New Aspercryptins, Lipopeptide Natural Products, Revealed by HDAC Inhibition in Aspergillus nidulans

Article abstract of DOI:10.1021/acschembio.6b00398

Unlocking the biochemical stores of fungi is key for developing future pharmaceuticals. Through reduced expression of a critical histone deacetylase in Aspergillus nidulans, increases of up to 100-fold were observed in the levels of 15 new aspercryptins, recently described lipopeptides with two noncanonical amino acids derived from octanoic and dodecanoic acids. In addition to two NMR-verified structures, MS/MS networking helped uncover an additional 13 aspercryptins. The aspercryptins break the conventional structural orientation of lipopeptides and appear "backward" when compared to known compounds of this class. We have also confirmed the 14-gene aspercryptin biosynthetic gene cluster, which encodes two fatty acid synthases and several enzymes to convert saturated octanoic and dodecanoic acid to α-amino acids.

Full text of DOI:10.1021/acschembio.6b00398

Subscriber access provided by UNIVERSITY OF LEEDS
Letter
New aspercryptins, lipopeptide natural products,
revealed by HDAC inhibition in Aspergillus nidulans
Matthew T Henke, Alexandra A. Soukup, Anthony Whitney Goering, Ryan
A. McClure, Regan J. Thomson, Nancy P. Keller, and Neil L. Kelleher
ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.6b00398 • Publication Date (Web): 16 Jun 2016
Downloaded from http://pubs.acs.org on June 19, 2016
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.
ACS Chemical Biology 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 10
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
New aspercryptins, lipopeptide natural products, revealed by
HDAC inhibition in Aspergillus nidulans
a
b
a
c
Matthew T. Henke , Alexandra A. Soukup , Anthony W. Goering , Ryan A. McClure , Regan J.
c
b,d,e
a,c,f*
Thomson , Nancy P. Keller and Neil L. Kelleher
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
a
c
f
Department of Molecular Biosciences, Department of Chemistry and Feinberg School of Medicine, Northwestern University,
Evanston, IL 60208, USA
b
d
e
Department of Genetics, Department of Bacteriology, and Department of Medical Microbiology and Immunology University
of Wisconsin, Madison, Wisconsin, 53706, USA
ABSTRACT: Unlocking the biochemical stores of fungi is key for developing future pharmaceuticals.
Through reduced expression of a critical histone deacetylase in Aspergillus nidulans, increases of up to
100ꢀfold were observed in the levels of 15 new aspercryptins, recently described lipopeptides with two
nonꢀcanonical amino acids derived from octanoic and dodecanoic acids. In addition to two NMRꢀverified
structures, MS/MS networking helped uncover an additional 13 aspercryptins. The aspercryptins break
the conventional structural orientation of lipopeptides and appear ‘backward’ when compared to known
compounds of this class. We have also confirmed the 14ꢀgene aspercryptin biosynthetic gene cluster,
which encodes two fatty acid synthases and several enzymes to convert saturated octanoic and dodecanoic
acid to αꢀamino acids.
INTRODUCTION
For decades the search for new natural products by screening for bioactivity has been hampered by
1
rediscovering the same compounds. To avoid the rediscovery of natural products, some laboratories have
decoupled discovery from screening for bioactivity by measuring accurate mass as the primary, highꢀ
throughput screen for the expression of new natural products. This new approach to molecular discovery
has been applied to bacteria but less so in fungi, which offer an enormous biosynthetic potential. The
genome of the mold Aspergillus nidulans contains more than 50 gene clusters annotated to be involved in
the biosynthesis of natural products. Yet just over 20 of these biosynthetic gene clusters (BGCs) have
2
3,4
5
6
associated natural products.
7
Several strategies have been developed to better harness the biosynthetic repertoire of fungi. One such
8ꢀ10
strategy involves the inhibition of histone deacetylase activity (HDACi) and has shown some promise.
HDACi increases global histone acetylation levels and can increase transcription of otherwise repressed
natural product BGCs. Our previous work used quantitative mass spectrometryꢀbased analyses to
compare the extracellular metabolomes of A. nidulans before and after both chemical and genetic
HDACi. We found that HDACi both upꢀregulated and downꢀregulated the expression of many
compounds. Among the upꢀregulated compounds were several not known to be produced by A. nidulans,
1
1
1
1
such as the fellutamides.
Continuing those efforts here, we have solved the structures of two new lipopeptides, aspercryptin A1
and A2, that are upꢀregulated by up to 90ꢀfold using the HDACiꢀbased strategy. The aspercryptins are sixꢀ
aminoꢀacid peptides containing two nonꢀcanonical αꢀamino acids derived from saturated C ꢀ and C ꢀfatty
acids, and a Cꢀterminal alcohol and related to two previous aspercryptins published by Chiang et al. in
2
8
12
12
016. Unlike the overwhelming majority of lipopeptides in the literature, the aspercryptins appear
backward’ and have their lipid tail at the Cꢀterminus. We also map the BGC responsible for the
‘
aspercryptins by genetic disruption and Northern blotting. It features a NRPS backbone gene encoding a
terminal reductase, atypical activation domains for unusual “fatty amino acid” substrates, and enzymes
putatively involved in fatty amino acid biogenesis. The aspercryptins join a small group of lipopeptide
natural products where incorporation of the lipid moiety occurs at sites other than the Nꢀterminus.
1
ACS Paragon Plus Environment

ACS Chemical Biology
Page 2 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
RESULTS AND DISCUSSION
Comparative metabolomics of HDAC-deficient A. nidulans reveals the aspercryptins. In previous
work, we generated a mutant of A. nidulans with constitutively lower levels of the HDAC RpdA
1
1
KD
(AN4493). Using differential metabolomics on this strain (rpdA , RAAS58.4) compared with wildtype
KD
fungus (Figure 1, left)), we detected many signals that were upꢀregulated in the rpdA mutant (Figure 1,
middle panels). Dereplication of these metabolites based on their accurate masses allowed us to cull the
list to only those that have not yet been characterized (Figure 1).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Several dozen of the putatively new metabolites displayed a high degree of similarity in their MS/MS
fragmentation spectra, suggesting structural and therefore biosynthetic relatedness (Figure 2). For the
KD
members of this compound family, all were upꢀregulated over a wide range from 2ꢀ to 130-fold in rpdA
vs. wildtype (data for one compound are shown in Figure 1, third panel). The structures of two of these
+
compounds, aspercryptin A1 (m/z 758.5386 [M+H] , C H N O , ꢀ0.02 ppm) and aspercryptin A2 (m/z
3
7
71
7
9
+
742.5440 [M+H] , C H N O , 0.45 ppm), (Figure 1, right panel) were determined completely by MS
37
71
7
8
with stable isotope feeding and extensive NMR (Supporting Information Table S1).
Figure 1. Workflow to compare the metabolomes of wildtype Aspergillus nidulans and a strain where the HDAC RpdA is
KD
constitutively repressed (rpdA ) leading to hyperacetylation in bulk chromatin. Knockdown of RpdA leads to many newly
observed metabolites (second panel), from which new metabolites can be viewed selectively (third panel) and targeted for
structure determination when they display unique mass signatures. Aspercryptins A1 and A2 are lipopeptides made by A.
nidulans (fourth panel at far right). Full NMR data can be found in Supporting Information. Stable isotope incorporation
experiments verify much of these structures (Supporting Information Figure S2). Annotated MS/MS spectra for both aspercryptin
A1 and A2 (Supporting Information Figures S3 and S4) are included for reference.
The aspercryptins are ‘backward’ lipopeptides. The aspercryptins are linear lipopeptides built from six
amino acids and appear to be the first example of peptide natural products with two lipid groups. They
also seem to be the only known lipopeptides with a lipid tail at the Cꢀterminus; thus they appear
‘backward’ to other lipopeptides. This is despite the overwhelming literature precedent for lipopeptides
having an Nꢀterminal lipid group.
Aspercryptin A1 and A2 differ from each other only at the Nꢀterminal residue ꢀ serine for aspercryptin
A1 and alanine for aspercryptin A2. The most striking feature of the aspercryptins is that of the six amino
acids, two are highly unusual and nonꢀproteogenic: 2ꢀaminoꢀoctanoic acid and 2ꢀaminoꢀdodecanol.
Stereochemical assignment of the aspercryptins was determined for each residue by derivatization with
Marfey’s reagent and comparison with standards using HPLCꢀMS (Supporting Information Figure S1).
For aspercryptin A1 the first two residues (serine and threonine) are epimerized to Dꢀserine and Dꢀalloꢀ
threonine while the remaining 4 residues are the Lꢀisomers. The stereocenters of aspercryptin A2 are the
same as those in aspercryptin A1; however, there is also an epimer of aspercryptin A2 such that the
alanine is the Lꢀisomer, thus we name this analogue epiꢀaspercryptin A2 (Supporting Information Figure
S1). Several of these monomers (threonine, isoleucine and serine) were also confirmed by metabolic
feeding of stable isotope analogues (Supporting Information Figure S2). Furthermore, stableꢀisotope
labeling shows that when fed d ꢀserine, labeled aspercryptin A1 only shows incorporation of 2 deuterons
3
(
Supporting Information Figure S2d). This further supports the epimerization of this residue by loss of the
deuteron on the alpha carbon.
MS/MS networking reveals a large family of aspercryptins. We observed that the MS/MS spectra of
KD
many other metabolites upꢀregulated by rpdA had striking similarity to those of aspercryptin A1 and
A2. We then turned to MS/MS networking to visualize the relatedness of these metabolites in the dataset.
2
ACS Paragon Plus Environment

Page 3 of 10
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
13
The ability for MS/MS networking to cluster biosynthetically related compounds is now established.
KD
When performing network analysis of the wildtype and rpdA extracts we saw several examples of
biosynthetically related natural products clustering together as expected; for example, the emericellamides
group tightly together based on MS/MS spectral similarity (Figure 2a, circled).
KD
One cluster of metabolites consistently upregulated in the rpdA extracts (Figure 2a, boxed)
contained aspercryptins A1 and A2 (Figure 2b). Using the MS/MS fragmentation patterns of aspercryptin
A1 and A2 (Supporting Information Figures S3 and S4) and the recently published aspercryptins B1 and
12
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
B3 as anchor points, we detected and have proposed putative structures for an additional 13
aspercryptins (Figure 2b, yellow circles, Table 1). These new aspercryptins fall into subꢀfamilies and we
KD
were readily able to quantify their abundance increases in the rpdA mutant (Supporting Information
Figure S5).
Figure 2. MS/MS networking to identify and characterize a large cluster of aspercryptins. (a) MS/MS networking clusters natural
products into molecular families. Here the metabolites made by wildtype A. nidulans (blue) are compared to those made by the
rpdA mutant (red). Some metabolites are expressed equally (grey). Some molecular families are seen in only one biological
state. The aspercryptins are boxed and the emericellamides are circled. (b) Aspercryptin A1 and A2 (larger yellow circles) fall
into an MS/MS cluster of metabolites mostly expressed by rpdA mutant. Comparing the MS/MS spectra of 5 NMRꢀelucidated
KD
KD
aspercryptins allows for characterization of an additional 13 aspercryptins in this molecular family for a total of 18 structurally
characterized aspercryptins, which includes aspercryptins B1 and B3 (yellow).
Table 1. The formulae and expression ratios for the 18 aspercryptins from Figure 2, including the five aspercryptins elucidated
by NMR (bold) and the 13 additional aspercryptins whose putative structures are supported by comparing MS/MS spectra.
NMR structures described in this work; NMR structures described by Chiang et al; ratio is an unresolved combination of
epimers
a
b
12 c
KD
KD
m/z
[M+H]+
molecular
formula
[rpdA
]
m/z
[M+H]+
molecular
formula
[rpdA
]
Name
Name
[wildtype]
[wildtype]
a
a
b
b
aspercryptin A1
aspercryptin A2
758.539
742.544
C H N O
20
aspercryptin B1
aspercryptin B2
aspercryptin B3
aspercryptin B4
aspercryptin C1
aspercryptin C2
aspercryptin C3
aspercryptin C4
aspercryptin C6
934.586
918.591
920.571
906.555
800.549
784.554
786.534
772.518
756.523
C H N O
60
130
50
120
20
60
3
3
7
71
7
9
47 79
7
12
11
12
12
10
c
C H N O
90
C H N O
47 79 7
3
7
71
7
8
a
c
epi-aspercryptin A2 742.544
C H N O
90
C H N O
46 77 7
3
7
71
7
8
aspercryptin A3
aspercryptin A4
aspercryptin A5
aspercryptin A6
aspercryptin A7
aspercryptin D1
744.523
730.508
728.528
714.513
671.507
814.566
C H N O
2
2
C H N O
45 75 7
3
6
69
7
9
C H N O
C H N O
39 73 7
3
5
67
7
9
C H N O
20
20
20
5
C H N O
39 73 7
36 69
7
8
9
C H N O
C H N O
38 71 7
3
5
67
7
8
10
10
C H N O
C H N O
2
3
4
66
6
7
37 69
7
C H N O
C H N O
9
20
40 75
7
10
37 69
7
The putative structures for 8 of the additional 13 are shown in Figure 3. A subset was chosen for
clarity in the main text (for a complete version of this chart see Supporting Information Figure S6). The
variations in the structures of the aspercryptins are standard for what is seen from NRPS pathways:
incorporation of alanine instead of serine, valine for isoleucine and a C fatty amino alcohol instead of
1
0
the C version. Taking the structure of aspercryptin A1 as the ‘canonical’ sequence of the aspercryptins,
12
3
ACS Paragon Plus Environment

ACS Chemical Biology
Page 4 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
we have made a hierarchical nomenclature system, where the letter represents the status of the Nꢀ
terminus: A for a free amine, B for cichorine capped, C for acetyl and D for propionyl; and where the
number represents ‘variants’ of the base amino acid sequence: 1 for canonical, 2 for serine to alanine, etc.
Isolation and full NMR structure determination for each of the aspercryptins are beyond the scope of this
work. While the de novo structure elucidation of natural products from analysis of MS and MS/MS data
alone is not possible, such analyses can readily be used for the detection and putative characterization of
highly similar structural analogues. Though the stereochemistry of the aspercryptins proposed by MS/MS
analysis is likely the same as those that have been experimentally determined, we have refrained from
assuming that this is indeed the case (Figure 3).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Putative structures for 8 of the 13 aspercryptins based on analysis of MS spectral differences. A complete version of
this chart with the 18 structurally characterized aspercryptins is shown in the Supporting Information. Heavily annotated MS/MS
spectra of aspercryptins solved by NMR (blue asterisk) were used as the basis for comparison to other metabolites (Supporting
Information Figures S3 and S4). Conservatively, stereochemical assignment has not been made for the MSꢀbased structures.
Confirmation of the aspercryptin biosynthetic gene cluster. Linking natural products to their gene
clusters is key to annotating the biosynthetic potential of phylogenetically diverse fungi. The genome of
12
A. nidulans contains a 14ꢀgene BGC (AN7884 to AN7872, Figure 4a and Supporting Information Table
1
2
S3) that was recently shown to produce aspercryptins B1 and B3. This BGC contains a sixꢀmodule
NRPS (AN7884) and two fatty acid synthase (FAS) subunits (AN7880 and AN7873). To experimentally
link all of the aspercryptins to this BGC, we deleted the NRPS gene, AN7884 and observed no detectable
levels of any of the aspercryptins (Figure 4b). We also generated an overexpression mutant of the nearby
transcription factor, AN7872, which led to increased transcript levels for many of the genes predicted to
KD
be in the cluster. This mirrored the overexpression of these genes in the rpdA strain (Supporting
Information Figure S7) and confirmed the predicted cluster boundaries. Further solidifying its role as the
transcription factor for the cluster, we also observed a >2ꢀfold increase in the levels of aspercryptin A1
upon overexpression of AN7872 (Figure 4c). Additionally deletion of the FAS gene AN7880 abolished
production of the aspercryptins. Levels of the aspercryptins could be rescued partially by supplementing
the media with octanoic and dodecanoic acids, which are likely the direct products of the FAS enzymes
(
Figure 4c).
4
ACS Paragon Plus Environment

Page 5 of 10
ACS Chemical Biology
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. Evidence for the association of the aspercryptins to their BGC. (a) The BGC responsible for the biosynthesis of the
aspercryptins contains a non-ribosomal peptide synthetase encoded by atnA (AN7884), which contains 6 adenylation domains
and a terminal reductase domain; the two fatty acid synthase subunits atnF (AN7880) and atnM (AN7873); a transcription
factor atnN (AN7872); two aminotransferases atnH (AN7878) and atnJ (AN7876); a cytochrome P450 atnE (AN7881); an
oxidoreductase atnD (AN11028); and 3 for transport atnC (AN11031) and atnF (AN7879) and resistance atnI (AN7877). (b)
KD
Deletion of the NRPS gene atnA in the background of rpdA showed it is necessary for biosynthesis of the aspercryptins (only
aspercryptin A1 is shown here for clarity). (c) Overexpression of the transcription factor atnN (OE atnN) led to a doubling of
aspercryptin A1 levels. Subsequent deletion of the FAS gene atnF (OE atnN, ꢁatnF) abolished levels of aspercryptins, which
could rescued by supplementing media with the fatty acids octanoic and dodecanoic acids (OE atnN, ꢁatnF + FAs).
The NRPS, the transcription factor, and the intervening genes have been named atnA (NRPS) through
atnN (transcription factor). Upon exploring the phylogenetic distribution of the atn BGC in the
Aspergillus genome repository (http://www.aspergillusgenome.org), we found the atn BGC present in six
other Aspergilli and to be most conserved in A. versicolor (NRPS genes 87% identical, Supporting
Information Figure S8). The BGC borders were confirmed by transcript mapping using Northern blots of
all genes in the cluster (Supporting Information Figure S7).
Proposed biosynthesis for the aspercryptins. Based on annotations of the BGC and the above
deletion experiments, we propose the following for aspercryptin biosynthesis (Supporting Information
Figure S9). The first step is the generation of the fatty acid precursors, octanoic and dodecanoic acids, by
the FAS subunits AtnF and AtnM (AN7880 and AN7873). A. nidulans has 3 other pairs of FAS genes.
One, fasA and fasB, is required for fatty acids involved in primary metabolism; deletion of either fasA or
14
fasB is lethal. The other 2 pairs of FAS genes make precursor fatty acids for natural products – pkiB and
1
5
pkiC for the polyketides made by the PKS pkiA (AN3386) and stcJ and stcK for sterigmatocystin
14
biosynthesis.
The fatty acid precursors are perhaps the most interesting aspect of the system and are likely
transformed into the corresponding αꢀamino fatty acids in three steps. First they are hydroxylated by the
cytochrome P450 AtnE (AN7881), then oxidized to the corresponding αꢀketo acids by the NAD(P)ꢀ
dependent oxidoreductase AtnD (AN11028), and finally converted to the αꢀamino fatty acids by the PLPꢀ
dependent aminotransferases AtnH or AtnJ (AN7878 and AN7876). Similar pathways to convert a fatty
acid or polyketide precursor to the corresponding αꢀamino acid have been proposed for the cyclosporins
16ꢀ18
and apicidins/HCꢀtoxin, and experimentally supported to varying degrees.
The only experimental
evidence to support this transformation pathway from these systems comes from the deletion of a
19
branched chain aminotransferase which eliminates HCꢀtoxin biosynthesis. A recent publication by
Chiang et al. also proposes this pathway for production of the fatty amino acids, and showed that deletion
5
ACS Paragon Plus Environment

ACS Chemical Biology
Page 6 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
of these genes abolished levels of aspercryptin B1 (except in the case of the aminotransferases, AtnH and
1
2
AtnJ, which could reasonably compensate for each other).
Unlike the other FAS genes in the A. nidulans genome, the aspercryptin FAS elements are
interrupted’ by the aminotransferase genes that we suggest are involved in generation of the αꢀamino
‘
fatty acids (Supporting Information Figure S10). Perhaps this chromosomal organization evolved to
decrease the likelihood that the gene cassette necessary for the fatty amino acid biosynthesis will be
separated by chromosomal reorganization. If they were not disrupted by the aminotransferases, the
expression of the FAS genes could conceivably result in unproductive synthesis of free mediumꢀchain
fatty acids that could disrupt membrane stability. In fact, addition of dodecanoic acid above 10 ꢂM for
rescue experiments resulted in little to no growth.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Once made, we propose that the αꢀamino fatty acids, 2ꢀaminoꢀoctanoic and 2ꢀaminoꢀdodecanoic acids,
are recognized, activated and covalently tethered to the NRPS AtnA by its fourth and sixth adenylation
domains (Supporting Information Figure S9). For typical lipopeptides, lipid moieties are added to the Nꢀ
2
0
terminus by an initial terminal condensation domain of the NRPS. Incorporation of the lipid group as an
αꢀamino fatty acid by an adenylation domain has been proposed for members of the apicidin/HCꢀtoxin
family and demonstrated in the biosynthesis of cyclosporin, where the NRPS has been shown to adenylate
and covalently bind to the polyketideꢀderived αꢀamino acid (4R)ꢀ4ꢀ[(E)ꢀ2ꢀbutenyl]ꢀ4ꢀmethylꢀLꢀthreonine
2
1,22
(
bmt).
Experiments are onꢀgoing to directly establish the activity of adenylation domains 4 and 6 of
AtnA in recognizing and activating 2ꢀaminoꢀoctanoic and 2ꢀaminoꢀdodecanoic acids, respectively. To
date, only a handful of αꢀamino fatty acids have been observed in cyclic lipopeptides (Supporting
Information Figure S11). Intriguingly all of these are derived from fungi, with the exception of the
23
piperazimycins. Perhaps this strategy for lipid incorporation into natural products is a fungal innovation
that spread into Streptomycetes.
In general, using the NRPSpredictor2 (modelled from bacterial NRPS) did not confidently predict the
amino acid substrates recognized by each of the six adenylation domains of AtnA (Supporting
Information Table S4). While Chiang et al. propose cichorineꢀserine as the monomer activated by the first
12
adenylation domain of AtnA for the biosynthesis of aspercryptin B1, we believe it to be serine, and that
the Nꢀterminal amine of mature aspercryptin A1 subsequently reacts during the biosynthesis of
24
cichorine to form aspercryptin B1. Additionally, despite AtnA having only one epimerase domain in the
threonine module, the first two amino acids of aspercryptin A1 are Dꢀserine and Dꢀalloꢀthreonine. This
suggests that serine is either loaded directly as Dꢀserine, or that the epimerase domain in the threonine
module epimerizes both Lꢀserine and Lꢀthreonine. Iterative epimerase domains are not without
25
precedent. Because we observed that the alanine residue of aspercryptin A2 exists as a mixture of Lꢀ and
Dꢀ enantiomers, we predict that the epimerase domain acts iteratively. Chronologically, Lꢀserine (for
aspercryptin A1) or Lꢀalanine (for aspercryptin A2) is activated by the first adenylation domain of AtnA,
and while serine is fully epimerized to the Dꢀenantiomer by the epimerase domain, alanine is only
partially converted to the Dꢀenantiomer. The complete conversion of serine and threonine and partial
conversion of alanine suggests that the sidechain hydroxyl of serine and threonine may aid in governing
substrate recognition by the epimerase domain.
The final step in the biosynthesis of the aspercryptins is the reduction of the Cꢀterminus to an alcohol.
Terminal reductase domains generally use the energy from NAD(P)H to install a reactive aldehyde that
26
25
serves as a warhead or as an intermediate that rearranges to yield a mature natural product. For the
aspercryptins, the removal of a chargeꢀbearing site from dominantly hydrophobic portion of the
compound is a potential reason for the installation of this alcohol at the Cꢀtermini of aspercryptins. In
fact, the putatively membraneꢀassociated nature of the aspercryptins may be relevant to their function and
is consistent with our ability to isolate these compounds mainly from the cell mass and little from the
extracellular medium. Many microbes use lipopeptides as a means to adhere to or move across surfaces,
27
and to establish biofilms. Because we could not assign an antiꢀbacterial activity, the aspercryptins may
serve a similar structural/motility function for A. nidulans. We also note that atnH and atnJ are induced
by ethanol (the other atn genes were not examined), which possibly reflects a role for this metabolite
28
during hypoxic stress.
CONCLUSION
Here we uncovered over a dozen family members of the recently discovered lipopeptides, the
aspercryptins, and mapped the complete gene cluster responsible for their biosynthesis. Instrumental in
the discovery of the aspercryptins was the marriage of HDAC inhibition with MSꢀbased metabolite
screening. By inhibiting HDAC function, we were able to tease the production of a new family of natural
6
ACS Paragon Plus Environment

Page 7 of 10
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
products. By taking advantage of MS/MS networking, we could characterize the structural variation of the
aspercryptins. Our successful use of HDACi in the discovery of the aspercryptins demonstrates its
potential to survey fungal extracts for new chemical matter, even in species that have been so intensively
studied, such as A. nidulans. Discovery of new natural product scaffolds from the microbial world at rates
far higher than in past decades now represents a promising path for reinvigorating the pipeline of
compounds flowing into pharmaceutical screening platforms.
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
METHODS
KD
11
Fungal Transformations. The generation of the rpdA strain was previously described. All other strains in this
study are described in Table S5. For construction of overexpression strains, 1 kb of AN7884 or AN7872 5’ flanking
regions and 3’ coding regions were amplified and fused to A. parasiticus pyrG (amplified from pJW24) and a 401
29,30
bp fragment of the alcA promoter using double joint PCR.
The resulting overexpression construct was
transformed into RJMP1.1 to create strains TAAS393.2 and TAAS394.2. For deletion constructs, 1 kb of flanking
regions were amplified and fused to A. fumigatus riboB using double joint PCR. The resulting knockout construct(s)
were transformed into RJMP1.1 and/or TAAS393.2 to create strains TAAS176.3 TAAS395.1, and TAAS217.1.
Transformants were examined for targeted replacement of the native loci by PCR and Southern blotting, and
expression levels were confirmed by northern analysis.
Growth and extraction of fungal strains for LC-MS/MS analysis. All strains were grown with initial inoculations
6
of 10 spores/mL in 250 mL GMM in 1L unbaffled flasks, grown in the dark at 37˚C for 4 days at 200 rpm. For the
KD
initial screening of the rpdA , overexpression and deletant mutants, extractions and analysis were performed as
previously described. For induction of overexpression strains, lactose minimal medium + 30 mM cyclopentanone
was used. For rescue experiments, media was supplemented with 10 µM octanoic and doceanoic acids in acetone.
1
1
However, for stable isotope incorporation experiments, 5 mL of GMM in 13 mL culture tubes were inoculated with
6
1
0 spores/mL and grown as described above, with the additional step of spiking with 1 mM sterileꢀfiltered stableꢀ
isotope amino acids after 48 hours growth. After a total of 4 days of growth, whole cultures were extracted with an
equal volume of ethyl acetate with sonication and vortexing for several minutes, followed by overnight incubation at
11
4
˚C. Organic layers were then dried down and analyzed as previously described.
KD
Isolation and structural determination of aspercryptins A1 and A2. For isolation of aspercryptin from rpdA
mutant, largeꢀscale growths (500 mL growths in 2L flask, total of 4 L) were performed as described above.
Mycelium was separated from the spent media with coffee filters. The spent media was extracted with
dichloromethane to generate an emulsion layer. The emulsion was dried down. The cellular material was washed
with excess methanol. Methanol extracts and emulsions from DCM extraction were pooled in methanol and dried
onto excess silica. Silica was washed with excess ethyl acetate and aspercryptins were eluted from silica with
methanol. Methanol fraction was pHed to ~8.5 and run over SAX resin (Dowex® 1X2 chloride form), flowthrough
contained aspercryptins. SAX Resin washed with MeOH (pH 5) and pooled with flowthrough. This was then
fractionated over preparative RPLC to afford ~4 mg of aspercryptin A1 and A2 in a ~2:1 mixture that could not be
chromatographically resolved.
All NMR experiments were performed in DMSOꢀd on an Agilent 600 MHz DD2 with HCN cryoprobe, except for
CꢀNMR, which was acquired with AVANCE III 500 MHz with direct cryoprobe.
6
13
31
Stereochemistry of amino acids was determined following a standard Marfey’s Test protocol. Briefly, 1 mg of a
2:1 mixture of aspercryptins A1 and A2 was hydrolyzed in 6N HCl overnight at 110ºC. The hydrolyzed mixture
~
was dried in vacuo, to which 100 ꢂL 1% Marfey’s reagent (FDAA) in acetone and 20 ꢂL 1M NaHCO were added.
3
Amino acid standards were derivatized as described above; 2.5 ꢂmoles were used of each DLꢀalanine, Dꢀserine, Lꢀ
serine, DLꢀthreonine, DLꢀalloꢀthreonine, DLꢀisoleucine, DLꢀaspartic acid (for asparagine), synthetic DLꢀ2ꢀaminoꢀ
octanoic acid and synthetic DLꢀ2ꢀaminoꢀdodecanol.
MS/MS networking to identify aspercryptin molecular family. All MS/MS spectra were preꢀprocessed by
removing the 25% lowest intensity peaks, applying a nonꢀlinear transformation to the peak intensities by taking their
square root, and normalizing peak intensities to the sum of the intensities of all remaining peaks in the spectrum.
Spectra with less than 10 remaining peaks, along with those where the base peak constituted more than 75% of the
total scan intensity were removed from the analysis. MS/MS spectra were compared using the cosine similarity
method. When comparing two spectra, if more than six matching peaks were found, the remaining unmatched peaks
were aligned by shifting their m/z by the difference in precursor mass. The resulting output is a cosine score between
0
and 1 that describes the similarity between two spectra, where 1 represents a perfect match. MS/MS spectra from
the same precursor were determined by a cosine similarity of >0.7 and a precursor match within 0.01 m/z. In cases
where multiple spectra from the same precursor were observed, the spectrum with the higher intensity was used and
the lower intensity spectrum was discarded.
7
ACS Paragon Plus Environment

ACS Chemical Biology
Page 8 of 10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Using the cosine comparisons, a network containing 1590 nodes and 3396 edges was constructed and visualized in
Cytoscape, where each node is a representative spectrum from a unique precursor, and each edge is a cosine
comparison with a value of 0.4 or greater.
Synthesis of lipid amino acid monomers. DL-2-amino-octanoic acid. ±ꢀ2ꢀbromoꢀoctanoic acid (1 g, 4.5 mmol)
was dissolved in 10 mL 1:1 H O:acetone. NaN (0.5 g, 7.7 mmol) was added. The solution was vigorously stirred
2
3
overnight at room temperature (reaction mostly complete after 1 hour) to yield ±ꢀ2ꢀazidoꢀoctanoic acid. The 2ꢀ
azidoꢀoctanoic acid was then hydrogenated in dry THF with 20% Pd/C under 1 atm H with constant stirring at room
2
32
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
temperature overnight. Mixture was filtered through celite to remove Pd/C. A small aliquot was purified by RPꢀLC
+ +
to yield 36 ꢂg of pure DLꢀ2ꢀaminoꢀoctanoic acid (m/z 160.1333 [M+H] C H NOH , 0.6 ppm). White waxy
,
8
17
1
13
solid. Hꢀ and CꢀNMR spectra can be found in Supporting Information. DL-2-amino-dodecanol. ±ꢀ2ꢀbromoꢀ
dodecanoic acid (1.25 g, 4.5 mmol) was dissolved in 10 mL 1:1 H O:acetone. NaN (0.5 g, 7.7 mmol) was added.
2
3
The solution was vigorously stirred overnight at room temperature (reaction mostly complete after 1 hour) to yield
ꢀ2ꢀazidoꢀdodecanoic acid. The 2ꢀazidoꢀdodecanoic acid was then reduced in dry THF with LiAlH (0.35 g, 9.34
±
4
33
mmol) with constant stirring on an ice bath for 2 hours. The reaction was quenched with 5% KHSO4 and extracted
with ethyl acetate. A small aliquot was purified by RPꢀLC to yield 70 ꢂg of DLꢀ2ꢀaminoꢀdodecanol (m/z 202.2166
+
+
1
13
[
M+H] C H NOH , 0.2 ppm). Offꢀwhite waxy solid. Hꢀ and CꢀNMR spectra can be found in Supporting
, 12 27
Information.
Bioactivity assays. Fifteen µg of purified aspercryptins A1 and A2 (~2:1) were tested in a standard disk diffusion
assay against M. luteus, E. coli, P. aeruginosa, S. epidermidis, K. pneumoniae, B. subtilis, A. nidulans and P.
citrinum.
ASSOCIATED CONTENT
Supporting Information. Figures are largely of structural validation of the aspercryptins, including
MS/MS spectra. This material is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Eꢀmail: nꢀkelleher@northwestern.edu
Author Contributions
MTH grew strains, analyzed extracts by LCꢀMS, determined the structures of aspercryptins A1 and A2 by NMR, the
other 13 aspercryptins by MS/MS and prepared the manuscript. AAS generated all mutants used in this study. AWG
performed MS/MS networking. RAM performed reduction reactions and Marfey’s derivatization. NPK annotated
the distribution of the atn gene cluster. AAS, NLK and NPK heavily revised several versions of the manuscript.
Notes
ACKNOWLEDGEMENTS
We would like to thank the National Institutes of Health R01 AT009143 (N. Kelleher), T32 GM07133 and NRSA
AI55397 (A. Soukup), and NIH 5PO1GM084077 (N. Keller). We also thank Y. Zhang of IMSERC at Northwestern
University for NMR assistance (supported by the NIH 1S10OD012016ꢀ01 and 1S10RR019071ꢀ01A1).
REFERENCES
(1) Baltz, R. H. (2006) Marcel Faber Roundtable: is our antibiotic pipeline unproductive because of starvation, constipation or lack of
inspiration? J. Ind. Microbiol. Biotechnol. 33, 507ꢀ513.
(2) Hoffmann, T.; Krug, D.; Huttel, S.; Muller, R. (2014) Improving natural products identification through targeted LCꢀMS/MS in an
untargeted secondary metabolomics workflow Anal. Chem. 86, 10780ꢀ10788.
(3) Sidebottom, A. M.; Johnson, A. R.; Karty, J. A.; Trader, D. J.; Carlson, E. E. (2013) Integrated metabolomics approach facilitates
discovery of an unpredicted natural product suite from Streptomyces coelicolor M145 ACS Chem. Biol. 8, 2009ꢀ2016.
(
4) Derewacz, D. K.; Covington, B. C.; McLean, J. A.; Bachmann, B. O. (2015) Mapping Microbial Response Metabolomes for Induced
Natural Product Discovery ACS Chem. Biol. 10, 1998ꢀ2006.
5) Khaldi, N.; Seifuddin, F. T.; Turner, G.; Haft, D.; Nierman, W. C.; Wolfe, K. H.; Fedorova, N. D. (2010) SMURF: Genomic mapping of
fungal secondary metabolite clusters Fungal Genet. Biol. 47, 736ꢀ741.
6) Yaegashi, J.; Oakley, B. R.; Wang, C. C. (2014) Recent advances in genome mining of secondary metabolite biosynthetic gene clusters
and the development of heterologous expression systems in Aspergillus nidulans J. Ind. Microbiol. Biotechnol. 41, 433ꢀ442.
7) Brakhage, A. A.; Schroeckh, V. (2011) Fungal secondary metabolites ꢀ strategies to activate silent gene clusters Fungal Genet. Biol. 48,
(
(
(
1
5ꢀ22.
(8) Fisch, K. M.; Gillaspy, A. F.; Gipson, M.; Henrikson, J. C.; Hoover, A. R.; Jackson, L.; Najar, F. Z.; Wagele, H.; Cichewicz, R. H. (2009)
Chemical induction of silent biosynthetic pathway transcription in Aspergillus niger J. Ind. Microbiol. Biotechnol. 36, 1199ꢀ1213.
9) Henrikson, J. C.; Hoover, A. R.; Joyner, P. M.; Cichewicz, R. H. (2009) A chemical epigenetics approach for engineering the in situ
(
biosynthesis of a cryptic natural product from Aspergillus niger Org. Biomol. Chem.7, 435ꢀ438.
8
ACS Paragon Plus Environment

Page 9 of 10
ACS Chemical Biology
1
2
3
4
5
6
7
8
9
(
10)Mao, X. M.; Xu, W.; Li, D.; Yin, W. B.; Chooi, Y. H.; Li, Y. Q.; Tang, Y.; Hu, Y. (2015) Epigenetic genome mining of an endophytic
fungus leads to the pleiotropic biosynthesis of natural products Angew. Chem. 54, 7592ꢀ7596.
11)Albright, J. C.; Henke, M. T.; Soukup, A. A.; McClure, R. A.; Thomson, R. J.; Keller, N. P.; Kelleher, N. L. (2015) Largeꢀscale
metabolomics reveals a complex response of Aspergillus nidulans to epigenetic perturbation ACS Chem. Biol. 10, 1535ꢀ1541.
12)Chiang, Y. M.; Ahuja, M.; Oakley, C. E.; Entwistle, R.; Asokan, A.; Zutz, C.; Wang, C. C.; Oakley, B. R. (2016) Development of Genetic
(
(
Dereplication Strains in Aspergillus nidulans Results in the Discovery of Aspercryptin Angew. Chem. 55, 1662ꢀ1665.
(13)Nguyen, D. D.; Wu, C. H.; Moree, W. J.; Lamsa, A.; Medema, M. H.; Zhao, X.; Gavilan, R. G.; Aparicio, M.; Atencio, L.; Jackson, C.;
Ballesteros, J.; Sanchez, J.; Watrous, J. D.; Phelan, V. V.; van de Wiel, C.; Kersten, R. D.; Mehnaz, S.; De Mot, R.; Shank, E. A.; Charusanti, P.;
Nagarajan, H.; Duggan, B. M.; Moore, B. S.; Bandeira, N.; Palsson, B. O.; Pogliano, K.; Gutierrez, M.; Dorrestein, P. C. (2013) MS/MS
networking guided analysis of molecule and gene cluster families Proc. Natl. Acad. Sci. U. S. A. 110, E2611ꢀ2620.
(14)Brown, D. W.; Adams, T. H.; Keller, N. P. (1996) Aspergillus has distinct fatty acid synthases for primary and secondaryꢃmetabolism
Proc. Natl. Acad. Sci. U. S. A. 93, 14873ꢀ14877.
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
(
15)Ahuja, M.; Chiang, Y. M.; Chang, S. L.; Praseuth, M. B.; Entwistle, R.; Sanchez, J. F.; Lo, H. C.; Yeh, H. H.; Oakley, B. R.; Wang, C. C.
2012) Illuminating the diversity of aromatic polyketide synthases in Aspergillus nidulans J. Am. Chem. Soc. 134, 8212ꢀ8221.
16)Offenzeller, M.; Santer, G.; Totschnig, K.; Su, Z.; Moser, H.; Traber, R.; SchneiderꢀScherzer, E. (1996) Biosynthesis of the unusual amino
(
(
acid (4R)ꢀ4ꢀ[(E)ꢀ2ꢀbutenyl]ꢀ4ꢀmethylꢀLꢀthreonine of cyclosporin A: enzymatic analysis of the reaction sequence including identification of the
methylation precursor in a polyketide pathway Biochemistry 35, 8401ꢀ8412.
(17)Bushley, K. E.; Raja, R.; Jaiswal, P.; Cumbie, J. S.; Nonogaki, M.; Boyd, A. E.; Owensby, C. A.; Knaus, B. J.; Elser, J.; Miller, D.; Di, Y.;
McPhail, K. L.; Spatafora, J. W. (2013) The genome of Tolypocladium inflatum: evolution, organization, and expression of the cyclosporin
biosynthetic gene cluster PLoS Genet. 9, e1003496.
(18)Niehaus, E. M.; Janevska, S.; von Bargen, K. W.; Sieber, C. M.; Harrer, H.; Humpf, H. U.; Tudzynski, B. (2014) Apicidin F:
characterization and genetic manipulation of a new secondary metabolite gene cluster in the rice pathogen Fusarium fujikuroi PloS One 9,
e103336.
(19)Cheng, Y.ꢀQ.; Ahn, J.ꢀH.; Walton, J. D. (1999) A putative branchedꢀchainꢀaminoꢀacid transaminase gene required for HCꢀtoxin
biosynthesis and pathogenicity in Cochliobolus carbonum Microbiology 145, 3539ꢀ3546.
(20)Chooi, Y. H.; Tang, Y. (2010) Adding the lipo to lipopeptides: do more with less Chem. Biol. 17, 791ꢀ793.
(21)Dittmann, J.; Wenger, R. M.; Kleinkauf, H.; Lawen, A. (1994) Mechanism of cyclosporin A biosynthesis. Evidence for synthesis via a
single linear undecapeptide precursor J. Biol. Chem., 269, 2841ꢀ2846.
22)Lawen, A.; Zocher, R. (1990) Cyclosporin synthetase. The most complex peptide synthesizing multienzyme polypeptide so far described J.
Biol. Chem., 265, 11355ꢀ11360.
23)Miller, E. D.; Kauffman, C. A.; Jensen, P. R.; Fenical, W. (2007) Piperazimycins:ꢃ Cytotoxic Hexadepsipeptides from a MarineꢀDerived
Bacterium of the Genus Streptomyces J. Org. Chem. 72, 323ꢀ330.
24)Sanchez, J. F.; Entwistle, R.; Corcoran, D.; Oakley, B. R.; Wang, C. C. (2012) Identification and molecular genetic analysis of the
(
(
(
cichorine gene cluster in Aspergillus nidulans MedChemComm, 3, 997ꢀ1002.
(25)Evans, B. S.; Ntai, I.; Chen, Y.; Robinson, S. J.; Kelleher, N. L. (2011) Proteomicsꢀbased discovery of koranimine, a cyclic imine natural
product J. Am. Chem. Soc. 133, 7316ꢀ7319.
(26)Shigemori, H.; Wakuri, S.; Yazawa, K.; Nakamura, T.; Sasaki, T.; Kobayashi, J. i. (1991) Fellutamides A and B, cytotoxic peptides from a
marine fishꢀpossessing fungus Penicillium fellutanum Tetrahedron, 47, 8529ꢀ8534.
(27)Raaijmakers, J. M.; De Bruijn, I.; Nybroe, O.; Ongena, M. (2010) Natural functions of lipopeptides from Bacillus and Pseudomonas: more
than surfactants and antibiotics FEMS Microbiol. Rev. 34, 1037ꢀ1062.
(
28)Shimizu, M.; Fujii, T.; Masuo, S.; Takaya, N. (2010) Mechanism of de novo branchedꢀchain amino acid synthesis as an alternative electron
sink in hypoxic Aspergillus nidulans cells Appl. Environ. Microbiol. 76, 1507ꢀ1515.
29)Yu, J. H.; Hamari, Z.; Han, K. H.; Seo, J. A.; ReyesꢀDomínguez, Y.; Scazzocchio, C. (2004) Doubleꢀjoint PCR: a PCRꢀbased molecular
tool for gene manipulations in filamentous fungi Fungal Genet. Biol. 41, 973ꢀ981.
30)Calvo, A. M.; Wilson, R. A.; Bok, J. W.; Keller, N. P. (2002) Relationship between secondary metabolism and fungal development
Microbiol. Mol. Biol. Rev. 66, 447ꢀ459.
(
(
(31)Bhushan, R.; Bruckner, H. (2004) Marfey's reagent for chiral amino acid analysis: a review Amino acids 27, 231ꢀ247.
(32)Corey, E. J.; Link, J. O. (1992) A general, catalytic, and enantioselective synthesis of αꢀamino acids J. Am. Chem. Soc. 114, 1906ꢀ1908.
(33)Boyer, J. H. (1951) Reduction of Organic Azides to Primary Amines with Lithium Aluminum Hydride J. Am. Chem. Soc. 73, 5865ꢀ5866.
9
ACS Paragon Plus Environment

MS/MS
Backward
lipopeptides
HDACi + LC/MS
ACS Chemical Biolo Pg ay ge 10 of 10
networ644.4king
6
60.4
6
3
4
1
8.374
77
8.581
646.415
6
62.389
88.24
4
9
6
.3
4
5
9
.
2
0
1
4
5
9
.202
6
76.40
5
5
594497.1946.36498
5
.1
2
7
7
.5
4
5
5
0
5.62 95 31
61
.3
6
5
6
.44
4
5
97.287
4
7
6
.
2
2
8
8
4
9.4
4
9
4
.
5
6
6
6
3
3
6
.
6
3
4
96
7.
5.87
4
15
6
38
7
.7
5.
24. 4
91
6
4
6
3
83.5
3
2
0
2
8.5
9
7
915.
8
06.539 353.102
51
9.33
631.439
5
6
4
3
.
3
4
8
7
497.267
2
8
4.29 44 52.2
4
9
92.345
6.48
452.322
5
78.3
455.33445 34 .9
5
3
2
3
9
6
3
4
.13
5
7
6
.
5
1
1
.
423
6
7650
3
0
5283 32.9. 112.9
.
3
843
7
14
8
47
0
1
.571
.
3
2
53 .
3
7932
2
9
1434
7
7
..2024
8
328.079
439.332
540.375
455.207 437
3
.
64
19
.
6
0
8
2
4
0
4
.
2
2
0
8
3.132
2
0233
8
3
3
7103
5
6
5
..73.82
1
0
.34.1.444232028056238. 772
1396
5
3.5.2
.
3
21
1
72421
4
325192.42.7
6
06083753.
2
37341.
8
.2
6
9.30
0
4
0
8
7
3428
0
83254
1
69120..28
6
13
3
3
0
763
5
2.
2
17..1621.8822. 102.052387759161
3.
.
1
0
0
8
2
3
328.09 327.15 378.227 377.197
4
53.32497.346
3
07.
6
2
4
2
3
7.26
7
7
9
2
.1
2
6
1
0
8
.0
4
3
6
8
2
4
8
1
9
9
2
.
.3
7
1
.
3
52
4
2.7
3
90
2
9
15
83
3
.5
0
6
.
5
3
0
2
.
5
1
3
2
8
2
0
2
2
4
.
7.
8
7
98
7
.1.212
3
9298 8. 1. 2
0
9
2
449.303
7
5
4
1
.
3
7
3
5
8
5
.
3
9
9
4
4
1
5
8
9
.
1
.
4
1
8
3
8
4
6
3
.
2
2
8
8
2
266.067
6.31259.0
3
6
0
5
9.2
2
7
5
9
1.04 72 1.271335.219
348.2
276.16
OH
1
6
540.39
295.188
231.065
258.28 239.129 292.071 380.199
363.171
O
H
O
H
O
96.363
H2
N
N
N
OH
3
8
N
N
H
N
H
O
1
2
3
4
5
6
882.507 477.26
3
91. 24 20 69 .8
2
13
4
795 .24
3
0. 91
8
59. 32 05
44
5
1
5
9..1
1
2
39 929.23 1078.26 965.298
HO
H
O
O
817.395
816.3
3
8
90.18
519.139
1073
1
.
0
3
7
1
4
8. 33 12 .24
519.139
9
519.5
9.1
7
96.36
795.3
7
5
8
9
1
7
4..3
37
6
9
87 17 57 .3 84 2
4
74.311
5
19.13
491.34
519.139
519.13
519.371
1
3
3
6
6
1
6
9. 11 03 39 8.2
889.23
1072.31
9
7
0
9
6
24
.6
2
8
6
6
.
4
5
2
2
.
5
7
2
8
4
0
.1.
3
1
774.25
95
7
6
371.1
7
5
0
8
2
.57
NH
2
8.35 813.36
9
5
519.139
850.25
77
407.209389.173
5
19.139
7
77.343
5
19.13
9
aspercryptin A1
4
08.317
406.17
989.581
249.
3
1
0
1
8
7
.
2
1
8
2
4
8
.
2
12
6
7
8
.103
496.34 520.34
4
09.303
644.46
851.55
386.14 446.
656.322
181
784.58 75 6
0
7
9
.5
8
8
.5
5
7
7
7
70.569758.57
782.57
0.56252 27 . 83 65 .66 0
7
8
. 470 .34.5
485.58
1
ACS Paragon P731
7
74.51
78
7
92.5 28 77.55
8
.
25 2
773.502671
l
.492
.5
0
9
u
744.51
.507 1490.01
s
747.002
830.55
E
786.6 70 51
7
4
72.585
7
8
7. 52 7. 5
78
7
1
8
6
5
4
8.5
7.
7
5
5
4
6
762.
.5
7
5
6
6
5
669 . 775
1
5
3
6
0
8
7
8
8
8
67
6
.74.7
4
9
4656.0751 287.5 5886 9.
5
1
5
0
4
.60
6 .145 .764 95 5
7
1
84.585
v
ironment
77 59 61
.. 54
7
5
5
. 4. 5
9
07
1
84
2
7
.
75
4
21
5
67 1. 4
9. 47 7
1485.98
7
4
3
.
4
9
747.603
7
8
6
. 5. 6
0
1
7
5
9
.
5
5
7
6
.
9
5
7
5
8
3
6
7
4
5
7
8
.
4
7
3
81
..5
9
9
4.
559814.559
2
3
26
775.38
776.52
7
57
7
4
1
.
51
2
2
7
5
8
.
5
7
6
5
9
8
69
7
84.58 37 58.
7
7
8
1
5
5
5
283
697.486699.502
743.5
742.535
784. 75 88 45 .5
7
8
8
5
2.5
7
7
82.554
782.569
782.57
OH
H
O
O
H
O
7
7
3
.
5
1
7
7
2
.
5
1
8
9
12.55
810.57
786.57
H
N
N
N
OH
2
65.253
2
N
H
O
8.249
799.57
798.5
7
48.602826.565
769.519 786.533
758.539 1320.6
8
5
94.591
372.311263.237 297.28
N
N
8
466.28
1000.6814.5688
9
6
94 42 .. 55 56
7
1
7
0
.
5
3
8
7
7
1
.
5
4
8
1
2
.
5
8
730.5
6
10
08
16.59
295.263
H
O
H
O
8
67.544
9
85
2
4
3..5
52
6
6
8
768.56
6
828.575
756.523
1000.6
299.223281.249 282.279
950.586 920.571
8
934
5
2
. 5. 5
8
89
1
71 8.59
8
2
00.55 801.5
7
5
72.51 97 42.544 714.511
1032.59
285.242
9
05.561 904.5 79 63 2.57 962.5811034.6
1060. 16 62 00.09
783.521
257.211
NH
2
9
33.57906.555 155 71 .2
0
5
5
4. 16 38 95
587.08
45.05
15
.9
5
243.196
1
618.1
1
2
544.363
4
aspercryptin A2