Manipulation of quorum sensing regulation in Pseudomonas fluorescens NCIMB 10586 to increase mupirocin production

Abstract

Transcription of the 74kb Pseudomonas fluorescens mupirocin [pseudomonic acid (PA)] biosynthesis cluster depends on quorum sensing-dependent regulation via the LuxI/LuxR homologues MupI/MupR. To facilitate analysis of novel PAs from pathway mutants, we investigated factors that affect mup gene expression. First, the signal produced by MupI was identified as N-(3-oxodecanoyl)homoserine lactone, but exogenous addition of this molecule did not activate mupirocin production prematurely
nor did expression of mupI in trans increase metabolite production. Second, we confirmed that mupX, encoding an amidase/hydrolase that can degrade N-acylhomoserine lactones, is also required for efficient expression, consistent with its occurrence in a regulatory module
linked to unrelated genes in P. fluorescens. Third, and most significantly, mupR expression in trans to wild type and mutants can increase production of antibiotic and novel intermediates up to 17-fold.

KeywordsAntibiotic–Polyketide–Quorum sensing–
N-acyl homoserine lactone

Full text

Page 1

APPLIED GENETICS AND MOLECULAR BIOTECHNOLOGY
Manipulation of quorum sensing regulation in Pseudomonas
fluorescens NCIMB 10586 to increase mupirocin production
Joanne Hothersall & Annabel C. Murphy & Zafar Iqbal & Genevieve Campbell &
Elton R. Stephens & Ji’en Wu & Helen Cooper & Steve Atkinson & Paul Williams &
John Crosby & Christine L. Willis & Russell J. Cox & Thomas J. Simpson &
Christopher M. Thomas
Received: 24 November 2010 /Revised: 18 January 2011 /Accepted: 21 January 2011 /Published online: 12 February 2011
# Springer-Verlag 2011
Abstract Transcription of the 74 kb Pseudomonas fluo-
rescens mupirocin [pseudomonic acid (PA)] biosynthesis
cluster depends on quorum sensing-dependent regulation
via the LuxI/LuxR homologues MupI/MupR. To facilitate
analysis of novel PAs from pathway mutants, we investi-
gatedfactors thataffectmup gene expression. First, the signal
produced by MupI was identified as N-(3-oxodecanoyl)
homoserine lactone, but exogenous addition of this molecule
did not activate mupirocin production prematurely nor did
expression of mupI in trans increase metabolite produc-
tion. Second, we confirmed that mupX, encoding an
amidase/hydrolase that can degrade N-acylhomoserine
lactones, is also required for efficient expression, consis-
tent with its occurrence in a regulatory module linked to
unrelated genes in P. fluorescens. Third, and most
significantly, mupR expression in trans to wild type and
mutants can increase production of antibiotic and novel
intermediates up to 17-fold.
Keywords Antibiotic.Polyketide.Quorum sensing.
N-acyl homoserine lactone
Introduction
The 74 kb mupirocin cluster (Figs. 1 and S1) from
Pseudomonas fluorescens NCIMB 10586 encodes polyke-
tide synthases (PKSs) for the biosynthesis of a mixture of
antibacterial pseudomonic acids (PAs), of which PA-A is
the major component (Fig. 2; El-Sayed et al. 2003).
Mupirocin blocks protein synthesis by targeting isoleucyl-
tRNA synthetase of Gram-positive bacteria including
methicillin-resistant Staphylococcus aureus. Approximately
half of the cluster is occupied by four type I PKS genes and
the rest by separate (type II) PKS or putative tailoring
function genes. In-frame deletion analysis of the cluster has
generated novel intermediates or shunt products (Cooper et
al. 2005a, b; Hothersall et al. 2007; Wu et al. 2007, 2008),
but yields are usually very low and the majority of mutants
produce no detectable products. Thus ways to increase
PA metabolite yield may facilitate analysis of products
from such mutants. In some other systems, over-
expression of pathway-specific regulators has resulted
in increased metabolite production, for example, mon-
ensin production was increased with over-expression of
monR (Oliynyk et al. 2003) and tylosin with tylS or tylR
(Stratigopoulos et al. 2004).
Quorum sensing (QS) regulation typically involves the
accumulation of a diffusible signalling molecule which at
a threshold concentration activates a transcriptional
regulator to promote gene transcription. We have previ-
ously shown that expression of the mupirocin cluster is
QS-regulated by homologues of the luxI/R genes, mupI
[N-acylhomoserine lactone (AHL) synthase] and mupR
Electronic supplementary material The online version of this article
(doi:10.1007/s00253-011-3145-2) contains supplementary material,
which is available to authorized users.
J. Hothersall:E. R. Stephens:H. Cooper:C. M. Thomas (*)
School of Biosciences, University of Birmingham,
Edgbaston,
Birmingham B15 2TT, UK
e-mail: c.m.thomas@bham.ac.uk
A. C. Murphy:Z. Iqbal:J. Wu:J. Crosby:C. L. Willis:
R. J. Cox:T. J. Simpson
School of Chemistry, University of Bristol,
Cantock’s Close,
Bristol BS8 1TS, UK
G. Campbell:S. Atkinson:P. Williams
School of Molecular Medical Sciences,
Centre for Biomolecular Sciences, University of Nottingham,
Nottingham NG7 2RD, UK
Appl Microbiol Biotechnol (2011) 90:1017–1026
DOI 10.1007/s00253-011-3145-2

Page 2

(transcriptional regulator), with the result that these genes
are only expressed during late exponential or stationary
phase (El-Sayed et al. 2001). Putative ‘lux’ boxes, which
should bind MupR and mediate activation of expression,
are located upstream of the mupA promoter at the start of
the cluster (El-Sayed et al. 2001) as well as internally at
divergent bidirectional promoters between mupF and
macpC (Hothersall et al. 2007; Fig. 1).
Exogenous addition of QS signalling molecules has been
previously used to increase antibiotic production in other
systems. For example, addition of either synthetic or broth-
extracted AHLs at concentrations above 100 μg/l resulted
in increased pyoluteorin and 2,4-diacetylphloroglucinol
(2,4-DAPG) antibiotic production by P. fluorescens S272
(Nakata et al. 1999). Similarly exogenous addition of AHLs
to Erwinia carotovora ssp. carotovora at the start of growth
resulted in early induction of carbapenem biosynthesis
(Williams et al. 1992).
We therefore set out to identify the specific AHL
produced by MupI. However, neither over-expression of
MupI nor exogenous addition of AHL was found to
increase PA metabolite production. We report in this paper
instead that manipulation of the QS regulatory system
through over-expression of mupR can increase PA metab-
olite yields from wild-type (WT) and tailoring mutants
10586ΔmupC and 10586ΔmupF.
KR
DH
KS ACPACPACPACPACPACP ACPACP ACP ACP ACP
KRKRKR
KR
KRMTMT
TE KSKSKS KSKS KS KS
DHDHDH
trans
ER
?
module 6
module1
module 3
module 4module 2
module 5
load/
transfer
mmpAmmpBmmpD
S
O
OH
S
O
OH
S
O
OH
OH
OH
S
O
OH
OH
OH
S
O
OH
OH
OH
S
O
C15 methyl addition at C3
macpC mupG mupH mupJ mupK
mupirocin
X
S
O
OH
1 kb
+
+
RI
AHL
mmpAmmpB
mmpC
mmpDmmpEmmpF
mupAB
C DE F G HJK
L MNOPQSTU V W R X I
macpAB CDE
Fig. 1 Mupirocin gene cluster organisation. a Summary of the 74 kb
mupirocin biosynthesis gene cluster. MupR (AHL responsive transcrip-
tional activator) and MupI (AHL synthase) are indicated by labelled
circles and AHL by closed circles. Proposed MupR-controlled
promoters are indicated by +. b Proposed scheme for mupirocin
biosynthesis. Ketosynthase (KS), dehydratase (DH), ketoreductase (KR),
C-methyl transferase (MT), acyl carrier protein (ACP), enoyl reductase
(ER), thioesterase (TE). Inactive DH domain crossed out
1018 Appl Microbiol Biotechnol (2011) 90:1017–1026

Page 3

Materials and methods
Bacterial strains, plasmids and growth conditions
The strains and plasmids used and created in this study are
listed in Table 1. The WT P. fluorescens mupirocin
producer NCIMB 10586 has previously been deposited
with the National Collection of Industrial, Food and Marine
Bacteria. The sequence of the mupirocin biosynthesis
cluster is available via GenBank accession no. AF318063.
All other strains and plasmids constructed during this study
are freely available upon request from the corresponding
author. Bacteria were grown on L-broth (LB) and L-agar
(LB supplemented with 1.5%, w/v agar) at 37 °C for
Escherichia coli and 30 °C for P. fluorescens. Plasmid
maintenance and selection of antibiotic-resistant transform-
ants medium were supplemented with appropriate antibiotic
concentrations as follows: 100 μg ampicillin ml-1, 50 μg
kanamycin ml-1and 10–25 μg tetracycline HCl ml-1.
DNA isolation and manipulation
E. coli DH5α was used for plasmid transformation and
propagation. Plasmid deoxyribonucleic acid (DNA) extraction
Table 1 Strains and plasmids used in this study
Strain or plasmidRelevant characteristicsReference or source
Pseudomonas fluorescens
NCIMB 10586
10586ΔmupC
10586ΔmupF
10586ΔmupI
10586ΔmupX
Escherichia coli
DH5α
WT mupirocin producer
mupC deletion mutant
mupF deletion mutant
mupI deletion mutant
mupX deletion mutant
(Whatling et al. 1995)
(Hothersall et al. 2007)
(Hothersall et al. 2007)
(El-Sayed et al. 2001)
(Cooper et al. 2005b)
F-f80dlacZΔM15 recA1 endA1 gyrA96 thi-1 hsdR17
supE44 relA1 deoR Δ(lacZYA-argF) U169
recA pro hsdR RP4-2-Tc::Mu-Km::Tn7
trpC2
(Hanahan 1983)
S17-1
Bacillus subtilis 1064
Plasmids
pAKE106
pGEMT-easy
pJH1
pJH10
pJH2
pJH210
pSB401
pSB536
pSB1075
pSSCX
(Simon et al. 1983)
(Moir et al. 1979)
Apr, Kmr, oriT, sacB, mupA fused with xylE
Apr, lacZα, T tailed PCR cloning vector
pJH10 containing mupI
Tcr; lacIq, tacp, IncQ replicon
pJH10 containing mupR
pJH10 containing mupR and mupX
pACYC184 containing luxRI′::luxCDABE
pUC19 containing ahyRI′:: luxCDABE
pUC18 containing luxRI′::luxCDABE
pJH10 containing mupX
(El-Sayed et al. 2001)
Promega
(El-Sayed et al. 2001)
(El-Sayed et al. 2003)
This study
This study
(Winson et al. 1998)
(Swift et al. 1997)
(Winson et al. 1998)
(Cooper et al. 2005b)
Fig. 2 Structure of mupirocin
(PA). PA consists of a C17unit
(monic acid) derived from an
unsaturated polyketide contain-
ing a tetrahydropyran ring, and a
C9saturated fatty acid
(9-hydroxynonanoic acid).
Mupirocin is a mixture of four
PAs A–D, with A being the
predominant form
Appl Microbiol Biotechnol (2011) 90:1017–10261019

Page 4

was performed essentially by the alkaline sodium dodecyl
sulphate method (Birnboim and Doly 1979) or with Wizard
Plus SV Minipreps DNA Purification Systems (Promega).
Extraction of DNA from agarose was performed using
GeneClean kit (Bio101). Polymerase chain reaction (PCR)
fragments were initially cloned into pGEMT-Easy (Promega)
for sequencing. Competent E. coli cells were transformed
with plasmid DNA using the method of Cohen et al. (1972).
DNA sequencing was carried out using the BigDye Termi-
nator kit (PE-ABI). The sequencing reactions were separated
on an ABI 3700 DNA Analyser.
Bi-parental mating
For bi-parental mating to mobilise suicide and expression
plasmidsfromE. coli S17-1 to P. fluorescens, a mixture of late
exponential phase E. coli S17-1 (0.5 ml) and P. fluorescens
(0.5 ml) was filtered on to a 0.45 μm sterile Millipore filter,
incubated overnight on L-agar at 30 °C, resuspended in sterile
saline solution and aliquots spread on M9 minimal medium or
L-agar supplemented with the appropriate antibiotic to select
for the presence of the plasmid and against E. coli S17-1.
Isolation and bioautography of AHLs
For AHL extraction, Pseudomonas strains were grown in
mupirocin production medium (MPM; Whatling et al. 1995)
for 40 h at 30 °C and E. coli strains in LB for 20 h at 37 °C.
AHLs were extracted essentially as previously described
(McClean et al. 1997), adjusting to pH 2.5 with HCl prior to
extraction with dichloromethane (7:3 supernatant:dichloro-
methane). Samples (1–5 μl) and standard (20–200 μM) were
resolved on C18 reversed phase thin layer chromatography
(TLC) plates in 60% methanol. The point to which the
solvent front-migrated was recorded. AHLs were then
detected by bioautography. The plate was air-dried to remove
residual methanol and then overlaid with L-agar seeded with
P. fluorescens ΔmupI, mupA::xylE, and incubated overnight
at 30 °C. The agar was sprayed with 0.3 M catechol and the
production of a yellow colour was noted. The Rfvalue of
detected compounds was calculated. This is the ratio of
compound migration to solvent front migration.
Synthesis and LC-MS/MS analysis of AHLs
N-butanoyl-L-homoserine lactone (C4-HSL), N-(3-
oxobutanoyl)-L-homoserine lactone (3-O-C4-HSL),
N-hexanoyl-L-homoserine lactone (C6-HSL) and N-(3-
oxohexanoyl)-L-homoserine lactone (3-O-C6-HSL),
N-octanoyl-L-homoserine lactone (C8-HSL), N-(3-
oxooctanoyl)-L-homoserine lactone (3-O-C8-HSL),
N-decanoyl-L-homoserine lactone (C10-HSL), N-(3-
oxodecanoyl)-L-homoserine lactone (3-O-C10-HSL),
N-dodecanoyl-L-homoserine lactone (C12-HSL), N-(3-
oxododecanoyl)-L-homoserine lactone (3-O-C12-HSL), N-
tetradecanoyl-L-\homoserine lactone (C14-HSL) and N-(3-
oxotetradecanoyl)-L-homoserine lactone (3-O-C14-HSL)
were synthesised as described before (Chhabra et al. 1993,
2005). Methods for liquid chromatography mass spectrom-
etry (LC-MS)/MS are as described in Ortori et al. (2007).
FT-ICR MS analysis of AHLs
Samples were diluted 1:1 into 2.5% formic acid (v/v). MS
experiments were conducted on a Thermo Finnigan LTQ
Fourier transform (FT) mass spectrometer (Thermo Fisher
Scientific, Bremen, Germany). Samples were introduced to
the mass spectrometer using an Advion Biosciences
Triversa Nanomate electrospray source (Advion Bioscien-
ces, Ithaca, NY). Data acquisition and analysis was
performed using the Xcalibur 2.0 software (Thermo Fisher
Scientific). Mass spectra were acquired at a resolution of
100,000 at m/z 400. Automatic gain control was used to
accumulate optimum numbers of ions (target value 3×105).
Fragmentation was induced by infrared multiphoton disso-
ciation (IRMPD) in the ion cyclotron resonance (ICR) cell.
Precursor ions were isolated in the linear ion trap prior to
transfer to the ICR cell. Photons for IRMPD were provided
by a 75 W built-in CO2 laser (Synrad, Mikilteco, WA,
USA). Precursor ions were irradiated for 100 ms. Each MS
and MS/MS scan comprised one microscan. Mass spectra
shown comprise ~100 co-added scans. Data were analysed
manually by use of Xcalibur software.
Exogenous supply of AHL and well bioassay
WT and 10586ΔmupI grown to OD6000.1 in 50 ml LB
were harvested by centrifugation (14,000 × g, 5 min) and
resuspended in 50 ml MPM AHL medium comprised of
90% filter-spent MPM medium from 16 h 10586ΔmupR
cultures and 10% fresh MPM medium. WT bacteria,
resuspended in fresh MPM medium, were included as a
control. These were incubated at 30 °C, 200 rpm and
samples taken at 1 h intervals for 3 h and a final sample
after 24 h. Cells were harvested as above and the
supernatant filter-sterilised (0.2 μm) and stored at -20 °C.
Alternatively, 50 ml MPM containing 2 μM 3-O-C10-HSL
was inoculated with 1% WT and 10586ΔmupI 16 h LB
cultures. Controls without exogenous AHL were included.
Samples were taken hourly for 6 h and again after 24 h as
above. Biological activity was monitored by loading
150 μl filtered supernatant into 8-mm-diameter wells cut
into LB seeded with 4% Bacillus subtilis 1064 16 h
culture and 5% triphenyltetrazolium chloride. Plates were
incubated at 37 °C for 16 h and clear zones of inhibition
measured.
1020 Appl Microbiol Biotechnol (2011) 90:1017–1026

Page 5

xylE assay
A xylE reporter vector pAKE106 (El-Sayed et al. 2001),
which has an internal fragment of mupA fused with a xylE
reporter gene, was used to monitor mupA expression.
Overnight, LB cultures were diluted 20-fold into 25 ml
MPM (Whatling etal. 1995) and incubated for 20 h at 30 °C,
200 rpm. Aliquots (1.2 ml) were pelleted at 15.000 × g for
10 min, resuspended in 0.5 ml sonication buffer, then
measured quantitatively for xylE activity and protein con-
centration as described previously (Zukowski et al. 1983).
Protein concentration was determined as described by
Gornall et al. (1949).
AHL degradation assays
To determine whether MupX was capable of degrading
AHLs, a 96-well microtitre plate assay was used in
combination with lux-based AHL biosensors (Winson et
al. 1998). A wide range of unsubstituted, 3-oxo and 3-
hydroxy AHLs dissolved in acetonitrile were added to a 96-
well microtitre plate and the solvent allowed to evaporate.
An overnight culture (100 μl) of either E. coli (pSCCX) or
the vector control E. coli (pJH10) grown in LB at 37 °C
were added to the AHL-coated wells and the plate
incubated at 37 °C overnight. To each well, 100 μl of the
appropriate AHL biosensor [E. coli (pSB536), E. coli
(pSB401) or E .coli (pSB1075); Swift et al. 1997; Winson
et al. 1998] grown in LB were added and incubated for 4 h
at 37 °C. Bioluminescence output from each well was
quantified using a LB980 photon video camera (EG &
E Berthold).
To demonstrate that MupX possesses amidase/acylase
activity, crude cell lysates prepared from E. coli (pSCCX)
and E. coli (pJH10) were incubated with C10-HSL, and any
free amines released were chemically trapped using dansyl
chloride as described previously (Uroz et al. 2005).
Construction of pJH2 and pJH210 expression vectors
mupR was amplified as a 717 bp fragment incorporating a
DraI site in front of the start codon (71,638 bp) and an SstI
site after the stop codon (72,342 bp). This was cloned into
pJH10 (El-Sayed et al. 2001), an IncQ broad host range
plasmid, digested with EcoRI which was blunt-ended by
filling in with Klenow, and SstI, to generate the vector pJH2
expressing MupR under the control of the tac promoter. A
1.7 kb PCR product encompassing the end of mupR from
an internal EcoRI site (72,218 bp) and the whole of mupX
was amplified generating a terminal SacI site after the
mupX stop codon (73,918 bp) and cloned into pJH2
digested with EcoRI and SacI to generate the vector
pJH210 expressing both mupR and mupX.
HPLC analysis of PA
Seed cultures grown for 16 h in MPM (Whatling et al.
1995) at 25 °C, 200 rpm were diluted 20-fold into 25 ml
MPM and incubated for 40 h at 22 °C, 200 rpm. Aliquots
(1 ml) were centrifuged at 15,000 × g for 10 min, and the
supernatant stored at -20 °C. Samples were filtered
(0.2 μm) prior to analysis. High-performance liquid
chromatography (HPLC) was performed using either
Gilson 712 or Unipoint LC system software, a reversed
phase C18 column (15 cm×4.6 mm, 5 μm), ultraviolet
detection at 233 nm, and a mobile phase water/acetonitrile
gradient [5–70% acetonitrile TFA (0.01%)] over 30 min at a
flow rate of 1 ml min-1.
Isolation and characterisation of mupiric acid, mupirocin H,
PA-A7, 1, and PA-B7, 2
Seed culture of P. fluorescens (pJH2) grown for 16 h in
50 ml of LB-medium supplemented with 10 μg/ml
tetracycline at 25 °C, 200 rpm was diluted 20-fold into 1 l
secondary stage medium [25 g/l soya flour, 2.5 g/l spray-
dried corn liqour, 5 g/l (NH4)2SO4, 0.5 g/l MgSO4∙7H2O,
1 g/l Na2HPO4, 1.5 g/l KH2PO4, 1 g/l KCl, 6.25 g/l CaCO3
and 4% glucose, adjusted to pH 7.5 with NaOH(aq)] in 10×
500 ml baffled conical flasks and grown at 22 °C, 250 rpm
for 50 h. Cells were removed by centrifugation at
7,500 rpm for 30 min. The supernatant was acidified to
pH 4.5 with dilute HCl, extracted with ethyl acetate (0.6
vol) twice, and the organic solvent was removed in vacuo
to give 580 mg crude extract. The crude extract (96 mg)
was purified by HPLC using a Phenomenex Luna C1810×
250 mm column (room temperature), 4 ml/min, A = H2O +
0.05% formic acid, B = MeOH, 0–5 min isocratic 25% B,
5–51 min gradient 25%–95% B. This yielded <0.2 mg of
mupiric acid (Rt25.4 min; Wu et al. 2008), 0.8 mg of
mupirocin H (Rt29.7 min; Wu et al. 2007), and 1.1 mg of 1
(Rt34.2 min). A separate purification of 98 mg of the crude
extract using a 48 min gradient of 50%–75% B on the same
system yielded 0.8 mg of impure 2 (Rt18.0 min). Further
purification using a Phenomenex Luna C184.6×250 mm,
5 μm column (40 °C), 1 ml/min A = H2O+0.05% formic
acid, B = MeOH, 0–2 min isocratic 20% B, 2–45 min
gradient 20%–95% B yielded ~0.3 mg pure 2 (Rt27.6 min).
1H and13C nuclear magnetic resonance (NMR) spectra
were determined on a Varian VNMRS500 spectrometer
and were assigned using1H–1H correlation spectroscopy,
DEPT, heteronuclear multiple-quantum correlation and
heteronuclear multiple-bond correlation (HMBC) spectra.
High-resolution electrospray ionisation mass spectra
(HRESIMS) were obtained on a Bruker Daltronics Apex
4e 7.0T FT-ICR mass spectrometer. Samples were introduced
via a nanospray (Nanomate) sample inlet as above.
Appl Microbiol Biotechnol (2011) 90:1017–10261021

Page 6

Results
Identification of AHLs produced by MupI
One possible way to upregulate the mupirocin pathway could
be to increase the levels of pathway-specific AHL either
through in trans expression of mupI or exogenous addition of
AHL. To identify the diffusible QS signal produced by MupI
supernatant extracts of WT P. fluorescens, NCIMB 10586
were analysed by reversed phase TLC against synthetic
C8-HSL, C10-HSL, 3-O-C10-HSL and C12-HSL standards.
The separated signal molecules could then be detected by
bioautography overlaying a mupA::xylE fusion reporter strain
defective in mupI. The WT extract contained a compound
that co-migrated with 3-O-C10-HSL and that diffused into
the overlay agar seeded with reporter strain and suppressed
the mupI mutation. This allowed expression of the mupirocin
cluster and promoted XylE activity from the xylE gene
inserted into the first open reading frame (ORF) mupA. The
bioautograph also showed that transcription of the mup
cluster can be activated by the synthetic standards C8-HSL,
C10-HSL, 3-O-C10-HSL and C12-HSL (Fig. 3a & b).
Supernatant extracts from E. coli with either expression
plasmid pJH1 (El-Sayed et al. 2001) containing mupI or
pJH10 with no insert were analysed by positive electrospray
FT-ICR MS (Fig. S2). A peak with m/zmeas 270.1699
(m/zcalc270.1700; C14H24O4N1), which corresponds to 3-
O-C10-HSL, was present in E. coli (pJH1) but not E. coli
(pJH10). FT-ICR MS analysis of chemically synthesised 3-
O-C10-HSL (Fig. S3) produced a peak at m/zmeas270.1701
which on fragmentation yielded two peaks: m/z 102.0550
(C4H8O2N1) corresponding to the homoserine lactone ring
cleaved between the amide bond and C1 of the acyl chain;
and m/z 169.1224 (C10H17O2) corresponding to the 3-
oxo-C10 acyl chain. The fragmented extract of E. coli
(pJH1) m/z 270.1699 peak yielded both these peaks, but
also produced an m/z 144.0655 peak (C6H10O2N1)
corresponding to an alternative fragmentation of the
AHL between the C2 and C3 of the acyl chain, which
yields N-acetylhomoserine lactone (Chhabra et al. 2005).
Solvent extracts from WT P. fluorescens and mupI mutant
were also analysed by positive electrospray FT-ICR MS.
The 3-O-C10-HSL, m/zmeas270.1699, was present in WT
but not the mupI mutant (Fig. S4). Further LC-MS/MS
spectrometry of WT P. fluorescens and E. coli (pJH1)
extracts showed that while 3-O-C10-HSL was the major
AHL; they also produce C8, C10 and C12-HSLs as the
unsubstituted, 3-oxo and 3-hydroxyl compounds (Fig. S5).
Provision of exogenous 3-O-C10-HSL does not increase
production of mupirocin
Attempts were first made to switch on production early by
supply of AHL either through growth in spent medium
containing AHLs or through addition of exogenous 2 μM 3-
O-C10-HSL at inoculation as described in ‘Materials and
methods’. Exogenous supply of 3-O-C10-HSL, however, did
not result in early or enhanced mupirocin production. Second,
we expressed mupI in trans to WT to investigate whether
higher levels of MupI could lead to increased mupirocin
production. Surprisingly, mupirocin production was not
increased and indeed may decrease slightly when higher
levels of IPTG were used to induce expression of mupI.
mupX is required for mup expression
P. fluorescens 2P24, a biocontrol strain producing polyketide
antibiotic 2,4-DAPG (Maurhofer et al. 2004), was recently
reported to encode close homologues of MupR, MupX and
MupI (PcoR 89% aa identity, PcoX 87% aa identity and PcoI
89% aa identity) with the same genetic organisation. pcoX is
intergenic between pcoR and pcoI and divergently orientated
with respect to pcoI (accession no. Wei and Zhang 2006).
This may suggest that mupX is part of a cassette that has been
acquired to regulate mupirocin biosynthesis. A role in
regulation would be consistent with our previous observation
that an in-frame deletion of mupX reduces mupirocin
production but otherwise leaves a normal product profile
(Cooper et al. 2005b). Introducing a xylE reporter gene fusion
into mupA, the first ORF of the mupirocin cluster, in WT and
Fig. 3 Bioautography of AHLs produced by P. fluorescens NCIMB
10586 and from E. coli expressing MupI. a WT, 10586ΔmupX and E.
coli expressing mupI from pJH1 each produce an AHL that co-
migrates with 3-O-C10-HSL. 10586ΔmupI does not produce this
AHL. b The mupirocin cluster is also activated by C8-HSL, C10-HSL
and C12-HSL
1022 Appl Microbiol Biotechnol (2011) 90:1017–1026

Page 7

in a mupX deletion mutant, showed that mupX is indeed
important for full expression of mupA (Fig. 4).
MupX degrades AHLs
Blast searches predicted MupX should possess amidase or
hydrolase activity (e.g. 45% aa identity to hydrolase from
Gemmatimonas aurantiaca T-27, BAH39344; 44% aa iden-
tity to amidase from Anaeromyxobacter dehalogens 2CP-C,
ABC83200). MupX may therefore influence mupirocin
production by modulating AHL levels and perhaps degrade
minor AHLs produced by MupI, eliminating those that might
block activation of MupR. The ability of E. coli expressing
mupX to degrade AHLs was analysed using a wide range of
chain lengths (from C4 to C14) with or without an acyl chain
3-oxo substituent. Surprisingly, while there was variation in
the activity, in general, MupX was active against each of the
AHLs examined (Fig. S5). MupX was more efficient at
degrading the N-3-oxoacyl-HSLs than the N-acyl-HSLs and
was more active against short-chain C4- to C8-HSLs than the
longer-chain C10- to C14-AHLs (data not shown).
To confirm that MupX has amidase/acylase activity, crude
cell extracts were prepared from E. coli pSCCX and the
vector control, pJH10, incubated with C10-HSL and any
homoserine lactone released captured by chemical trapping of
its free amine with dansyl chloride (Uroz et al. 2005). HPLC
analysis revealed a reduction in C10-HSL levels alongside a
concomitant increase in dansylated homoserine lactone levels
for the MupX containing extract (data not shown). No
reduction in C10-HSL or increase in dansylated homoserine
lactone levels were observed for the vector control.
When expressed in trans in the WT P. fluorescens from
pSCCX (Cooper et al. 2005b), MupX did not reduce mupA::
xylE transcription nor reduce mupirocin nor AHL production.
Conversely, comparison of AHL production by bioautogra-
phy showed that WT and a mupX mutant produced similar
amounts of a compound with an Rfvalue similar to that of the
3-O-C10-HSL standard which suppressed the mupI mutation
(Fig. 3). The identity of this compound was confirmed as 3-
O-C10-HSL by LC-MS/MS. These results indicate that
MupX does not work via AHL degradation.
trans MupR increases PA metabolite production
To determine if we could increase PA-A production through
increased expression of the QS transcriptional regulator
MupR, pJH2 containing mupR and pJH210 containing both
mupR and mupX were introduced into WT P. fluorescens.
HPLC analysis of PA-A production showed that the
presence of either plasmid increased production of PA-A
fourfold to fivefold (Fig. 5).
Expression of mupR in trans to WT has also enabled the
detection of minor shunt products of the mupirocin biosyn-
thesis pathway, mupiric acid and mupirocin H, previously
only identified in deletion mutants (Wu et al. 2008), proving
that these are genuine natural products of the PA biosynthetic
pathway and not simply by-products of genetic mutation. In
addition, related metabolites present in such trace quantities
that have not been observed previously were isolated from
this upregulated WT—PA- A7, 1 and PA-B7, 2 (Fig. 6).
Metabolite 1 was shown by HRESIMS,1H, DEPT and 2D
NMR experiments to be a homologue of PA-A with a
truncated C7side chain in place of the C9side chain reported
for all PAs to date (Table S1). The low titre obtained of 2
limited the data obtainable via 2D NMR experiments (key
correlations for 1 and 2 shown in Fig. 6); however,
comparison with NMR data for PA-B and HRESIMS allowed
the structure to be confidently assigned as a homologue of
PA-B with a truncated C7side chain (Table S1). Comparison
of all coupling constants in 1 and 2 with those of PA-A and
PA-B confirmed that they have the same relative stereo-
chemistries as the previously reported PAs.
We tested whether in trans mupR increased production of
novel PA intermediates in mup mutants. Previously, a mupC
in-frame deletion was shown to reduce PA-A production by
HPLC and to produce a new peak with retention time
22.1 min determined to be a novel PA mupirocin C
(Hothersall et al. 2007). The mupirocin C yield was
extremely low, but when mupR was expressed in trans, the
peak area increased 17-fold (Fig. 5). Similarly, a deletion
mutant of mupF produces a number of novel metabolites at
extremely low levels (Fig. 5, peaks 20, 21.4 and 22 min).
The major metabolite was deduced to be mupirocin F
(Hothersall et al. 2007). Expressing mupR in trans improved
metabolite yields, for example, the 22 min peak was
increased fourfold. Expression of mupR in trans therefore
provides a general strategy to increase yields of PA.
Fig. 4 Knockout mutagenesis of mupR and mupX reduces mupA
expression as shown by levels of a xylE reporter fused with mupA.
n=14 for WT, otherwise n=6, and error bars represent standard
deviation
Appl Microbiol Biotechnol (2011) 90:1017–10261023

Page 8

Discussion
We have shown that P. fluorescens NCIMB 10586 produces
primarily 3-O-C10-HSL together with a range of minor
AHLs (including C8-HSL, C10-HSL and C12, unsubstituted,
3-oxo and 3-hydroxyl compounds). 3-O-C10-HSL is suffi-
cient for expression of the mupirocin cluster. MupI and MupR
amino acid sequences are consistent with their cognate signal
molecule being a long-chain AHL oxidised at the C3 position
(Collins et al. 2005; Gould et al. 2006; Watson et al. 2002).
P. fluorescens is reported to produce a wide range of AHLs in
a strain-specific manner (Cha et al. 1998; Laue et al. 2000;
Shaw et al. 1997). However, it is common for a single species
to produce numerous AHL signals, and a single LuxI
homologue can produce AHLs with different chain lengths
(Parsek et al. 1999; Schaefer et al. 1996).
A close homologue of MupI, PcoI (89% identity) from P.
fluorescens 2P24, also produces several QS signals, two of
which, when separated by TLC, co-migrated with 3-O-C6-
HSL and 3-O-C8-HSL but were not characterised further
(Wei and Zhang 2006). P. fluorescens 2P24 is a biocontrol
strain producing polyketide antibiotic 2,4-DAPG (Maurhofer
et al. 2004) but does not produce mupirocin. P. fluorescens
2P24 also has close homologues of mupR and mupX, pcoR
(89% identity) and pcoX (87% identity), respectively. The
pcoIXR cluster has the same organisation as the mupRXI
cluster, and thus this may represent a novel regulatory
module characterised by the extra mupX/pcoX gene which
has been acquired and specifically adapted by P. fluorescens
NCIMB 10586 to control mupirocin biosynthesis.
Biologically, AHLs are inactivated by lactonases cleav-
ing the lactone ring (Dong et al. 2002) and by acylases
cleaving the amide bond (Lin et al. 2003). MupX, showing
homology to amidases and hydrolases, might cleave the
AHL amide bond and thus degrade AHLs (endogenous or
exogenous). Speculation on the natural roles of such
enzymes includes exploitation of AHLs as carbon sources
(Leadbetter and Greenberg 2000) and competitive interfer-
ence with QS in other bacteria (Dong et al. 2004). In
Pseudomonas aeruginosa, PvdQ, which was originally
identified as an AHL acylase/amidase, in fact, functions in
the periplasmic maturation of the fluorescent siderophore,
pyoverdine (Koch et al. 2010).
When homologues of these degrading enzymes are also
found in strains that produce AHLs, their roles may be to
ensure the correct timing of a quorate response (Zhang et al.
2002) or to maintain correct ratios of different AHL signal
molecules (Huang et al. 2003). It is conceivable that the
variety of minor AHL signals produced by P. fluorescens
NCIMB 10586 could act as inhibitors of MupR and be
degraded by MupX. However, the lack of substrate
specificity and the fact that over-expression of MupX in
Fig. 6 Key HMBC correlations observed for novel PA shunt products
1 and 2
Fig. 5 Expression of mupR from pJH2 or mupR and mupX together
from pJH210 in trans increased PA-A production from WT to fivefold
and novel intermediate production from mutants of mupC and mupF
to 17-fold and fourfold, respectively
1024 Appl Microbiol Biotechnol (2011) 90:1017–1026

Page 9

trans did not increase or decrease mup expression while a
mupX knockout decreased expression suggests a more subtle
role. For example, it may remove bound 3-O-C10-HSL from
MupR to ensure mup expression is only maintained when the
population and levels of de novo signal are quorate.
Over-expression of mupR in trans to WT increased PA-A
production fivefold, while in the ΔmupC tailoring region
mutant novel metabolite (mupirocin C) production was
increased 17-fold, and in the mupF mutant, mupirocin F
production went up fourfold. Isolation and characterisation
of these novel metabolites and proposed activities of MupC
and MupF are described elsewhere (Hothersall et al. 2007).
Over-expression of mupR also enabled the observation of
the minor shunt products, mupiric acid and mupirocin H,
from WT P. fluorescens, whereas, previously, these had
only been seen in deletion mutants. This supports our
postulation that their production in deletion mutants is via
adventitious release at labile points along the biosynthetic
pathway and indicates that these alternative release pro-
cesses can compete, albeit very poorly, with the main flux
through the pathway. The novel truncated PA homologues,
1 and 2, also appear to represent alternative selectivity by
the PKS resulting in early release of the growing polyke-
tide. They may also provide evidence for the stepwise
biosynthesis of 9-hydroxynonanoic acid on a monic acid
precursor rather than its separate biosynthesis and then
esterification to monic acid.
The major conclusion from this work is that expression of
the positive transcriptional regulator, MupR, is a rate-limiting
step in mupirocin production. Over-expression of luxR/mupR
homologues and associated genes in their original strain may
therefore provide a general way of increasing yields of
metabolites engineered to be under QS control.
Acknowledgement
P15257 and P07071, BBSRC/EPSRC grant E021611 employing JH,
ES and AM, as well as EPSRC grant S78124 that employed JW. GC
was supported by a University of Nottingham Studentship and work in
PW’s laboratory was supported by BBSRC. DNA sequencing was
performed by the JIF-funded Genomics Laboratory in the School of
Biosciences (6/JIF13209). HPLC/LCMS equipment was funded by
EPSRC grant EP/F066104). We thank Miguel Camara at the
University of Nottingham for his input and comments and Ram
Chhabra and Alex Truman for AHL synthesis.
This work was supported by BBSRC grants
References
Birnboim HC, Doly J (1979) A rapid alkaline extraction procedure for
screening recombinant plasmid DNA. Nucleic Acids Res
7:1513–1523
Cha C, Gao P, Chen YC, Shaw PD, Farrand SK (1998) Production of
acyl-homoserine lactone quorum-sensing signals by Gram-
negative plant-associated bacteria. Mol Plant Microbe Interact
11:1119–1129
Chhabra SR, Stead P, Bainton NJ, Salmond GPC, Stewart GSAB,
Williams P, Bycroft BW (1993) Autoregulation of carbapenem
biosynthesis in Erwinia carotovora ATCC 39048 by analogues of
N-(3-oxohexanoyl)-L-homoserine lactone. J Antibiot 46:441–454
Chhabra SR, Philipp B, Eberl L, Givskov M, Williams P, Camara M
(2005) Extracellular communication in bacteria. In: Schulz S (ed)
Chemistry of pheromones and other semiochemicals. Springer,
pp 279–315
Cohen SN, Chang ACY, Hsu H (1972) Non-chromosomal antibiotic
resistance in bacteria; genetic transformation of Escherichia coli
by R factor DNA. Proc Natl Acad Sci USA 69:2110–2114
Collins CH, Arnold FH, Leadbetter JR (2005) Directed evolution of
Vibrio fischeri LuxR for increased sensitivity to a broad spectrum
of acyl-homoserine lactones. Mol Microbiol 55:712–723
Cooper SM, Cox RJ, Crosby J, Crump MP, Hothersall J, Laosripai-
boon W, Simpson TJ, Thomas CM (2005a) Mupirocin W, a novel
pseudomonic acid produced by targeted mutation of the
mupirocin biosynthetic gene cluster. Chem Commun 9:1179–
1181
Cooper SM, Laosripaiboon W, Rahman AS, Hothersall J, El-Sayed
AK, Winfield C, Crosby J, Cox RJ, Simpson TJ, Thomas CM
(2005b) Shift to pseudomonic acid B production in P. fluorescens
NCIMB 10586 by mutation of mupirocin tailoring genes mupO,
mupU, mupV, and macpE. Chem Biol 12:825–833
DongYH,GustiAR,ZhangQ,XuJL,ZhangLH(2002)Identificationof
quorum-quenching N-acyl homoserine lactonases from Bacillus
species. Appl Environ Microbiol 68:1754–1759
Dong Y-H, Zhang X-F, Xu J-L, Zang Z-H (2004) Insecticidal Bacillus
thuringiensis silences Erwinia carotovora virulence by a new
form of microbial antagonism, signal interference. Appl Environ
Microbiol 70:954–960
El-Sayed AK, Hothersall J, Thomas CM (2001) Quorum-sensing-
dependent regulation of biosynthesis of the polyketide antibiotic
mupirocin in Pseudomonas fluorescens NCIMB 10586. Micro-
biology 147:2127–2139
El-Sayed AK, Hothersall J, Cooper SM, Stephens E, Simpson TJ,
Thomas CM (2003) Characterisation of the mupirocin biosyn-
thesis gene cluster from Pseudomonas fluorescens NCIMB
10586. Chem Biol 10:419–430
Gornall AG, Bardawill CJ, David MM (1949) Determination of serum
proteins by means of the biuret reaction. J Biol Chem 177:751–
766
Gould TA, Herman J, Krank J, Murphy RC, Churchill MEA (2006)
Specificity of acyl-homoserine lactone synthases examined by
mass spectrometry. J Bacteriol 188:773–783
Hanahan D (1983) Studies on transformation of Escherichia coli with
plasmids. J Mol Biol 166:557–580
Hothersall J, Wu J-E, Rahman AS, Shields JA, Haddock J, Johnson N,
Cooper SM, Stephens E, Cox RJ, Crosby J, Willis CL, Simpson
TJ, Thomas CM (2007) Mutational analysis reveals that all
tailoring region genes are required for production of polyketide
antibiotic mupirocin by Pseudomonas fluorescens: pseudomonic
acid B biosynthesis precedes pseudomonic acid A. J Biol Chem
282:15451–15461
Huang JJ, Han JI, Zhang LH, Leadbetter JR (2003) Utilization of acyl-
homoserine lactone quorum signals for growth by a soil
Pseudomonad and Pseudomonas aeruginosa PAO1. Appl Envi-
ron Microbiol 69:5941–5949
Koch G, Jimenez PN, Muntendam R, Chen YX, Papaioannou E, Heeb
S, Camara M, Williams P, Cool RH, Quax WJ (2010) The
acylase PvdQ has a conserved function among fluorescent
Pseudomonas spp. Environ Microbiol Rep 2:433–439
Laue BE, Jiang Y, Chhabra SR, Jacob S, Stewart GSAB, Hardman A,
Downie JA, O’Gara F, Williams P (2000) The biocontrol strain
Pseudomonas fluorescens F113 produces the Rhizobium small
bacteriocin, N-(3-hydroxy-7-cis-tetradecenoyl) homoserine lac-
Appl Microbiol Biotechnol (2011) 90:1017–10261025

Page 10

tone, via HdtS, a putative novel N-acylhomoserine lactone
synthase. Microbiology 146:2469–2480
Leadbetter JR, Greenberg EP (2000) Metabolism of acyl-homoserine
lactone quorum-sensing signals by Variovorax paradoxus. J
Bacteriol 182:6921–6926
Lin YH, Xu JL, Hu J, Wang LH, Ong SL, Leadbetter JR, Zhang LH
(2003) Acyl-homoserine lactone acylase from Ralstonia strain
XJ12B represents a novel and potent class of quorum-quenching
enzymes. Mol Microbiol 47:849–860
Maurhofer M, Baehler E, Notz R, Martinez V, Keel C (2004) Cross
talk between 2, 4-diacetylphloroglucinol-producing biocontrol
Pseudomonads on wheat roots. Appl Environ Microbiol
70:1990–1998
McClean KH, Winson MK, Fish L, Taylor A, Chhabra SR, Camara M,
Daykin M, Lamb JH, Swift S, Bycroft BW, Stewart GS, Williams P
(1997) Quorum sensing and Chromobacterium violaceum: exploi-
tation of violacein production and inhibition for the detection of N-
acylhomoserine lactones. Microbiology 143:3703–3711
Moir A, Lafferty E, Smith DA (1979) Genetic analysis of spore
germination mutants of Bacillus subtilis 168, the correlation of
phenotype with map location. J Gen Microbiol 111:165–168
Nakata K, Yoshimoto A, Yamada Y (1999) Promotion of antiobiotic by
highethanol,highNaClconcentration,orheatshockinPseudomonas
fluorescens S272. Biosci Biotechnol Biochem 63:293–297
Oliynyk M, Stark CBW, Bhatt A, Jones MA, Hughes-Thomas ZA,
Wilkinson C, Oliynyk Z, Demydchuk Y, Staunton J, Leadlay PF
(2003) Analysis of the biosynthetic gene cluster for the polyether
antibiotic monensin in Streptomyces cinnamonensis and evidence
for the role of monB and monC genes in oxidative cyclization.
Mol Microbiol 49:1179–1190
Ortori CA, Atkinson S, Chhabra SR, Camara M, Williams P, Barret
DA (2007) Comprehensive profiling of N-acylhomoserine lac-
tones produced by Yersinia pseudotuberculosis using liquid
chromatography coupled to hybrid quadrupole-linear ion trap
mass spectrometry. Anal Bioanal Chem 387:497–511
Parsek MR, Val DL, Hanzelka BL, Conran JE Jr, Greenberg EP
(1999) Acyl homoserine-lactone quorum-sensing signal genera-
tion. Proc Natl Acad Sci USA 1999:4360–4365
Schaefer AL, Val DL, Hanzelka BL, Conran JE Jr, Greenberg EP
(1996) Generation of cell-to-cell signals in quorum sensing: acyl
homoserine lactone synthase activity of a purified Vibrio fischeri
LuxI protein. Proc Natl Acad Sci USA 93:9505–9509
Shaw PD, Ping G, Daly SL, Cha C, Cronan JE Jr, Rinehart KL, Farrand
SK (1997) Detecting and characterizing N-acylhomoserine lactone
signal molecules by thin-layer chromatography. Proc Natl Acad Sci
USA 94:6036–6041
Simon R, Priefer U, Puhler A (1983) A broad host range mobilization
system for in vivo genetic-engineering-transposon mutagenesis in
Gram-negative bacteria. Biotechnol NY 1:784–791
Stratigopoulos G, Bate N, Cundliffe E (2004) Positive control of tylosin
biosynthesis: pivotal role of TylR. Mol Microbiol 54:1326–1334
Swift S, Karlyshev AV, Fish L, Durant EL, Winson MK, Chhabra SR,
Williams P, MacIntyre S, Stewart G (1997) Quorum sensing in
Aeromonas hydrophila and Aeromonas salmonicida: identifica-
tion of the LuxRI homologs AhyRI and AsaRI and their cognate
N-acylhomoserine lactone signal molecules. J Bacteriol
179:5271–5281
Uroz S, Chhabra SR, Camara M, Williams P, Oger P, Dessaux Y
(2005) N-acylhomoserine lactone quorum-sensing molecules are
modified and degraded by Rhodococcus erythropolis W2 by both
amidolytic and novel oxidoreductase activities. Microbiology
151:3313–3322
Watson WT, Minogue TD, Val DL, von Bodman SB, Churchill MEA
(2002) Structural basis and specificity of acyl-homoserine lactone
signal production in bacterial quorum sensing. Mol Cell 9:685–
694
Wei HL, Zhang LQ (2006) Quorum-sensing system influences root
colonization and biological control ability in Pseudomonas fluo-
rescens 2P24. Antonie Leeuwenhoek 89:267–280
Whatling CA, Hodgson JE, Burnham MKR, Clarke NJ, Franklin
FCH, Thomas CM (1995) Identification of a 60 kb region of the
chromosome of Pseudomonas fluorescens NCIB 10586 required
for the biosynthesis of pseudomonic acid (mupirocin). Microbi-
ology 141:973–982
Williams P, Bainton NJ, Swift S, Chhabra SR, Winson MK, Stewart
G, Salmond GPC, Bycroft BW (1992) Small molecule-mediated
density-dependent control of gene-expression in prokaryotes—
bioluminescence and the biosynthesis of carbapenem antiobiotics.
FEMS Microbiol Lett 100:161–167
Winson MK, Swift S, Fish L, Throup JP, Jorgensen F, Chhabra SR,
Bycroft BW, Williams P, Stewart G (1998) Construction and
analysis of luxCDABE-based plasmid sensors for investigating
N-acyl homoserine lactone-mediated quorum sensing. FEMS
Microbiol Lett 163:185–192
Wu J, Cooper SM, Cox RJ, Crosby J, Hothersall J, Simpson TJ,
Thomas CM, Willis CL (2007) Mupirocin H, a novel metabolite
resulting from mutation of the HMG-CoA synthase analogue,
mupH in Pseudomonas fluorescens. Chem Commun 20:2040–
2042
Wu J, Hothersall J, Mazzetti C, O’Connell Y, Shields JA, Rahman AS,
Cox RJ, Crosby J, Simpson TJ, Thomas CM, Willis CL (2008) In
vivo mutational analysis of the mupirocin gene cluster reveals
labile points in the biosynthetic pathway: the “Leaky Hosepipe”
mechanism. Chembiochem 9:1500–1508
Zhang HB, Wang LH, Zhang LH (2002) Genetic control of quorum-
sensing signal turnover in Agrobacterium tumefaciens. Proc Natl
Acad Sci USA 99:4638–4643
Zukowski MM, Gaffney DF, Speck D, Kauffman M, Findelli A,
Wisecup A, Lecocq JP (1983) Chromogenic identification of
genetic regulatory signals in Bacillus subtilis based on expression
of a cloned Pseudomonas gene. Proc Natl Acad Sci USA
80:1101–1105
1026Appl Microbiol Biotechnol (2011) 90:1017–1026