Directed Accumulation of Anticancer Depsipeptides by

Jul 6, 2018 - Research Center for Marine Drugs, State Key Laboratory of Oncogenes and ... (3-FAS) biosynthesis in the antimycin biosynthetic pathway...
0 downloads 0 Views 918KB Size
Subscriber access provided by Technical University of Munich University Library

Article

Directed Accumulation of Anticancer Depsipeptides by Characterization of Neoantimycins Biosynthetic Pathway and an NADPH-dependent Reductase Yongjun Zhou, Xiao Lin, Simon R. Williams, Liyun Liu, Yaoyao Shen, ShuPing Wang, Fan Sun, Shihai Xu, Hai Deng, Peter F Leadlay, and Hou-Wen Lin ACS Chem. Biol., Just Accepted Manuscript • DOI: 10.1021/acschembio.8b00325 • Publication Date (Web): 06 Jul 2018 Downloaded from http://pubs.acs.org on July 7, 2018

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 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 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.

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 27 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

ACS Chemical Biology

1

Directed Accumulation of Anticancer Depsipeptides by Characterization

2

of Neoantimycins Biosynthetic Pathway and an NADPH-dependent

3

Reductase

4

# # Yongjun Zhou1 , Xiao Lin2 , Simon R. Williams3, Liyun Liu1, Yaoyao Shen1,

5

Shu-Ping Wang 1, Fan Sun 1, Shihai Xu2, Hai Deng4, Peter F. Leadlay5,

6

Hou-Wen Lin1*

7

1

Research Center for Marine Drugs, State Key Laboratory of Oncogenes and

8

Related Genes, Department of Pharmacy, Ren Ji Hospital, School of Medicine,

9

Shanghai Jiao Tong University, Shanghai 200127, China 2

College of Pharmacy, Jinan University, Guangzhou 510632, China

11

3

Department of Chemistry, University of Cambridge, Cambridge, CB2 1EW,

12

UK

13

4

Department of Chemistry, University of Aberdeen, Aberdeen AB24 3UE, UK

14

5

Department of Biochemistry, University of Cambridge, Cambridge CB2 1GA,

15

UK

10

16



The authors contributed equally to this work.

17

* To whom correspondence should be addressed. E-mail:

18

[email protected]

19 20

ABSTRACT: Neoantimycins (NATs) are members of antimycin-type of

21

depsipeptides with outstanding anticancer activities. We isolated NAT-A (1)

22

and -F (2) from the fermentation extract of Streptomyces conglobatus. The 1

ACS Paragon Plus Environment

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

Page 2 of 27

23

NAT biosynthetic gene cluster (nat BGC) was identified by genome

24

sequencing and bioinformatics analysis. nat BGC includes two nonribosomal

25

peptide synthetases (NRPSs) and one polyketide synthase (PKS) genes, and

26

a gene cassette (ten genes), of which the encoded enzymes share high

27

homologs to the ones responsible for 3-formamidosalicylate (3-FAS)

28

biosynthesis in antimycin biosynthetic pathway. Heterologous expression of

29

the partial nat BGC without the 3-FAS gene cassette in the antimycin producer,

30

Streptomyces albus J1074, results in the production of 1 and 2, suggesting

31

that the nat BGC indeed directs NATs biosynthesis. Targeted in-frame deletion

32

of the reductase gene (natE) abolished the production of 1 and 2 but

33

accumulated two NAT derivatives, the known NAT-H (3) and a new NAT-I (4).

34

Biochemical verification demonstrated that the recombinant NatE indeed

35

catalyzes an NADPH-dependent reaction of 3 or 4 to 1 or 2, respectively.

36

Compound 3 presented significantly stronger activities against eight cancer

37

cell lines than the ones using cisplatin, the clinical chemotherapy medicine. In

38

particular, 3 displayed 559 and 57-fold higher activities toward human

39

melanoma and cervix epidermoid carcinoma cells, respectively, compared with

40

cisplatin. The new derivative, 4, was 1.5 to 10.9-fold more active than cisplatin

41

toward five cancer cell lines. The evaluation of NATs biosynthesis depicted

42

here will pave the way to generate new NAT derivatives through rational

43

pathway engineering.

44 2

ACS Paragon Plus Environment

Page 3 of 27 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

ACS Chemical Biology

45

INTRODUCTION

46

Antimycin-type depsipeptides are a group of natural products sharing a

47

common structural skeleton consisting of a macrocyclic ring with an amide

48

linkage to a 3-formamidosalicylate (3-FAS) unit. They present great structural

49

diversity due to the variation in the size of macrocycle ring and substitution on

50

the ring.1 NATs are members of antimycin-type depsipeptides, featuring a

51

15-membered tetraester ring, in contrast to the 9-membered ring of antimycins

52

(Figure 1a), which have received considerable interest due to their diverse

53

biological activities. 2-4 Since the first isolation of NAT-A (1) (Figure 1a), from

54

Streptomyces orinoci in 1967, there have been twelve NAT derivatives

55

discovered from various Streptomyces strains. 5-7 The structural variations in

56

the NAT family are mainly derived from the different chain length of alkyl

57

groups at C4 and C9 and the hydroxyl or keto group at C1 in macrocyclic ring

58

(Figure 1a). Derivatives with an alternative start unit such as benzoic acid,

59

3-hydroxybenzoic acid or 3-aminosalicyclic acid instead of 3-FAS are also

60

known. 8, 9

61

It was recently discovered that NATs possess significant anticancer activity

62

and this has led to an increase interest in these natural products. The NAT

63

derivatives, prunustatin A, JBIR-04, and JBIR-05 were characterized as

64

down-regulators of GRP-78, which functions as a molecular chaperone in the

65

endoplasmic reticulum to promote protein folding, it thus contributes to

66

chemotherapy resistance and hypoglycemic stress.

8, 10, 11

Compounds 1, 2, 3

ACS Paragon Plus Environment

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

Page 4 of 27

67

and 3 (Figure 1a) have been shown to be potent inhibitors of GTPase K-Ras

68

plasma membrane (PM) localization. K-Ras acts as a key molecular switch to

69

regulate cell growth, proliferation, and differentiation, and the inhibition of Ras

70

PM localization will block all oncogenic activity.7 It was also reported that the

71

activities of NATs were unaffected by P-glycoprotein mediated drug efflux in

72

human colon cancer cells, a mechanism which is commonly found to confer

73

chemoresistance in cancer.7

74

Despite of outstanding anticancer activities of NATs recently discovered,

75

there were very few investigation reported on the biosynthesis of these

76

valuable compounds. The nat BGC from S. orinoci was reported solely based

77

on bioinformatics prediction, 6 and there exist obvious gaps in the NRPS gene

78

due to the poor sequencing results. The lack of knowledge of NAT biosynthetic

79

pathways has limited attempts to generate further derivatives with enhanced

80

anticancer activities through pathway engineering or synthetic biology. During

81

our studies of the biosynthesis of the anticancer compound conglobatin from S.

82

conglobatus, 12 two known NATs, 1 and 2, was isolated from this strain. Herein,

83

we report the identification and characterization of the nat BGC from S.

84

conglobatus. Heterologous expression of the partial nat BGC (natA-E) without

85

the 3-FAS gene cassette in the producer of antimycin, S. albus J1074, which

86

naturally possesses the 3-FAS precursor pathway, resulted in the production

87

of two NATs, demonstrating the identity of the gene cluster. A new

88

ketoreductase NatE

was

verified to catalyze an NADPH-dependent 4

ACS Paragon Plus Environment

Page 5 of 27 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

ACS Chemical Biology

89

ketoreduction at C1 in the macrocyclic ring of NATs at the late modification

90

stage of the biosynthesis. Finally, the new NAT derivative, 4, accumulated in

91

the ∆natE mutant and three known NATs, 1-3, displayed significantly stronger

92

anticancer activities against eight cancer cell lines compared to the currently

93

used drug, cisplatin. Taken together, the understanding of the NAT production

94

in this study can facilitate to further explore the chemical space of bioactive

95

NATs.

96 97

RESULTS AND DISCUSSION

98

Discovery and Identification of NATs from S. conglobatus. Two NAT

99

derivatives were isolated from the fermentation extract of S. conglobatus, the

100

producer of the anticancer compound conglobatin.12 Comparison of the high

101

resolution

102

comprehensive NMR data (Supplementary Table 4 and 5) and optical rotation

103

measurements (Supplementary 1.6) with literature values demonstrated that

104

these two NAT derivatives are two the known compounds 1 and 2 (Figure 1a).7

105

in silico Analysis of NAT BGC. The nat BGC was previously identified

mass

spectroscopy

(HR-MS)

(Supplementary

6

Table

3),

106

based on bioinformatics analysis in the genome of S. orinoci,

107

DNA sequence of nat BGC is not accessible in public database and the

108

analysis of the gene cluster is obviously flawed due to the incomplete

109

sequencing. Specifically, there are two genes essential for the starter unit

110

3-FAS biosynthesis in antimycin-type depsipeptides

1

however, the

were missing: an 5

ACS Paragon Plus Environment

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

Page 6 of 27

111

orthologue of antG encoding a discrete peptidyl carrier protein (ACP), and an

112

orthologue of antI encoding a subunit of the multicomponent oxygenase that

113

converts anthranilic acid to 3-aminosalicylate. In addition, the last module of

114

the NRPS-PKS assembly line lacked a typical condensation domain and the

115

module was abnormally encoded by two genes transcribed with opposite

116

direction, in which natD encoded an adenylation domain and natE encoded the

117

multidomain of ketoreduction-thiolation-release/cyclization.

118

Those peculiarities of the reported nat BGC prompted us to identify the nat

119

BGC in the new NAT producer S. conglobatus. In silico analysis of the

120

annotated genome sequence allowed the identification of the nat BGC that

121

could direct the biosynthesis of 1 and 2. The nat BGC (GenBank MH128122)

122

spans ~40 kbp of genomic DNA and contains 17 open reading frames (ORFs)

123

with predicted functions that were assigned based on homology analysis

124

(Figure 2 and Supplementary Table 1). The nat BGC encodes two multidomain

125

NRPSs (NatB and NatD) and one multidomain PKS (NatC). NatB possesses

126

the typical arrangement of C1-A1-T1-C2-A2-KR1-T2-C3-A3-KR2-T3 and NatD has

127

the arrangement of C4-A4-KR3-T4-TE. NatC is organized as KS-AT-MT-ACP.

128

As expected, a set of genes (natF-L, N, O, and Q) responsible for 3-FAS

129

biosynthetic genes is located at the right side of the gene cluster. In addition,

130

three genes that encode reductase (natE), type-II thioesterase (natR) and

131

regulatory (natA), respectively, are also included in the gene cluster. Notably,

132

the architecture of the nat BGC from S. orinoci NRRL B-3379

13

reported 6

ACS Paragon Plus Environment

Page 7 of 27 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

ACS Chemical Biology

133

recently is identical to the one from S. conglobatus with an exception that an

134

extra MbtH-like protein gene is inserted between natD and natE (the

135

orthologue of natF in the nat BGC from S. orinoci).

136

Based on bioinformatics analysis and our current understanding of the

137

antimycin-type natural products, 1 the NAT biosynthetic pathway is proposed

138

(Figure 2). The A1 domain activates and loads threonine onto T1, followed by

139

condensation with one 3-FAS tethered on NatG by C1 to form the first amide

140

bond. The A2 activates and loads 2-oxo-3-methylbutanoic acid onto T2, which

141

is stereoselectively reduced by KR1, followed by condensation with the

142

growing dipeptide chain by C2. The A3 incorporates phenylpyruvic acid on T3,

143

which is subsequently reduced by KR2 and condensed by C3 with the growing

144

aminoacyl thioester tethered on T2. The AT domain of the PKS module

145

transfers malonyl unit to ACP followed with demethylation by MT domain.

146

Then KS domain catalyzes the Claisen condensation reaction between

147

dimethyl malonyl ACP and the aminoacyl thioester attached to T3. Finally, the

148

A4 domain of NatD activates and loads 2-oxo-3-methylbutanoic or

149

2-oxo-3-methylpentanoic acid onto T4, which is then stereospecifically reduced

150

by KR3 domain followed by condensation with the aminoacyl thioester

151

attached to NatC-ACP prior to the regiospecific macrolactone cyclization to

152

release a 15-membered tetralactone product catalyzed by the last TE domain

153

of NatD. The putative NAD(P)H dependent ketoreductase (NatE) may mediate

154

the ketoreduction at C1 to generate the final products of 1 and 2. The type II

13

7

ACS Paragon Plus Environment

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

Page 8 of 27

155

thioesterase (NatR) could play the proofreading role during the NAT

156

biosynthesis by hydrolytically removing aberrant intermediates.14

157

Heterologous Expression of nat BGC. To confirm whether the nat gene

158

cluster directs the biosynthesis of NATs, heterologous expression strategy was

159

carried out. We chose S. albus J1074, the producer of antimycins, 15 as the

160

heterologous host because the strain contains all of necessary genes for the

161

production of 3-FAS, the important precursor to NATs. A partial nat gene

162

cluster consisting of natA-E was cloned using our previously developed

163

strategy, one-step cloning of gene cluster (Figure 1b)

164

integrative vector to generate the construct pRJ4, followed by conjugation into

165

S. albus J1074. The resulting strain pRJ4/J1074 was fermented, followed by

166

organic extraction. The HPLC-MS analysis of the corresponding extract

167

allowed the identification of the expected peaks of compound 1, [M+H]+ m/z

168

699.3, and compound 2, [M+H] + m/z 685. 3 (Figure 1c), which were further

169

confirmed by MS/MS fragmentation analysis (Accurate-Mass Q-TOF)

170

(Supplementary Figure 1). Taken together, our results confirm that these five

171

genes (natA-E), with the help of the 3-FAS precursor biosynthetic pathway

172

supplied by the host strain, are sufficient to fulfill the production of 1 and 2 in

173

the corresponding mutant strain.

12

into a phiC31

174

In-frame Deletion of natE Encoding Ketoreductase. To further explore

175

the origin of the C1-hydroxyl group in the macrolactone skeleton of 1 and 2, we

176

set out to investigate the possible role of natE, encoding a ketoreductase, in 8

ACS Paragon Plus Environment

Page 9 of 27 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

ACS Chemical Biology

177

NATs biosynthesis. It was hypothesized that the keto moiety at C1 should be

178

maintained after the NRPS-PKS assembly line since there is not

179

ketoreductase domain found in the PKS module of NatC. It is likely that the

180

stand-alone ketoreductase NatE may be responsible for the reduction step

181

either before or after chain release from the hybrid NRPS-PKS. NatE shows

182

the most similarity to atypical short-chain dehydrogenases/reductases (SDRs),

183

which have a putative glycine-rich NAD(P)-binding motif, GGXXXXG, and an

184

active site motif of YXXXK with an upstream Ser conservative residue

185

according to protein sequence alignment (Supplementary Figure 2).

186

Interestingly, NatE shows little sequence similarity to other characterized

187

NAD(P)H-dependent ketoreductases that are involved in the late modification

188

step of natural product biosynthesis. For example, NatE shares only 14%

189

sequence identity to either LanV that catalyzes the ketoreduction at C‑6 of

190

landomycin 16 or SimC7 that catalyzes the conversion of 7-oxo-simocyclinone

191

D8 into simocyclinone D8.

192

NAD(P)H-dependent ketoreductase catalyzing the reduction of the C-1

193

carbonyl intermediate, 3 or 4, to generate 1 or 2. To verify this, natE was

194

truncated by in-frame deletion of 51 bp, which includes the putative NAD(P)H

195

binding site of GGTAFLG at the N-terminal of NatE (Figure 3a) to generate the

196

mutant strain RJ8. HPLC-MS analysis of the fermentation extract showed that

197

RJ8 lost the production of 1 and 2, but accumulated two new products with

198

mass [M+H]+ signals of m/z 697.3 for compound 3 and m/z 683.3 for

17

We therefore proposed that NatE may be a new

9

ACS Paragon Plus Environment

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

199

Page 10 of 27

compound 4, 2 Da less than the mass of 1 and 2, respectively (Figure 3b).

200

Isolation and Structural Elucidation of 3 and 4 Accumulated inΔ ΔnatE.

201

To confirm the identities of compounds 3 and 4 produced by ΔnatE mutant

202

RJ8, a 12 L fermentation was performed and the broth was extracted with ethyl

203

acetate. The organic extract was purified by consecutive chromatographic

204

fractionations to produce pure compounds 3 and 4 with the yield of 400 mg

205

and 62 mg, respectively.

206

The NMR spectra of 3 and 4 gave the representative features of the NAT

207

skeleton (Supplementary Information 1.6). All the proton resonances were

208

assigned to their respective carbons via the HSQC spectra, and identification

209

and assembly of the skeletal units was assisted by diagnostic COSY and

210

HMBC correlations (Figure 4). The proton-proton correlations observed in the

211

1

212

them were elucidated from HMBC correlations, leading to the formation of a

213

15-membered macrolactone ring. The chemical structure of 3 was confirmed

214

to be identical to NAT-H by comparative analysis of the 1H and

215

(Supplementary Table 6b) and [α]D measurements previously reported

216

(Supplementary Information 1.6). The structure of compound 4 was elucidated

217

to be a new NAT derivative, which is structurally identical to compound 3

218

except for the absence of a methylene resonance at δC 24.7 (C-32) in 3

219

(Figure 1a). This is supported by the evidences that both the H-35 (δH 0.86)

220

and C-35 (δC 10.7) resonances in the NMR spectra of 3 are absent in that of 4,

H-1H COSY revealed six isolated spin systems, and the connections between

13

C NMR data 7

10

ACS Paragon Plus Environment

Page 11 of 27 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

ACS Chemical Biology

221

and that the formula of 3 bears one more methylene group than that of 4

222

according to Accurate-Mass Q-TOF analysis (Supplementary Table 3).

223

Furthermore, the consistency of the 1H,

224

chiral atoms in 4 with those in 3 (Supplementary Table 7, 6a) as well as their

225

accordant [α]D values (Supplementary Information 1.6) indicate their

226

configurational identity.

13

C, and NOESY NMR data for the

227

in vitro Characterization of NatE. Given that 3 and 4 were not further

228

processed in the mutant strain, it is postulated that they might be the

229

immediate substrates for NatE. To verify this hypothesis, we set out to

230

investigate the role of NatE in the biosynthesis of 1 and 2. Overexpression of

231

natE in E. coli BL21 (DE3) allowed for the purification of its encoded protein to

232

near homogeneity. The resultant recombinant protein NatE with a C-terminal

233

His-tag was estimated to have a molecular weight of 38.9 kDa as determined

234

by SDS-PAGE analysis (Supplementary Figure 3). Upon incubation of

235

compound 3 or 4 (0.4 mM) with the recombinant NatE (1 µM), and NADPH (1

236

mM), the corresponding reduced product 1 or 2 was formed as observed in the

237

HPLC-MS analysis. Based on the HPLC observation, NatE could catalyze

238

approximately 80% conversion of 3 to 1 or 4 to 2 within 30 min. If NADPH was

239

omitted, no conversion of 1 or 2 was observed (Figure 5). Taken together,

240

these results indicate that NatE is an NADPH-dependent ketoreductase that

241

specifically reduces the C1-keto group into a C1-hydroxyl group at the last

242

modification stage in the biosynthesis of 1 and 2. 11

ACS Paragon Plus Environment

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

243

Anticancer

Page 12 of 27

Activity Evaluation of Four NATs. Aware that the

244

cytotoxicities of compounds 1, 2, and 3 had previously only been evaluated in

245

human colon and lung cancer cells,6, 7 we expanded the evaluation spectrum

246

using eight human cancer cell lines with cisplatin, the clinical chemotherapy

247

medicine, as a positive control (Table 1). Notably, compound 3 consistently

248

showed stronger inhibition than that of cisplatin against all cell lines, and

249

exhibited 559 and 57-fold higher IC50 values toward human melanoma (2.9 nM)

250

and cervix epidermoid carcinoma (12.0 nM) cells, respectively. The new NAT

251

derivative, 4 was 1.5 to 10.9-fold more active than cisplatin toward five cancer

252

cell lines. Interestingly, 4 was found to be consistently less active than 2, which

253

only differs from 4 in the C1-hydroxyl. However, the correlation between the

254

C1-hydroxyl and the activity increment was not well reflected by the activity of

255

1 and 3 against cervix epidermoid, hepatic and ovarian cancer cells. Moreover,

256

both 1 and 2 gave much better performance than cisplatin toward seven cell

257

lines, the exception being human hepatic carcinoma cells. The broad-spectrum

258

and significant cytotoxic activities of NATs encourage us to carry out further

259

exploration, either in the study of their anti-tumor pharmacology or engineering

260

of their biosynthetic pathway to produce new derivatives.

261

Conclusions. We identified the nat BGC in S. conglobatus. Heterologous

262

expression of natA-E with the native 3-FAS pathway in S. albus J1074 has

263

been confirmed to be sufficient to promote the production of NATs. In-frame

264

deletion of natE resulted in the loss of the production of 1 and 2 but the 12

ACS Paragon Plus Environment

Page 13 of 27 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

ACS Chemical Biology

265

accumulation of 3 and 4. Biochemical characterization confirmed that the

266

ketoreductase NatE catalyzed the conversion of 3 or 4 to 1 or 2, respectively.

267

The same production level of NATs in natE mutant and wide type strains, and

268

the efficient reduction conversion catalyzed by NatE to the substrates 3 and 4

269

implies that NatE should play the role in the late modification step of NAT

270

biosynthesis. A comprehensive understanding of the NAT biosynthesis

271

presented herein could pave the way to generate new NAT derivatives with

272

improved bioactivities. The plastic property of the NAT biosynthetic pathway is

273

implied by the diverse structures of six NAT derivatives produced by S. orinoci,

274

6

which include alkyl group variations at C4 and C9 positions of NAT and

275

alternative start units linked to the macrolactone. This flexibility implies the

276

feasibility of producing more valuable NAT analogs by engineering of the

277

biosynthetic pathway. Further explorations could be carried out towards 1)

278

directed accumulation of the NAT derivative with the different carbon alkyl

279

chains at C9, which noticeably influences anticancer activity, e.g. compound 1

280

shows more than 20-fold stronger inhibition than 2 against human colon

281

carcinoma cells. This engineering could be achieved by modifying the

282

substrate specificities of the last A4 domain; 2) diversification of the alkyl

283

groups at C11 by changing the substrate specificity residues of NatC-AT

284

according to the information recently obtained from SpnD-AT of splenocin

285

biosynthesis.

286

derived from the antimycin or splenocin biosynthetic pathway may achieve this

18

The introduction of the new extender unit biosynthetic genes

13

ACS Paragon Plus Environment

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

Page 14 of 27

287

modification. 19, 20 3) generation of new NATs with substituted start units by

288

over-expression of natG and natF in coupled with feeding different benzoic

289

acids with different substitutions through precursor-directed biosynthesis.21

290 291

METHODS

292

Cloning and Heterologous Expression of nat BGC. The genomic DNA of

293

S. conglobatus was digested with BspHI at 37 ℃ for 5 hr before adding 0.6

294

volumes of chilled isopropanol to precipitate the DNA by centrifugation for 10

295

min. After washing with 80% ethanol, the DNA was dissolved in 20 µL water.

296

The 5961 bp PCR product of vector fragment was generated by using the

297

template of plasmid pIB139

298

of pIB139-nat-S and pIB139-nat-A (Supplementary Table 2), which contain 27

299

bp and 28 bp overlap respectively with the ends of the 29.4-kbp target DNA.

300

For DNA assembly, 2 µL of vector DNA (151 ng/µL) and 8 µL of digested

301

genomic DNA (323 ng/µL) were added into 10 µL of Gibson DNA assembly

302

solution. 23 The reaction was carried out at 50 ℃ for 1 hr, and then 10 µL of the

303

reaction was used for calcium-assisted transformation of E.coli DH10B. Colony

304

PCR was used to screen the target clone with the three pairs of primers

305

located, respectively, at the flanking and middle area of target fragment

306

(Supplementary Table 2). The obtained plasmid pRJ4 was mobilized into S.

307

albus J1074 by conjugation. The resultant strain pRJ4/J1074 was fermented

308

by liquid medium and the broth was extracted by using resin absorption. The

22

linearized by NdeI and EcoRV, and the primers

14

ACS Paragon Plus Environment

Page 15 of 27 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

ACS Chemical Biology

309

dry extract was re-dissolved into 3 ml methanol and 20 µL was subjected to

310

HPLC-MS analysis.

311

Deletion of natE, Plasmid Construction and Mutant Screening. For

312

in-frame deletion of natE by homologous recombination, two PCR products of

313

1105 bp and 1089 bp, flanking the deletion area, were amplified from genomic

314

DNA by using two pairs of primers, n7-L-F and n7-L-R, n7-R-F and n7-R-R

315

(Supplementary Table 2). The two PCR fragments overlapped each other by

316

48 bp have 25 bp and have 26 bp respectively overlapping with a E.

317

coli-Streptomyces shuttle plasmid pRJ2 linearized by XbaI and EcoRI. The two

318

PCR products and the vector fragment were assembled by Gibson assembly

319

reaction using the same condition as above. The resultant plasmid pRJ28 with

320

hygromycin resistance was introduced into a mutant of S. conglobatus losing

321

conglobatin production (unpublished data), with the helper plasmid pUZ8002 in

322

E. coli ET12567 by conjugation. The single colonies of the exconjugants

323

re-inoculated on SFM agar plate for three rounds were used for screening of

324

double-crossover mutants with the primers of nat7D-T-F and 1NL-R

325

(Supplementary Table 2). The target mutant RJ8 was further confirmed by

326

sequencing of the PCR product.

327

Purification of NAT Derivatives. To isolate compounds 1 and 2 produced

328

by S. conglobatus, and 3 and 4 produced by RJ8, a 12 L fermentation broth of

329

each strain was extracted with ethyl acetate. The organic extract was

330

concentrated in vacuo to generate 8-11 g of syrup, which was fractionated 15

ACS Paragon Plus Environment

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

Page 16 of 27

331

sequentially on silica gel and then an ODS column, tracing the target

332

molecules using HPLC-MS analysis. Finally, the target fractions were isolated

333

by preparative HPLC to yield 1 and 3, both as yellow amorphous solids.

334

(Supplementary Information 1.4)

335

NatE Protein Expression, Purification, and Activity Assay. The 1144 bp

336

PCR product of natE gene was amplified from genomic DNA with primers

337

nat7P-S and nat7P-A (Supplementary Table 2). The C-terminal 6×His tag was

338

introduced by the primer nat7P-A. The PCR product was digested with NdeI

339

and XhoI before introducing to the same sites of pET29a. The resulting

340

plasmid pRJ35 was introduced into E. coli BL21 (DE3) plysS for protein

341

expression and purification (Supplementary Information 1.3). The enzyme

342

activity assay was carried out respectively with the substrates of compounds 3

343

and 4 produced by RJ8. Compounds 1 and 2 isolated from S. conglobatus

344

were used as standards for verification of the products generated in the

345

enzyme assay. The in vitro enzymatic assay was carried out by combining 1

346

µM NatE, 0.4 mM substrate, 1 mM NADPH, and 5 µL DMSO in 50 µL of 50 mM

347

Tris.HCl (pH8.0) buffer. The reaction was quenched with 450 µL acetonitrile

348

after 30 min incubation at 30 ℃ and the sample was centrifuged at maximum

349

speed for 5 min before injecting 20 µL for HPLC-MS analysis.

350

Cell Cultures and Cell Viability Assay. The cancer cells used in this study

351

were all obtained from Shanghai Institute of Cell Biology, Chinese Academy of

352

Sciences. The mediums (Gibco, USA) used for cell cultivation were 16

ACS Paragon Plus Environment

Page 17 of 27 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

ACS Chemical Biology

353

RPMI-1640 for Hey, MNK28, U343 and SW1990 cells, DMEM for HepG2 and

354

A375 cells, DMEM/F12 for Caco-2 cells, and MEM for HeLa cells. All the

355

mediums were supplemented with 10% fetal bovine serum (FBS, Gibco, USA),

356

100 U/mL penicillin and 100 µg/mL streptomycin. Cells were maintained in a

357

humidified atmosphere with 5% CO2 at 37 ˚C. Half maximal inhibitory

358

concentration (IC50) was determined by a CCK8 assay (Dojindo, Tokyo, Japan).

359

Briefly, cells were seeded in 96-well plates at a density of 3×103 cells/well and

360

incubated for 24 h at 37 °C. Then, the cells were treated with various

361

concentrations of compounds (0, 0.3, 1, 3, 10, 30, 100, 300, 1000, and 10000

362

nM) and incubated for 72 h. Afterwards, 10 uL WST-CCK8 reagent was added

363

to each well before a further incubation for 0.5 h to 4 h at 37 °C. Finally, the

364

absorbance of each well was measured at 450 nm using microplate reader

365 366

AUTHOR INFORMATION

367

Correspondence Authors

368

*E-mail: [email protected]

369

Notes

370

The authors declare no competing financial interest.

371

ACKNOWLEDGEMENTS

372

This work was sponsored by the National Science Foundation of China (Nos.

373

31670096, 31628001, 41476121, U1605221, 41729002, 81502936, and

374

81402844) and Shanghai Pujiang Program (16PJ1405800). S.R. Williams 17

ACS Paragon Plus Environment

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

Page 18 of 27

375

thanks I. Paterson and D. Spring for their support.

376

ASSOCIATED CONTENT

377

Supporting Information

378

Supporting Information Available: This material is available free of charge via

379

the Internet.

380

General methods, Supplementary Figure 1-7 and Table 1-7 (PDF).

381 382

REFERENCES

383

1. Liu, J., Zhu, X., Kim, S. J., and Zhang, W. (2016) Antimycin-type

384

depsipeptides:

discovery,

biosynthesis,

385

bioactivities. Nat Prod Rep. 33, 1146-1165.

chemical

synthesis,

and

386

2. Yan, Y., Zhang, L., Ito, T., Qu, X., Asakawa, Y., Awakawa, T., Abe, I., and

387

Liu, W. (2012) Biosynthetic pathway for high structural diversity of a

388

common dilactone core in antimycin production. Org Lett. 14, 4142-4145.

389

3. Sandy, M., Rui, Z., Gallagher, J., and Zhang, W. (2012) Enzymatic

390

Synthesis of Dilactone Scaffold of Antimycins. ACS Chem. Biol. 7,

391

1956−1961

392

4. Yan, Y., Chen, J., Zhang, L., Zheng, Q., Han, Y., Zhang, H., Zhang, D.,

393

Awakawa, T., Abe, I., and Liu, W. (2013) Multiplexing of Combinatorial

394

Chemistry in Antimycin Biosynthesis: Expansion of Molecular Diversity and

395

Utility. Angew Chem Int Ed Engl. 52, 12308-12312.

396

5. Vanner, S. A., Li, X., Zvanych, R., Torchia, J., Sang, J., Andrews, D. W., and 18

ACS Paragon Plus Environment

Page 19 of 27 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

ACS Chemical Biology

397

Magarvey, N. A. (2013) Chemical and biosynthetic evolution of the

398

antimycin-type depsipeptides. Mol Biosyst. 9, 2712-2719.

399

6. Li, X., Zvanych, R., Vanner, S. A., Wang, W., and Magarvey, N. A. (2013)

400

Chemical variation from the neoantimycin depsipeptide assembly line.

401

Bioorg Med Chem Lett. 23, 5123-5127.

402

7. Salim, A. A., Cho, K. J, Tan, L., Quezada, M., Lacey, E., Hancock, J. F., and

403

Capon, R. J. (2014) Rare Streptomyces N-formyl amino-salicylamides

404

inhibit oncogenic K-Ras. Org Lett. 16, 5036-5039.

405

8. Izumikawa, M., Ueda, J. Y., Chijiwa, S., Takagi, M., and Shin-ya, K. (2007)

406

Novel GRP78 molecular chaperone expression down-regulators JBIR-04

407

and -05 isolated from Streptomyces violaceoniger. J Antibiot (Tokyo). 60,

408

640-644.

409

9. Lim, C. L., Nogawa, T., Okano, A., Futamura, Y., Kawatani, M., Takahashi,

410

S., and Ibrahim, D., and Osada, H. (2016) Unantimycin A, a new

411

neoantimycin analog isolated from a microbial metabolite fraction library. J

412

Antibiot (Tokyo). 69, 456-458.

413

10. Umeda, Y., Furihata, K., Sakuda, S., Nagasawa, H., Ishigami, K.,

414

Watanabe, H., Izumikawa, M., Takagi, M., Doi, T., Nakao, Y., and Shin-ya,

415

K. (2007) Absolute structure of prunustatin A, a novel GRP78 molecular

416

chaperone down-regulator. Org Lett. 9, 4239-4242.

417 418

11. Manaviazar, S., Nockemann, P., and Hale, K. J. (2016) Total Synthesis of the

GRP78-Downregulatory

Macrolide

(+)-Prunustatin

A,

the 19

ACS Paragon Plus Environment

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

Page 20 of 27

419

Immunosuppressant (+)-SW-163A, and a JBIR-04 Diastereoisomer That

420

Confirms JBIR-04 Has Nonidentical Stereochemistry to (+)-Prunustatin A.

421

Org Lett.18, 2902-2905.

422

12. Zhou, Y., Murphy, A. C., Samborskyy, M., Prediger, P., Dias, L. C., and

423

Leadlay, P. F. (2015) Iterative Mechanism of Macrodiolide Formation in the

424

Anticancer Compound Conglobatin. Chem Biol. 22, 745-754.

425

13. Skyrud W., Liu J., Thankachan D., Cabrera M., Seipke R.F., Zhang W.

426

(2018) Biosynthesis of the 15-Membered Ring Depsipeptide Neoantimycin.

427

ACS Chem Biol. 13, 1398-1406.

428

14. Zhou, Y., Meng, Q., You, D., Li, J., Chen, S., Ding, D., Zhou, X., Zhou, H.,

429

Bai, L., and Deng, Z. (2008) Selective removal of aberrant extender units

430

by a type II thioesterase for efficient FR-008/candicidin biosynthesis in

431

Streptomyces sp. strain FR-008. Appl Environ Microbiol.74, 7235-7242.

432

15. Olano, C., García, I., González, A., Rodriguez, M., Rozas, D., Rubio, J.,

433

Sánchez-Hidalgo, M., Braña, A. F., Méndez, C., and Salas, J. A. (2014)

434

Activation and identification of five clusters for secondary metabolites in

435

Streptomyces albus J1074. Microb Biotechnol. 7, 242-256.

436

16. Paananen, P., Patrikainen, P., Kallio, P., Mäntsälä, P., Niemi, J., Niiranen, L.,

437

and Metsä-Ketelä, M. (2013) Structural and functional analysis of

438

angucycline C-6 ketoreductase LanV involved in landomycin biosynthesis.

439

Biochemistry.52, 5304-5314.

440

17. Schäfer, M., Le, T. B., Hearnshaw, S. J., Maxwell, A., Challis, G. L., 20

ACS Paragon Plus Environment

Page 21 of 27 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

ACS Chemical Biology

441

Wilkinson,

B.,

and

Buttner,

M.

J.

(2015)

SimC7

Is

a

Novel

442

NAD(P)H-Dependent Ketoreductase Essential for the Antibiotic Activity of

443

the DNA Gyrase Inhibitor Simocyclinone. J Mol Biol. 427, 2192-2204.

444

18. Li, Y., Zhang, W., Zhang, H., Tian, W., Wu, L., Wang, S., Zheng, M., Zhang,

445

J., Sun, C., Deng, Z., Sun. Y., Qu, X., Zhou, J. (2018) Structural Basis of a

446

Broadly Selective Acyltransferase from the Polyketide Synthase of

447

Splenocin. Angew Chem Int Ed Engl. 57, 5823-5827.

448

19. Chang, C., Huang, R., Yan, Y., Ma, H., Dai, Z., Zhang, B., Deng, Z., Liu, W.,

449

and Qu, X. (2015) Uncovering the formation and selection of

450

benzylmalonyl-CoA from the biosynthesis of splenocin and enterocin

451

reveals a versatile way to introduce amino acids into polyketide carbon

452

scaffolds. J Am Chem Soc. 137, 4183-4190.

453

20. Zhang, L., Mori, T., Zheng, Q., Awakawa, T., Yan, Y., Liu, W., and Abe, I.

454

(2015) Rational Control of Polyketide Extender Units by Structure-Based

455

Engineering of a Crotonyl-CoA Carboxylase/Reductase in Antimycin

456

Biosynthesis. Angew Chem Int Ed Engl. 54, 13462-13465.

457

21. Schoenian, I., Paetz, C., Dickschat, J. S., Aigle, B., Leblond, P., and

458

Spiteller, D. (2012) An unprecedented 1, 2-shift in the biosynthesis of the

459

3-aminosalicylate moiety of antimycins. Chembiochem. 13, 769-773.

460

22. Wilkinson, C. J., Hughes-Thomas, Z. A., Martin, C. J., Böhm, I., Mironenko,

461

T.,Deacon, M., Wheatcroft, M., Wirtz, G., Staunton, J., and Leadlay, P.F.

462

(2002).

Increasing

the

efficiency

of

heterologous

promoters

in 21

ACS Paragon Plus Environment

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

463 464

Page 22 of 27

actinomycetes. J. Mol. Microbiol. Biotechnol. 4, 417–426. 23. Gibson, D. G., Young, L., Chuang, R. Y., Venter, J. C., Hutchison, C. A., 3rd,

465

and Smith, H. O. (2009) Enzymatic assembly of DNA molecules up to

466

several hundred kilobases. Nat. Methods. 6, 343–345.

467 468

FIGURE LEGENDS

469

Figure 1. The chemical structures and gene clusters of antimycins and

470

neoantimycins, and heterologous expression of nat gene cluster. (a) The

471

3-FAS moiety is marked with shadow in the structures of the compounds. (b)

472

The genes with different functions are colored with redness. A 29.4 kb

473

fragment consisting of natABCDE was cloned for heterologous expression in

474

the antimycin producer S. albus J1074. The overlapping length between the

475

insert and vector fragments used in Gibson assembly is displayed. (c)

476

Heterologous expression of nat gene cluster was confirmed by HPLC-MS and

477

comparison with the authentic compounds produced by S. conglobatus. The

478

data is shown with the mass extraction of m/z 699.3 for 1 (i) and m/z 685.3 for

479

2 (ii).

480

Figure 2. Proposed NAT biosynthetic pathway.

481

Figure 3. In-frame deletion of natE gene in nat gene cluster (a) and HPLC-MS

482

analysis of the fermentation extract of the mutant RJ8 and WT (wide type)

483

strains. The HPLC-MS data is shown with the mass extraction of m/z 683.3 for

484

4 (i), m/z 697.3 for 3 (ii), m/z 685.3 for 2 (iii) and m/z 699.3 for 1 (iv).

485

Figure 4. The key correlations in 2D NMR of compounds 3 and 4. 22

ACS Paragon Plus Environment

Page 23 of 27 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

ACS Chemical Biology

486

Figure 5. HPLC analysis of the products generated in NatE enzymatic assay.

487

The enzyme activity assay was carried out respectively with the substrates of

488

compounds 3 (a) and 4 (b) produced by RJ8. The compounds 1 and 2 isolated

489

from S. conglobatus were used as standards for verification of the products of

490

the enzyme assay. The HPLC-MS data is displayed with total ion current.

491 492

TABLE

493

Table 1. Cytotoxic activity with IC50 values (nM) of compounds 1-4 Compounds Cell lines A375

Caco2 Hela

3

3.7

24.9

2.9

149.4

1622.0

4.0

82.7

180.2

571.5

862.1

23.6

53.2

12.0

432.3

684.2

2695.0

5764.0

820.9

7299.0

1156.0

296.5

158.1

104.5

794.9

741.7

f

50.0

59.9

223.3

942.5

862.1

g

16.4

14.3

28.0

153.3

434.3

15.9

15.2

11.8

184.9

379.1

b

c

HepG2 Hey

2

a

d

e

MNK28 Sw1990 U343

h

1

4

Cisplatin

494

a

human melanoma cell, b human colon carcinoma cell, c human cervix

495

epidermoid carcinoma cell, d human hepatic carcinoma cell, e human ovarian

496

cancer cell, f human gastric carcinoma cell, g human pancreatic carcinoma cell,

497

h

human glioma cell.

498 499 500 501 23

ACS Paragon Plus Environment

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

502

Page 24 of 27

Figure. 1

503

504 505 506 507 508

24

ACS Paragon Plus Environment

Page 25 of 27 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

ACS Chemical Biology

509

Figure 2.

510 NatB

511

M1

NatG C1

C2

A1

KR 1

A2

O

O

HO

NHCHO

MT

AT

C4

ACP

O

O

O

O

O

O

3, R= CH3; 4, R= H

O

NADPH

O

O

OH

O

NatE

O

O HO

OHCHN

O

NADP+

O

O

O

OH NHCHO OH H N

OHCHN

HN HO

NH 2

T

S

S

O

OHCHN

NatG T

O O

517 O

518 519

NatH-L

NatF

O

O

NH 2

NH 2

O OH NH2

NatQ NatO ?

O

O

R

O

OH

516

O

1

O

OH

HO

S

O O

O

O

HN

515

O

O

O

HN

T

O

S

1

O

OHCHN

TE

R

O

NHCHO OH H N

T4

S

O

OH

KR 3

A4

O

HN

NatO

M5

S

O

HO

M4

O

O O

KS

NatD

T3

O

OHCHN

514

KR 2

A3

S

OH

HN

OH

C3

T2

S

S

513

M3

T1

T

512

M2

NatC

NH 2

O O O O

O

O

COOH

NatN

NH 2

COOH NHCHO

1

O

OH

R

1, R= CH3; 2, R= H NH

C, condensation; A, adenylation; T, thiolation; KR, ketoreduction KS, ketosynthase; AT, acyltransferase; MT, methyltransferase; TE, thioesterase

520 521

Figure 3.

522

523 524 25

ACS Paragon Plus Environment

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

525

Page 26 of 27

Figure 4.

526 HN

CHO HN OH H N

527 528

O

529

O

CHO OH H N

O O O O

O

O

O O O

O

O O O O

O O O

O

530

4

3 531

COSY

532

HMBC

Diagnostic 2D NMR (CDCl3) correlations for 3 and 4 533 534 535

Figure 5.

536

537 538 539 26

ACS Paragon Plus Environment

Page 27 of 27 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

ACS Chemical Biology

540

Graphical Table of Contents

541

M1 C1

M2

A1

C2

KR1

A2

T1

T0 S

S

OH HO NHCHO OHCHN

C3

KR2

A3

T2

S

O

O

O

OHCHN

S

O O

O O

O

OH O

O

OH

O

HN HO

O

1

O

HO

TE

R

O

HN

KR3

A4

T4

O O

OH

O

C4

ACP

O

O O

MT

AT

S O

HN

KS

M5

T3

S

OH

O

O

M4

M3

O

O HN HO

OHCHN

O

O

O

OH

OHCHN

HN HO NHCHO OH H N O O

O O O O

O

1

O O R

NADPH

NatE

NADP +

NHCHO OH H N O O

O

OHCHN

O O O O

O

1

O

O

O

OH

R

27

ACS Paragon Plus Environment