z 443 [M −
–
H]–; HRESI(−)MS m/z 443.1505 [M − H]– (calcd for C27H23O6 , 443.1500).
X-ray Crystallographic Analysis of 1ꢀ2MeOHꢀ0.5H2O. The single crystal of
1ꢀ2MeOHꢀ0.5H2O was obtained from MeOH by slow evaporation at r.t. Data were collected at
190 K using an Oxford Diffraction Gemini CCD diffractometer with Cu Kα radiation and the
crystal was cooled with an Oxford Cryosystems Desktop Cooler. Data reduction was performed
with the CrysAllisPro program (Oxford Diffraction vers. 171.34.40). The structure was solved by
direct methods with SHELXS86 and refined with SHELX97.15 The thermal ellipsoid diagram
was produced with ORTEP316 and all calculations were performed within the WinGX package.17
C30H31NO10ꢀ2(CH3OH)ꢀ0.5H2O, M = 638.65, monoclinic, a = 30.1333(7), b = 7.7644(2), c =
12.9297(3) Å, V = 2912.26(12) Å3, T = 190(2) K, space group C2, Z = 4, 16514 reflections
measured, 4539 unique (Rint = 0.0413) which were used in all calculations. The final R(obs. data)
ACS Paragon Plus Environment
24

Page 25 of 26
The Journal of Organic Chemistry
1
2
3
4
was 0.0466, goodness of fit 1.069. CCDC number 1429664. The absolute structure was
5
6
confirmed by the methodology of Hooft et al.18
7
8
9
Associated Content
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information
1H and 13C NMR spectra and tabulated 1D and 2D NMR data for 1−9, 11−15, taxonomy of
Paecilomyces sp., optical rotation study of 1, acid and heat stability of 1, 3 and 9, HPLC
chromatograms, bioassay data and Xꢀray crystallography of 1. This material is available free of
charge via the Internet at http://pubs.acs.org/.
Author Information
Corresponding Author
*Eꢀmail: r.capon@uq.edu.au
Notes
The authors declare no competing financial interest.
Acknowledgments
This research was funded in part by the Institute for Molecular Bioscience, UQ and the
Australian Research Council (DP120100183). Z. S. acknowledges UQ for a Postgraduate
Research Scholarship. We acknowledge E. Lacey (BioAustralis) for a sample of viridicatumtoxin
A. D. Paterson & H. Zowawi (UQ Centre for Clinical Research) for multidrugꢀresistant bacteria
isolates, X. Xiao (IMB) for advice with synthetic chemistry, and A. Jarrad (IMB) for bioassay
support.
References
ACS Paragon Plus Environment
25

The Journal of Organic Chemistry
Page 26 of 26
1
2
3
4
5
6
(1) Hutchison, R. D.; Steyn P. S.; Van Rensburg, S. J. Toxicol. Appl. Pharmacol. 1973
507–509.
, 24,
7
8
9
(2) Zheng, C. J.; Yu, H. E.; Kim, E. H.; Kim, W. G. J. Antibiot. 2008, 61, 633–637.
(3) Nicolaou, K. C.; Nilewski, C.; Hale, C. R.; Ioannidou, H. A.; ElMarrouni A.; Koch, L. G.
Angew. Chem. Int. Ed. Engl. 2013, 52, 8736–8741.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(4) Nicolaou, K. C.; Hale, C. R.; Nilewski, C.; Ioannidou, H. A.; ElMarrouni, A.; Nilewski, L.
G.; Beabout, K.; Wang, T. T.; Shamoo, Y. J. Am. Chem. Soc. 2014, 136, 12137–12160.
(5) Inokoshi, J.; Nakamura, Y.; Hongbin, Z.; Uchida, R.; Nonaka, K.; Masuma R.; Tomoda, H. J.
Antibiot. 2013, 66, 37–41.
(6) Chooi, Y. H.; Wang, P.; Fang, J.; Li, Y.; Wu, K.; Wang P.; Tang, Y. J. Am. Chem. Soc. 2012,
134, 9428–9437.
(7) Chooi, Y. H.; Hong, Y. J.; Cacho, R. A.; Tantillo D. J.; Tang, Y. J. Am. Chem. Soc. 2013,
135, 16805–16808.
(8) Kabuto, C.; Silverton, J. V.; Akiyama, T.; Sankawa, U.; Hutchison, R. D.; Steyn P. S.;
Vleggaar, R. J. Chem. Soc. Chem. Commun. 1976, 728–729.
(9) Silverton, J. V.; Kabuto C.; Akiyama, T. Acta Cryst. 1982, B38, 3032–3037.
(10) De Jesus, A. E.; Hull, W. E.; Steyn, P. S.; van Heerden, F. R.; Vleggaar, R. J. Chem. Soc.
Chem. Commun. 1982, 902–904.
(11) Horak, R. M.; Maharaj, V. J.; Marais, S. F.; van Heerden, F. R.; Vleggaar, R. J. Chem. Soc.
Chem. Commun. 1988, 1562–1564.
(12) Chooi, Y. H.; Cacho R.; Tang, Y. Chem. Biol. 2010, 17, 483–494.
(13) Casy A. F.; Yasin, A. Magn. Reson. Chem. 1985, 23, 767–770.
(14) Bendele, A. M.; Carlton, W. W.; Nelson G. E.; Peterson, R. E.; Grove, M. D. Toxicol. Lett.
1984, 22, 287–291.
(15) Sheldrick, G. M. Acta Crystallogr. A. 2008, 64, 112–122.
(16) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(17) Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837–838.
(18) Hooft, R. W.; Straver, L. H.; Spek, A. L. J. Appl. Crystallogr. 2008, 41, 96–103.
ACS Paragon Plus Environment
26