Structures and Biosynthetic Pathway of Pulvomycins B-D: 22-Membered Macrolides from an Estuarine Streptomyces sp.

Membered Macrolides from an Estuarine Streptomyces sp.

Letter

Structures and Biosynthetic Pathway of Pulvomycins B−D: 22-

⊥

Kyuho Moon, Jinsheng Cui, Eunji Kim, Evan Setiawan Riandi, So Hyun Park, Woong Sub Byun,

Youngju Kal, Jun Young Park, Sunghoon Hwang, Daniel Shin, Jeongyoon Sun, Ki-Bong Oh,

Sangwon Cha, Jongheon Shin, Sang Kook Lee, Yeo Joon Yoon, and Dong-Chan Oh *

Cite This: https://dx.doi.org/10.1021/acs.orglett.0c01249

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: Pulvomycins B−D (1−3) were discovered from an estuarine

Streptomyces strain along with the known pulvomycin (4). The 22-membered

macrocyclic lactone structures of 1−3 were determined based on the interpretation

of NMR, UV, and MS data, the modiﬁed Mosher’s method, and Kishi’s bidentate

chiral solvent NMR spectroscopy. Genomic analysis of the bacterial strain revealed

the biosynthetic gene cluster of pulvomycin and enabled us to propose the trans-

acyltransferase polyketide biosynthetic pathway. Pulvomycin D displayed potent cytotoxic activity against various cancer cell lines.

ulvomycin is an interesting antibiotic discovered from

structure elucidation of 1 because its UV spectrum (λ_m_a_x: 278,

316, and 374 nm) displayed an additional absorption

maximum at 374 nm compared with that of pulvomycin

(λ_m_a_x: 278 and 316 nm), indicating a deviation in the

conjugated double-bond systems.

Streptomyces sp. with a long history since its ﬁrst report by

J. T. Baker Chemical Company in the United States in 1957.

It was rediscovered in 1963 and 1976 and is also known as

labilomycin. The planar structure of pulvomycin was revised

as a novel 22-membered macrocyclic lactone in 1985.

Pulvomycin has a unique mode of action preventing bacterial

translational elongation factor Tu (EF-Tu)·GTP from

complexing with aminoacyl-tRNA. The stereochemistry of

pulvomycin was determined by the X-ray crystallographic

analysis of its complex with Thermus thermophilius EF-Tu in

006. A synthetic study on pulvomycin was reported, but the

total synthesis has not yet been accomplished. Surprisingly,

despite the attractive structure and bioactivity of pulvomycin,

neither additional members of this class nor its biosynthetic

pathway have been reported, even though an early isotope

labeling study proposed that pulvomycin is a polyketide-

derived molecule in 1995.

During our chemical studies of marine-derived actino-

mycetes, Streptomyces sp. HRS33, isolated from the estuary

between the Yellow Sea and the Han River in the Republic of

Korea, produced pulvomycin. A chemical examination of the

strain identiﬁed three new members of the pulvomycin class,

pulvomycins B−D (1−3), prompting the study of their

structures and biosynthesis. Herein we report the structure

elucidation with the comprehensive stereochemical determi-

nation, the bioactivity, and the biosynthetic pathway of

pulvomycins.

The analysis of the H, C, and HSQC NMR spectra

enabled us to assign all of the C− H one-bond correlations

of 1. The structural fragments were connected by the analysis

of the COSY and HMBC data (Figure 1). First, the C-13 to C-

4 chain structure bearing two aliphatic methyl groups (C-44

Received: May 15, 2020

Pulvomycin B (1) was isolated as a yellow powder. The

molecular formula of 1 was determined to be C H O based

66 13

on its HRFAB mass spectrometric data and H and C NMR

data (Table S1). Even though the molecular formula of

pulvomycin B (1) is identical to that of pulvomycin (4), a

comprehensive spectroscopic analysis was required for the

XXXX American Chemical Society

Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters

Letter

To determine the remaining stereogenic center, we utilized

the chemical shift behaviors of the adjacent carbons in Kishi’s

bidentate chiral solvents [(R,R)-,and (S,S)-bis-α-methylbenzyl-

amine-p-Me (BMBA)]. The Δδ

values of the

(

R,R)‑(S,S)

chemical shifts for C-22 and C-24 enabled us to assign the

absolute conﬁguration of C-23 as S (Figure 3). Finally, the

absolute conﬁgurations of pulvomycin B (1) were determined

as 3S, 13S, 21S, 22S, 23S, 24S, 32S, 33S, 35R, 36R, 37S, 38R,

and 39 R .

Figure 1. Key 2D NMR correlations of 1.

and C-45) was assembled based on an array of COSY

correlations among the protons in this spin system. The second

spin system, from C-26 to C-34, was constructed by

consecutive COSY correlations from H-26 to H -34. The

COSY correlations from the anomeric proton H-35 to the

methyl protons H -40 elucidated a sugar moiety. In addition,

the H -2/H-3 COSY correlation secured the C-2−C-3

connectivity. The C-8−C-9−C-10 partial structure was

deciphered based on the H-8/H-9 and H-9/H-10 COSY

correlations. Oleﬁnic methines C-5 and C-6 were connected by

the H-5/H-6 three-bond coupling.

Figure 3. Chemical shift diﬀerences of the indicated carbons of

pulvomycin B (Δδ₍

in a 5:2 mixture of (R,R)- or (S,S)-BMBA

R,R)−(S,S)

and CDCl ).

The analysis of HMBC data further connected these partial

structures and constructed the gross structure of pulvomycin B

Pulvomycin C (2) was isolated as a yellow powder. Its

molecular formula is identical to that of 1 based on high-

resolution fast-atom bombardment mass spectroscopy

(HRFABMS) and NMR data (Table S1). The UV spectrum

of 2 was also quite similar to that of 1, indicating the

conservation of the triene, trienone, and tetraenone chromo-

phores. Further investigation of its 1D and 2D NMR spectra

identiﬁed that pulvomycin C (2 ) is a geometric isomer of 1

with 6Z conﬁguration. (See the Supporting Information.)

Pulvomycin D (3), puriﬁed as a yellow powder, possessed

(

1), as shown in Figure 1. The geometries of the double bonds

were determined from their H− H coupling constants and

ROESY correlations to be 4E, 6E, 8E, 10E, 14E, 16E, 18E, 26E,

8E, and 30E. The sugar was established as labilose by

analyzing the one-bond C−H coupling constant of the

anomeric position and the H− H vicinal coupling constants

and ROESY NMR correlations. (See the Supporting

Information.)

The relative conﬁgurations of the stereogenic centers at C-

1, C-22, C-23, and C-24 were established as 21S*, 22 S *,

the molecular formula C47H O13 based on its HRMS and

3 S*, and 24S* by J-based conﬁguration analysis (Figure

NMR data (Table S1 ). The UV spectrum of 3 displayed only

one absorption maximum (at 314 nm), whereas those of 1 and

2 exhibited three maxima, requiring a more comprehensive

spectroscopic analysis for structure elucidation. A detailed

S1 ). Although the combination of the observed coupling

constants around C-32 and C-33 was most similar to that of

the rotamer shown in Figure S1d, some of the J values were

intermediate values, requiring an additional spectroscopic

analysis for their relative conﬁgurations. A careful analysis of

the ROESY correlations coupled to density functional theory

examination of the H− H COSY and HMBC spectra revealed

that pulvomycin D (3) shares the same C-14 to C-34 fragment,

including the C-44 and C-45 aliphatic methyl groups and the

labilose sugar moiety. However, this compound was distinct

from 1 and 2 by the substructure from C-1 to C-13. The

COSY NMR correlations connected C-2−C-3, C-5−C-6, and

C-8−C-9−C-10 substructures. The constructed chain span-

ning from C-2 to C-11 was ﬂanked by ester carbon C-1 and

(

DFT) modeling supported the 32S* and 33S* conﬁgurations,

whereas the other diastereomer (32S *,33 R *) could not explain

the observed ROESY correlations (Figure S2).

The absolute conﬁgurations of the stereogenic centers at C-

, C-13, C-32, and C-37, each bearing a secondary alcohol,

ketone group C-12 based on the H-2/C-1 and H -43/C-12

were determined using the modiﬁed Mosher’s method. The

hydroxy groups at C-3, C-13, C-32, and C-37 were esteriﬁed

with R- and S-α-methoxy-α-(triﬂuoromethyl) phenylacetyl

chloride (MTPA-Cl) to the tetra-S- and R-MTPA esters (1a

^H^M^B^C^c^r^o^s^s^-^p^e^a^k^s^.^T^h^e^l^a^s^t^c^a^r^b^oⁿ^y^l^c^a^r^b^oⁿ⁽^C^-¹³⁾^a^t^δC

00.1 was assigned adjacent to C-14 by the two-bond C−H

correlation from H-14 to C-13. The connectivity between the

two carbonyl carbons, C-12 and C-13, was conﬁrmed by the

and 1b). The calculated Δδ values established the absolute

S−R

conﬁguration as 3S, 13S, 32S, and 37S (Figure 2). Possibly

because of steric hindrance, the hydroxy group at C-23 was not

derivatized by MTPA-Cl.

JC12H14 coupling. The macrolactone structure was also

conﬁrmed based on the H-21/C-1 HMBC correlation. The

geometries of the oleﬁns in 3 were determined as 3Z, 6Z, 8E,

0E, 14E, 16E, 18E, 26E, 28E, and 30E by their J and

ROESY correlations. To determine the relative stereo-

chemistry of 3, its ROESY spectrum and J-based relative

conﬁgurations were considered, as shown for pulvomycin B

(

1), and this conﬁrmed the con ﬁ gurations of 21S*, 22S*,

3S*, 24S*, 32S* and 33S* (Figure S31). The absolute

conﬁguration of pulvomycin D (3) was established based on

the analysis of a combination of H chemical shifts of bis-S-

and R-MTPA esters (3a and 3b) (Figure S36 ) and the

sequences of the ketoreductase (KR) domains (Table S3 and

Figure S48c).

Figure 2. Δδ values (ppm) of the tetra-S- and R-MTPA esters (1a

S−R

and 1b) of 1 in CD CN.

Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters

Letter

Figure 4. Proposed biosynthetic pathway of pulvomycins. The bimodules 2/3 and 8/9 are represented by light-orange circles. The putative β-

branching enzymes and associated modules 12 and 14 are represented by light-red circles. Two discrete AT and AH domains, PulI and PulO, and

an unknown trans-acting DH domain in module 4 are indicated. The proposed function of PulC is shown between dotted parentheses. The

enzymes and locations putatively involved in the modiﬁcation steps are indicated by blue-gray and dark-orange blocks with arrows. Domain

abbreviations are as follows: ACP, acyl carrier protein; AH, acyl hydrolase; AT, trans-acyltransferase; cMT, C-methyltransferase; dehydrogenase,

short-chain dehydrogenase; DH, dehydratase; ECH, enoyl-CoA hydratase; GT, glycosyltransferase; HCS, hydroxymethylglutaryl-CoA synthase;

KR, ketoreductase; KS, ketosynthase; KS0, decarboxylating KS; KS*, KS-like enzyme; P450, cytochrome P450; TE, thioesterase.

The C-6 to C-19 fragment of pulvomycin D (3) is fully

conjugated with 14 sp carbons and a trienone from C-25 to C-

1, which would likely display two diﬀerent λ_m_a_x; however, the

polyketide synthase (trans-AT PKS) is involved in their

biosynthesis. The analysis of the draft genome sequence of

Streptomyces sp. HRS33 using antiSMASH showed that only

one trans-AT PKS-coding gene cluster is present in the

Streptomyces sp. HRS33 genome. The continuous 103.7 kb

DNA sequence, designated the pul gene cluster, harbors 35

open reading frames (orf’s) that putatively encode four PKSs

that lack integral AT domains, 2 trans-ATs, a β-branching

cassette, 6 enzymes involved in the biosynthesis and the

attachment of sugar moiety, 3 modiﬁcation enzymes, and

others (Figure S47 and Table S2 ). On the basis of the detailed

sequence analysis of the PKS and other genes in the pul cluster,

we propose the trans-AT PKS biosynthetic pathway for

pulvomycins. (See Figure 4 and the Supporting Information.)

The core structure of 3 and 4 bears an unusual β,γ-double

bond at C-3/C-4. It is predicted that the β,γ-double bond is

UV spectrum showed only one λ_m_a_xat 314 nm, corresponding

to a tetraenone chromophore. The energy-minimized structure

of the macrolide part of 3 clearly showed distortion by 80°

between the two ketone groups at C-12 and C-13, breaking the

coplanarity of the 14 conjugated sp carbons into two trienone

chromophores. This resulted in one λ _m _a _x at 314 nm from the

three trienone moieties (Figure S38 ). The calculated UV

spectrum based on the energy-minimized structure consistently

displayed one UV absorption maximum (Figure S39 ).

The discovery of these new pulvomycin class metabolites

and their intriguing structures led us to investigate the

biosynthesis of the pulvomycin family. The β-branched methyl

groups in pulvomycins suggested that a trans-acyltransferase

Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters

The Supporting Information is available free of charge at

Letter

incorporated by module 16. Such shifted double bonds are

found in some polyketides synthesized by trans-AT PKSs and

are known to be formed through either direct β,γ-dehydration

by the DH domain or a double-bond shift by the DH*

ASSOCIATED CONTENT

■

sı Supporting Information

https://pubs.acs.org/doi/10.1021/acs.orglett.0c01249.

5−17

domain.

Because PulH does not contain the double-bond-

shifting DH* domains (Figure S48d), the β,γ-double bond at

the C-3/C-4 position is proposed to be installed directly by the

action of the module 16-encoded DH (Figure 4). One notable

structural diﬀerence between 1/2 and 3/4 is the position of the

double bonds and hydroxy groups across C-3, C-4, and C-5. A

Detailed experimental procedures, NMR spectral data,

and genetic analysis for 1−4 (PDF)

AUTHOR INFORMATION

-hydroxyacyl-CoA dehydrogenase homologue encoded by

■

pulC was predicted to be involved for the conversion of the

pulvomycin D chain into the B chain. PulC belongs to the fadJ

superfamily, and its N-terminal region shows homology to the

C-terminal part of CorN, an enoyl-CoA hydratase-like enzyme

from the corallopyronin cluster, which was proposed to

remove the hydroxy group by dehydration during chain

Corresponding Author

Dong-Chan Oh − Natural Products Research Institute, College

of Pharmacy, Seoul National University, Seoul 08826, Republic

of Korea; orcid.org/0000-0001-6405-5535;

Email: dongchanoh@snu.ac.kr

assembly. Thus it is hypothesized that PulC is involved in the

addition of water to the double bond at the β-position followed

by the migration of the double bond at C-3/C-4 to C-4/C-5

along with the elimination of the hydroxy group at C-5. During

this process, we assume that both the cis (6E) and trans (6Z)

isomers, 5 and 6, are simultaneously formed by PulC-mediated

dehydration at C-5 due to the promiscuity of this enzyme.

Post-PKS tailoring steps are expected to occur in

prepulvomycins 5, 6, and 7. Two putative cytochrome P450s

encoded by pulD and pulP were found in the pul cluster.

Phylogenetic analysis of two P450 enzymes found in the pul

cluster suggested that PulD and PulP are responsible for the

formation of the C-12-carbonyl group and the C-32-hydroxy

Authors

Kyuho Moon − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Korea; College of Pharmacy, Chonnam National University,

Gwangju 61186, Republic of Korea

Jinsheng Cui − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Eunji Kim − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Evan Setiawan Riandi − Natural Products Research Institute,

College of Pharmacy, Seoul National University, Seoul 08826,

Republic of Korea

So Hyun Park − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Woong Sub Byun − Natural Products Research Institute,

College of Pharmacy, Seoul National University, Seoul 08826,

Republic of Korea

Youngju Kal − Department of Chemistry, Dongguk University,

Seoul 04620, Republic of Korea

Jun Young Park − Department of Chemistry, Dongguk

University, Seoul 04620, Republic of Korea

Sunghoon Hwang − Natural Products Research Institute,

College of Pharmacy, Seoul National University, Seoul 08826,

Republic of Korea

Daniel Shin − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Jeongyoon Sun − Department of Agricultural Biotechnology,

College of Agriculture & Life Sciences, Seoul National

University, Seoul 08826, Republic of Korea

Ki-Bong Oh − Department of Agricultural Biotechnology,

College of Agriculture & Life Sciences, Seoul National

University, Seoul 08826, Republic of Korea

group, respectively (Figure S50). Moreover, it is predicted

that a pulQ-encoded glycosyltransferase appends a di-O-

methyl-D-fucose, which is putatively biosynthesized by ﬁ ve

enzymes, PulA, PulB, PulR, PulS, and PulT (Figure S47, Table

S2), on the macrolactone (Figure 4). These modiﬁcation

steps could convert 5, 6, and 7 to 1, 2, and 4, respectively.

Unlike 4, pulvomycin D (3) has a carbonyl group at C-13

instead of a hydroxy group. A short-chain dehydrogenase

encoded by batM is responsible for the reoxidation of hydroxy

group in a late stage of the synthesis of kalimantacin.

Interestingly, two genes homologous to batM, pulZ1 , and

pulZ2 are located ∼56 kb away from pul PKS genes (Figure

S47 ). Either of these two or another gene is expected to

generate the C-13 ketone for 3.

Pulvomycins were initially evaluated in an antibacterial assay

against pathogenic bacteria (Table S4). Pulvomycin exhibited

antibacterial activity against Staphylococcus aureus, Enterococcus

faecalis, E. faecuium, and Salmonella enterica (MIC = 1−2 μg/

mL). In the antifungal assay, only pulvomycin B inhibited the

pathogenic fungi Candida albicans, Trichophyton rubum, and T.

mentagrophytes (MIC = 32 μg/mL) (Table S5).

The anticancer activity of pulvomycin has been limitedly

studied. Only low cytotoxicity against HeLa, Yoshida sarcoma,

Sangwon Cha − Department of Chemistry, Dongguk University,

Seoul 04620, Republic of Korea; orcid.org/0000-0003-

and Ehrlich carcinoma was reported in 1964. Thus the

cytotoxicities of the pulvomycins were then tested against

human cancer cell lines. Pulvomycin D (3), which bears an 1,2-

diketone functional group, was the most potent inhibitor of

colon cancer (HCT116), stomach cancer (SNU638), liver

cancer (SK-Hep-1), and breast cancer (MDA-MB-231) cell

lines (IC₅₀= 0.21−0.40 μM), whereas 1 displayed weak

cytotoxicity (IC₅₀= 3.7−25 μM) (Table S6).

819-3417

Jongheon Shin − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Korea; orcid.org/0000-0002-7536-8462

Sang Kook Lee − Natural Products Research Institute, College of

Pharmacy, Seoul National University, Seoul 08826, Republic of

Korea; orcid.org/0000-0002-4306-7024

Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters

Complete contact information is available at:

Letter

Yeo Joon Yoon − Natural Products Research Institute, College

of Pharmacy, Seoul National University, Seoul 08826, Republic

of Korea; orcid.org/0000-0002-3637-3103

(20) Xu, Z.; Jakobi, K.; Welzel, K.; Hertweck, C. Chem. Biol. 2005,

(

2, 579−588.

21) Mattheus, W.; Gao, L. J.; Herdewijn, P.; Landuyt, B.;

Verhaegen, J.; Masschelein, J.; Volckaert, G.; Lavigne, R. Chem.

https://pubs.acs.org/10.1021/acs.orglett.0c01249

Biol. 2010, 17, 149−159.

(22) Takeuchi, T.; Nitta, K.; Koyama, G.; Hori, M.; Umezawa, H. J.

Antibiot., Ser. A 1964, 17, 124−126.

Author Contributions

^⊥K.M. and J.C. contributed equally to this work.

Notes

The authors declare the following competing ﬁnancial

interest(s): The authors submitted a patent application for

pulvomycin D and its anticancer activity.

ACKNOWLEDGMENTS

■

This work was supported by the National Research

Foundation of Korea Grant funded by the Korean Government

(

Ministry of Science and ICT/2018R1A4A1021703 and

019R1A2B5B03069338) and by the Collaborative Genome

Program of the Korea Institute of Marine Science and

Technology Promotion (KIMST) funded by the Ministry of

Oceans and Fisheries (MOF) (no. 20180430).

REFERENCES

■

(

1) Zief, M.; Woodside, R.; Schmitz, H. Antibiot. Chemother. 1957, 7,

84−386.

(

2) Akita, E.; Maeda, K.; Umezawa, H. J. Antibiot., Ser. A 1963, 16,

47−151.

3) Schwartz, J. L.; Tishler, M.; Arison, B. H.; Shafer, H. M.; Omura,

S. J. Antibiot. 1976, 29, 236−241.

4) Smith, R. J.; Williams, D. H.; Barna, J. C. J.; McDermott, I. R.;

Haegele, K.; Piriou, F.; Wagner, J.; Higgins, W. J. Am. Chem. Soc.

985, 107, 2849−2857.

5) Wolf, H.; Assmann, D.; Fischer, E. Proc. Natl. Acad. Sci. U. S. A.

978, 75, 5324−5328.

6) Parmeggiani, A.; Krab, I. M.; Okamura, S.; Nielsen, R. C.;

(

Nyborg, J.; Nissen, P. Biochemistry 2006, 45, 6846−6857.

(

7) Bo

8) Priestley, N. D.; Gro

9) Pretsch, E.; Buhlmann, P.; Aﬀolter, C. Structure Determination of

rding, S.; Bach, T. Chem. Commun. 2014, 50, 4901−4903.

ger, S. J. Org. Chem. 1995, 60, 4951−4953.

Organic Compounds-Tables of Spectral Data; Springer: New York,

000; p 153.

10) Matsumori, N.; Kaneno, D.; Murata, M.; Nakamura, H.;

Tachibana, K. J. Org. Chem. 1999, 64, 866−876.

11) Freire, F.; Seco, J. M.; Quinoa, E.; Riguera, R. J. Org. Chem.

005, 70, 3778−3790.

12) Kobayashi, Y.; Hayashi, N.; Kishi, Y. Org. Lett. 2002, 4, 411−

14.

(

13) Helfrich, E. J.; Piel, J. Nat. Prod. Rep. 2016, 33, 231−316.

14) Blin, K.; Wolf, T.; Chevrette, M. G.; Lu, X.; Schwalen, C. J.;

Kautsar, S. A.; Suarez Duran, H. G.; de Los Santos, E. L. C.; Kim, H.

U.; Nave, M.; Dickschat, J. S.; Mitchell, D. A.; Shelest, E.; Breitling,

R.; Takano, E.; Lee, S. Y.; Weber, T.; Medema, M. H. Nucleic Acids

Res. 2017, 45, W36−W41.

(

15) Moldenhauer, J.; Go

Schneider, K.; Sussmuth, R. D.; Engeser, M.; Gross, H.; Bringmann,

G.; Piel, J. Angew. Chem., Int. Ed. 2010, 49, 1465−1467.

16) Kusebauch, B.; Busch, B.; Scherlach, K.; Roth, M.; Hertweck,

C. Angew. Chem., Int. Ed. 2010, 49, 1460−1464.

17) Erol, O.; Schaberle, T. F.; Schmitz, A.; Rachid, S.; Gurgui, C.;

El Omari, M.; Lohr, F.; Kehraus, S.; Piel, J.; Muller, R.; Konig, G. M.

ChemBioChem 2010, 11, 1253−1265.

18) Pogorevc, D.; Panter, F.; Schillinger, C.; Jansen, R.; Wenzel, S.

C.; Muller, R. Metab. Eng. 2019, 55, 201−211.

19) Greule, A.; Stok, J. E.; De Voss, J. J.; Cryle, M. J. Nat. Prod. Rep.

018, 35, 757−791.

tz, D. C.; Albert, C. R.; Bischof, S. K.;

(