Indole-3-Acetic Acid Produced by Burkholderia heleia Acts as a Phenylacetic Acid Antagonist to Disrupt Tropolone Biosynthesis in Burkholderia plantarii Mengcen Wang 1,2 Seiji Tachibana 1 Yuta Murai 1,3 Li Li 1 Sharon Yu Ling Lau 1 Mengchao Cao 2 Guonian Zhu 2 Makoto Hashimoto 1 Yasuyuki Hashidoko 1 1 Research Faculty of Agriculture, Hokkaido University, Kita 9, Nishi 9, Kita-ku, Sapporo 060-8589, Japan. 2 Institute of Pesticide and Environmental Toxicology, Zhejiang University, No. 268 Kaixuan Road, Hangzhou 310029, China. 3 Frontier Research Center for Post-Genome Science and Technology, Faculty of Advanced Life Sciences, Hokkaido University, Kita 10, Nishi 8, Kita-ku, Sapporo 060-0810, Japan.

Table S1. Isolation and identification of indole-3-acetic acid as B. heleia PAK1-2-derived active compound that represses tropolone production in B. plantarii

The major component repressing tropolone production in fraction 2 (25-50% MeOH) was obtained as a colourless powder (21 mg) with preparative thin-layer chromatography to give spectroscopic data. FD-MS (m/z, rel. int. %): 175.1 (100, [M]+); FD-HR-MS: found 175.0615 (C10H9NO2, calcd. 175.0633); EI-MS (m/z, rel. int. %): 175 (14, [M]+), 130 (100, [M-COOH]+), 103 (10), and 77 (14). 1H-NMR (NON, H-H COSY) and 13C-NMR (BCM, DEPT, HMQC and HMBC) spectra of the tropolone production-repressing principle from B. heleia PAK1-2 was measured in methanol-d4 to be assigned it as indole-3-acetic acid (IAA). Multiplicities of proton and carbon are deduced from DEPT and HMQC data, while C-H correlations including the partial structures are deduced from HMQC.

δH (multiplicity, J in Hz) δC (multiplicity) C-H substitution

The sequence of the 1.4-kbp 16S rRNA gene region (from 79 to 1464 for Escherichia coli 16S ribosomal RNA) from isolate PKA1-2 was subjected to phylogenetic analysis among representative members of genus Burkholderia at the same region using MEGA 6.06 (Wineskin). Multiple alignments of the data collected from NCBI database were run using Clustal W and the phylogenetic tree was constructed using neighbour-joining methods with 1000 bootstrap replicates. Bar, 0.02 nucleotide substitutions per nucleotide site. E. coli is used as an outgroup. Note that B. plantarii and B. heleia are relatively far from each other among the members of genus Burkholderia.

Figure S1. Phylogenetic analysis of the 16S rRNA gene of genus Burkholderia and isolate PAK1-2 using neighbor-joining tree

90

55

93

18

13

4

100

5

56

51

71

31

61

20 99

37

20

97

53

100

78

94

67

89

95

8

63

0.02

91 100

100

20

100

59

83 95 66

89

97

B. tropica MTo293 (AY128103)

B. sordidicola BLN14 (GQ181053) B. sordidicola SF-E2 (LC008478)

B. glathei ATCC29195T (AB021374)

B. sartisoli RG6-9 (EF635856.1) B. phenazinium SF-C1 (LC008479)

B. sabiae Br3407 (NR043180)

B. sediminicola HU265W (NR044383.1)

B. xenovorans LB400 (CP008762.1)

B. megapolitana A3 (NR042594.1)

B. fungorum UFLA04219 (GU144371.1)

B. phytofirmans RFNB13 (FJ266321.1)

B. kururiensis PR1 (JX083379.1) B. tuberum STM6020 (FN908405.1)

B. tropica SCu662 (FJ436048.1)

B. unamae MCo762 (AY221955.1) B. bannensis E25 (NR113178 )

B. mimosarum PTK1 (AY752955.1) B. silvatlantica AB219 (HQ706106)

B. heleia PKA1-2 (AB787501) B. heleia NBRC 101817T (NR113025.1)

B. kirkii 835429 (AF475066.1)

B. rhizoxinica H2199 (HQ005408.1) B. endofungorum HKI456 (NR042584.1)

B. thailandensis (DQ388537) B. pseudomallei V681 (AF093049.1) B. mallei KC1092 (CP009942.1)

B. gladioli ATCC 10248 (CP009323.1) B. gladioli NIAS 1065 (AB024491.1)

B. plantarii NBRC 104885 (AB682219.1) B. glumae LMG 2196 (CP009435.1)

B. multivorans AU0127 (AY486372.1)

B. oklahomensis E0147 (DQ108391.1)

B. ambifaria N422 (AB568317.1) B. cenocepacia LC1 (JN009770.1) B. cepacia NBRC14074T (NR113645) B. cepacia ATCC39277 (AY741347.1)

B. andropogonis 6269 (DQ786950.1)

B. phenazinium ATCC33666 (NR112068.1)

B. phytofirmans PsJN (CP001052.1)

Escherichia coli (J01859.1/AF233451.1 ) B. norimbergensis R2 DMS 11628 (NR036812.1)

B. heleia PAK1-2 �

B. plantarii�

Inhibitory effect of B. heleia PAK1-2 was observed on a B. plantarii-impregnated plate: white arrow indicates the growth background of B. plantarii and black arrow indicates the growth of B. heleia PAK1-2 that were point-inoculated in the center of the plate, while the red arrow indicates the halo due to inhibition of B. plantarii growth.

Figure S2. Colony outgrowth of B. plantarii cultured with B. heleia PAK1-2

Figure S3. Gene expression analysis of AHLs-QS system in B. plantarii exposed to IAA by real time qRT-PCR�

plaI gene � plaR gene �

0.0    

0.5    

1.0    

1.5    

0.0    

0.5    

1.0    

1.5    

Rel

ativ

e ex

pres

sion�

Bar, ± SD (n = 3)

control� control�IAA � IAA �

Expression of plaI and plaR genes in B. plantarii was not significantly inhibited by IAA in contrast to control (P> 0.05). Values are means ± SD (shown by error bars) (n = 3).

Bar, ± SD (n = 3)

Figure S4. Metabolic profile of B. plantarii and characterization of dominant metabolite by GC-MS

Direct EI-MS (A) and FI-MS (B) spectra of a mixture of metabolites extracted with EtOAc from culture fluid of B. plantarii at early culturing stage (24 h). An ion monitoring profile of the mixture in GC-MS (C). EI-MS (D) and FI-MS (E) spectra of the dominant metabolite detected at tR 4.19 min by GC-MS.

A portion of the EtOAc solute (2.5 ml) was concentrated and re-dissolved in methanol-d4 for 1H-NMR analysis without any purification process. In 1H-NMR spectrum of the mixture, major constituent was structurally identical to phenylacetic acid (PAA).

Figure S5. 1H-NMR spectroscopic analysis for a mixture of EtOAc soluble secondary metabolites by B. plantarii

(ppm)

COOH

phenylace,c  acid � tropolone�

Deuterated PAA synthesized was quantified using GC-MS/MS in comparison with authentic PAA (A); this revealed that the benzene ring was substituted by 1 to 5 of deuterium atoms (B-F). Also authentic tropolone (G) was compared with deuterated tropolones converted by B. plantarii from these deuterated PAAs. In the GC-MS/MS, deuterated tropolones maintained 2 to 4 deuterium atoms on the tropolone ring (H-J).

Figure S6. Identification of isotopomeric tropolones from deuterated PAA by GC-MS/MS

(A)    PAA �

(C)    [ring-­‐2H2]PAA�

(E)    [ring-­‐2H4]PAA� (F)    [ring-­‐2H5]PAA�

(G)    tropolone� (H)    [ring-­‐2H2]tropolone�

(I)    [ring-­‐2H3]tropolone� (J)    [ring-­‐2H4]tropolone�

OH

O*

*

* *

OH

O*

* *

OH

O

*

* OH

O

COOH* *

*

*

COOH

* * *

*

OH

O

COOH

*

* COOH

*

*

*

(B)    [ring-­‐2H1]PAA�

(D)    [ring-­‐2H3]PAA�

*

*

O

OH

*

*

O

OH

* O

OH*

* *

* OH

O*

*

*

O

OH*

*

*

*

COOH COOH*

1 2

1′ 2′

4′

3′

5′ 6′

0.0

0.2

0.4

0.6

0.8

1.0

0.0    

0.5    

1.0    

1.5    

2.0    

2.5    

0.0

0.5

1.0

1.5

2.0

Trop

olon

e pr

oduc

tion

(mM

)�

PAA

prod

uctio

n (m

M)� C

ell growth (O

D660 )

B�A�

a �

b � b �c�

cd �d �

e �** �

Bar, ± SD (n = 3) Bar, ± SD (n = 3)

PAA production by B. plantarii following exposure to IAA and other indole derivatives(columns) and cell growth (plots, ◆) quantified from B. plantarii PD broth cultures containing 200 µM indole derivative (A). ** P< 0.01 by Student’s-t test. Tropolone production by B. plantarii after exposure to IAA (B). Diverse alphabet indicates significant difference as determined by Student-Newman-Keuls test. Values (mM) in graphs A and B are means ± SD (shown by error bars). Chemical structures of indole derivatives tested are shown in panel C.

Figure S7. Effect of indole and its derivatives on phenylacetic acid and tropolone production by B. plantarii

Indole-3-acetic acid Indole 2-methylindole

Indole-3-ethanol Indole-3-aldehyde Indole-3-acetonitrile

C �