Direct cloning, genetic engineering, and heterologous expression of the syringolin biosynthetic gene cluster in E. coli through Red/ET recombineering.

Abstract

The reconstruction of a natural product biosynthetic pathway from bacteria in a vector and subsequent heterologous expression in a technically amenable microbial system represents an efficient alternative to empirical traditional methods for functional discovery, yield improvement, and genetic engineering to produce "unnatural" derivatives. However, the traditional cloning procedure based on genomic library construction and screening are complicated due to the large size (>10 kb) of most biosynthetic pathways. Here, we describe the direct cloning of a partial syringolin biosynthetic gene cluster (sylCDE, 19 kb) from a digested genomic DNA mixture of Pseudomonas syringae into a plasmid in which sylCDE is under the control of an inducible promoter by one step linear-plus-linear homologous recombination (LLHR) in Escherichia coli. After expression in E. coli GB05-MtaA, two new syringolin derivatives were discovered. The complete syringolin gene cluster was assembled by addition of sylAB and exchange of a synthetic bidirectional promoter against the native promoter to drive sylB and sylC expression by using Red/ET recombineering. The varying production distribution of syringolin derivatives showed the different efficiencies of native and synthetic promoters in E. coli. The successful reconstitution and expression of the syringolin biosynthetic pathway shows that Red/ET recombineering is an efficient tool to clone and engineer secondary metabolite biosynthetic pathways.

Full text

Page 1

DOI: 10.1002/cbic.201200310
Direct Cloning, Genetic Engineering, and Heterologous
Expression of the Syringolin Biosynthetic Gene Cluster in
E. coli through Red/ET Recombineering
Xiaoying Bian,[a]Fan Huang,[b, c]Francis A. Stewart,[b]Liqiu Xia,[c]Youming Zhang,*[d]and
Rolf M?ller*[a]
Introduction
Naturally occurring secondary metabolites have been and will
continue to be an indispensable source of new chemical enti-
ties and lead compounds for drugs and agrochemicals for the
foreseeable future.[1]Most of the therapeutically and agricultur-
ally useful natural products from bacteria are biosynthesized
by multifunctional megasynthases, the polyketide synthases
(PKSs)[2]and the nonribosomal peptide synthetases (NRPSs),[3]
or hybrids thereof.[4]During past decades, numerous PKS and
NRPS biosynthetic pathways have been cloned, sequenced,
and even reconstituted for heterologous expression in altera-
tive hosts (e.g., E. coli,[5,6]Streptomycetes,[7,8]or Pseudomo-
nads[9–11]) to improve fermentation or to generate new natural
or “unnatural” products for further evaluation as drug leads.
Certainly, reassembling biosynthetic gene clusters in a versatile
vector for heterologous expression in suitable hosts represents
a promising avenue to investigate the biosynthetic function of
unknown gene clusters or to produce new derivatives based
on the pharmacologically active compounds by molecular en-
gineering or combinatorial biosynthesis approaches.
The genetic engineering of natural product biosynthetic
pathways, especially type I PKS/NRPS gene clusters, is difficult
through conventional DNA engineering technology because of
their sizes. Engineering becomes possible through in vivo ho-
mologous recombination-based Red/ET recombineering[12–14]in
E. coli, which is independent of the location of restriction sites
and the size of DNA fragments. This technology has extraordi-
narily advanced the genetic manipulation of complex prokary-
otic biosynthetic pathways by omitting many steps in standard
restriction/ligation genetic modification. Several complete bio-
synthetic pathways from fastidious bacteria have been re-
assembled and engineered by construction and screening of
BAC or cosmid libraries, followed by Red/ET recombineering-
mediated stitching and modification of biosynthetic pathways
for the purpose of heterologous biosynthesis in more techni-
cally amenable microbial systems.[9,10,15–18]These recombineer-
ing approaches relied on a linear-plus-circular DNA molecule
homologous recombination (LCHR). Recently, LLHR, mediated
by the full-length Rac prophage protein RecE and its partner
The reconstruction of a natural product biosynthetic pathway
from bacteria in a vector and subsequent heterologous expres-
sion in a technically amenable microbial system represents an
efficient alternative to empirical traditional methods for func-
tional discovery, yield improvement, and genetic engineering
to produce “unnatural” derivatives. However, the traditional
cloning procedure based on genomic library construction and
screening are complicated due to the large size (>10 kb) of
most biosynthetic pathways. Here, we describe the direct clon-
ing of a partial syringolin biosynthetic gene cluster (sylCDE,
19 kb) from a digested genomic DNA mixture of Pseudomonas
syringae into a plasmid in which sylCDE is under the control of
an inducible promoter by one step linear-plus-linear homolo-
gous recombination (LLHR) in Escherichia coli. After expression
in E. coli GB05-MtaA, two new syringolin derivatives were dis-
covered. The complete syringolin gene cluster was assembled
by addition of sylAB and exchange of a synthetic bidirectional
promoter against the native promoter to drive sylB and sylC ex-
pression by using Red/ET recombineering. The varying produc-
tion distribution of syringolin derivatives showed the different
efficiencies of native and synthetic promoters in E. coli. The
successful reconstitution and expression of the syringolin bio-
synthetic pathway shows that Red/ET recombineering is an
efficient tool to clone and engineer secondary metabolite bio-
synthetic pathways.
[a] X. Bian,+Prof. Dr. R. M?ller
Helmholtz-Institut f?r Pharmazeutische Forschung Saarland
Helmholtz Zentrum f?r Infektionsforschung und
Pharmazeutische Biotechnologie, Universit?t des Saarlandes
Postfach 151150, 66041 Saarbr?cken (Germany)
E-mail: rom@helmholtz-hzi.de
[b] F. Huang,+Prof. F. A. Stewart
Department of Genomics, Biotech, TU Dresden
01307 Dresden (Germany)
[c] F. Huang,+Prof. L. Xia
Department of Molecule Microbiology, College of Life Science
Hunan Normal University
410081 Changsha (P. R. China)
[d] Dr. Y. Zhang
Gene Bridges GmbH, Building C2 3, Saarland University
66123 Saarbr?cken (Germany)
E-mail: youming.zhang@genebridges.com
[+ +] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201200310.
1946
? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2012, 13, 1946–1952

Page 2

RecT, was elaborated and applied to clone large gene clusters
directly from genomic DNA, thereby avoiding construction and
screening of genomic libraries.[19]Ten hidden biosynthetic
pathways were cloned from digested genomic DNA of Photo-
rhabdus luminescens by using this method, and two of these
were successfully expressed in E. coli to discover their prod-
ucts.[19]Establishment of this efficient cloning approach toward
large biosynthetic pathways could greatly promote the course
of genome mining and combinatorial biosynthesis.
Syringolin belongs to syrbactin, which stands for a new class
of potent proteasome inhibitors acting by the irreversible and
covalent attachment between the a,b-unsaturated carbonyl
group of the syrbactin macrocyclic core and the active site
threonine 1 of the 20S eukaryotic proteasome through a Mi-
chael-type 1,4-addition.[20,21]Consistent with their proteasome
inhibitory activity, studies proved that syrbactin induced apop-
tosis in neuroblastoma, ovarian, and leukemic cancer cells.[22–25]
Syringolins, a family of six structures closely related to NRPS–
PKS hybrid molecules, were first isolated from plant pathogen-
ic bacterium Pseudomonas syringae pv. syringae B301D-R in a
special medium but were not produced in the rich LB medi-
um.[26,27]The major variant, syringolin A (SylA, 1; Scheme 1) has
a structure which contains a 12-membered ring comprising
two nonproteinogenic amino acids (5-methyl-4-amino-2-hexe-
noic acid and 3,4-dehydrolysine) and an exocyclic N-terminal
acylation through a valine-ureido-valine linkage which is creat-
ed by head-to-head condensation of two amino acids.[26]The
minor variants, syringolin B–F (SylB–F, 2–6) differ from SylA by
replacement of 3,4-dehydrolysine with lysine and/or replace-
ment ofoneor bothvaline
(Scheme 1).[27]It has been shown that the biosynthesis of syrin-
golin is catalyzed by a combined megasynthetase consisting of
PKS and NRPS and involves five genes (sylA–sylE).[28]The gene
sylC encodes an NRPS which consists of single C, A, and T do-
mains, and it has been demonstrated in vivo and in vitro that
residues withisoleucine
this enzyme can iteratively activate two amino acid monomers
(valine and/or isoleucine) and create the ureido dipeptide
group by capture of bicarbonate/CO2. The gene sylD encodes
a hybrid protein containing both typical NRPS and PKS mod-
ules that activate and condense two amino acids, 3,4-dehydro-
lysine/lysine and valine, and one malonyl-CoA to form the
macrocyclic moiety of syringolin.[28–30]Genes sylA, sylB, and sylE
encode a putative transcription activator, a desaturase-like pro-
tein which is supposed to introduce a double bond into lysine,
and the exporter of syringolin, respectively (Scheme S1).[28]Sy-
ringolin A has been heterologously expressed in Pseudomonas
putida, which is close to the native producer strain P. syringae,
to confirm that genes sylA–E are sufficient to execute the bio-
synthesis of syringolin A.[30]
Herein, we report the direct cloning of the syringolin gene
cluster from genomic DNA of P. syringae by RecE/RecT-mediat-
ed LLHR in E. coli. As a result of heterologous expression of
sylCDE in E. coli GB05-MtaA,[19]two new syringolin members,
syringolin G and H, were produced and characterized. The
intact syringolin biosynthetic pathway was assembled by addi-
tion of sylAB to form the identical gene organization as found
in the native producer. In addition, the promoter region be-
tween sylB and sylC was changed to a synthetic bidirectional
promoter to interrogate the production of syringolin variants
in E. coli.
Results and Discussion
Direct cloning of the partial syringolin gene cluster sylCDE
Red/ET recombineering is an in vivo homologous recombina-
tion-based genetic engineering method employed primarily in
E. coli by using short homology arms, catalyzed by two equiva-
lent phage protein pairs, Reda/Redb from l phage and RecE/
RecT from Rac prophage, which are located in the E. coli K12
chromosome.[14,31,32]This technique not only greatly facilitated
genetic engineering of large complex biosynthetic pathways,
but also can be applied to clone large gene clusters from
a complex DNA source into a plasmid by LLHR.[19]In this inves-
tigation, we directly cloned sylCDE (19 kb) from the digested
genomic DNA of P. syringae by one round of LLHR with an effi-
ciency of 3/12 (Scheme 2). All other colonies examined were
empty circularization of linear vectors. In this construct, sylCDE
was under the control of a tetracycline-inducible promoter
rather than the native promoter from P. syringae. This direct
cloning method can be used to clone large chosen DNA re-
gions (10–52 kb)[19]from DNA mixtures directly into a plasmid
in E. coli. This leads to many fewer mutations and much longer
target DNA fragments than could possibly be achieved by PCR,
because their cloning is dependent on the E. coli replication
machinery and not on PCR, which is prone to errors. Obviously,
the method is much easier than genomic library construction
and screening, and is achieved by single- or two-step recombi-
nation reactions.
Scheme 1. Structures of six known syringolins, as well as the newly discov-
ered syringolins G and H.
ChemBioChem 2012, 13, 1946–1952? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
1947
The Syringolin Biosynthetic Gene Cluster

Page 3

Identification of previously unknown natural products
syringolin G and H by heterologous expression of sylCDE
The construct pASK-Ptet-sylCDE
harbors the sylCDE gene but
lacks sylB, which encodes a desa-
turase that is thought to con-
verse lysine to 3,4-dehydroly-
sine;[28,33]this construct was in-
troduced into the heterologous
host E. coli GB05-MtaA by elec-
troporation. The transformants
were verified by restriction di-
gestion analysis, cultivated in LB
medium, and induced by tetra-
cycline. The metabolite profiles
of the mutant strains were ana-
lyzed by HPLC-MS and com-
pared to those of the natural sy-
ringolin producer P. syringae. As
anticipated, all known syringo-
lins A, B, C, D, E, and F (SylA–
SylF, 1–6) were detected in the
extract of the natural producer
(Figure S1) and none were de-
tected in E. coli GB05-MtaA (Fig-
ure 1B). The lysine-desaturated
syringolins (1, 3, 4, and 6, in
which R3–R4is a double bond)
were completely eradicated as
proposed in the tetracycline-induced extract of heterologous
host-carrying plasmids, but the lysine-saturated syringolins
Scheme 2. Direct cloning and heterologous expression of the partial syringolin biosynthetic gene cluster sylCDE to discover new derivatives. Genomic DNA
was completely digested with DraI and NotI to release the DNA fragments containing the target biosynthetic pathway. The linear cloning vector flanked with
homology arms to the targeted gene was generated by PCR. The predigested genomic DNA mixture was mixed with a linear cloning vector and cotrans-
formed into recombineering proficient E. coli YZ2005 cells by electroporation. The homologous recombination between two linear DNA molecules, the clon-
ing vector and the targeted sylCDE genes, can be completed by using RecE/RecT in E. coli. The resulting recombinants carrying sylCDE under the control of
a Ptet-inducible promoter were identified by selection for the ampicillin resistance gene present on the linear cloning vector and subsequent DNA restriction
analysis and sequencing. The verified plasmid pASK-Ptet-sylCDE was electroporated into the heterologous host E. coli GB05-MtaA for expression and the cor-
responding natural products were identified by HPLC/MS analysis. ampR: ampicillin resistance gene, tetR: tetracycline-binding promoter repressor, ori: origin
of replication, Ptet: tetracycline-inducible promoter.
Figure 1. HPLC/MS analysis (base peak chromatograms (BPCs) m/z 494–525 + All MS) of extracts from fermenta-
tion of strains A) E. coli GB05-MtaA/pASK-Ptet-sylCDE and B) E. coli GB05-MtaA. The lysine-saturated syringolins: sy-
ringolin B (2, m/z 496 [M+ +H]+), syringolin E (5, m/z 510 [M+ +H]+), syringolin G (7, m/z 524 [M+ +H]+) and syringo-
lin H (8, m/z 510 [M+ +H]+) are plotted in chromatogram A. C) The MS fragmentation pattern and D) MS2 spectra
of lysine-saturated syringolins are also shown. In (D), the mass spectra are labeled to indicate the fragments lost
to generate each daughter peak, according to (C). The peak indicated by an asterisk represents unknown substan-
ces.
1948
www.chembiochem.org
? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, WeinheimChemBioChem 2012, 13, 1946–1952
Y. Zhang and R. M?ller et al.

Page 4

(SylB, 2, and SylE, 5) were observed (Figure 1A). Moreover, two
new syringolin candidates, which were designated as syringol-
in G (SylG, 7, m/z 524 [M+ +H]+) and syringolin H (SylH, 8, m/z
510 [M+ +H]+) (Figure 1), were also detected in the E. coli GB05-
MtaA/pASK-Ptet-sylCDE, differing from 6 and 3 by two atomic
mass units, respectively. This finding is consistent with the ex-
pected addition of two hydrogen atoms due to the absence of
sylB (Schemes 1 and S1, and Figure 1D). Both new syringolins
were identified as lysine-saturated syringolins because they
were present together with 2 and 5 when sylB was absent.
Both 7 and 8 can also be detected in the syringolin natural
producer but overlap with their corresponding lysine-saturated
products, 6 and 3, and their yields are quite low in the native
producer compared to 6 and 3 (Figure S1). Feeding of com-
mercially available labeled l-leucine and l-isoleucine to the LB
culture was used to establish the relationship with syringolins
and, more importantly, to confirm that either one or two l-iso-
leucine residues, rather than l-leucine, were incorporated into
7 or 8, respectively (Figure S4). This finding is in agreement
with the structure proposed here for 6 and 3. High-resolution
MS was performed to suggest the elemental composition as
C26H46N5O6 for 7 (m/z 524.3452 [M+ +H]+, calcd: 524.3443;
Dppm=1.715), with a two hydrogen atom (2H) difference
compared to 6 (C26H44N5O6; m/z 522.3292 [M+ +H]+, calcd:
522.3286; Dppm=1.052), and C25H44N5O6for 8 (m/z 510.3297
[M+ +H]+, calcd: 510. 3286; Dppm=2.037) with a 2H difference
compared to 3 (C25H42N5O6; m/z 508.3136 [M+ +H]+, calcd: 508.
3130; Dppm=1.317). Because the production yield of the new
derivatives was too low for NMR studies, 7 and 8 were further
characterized by MS/MS fragmentation in positive ionization
mode together with subsequent comparison of the MS2 fin-
gerprints[27]with those obtained for other lysine-saturated sy-
ringolins (2 and 5) or corresponding lysine-desaturated syrin-
golins (6 and 3). From the MS2 positive mode spectra (Fig-
ure 1D, Figure S3, Figure S5–S8, Table S1), the MS fragmenta-
tion pattern of 7 was highly similar to 6, while 8 was quite sim-
ilar to 3, except for a gap of two atomic mass units in most of
the fragment ions, and showed the addition of two hydrogen
atoms which are most likely located on the macrocyclic core.
The exocyclic moieties of 7 and 8 are identical to 6 and 3, re-
spectively, with two isoleucines at the ring-proximal and distal
positions for 7 and one isoleucine at the ring-proximal position
and one valine at the distal position for 8. The 12-membered
ring contains two carbon–carbon double bonds: one in the
3,4-dehydrolysine residue moiety that was proposed to be
generated by SylB,[28]and the other in the 5-methyl-4-amino-2-
hexenoic acid residue which is biosynthesized by the second
NRPS module (C-A-PCP, valine) and the PKS module (KS-AT-DH-
KR-ACP, malonyl-CoA) of SylD.[28]The genetic function of sylB[28]
and MS fragmentation analysis of all syringolin derivatives
(Table S1) strongly suggest structures of 7 and 8 (Scheme 1).
The discovery of syringolin G and H not only enrich the syrin-
golin family but also confirm the function of sylB in vivo. Until
now, the natural syringolin family consisted of eight members
with three modifications: 3,4-dehydrolysine or lysine at the tri-
peptide macrocycle, valine or isoleucine at either the macrocy-
cle-adjoining position or distal position (Scheme 1). Additional-
ly, the peak shown with an asterisk in Figure 1A increased in
comparison to its counterpart in Figure 1B. We have analyzed
this peak, showing m/z 489 or 499 [M+ +H]+, and could not
correlate it to tetracycline or syringolin derivatives. We believe
that it represents some unknown compounds from the
medium.
Assembly and expression of syringolin gene cluster
sylAB was inserted upstream of sylC of plasmid pASK-Ptet-
sylCDE (Scheme 3A)by LLHR
(Scheme 3B) containing the entire syl gene cluster. The genetic
toform pASK-sylABCDE
organization and promoters are identical to the original pro-
ducer P. syringae. This plasmid was transformed into E. coli
GB05-MtaA for expression of syringolins. All eight syringolins
can be found in these extracts from M9 medium (Figure 2A)
and LB (trace amount, data not shown). The yield in minimal
M9 medium was much higher than that in rich LB medium,
which is consistent with the fact that in the original producer
P. syringae, the syringolins can only be generated in a nutrient-
poor medium (SRMAF) but not in LB medium.[26]Nevertheless,
the relative yield of lysine-desaturated syringolins was relative-
ly lower than that of the corresponding lysine-saturated syrin-
Scheme 3. Reassembly and engineering of syringolin gene cluster through
Red/ET recombineering. Firstly, sylAB was inserted into pASK-Ptet-sylCDE (A)
by LLHR to delete the Ptet promoter to form pASK-sylABCDE (B), in which
the syl gene organization is identical to the native producer. Next, the native
promoter region between sylB and sylC was replaced with a bidirectional
synthetic promoter cassette by selection for zeocin resistance to obtain con-
struct C. The sylB-bizeo cassette amplified from C was inserted into A to give
rise to pASK-sylB-bizeo-sylCDE (D) in which sylB and sylC were placed under
the control of a bidirectional synthetic promoter. zeoR: zeocin resistance
gene, Red/ET: Red/ET recombineering, LCHR: linear-plus-circular homologous
recombination, LLHR: linear-plus-linear homologous recombination.
ChemBioChem 2012, 13, 1946–1952? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
1949
The Syringolin Biosynthetic Gene Cluster

Page 5

golins (Figure 2A), which contradicts the distribution in the
native producer (Figure S1). It seems that sylB in the heterolo-
gous host does not efficiently convert its substrate, which may
be due—at least in part—to its native promoter being less
active in E. coli. Thus, we attempted to introduce a synthetic
bidirectional promoter between sylB and sylC to investigate
the relative abundance of syringolins. The syringolin gene clus-
ter was reconstituted by deletion of the regulator sylA as well
as by insertion of a bidirectional promoter between sylB and
sylC, as sylB and sylC are organized in different directions, to
obtain pASK-sylB-biZeo-sylCDE (Scheme 3D). As expected, sy-
ringolins A–H were detected in both LB and M9 medium (Fig-
ure S2, Figure 2), and the increasing proportion of lysine-desa-
turated syringolins also shows the potent function of the syn-
thetic promoter driving sylB. The relative abundance of lysine-
desaturated syringolins is much higher than lysine-saturated
syringolins, similar to the ratio observed in the native syringo-
lin producer. However, the distribution of major syringolin var-
iants was altered in these two media. In M9 medium, SylA (1)
and SylB (2) dominate, while the ratio of SylF (6) to SylG (7)
was greatly increased in LB medium (Figure S2). This finding
might be explained by the different relative concentrations be-
tween valine and isoleucine in these two media. One report
shows that the intracellular metabolite concentration of valine
is far higher than that of isoleucine together with leucine in
glucose-fed M9 medium of exponentially growing E. coli,[34]
which fits well with our result that the syringolins A and B
dominate in M9 medium. Production of syringolins under stan-
dard conditions was comparable in the heterologous host
E. coli and in the native producer in SRMFmedium (analyzed
by comparison of the peak areas in HPLC-MS analysis). These
results clearly demonstrate the
successful heterologous expres-
sion of the syringolin family of
natural products in E. coli.
Conclusions
Traditionally, large natural prod-
uct biosynthetic gene clusters
require reconstruction from sev-
eral cosmids; this is time con-
suming due to the required
screening process from a ge-
nomic library plus the subse-
quent cloning steps. This work
is a prerequisite for heterolo-
gous expression in more suita-
ble host strains to scrutinize
their function or generate “un-
natural”natural
combinatorial biosynthesis. Our
LLHR-mediated straightforward
strategy to clone biosynthetic
gene clusters directly from the
predigested genomic DNA has
evident merits over previous ap-
productsby
proaches. We used this method to directly clone a partial syrin-
golin gene cluster (sylCDE) under the control of an inducible
promoter by one round of LLHR. Subsequently, the clone was
expressed in E. coli, and this led to the discovery of two new
syringolin family members. The complete syringolin gene clus-
ter was next assembled by insertion of sylB and promoter ex-
change, and its expression shows the diverse distribution of sy-
ringolin variants and the different efficiency of sylB under the
control of native or synthetic promoters. The Red/ET-mediated
recombineering strategy for direct cloning (LLHR) and engi-
neering (LCHR) furnishes a general tool for reconstituting large
gene clusters and, if coupled with suitable heterologous ex-
pression hosts, provides an effective alternative to investigate
or engineer known and unknown biosynthetic pathways dis-
covered by genome-scale sequencing projects, especially from
slow growing bacteria and those for which genetics are only
poorly established.
Experimental Section
Bacterial strains and culture condition: P. syringae van Hall 1902
DSM1241 and DSM1242 were purchased from Deutsche Sammlung
von Mikroorganismen und Zellkulturen GmbH (DSMZ, Braunsch-
weig) and grown in LB medium. For syringolin production, P. syrin-
gae strains were cultivated in SRMF(no arbutin) medium[26]supple-
mented with 2% XAD-16 adsorber resin for 3 days at 308C. The
E. coli strains used in this study for propagation or recombineering
were grown at 378C in LB medium amended with the suitable
antibiotics (ampicillin [Amp], 100 mgmL?1; gentamycin [Genta],
2 mgmL?1; and zeocin [Zeo], 15 mgmL?1). The heterologous host
for syringolin (syl) gene cluster expression was E. coli GB05-MtaA,[19]
which was grown at 308C in LB medium or M9 medium[35]contain-
Figure 2. Comparative chromatograms of HPLC-MS analysis (extracted ion chromatograms (EICs)) of syringolins
from extracts of A) E. coli GB05-MtaA/pASK-sylABCDE and B) E. coli GB05-MtaA/pASK-sylB-bizeo-sylCDE in M9
medium. SylA (1), EIC, m/z 494 [M+ +H]+; SylB (2), EIC, m/z 496 [M+ +H]+, SylC (3) and SylD (4), EIC, m/z 508 [M+ +H]+;
SylH (8) and SylE (5), EIC, m/z 510 [M+ +H]+; SylF (6), EIC, m/z 522 [M+ +H]+; and SylG (7), EIC, m/z 524 [M+ +H]+.
1950
www.chembiochem.org
? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2012, 13, 1946–1952
Y. Zhang and R. M?ller et al.

Page 6

ing Genta (2 mgmL?1) and Amp (100 mgmL?1) for pASK-Ptet-sylCDE
and pASK-sylABCDE or zeocin (13 mgmL?1) for pASK-sylB-biZeo-
sylCDE after transformation with the syl gene cluster.
Direct cloning and genetic engineering of the syringolin gene
cluster: Primarily, the genomic DNA of P. syringae DSM1241 was
completely digested with DraI and NotI. In parallel, the linear clon-
ing vector pASK-amp-Ptet, flanked with homology arms to target
genes, was amplified by PCR using pASK3 (IBA GmbH, Gçttingen,
Germany) as a template and primers sylDIRn5 (5’-TGCTG TGCGC
CGGCC CGCAT AGCGA GATCT TGAGA GCGTG ACGGT GCGGT ATTAT
TTTTC TCAAA ACTGT TCATG ACGGC CTCGG ATAAC ATATT TGATC
CTCGT TATCT AG-3’) and sylDIRn3 (5’-TGCGG CGAGC TTTGC ATGAC
CCAGT GCAGT ACGTC CGGGT CCAGC AGATG CCATT CGCGA
CGCGC CTTGA CCACC GTGCC GACAC GCGGC CAAGC TTGAC
CTGTG AAGTG AAAAA TG-3’; underlined sequences are homology
arms). Then, 5 mg of digested genomic DNA were mixed with
0.5 mg of linear cloning vector and cotransformed into 30 mL of
YZ2005[12]competent cells by electroporation (Scheme 2). Re-
combinants carrying pASK-Ptet-sylCDE were identified by selection
for Amp resistance (bla gene present on the linear cloning vector)
and subsequent DNA restriction analysis and sequencing. To con-
struct the plasmid containing the complete syl gene cluster, the
PCR product containing sylA and sylB flanked with homology arms,
together with linearized pASK-Ptet-sylCDE by XbaI, were coelectro-
porated into YZ2005[12]competent cells (Scheme 3B). Recombi-
nants were selected from LB plates with Amp. The oligonucleotides
used for insertion of sylAB are listed below.
sylAB-GB5 (5’-TCGAA TGGCC AGATG ATTAA TTCCT AATTT TTGTT
GACAC TCTAT CATTG ATAGA GACAC ACATG AGGTT ATCCA TC-3’)
and sylAB-GB3 (5’-TGAGA GCGTG ACGGT GCGGT ATTAT TTTTC-3’)
Furthermore, we also constructed pASK-sylB-biZeo-sylCDE in which
sylB and sylC were placed under the control of a strong bidirection-
al synthetic promoter, biZeo, and the transcription activator gene
sylA was deleted. The Zeo resistance gene was inserted into the
middle of two opposite synthetic promoters, state69[36]and
CP25,[37]by Red/ET recombination to form a biZeo cassette
(Scheme S2). Using this plasmid as a template to generate a biZeo
fragment with a 5’-end homology arm to the 5’-end of sylB and
a 3’-end homology arm to the 5’-end of sylC by PCR, it was then
electroporated into recombineering-proficient competent cells
(GB05-red)[19]to form pASK-sylAB-biZeo-sylCDE (Scheme 3C). The
sylB-biZeo cassette flanked with homology arms, which was gener-
ated by PCR using the linearized pASK-sylAB-biZeo-sylCDE with
EcoRI and KpnI as a template, was transformed into GB05-red bear-
ing pASK-Ptet-sylCDE. The reconstituted syringolin gene cluster
was selected by Zeo and analyzed by DNA restriction digestion
(Scheme 3D). The oligonucleotides used for zeoRPCR were zeo-kil5
(5’-TCAAT ACAGT ATAGA ACAAA TTTGC AAATT TGGCA AGAGG
CGAGC ACGTG TTGAC AATTA ATC-3’) and zeo-kil3 (5’-TCACT
ACATG TCAAG AATAA ACTGC CAAAG CATAA TGGGA TCAGT CCTGC
TCCTC GGCCA CG-3’). The biZeo DNA fragment was generated by
using oligonucleotides biZeosyl5 (5’-TCAGT TCAAG TTTTA GTGCA
TTTGT ACAGT GAAAT CGTTG GACGG TCTGC ATAAC TCGAT CCTTA
TAAAA TG-3’) and biZeosyl3 (5’-TCTTG AGAGC GTGAC GGTGC
GGTAT TATTT TTCTC AAAAC TGTTC ATATG ATATC CTCTT AACAG-3’).
The sylB-biZeo cassette was generated by using oligonucleotides
sylBzeo-C5 (5’-AAGTG AAATG AATAG TTCGA CAAAA ATCTA GATAA
CGAGG ATCAA ACCCA TGACG ACTGG GTTGA G-3’) and sylBzeoCl
(5’-TGCGG TATTA TTTTT CTCAA AACTG TTCAT GACGG CCTCG
GATAA CATAT GATAT CCTCT TAACA GTAC-3’).
Construction of the heterologous expression host E. coli GB05-
MtaA: To construct an E. coli GB2005 strain[15,38]constitutively ex-
pressing mtaA, plasmid pR6K-Tps-Genta (Scheme S3) was linearized
by restriction digestion, resulting in the 3.5 kb linear vector R6K-
Tps-Genta including the R6K replication origin, the Genta resist-
ance gene, and the MycoMar transposase gene. Meanwhile, an
mtaA gene flanked with homology arms to the linear vector was
generated by PCR using primers (sequence for homologous arm
for recombineering is underlined) MtaA-Tps5 (5’-AGACC CACTT
TCACA TTTAA GTTGT TTTTC TAATC CGCAT ATGAT CAATT CGCAA
ACCGC CTCTC CCCGC GCGTT G-3’), MtaA-Tps3 (5’-TACGC GAACG
CGAAG TCCGA CTCTA AGATG TCACG GAGGT TCAAG TTACC TAACC
AGTGC AACGA AAGCA ATACC) and pSUMtaA[39]as a template. Plas-
mid pSUMtaA contains the mtaA gene, encoding a broad substrate
phosphopantetheinyl transferase from Stigmatella aurantiaca DW4/
3-1.[39,40]The 1.1 kb mtaA PCR product and the 3.5 kb linear vector
R6K-Tps-Genta were cotransformed into YZ2005-pir competent
cells in which recE/recT genes were located in the chromosome
under the control of native promoter and contained pir-116 (high
copy mutant of pir) gene for propagation of an R6K-derived plas-
mid.[12,41]After selection from an LB plate containing Genta, a
4.0 kb plasmid (pR6K-Tps-MtaA-Genta) was generated and verified
by restriction analysis and sequencing. This plasmid can only repli-
cate in a strain expressing Pir. When this plasmid was transformed
into E. coli GB2005,[15,38]which has no pir gene, it cannot replicate,
but the transposase inserts the mtaA-genta cassette into the chro-
mosome to form E. coli GB05-MtaA (Scheme S4).
Expression and analysis of syringolin production: Plasmids har-
boring a partial, complete, or reconstituted syl gene cluster were
introduced into expression host E. coli GB05-MtaA by electropora-
tion. The resulting mutants (E. coli GB05-MtaA/pASK-Ptet-sylCDE,
E. coli GB05-MtaA/pASK-sylABCDE, and E. coli GB05-MtaA/pASK-
sylB-biZeo-sylCDE) were cultivated in 300 mL shake flasks contain-
ing 50 mL minimal medium M9 or LB medium supplemented with
suitable antibiotics. The medium was inoculated with 0.5 mL over-
night culture and incubated at 308C on a rotary shaker. After
induction (tetracycline, final concentration 0.5 mgmL?1, only for
pASK-Ptet-sylCDE, 6 h) and addition of XAD adsorber resin (2%,
24 h), incubation was continued for another 2 days. Cells and XAD
were harvested by centrifugation and extracted with acetone and
MeOH. The extracts were evaporated and redissolved into 0.5 mL
MeOH; 5 mL of solution was analyzed by HPLC-MS. Analysis was
performed on an Agilent 1100 series solvent delivery system that
was equipped with a photodiode array detector and coupled to
a Burker HCTultra ion trap mass spectrometer. Chromatographic
conditions were as follows: Luna RP-C18 column, 100?2 mm,
2.5 mm particle size, and precolumn C18, 8?3 mm, 5 mm. Solvent
gradient (with solvents A [water and 0.1% formic acid] and B
[CH3CN and 0.1% formic acid]): 20% B from 0 to 20 min, 20% B–
95% B within 10 min, followed by 5 min with 95% B at a flow rate
of 0.4 mLmin?1. Detection was carried out in positive ion mode,
auto MSn. Syringolins were identified by comparison to the reten-
tion times and the MS2 data of syringolins identified from the orig-
inal producer.[26,27]Relative productions of syringolins were calculat-
ed by peak areas of the extracted ion chromatograms (EICs) of
each derivative.
Syringolins G and H were identified by comparison to retention
times and MS or MS2 data of other syringolins. High-resolution
mass spectrometry was performed on an Accela UPLC-system
(Thermo-Fisher) coupled to a linear trap-FT-Orbitrap combination
(LTQ-Orbitrap), operating in positive ionization mode. Separation
was achieved on a Waters BEH RP-C18column (50?2.1 mm; 1.7 mm
ChemBioChem 2012, 13, 1946–1952? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim
www.chembiochem.org
1951
The Syringolin Biosynthetic Gene Cluster

Page 7

particle diameter; flow rate 0.6 mLmin?1, Waters), with a mobile
phase of water/CH3CN (each containing 0.1% formic acid) and
a gradient from 5%–95% CH3CN over 9 min. UV and MS detection
were performed simultaneously. Coupling of HPLC to MS was sup-
ported by an Advion Triversa Nanomate nano-ESI system attached
to a Thermo Fisher Orbitrap. Mass spectra were acquired in cent-
roid mode ranging from 200–2000 m/z at a resolution of R=
30000. Feeding experiments of commercially available l-isoleu-
cine-15N (Sigma–Aldrich) and l-leucine-5,5,5-d3(Deutero, Kastellaun,
Germany) in LB medium (1 mgmL?1) were performed to provide
direct evidence that isoleucine rather than leucine is incorporated
into those two compounds.
Acknowledgements
X.B. is grateful to the China Scholarship Council (CSC) for a PhD
scholarship. Research in the laboratory of R.M. was funded by
the DFG and BMBF. Y.Z. in Gene Bridges was partially funded by
the BMBF (MiPro). The authors thank Thorsten Klefisch and Eva
Luxenburger in the laboratory of R.M. for LC/MS measurements.
Keywords: biosynthesis · gene expression · natural products ·
Red/ET recombineering · syringolin
[1] R. H. Baltz, Nat. Biotechnol. 2006, 24, 1533–1540.
[2] J. Staunton, K. J. Weissman, Nat. Prod. Rep. 2001, 18, 380–416.
[3] M. A. Marahiel, T. Stachelhaus, H. D. Mootz, Chem. Rev. 1997, 97, 2651–
2674.
[4] M. A. Fischbach, C. T. Walsh, Chem. Rev. 2006, 106, 3468–3496.
[5] B. A. Pfeifer, S. J. Admiraal, H. Gramajo, D. E. Cane, C. Khosla, Science
2001, 291, 1790–1792.
[6] K. Watanabe, K. Hotta, A. P. Praseuth, K. Koketsu, A. Migita, C. N. Boddy,
C. C. Wang, H. Oguri, H. Oikawa, Nat. Chem. Biol. 2006, 2, 423–428.
[7] W. S. Jung, S. K. Lee, J. S. J. Hong, S. R. Park, S. J. Jeong, A. R. Han, J. K.
Sohng, B. G. Kim, C. Y. Choi, D. H. Sherman, Y. J. Yoon, Appl. Microbiol.
Biotechnol. 2006, 72, 763–769.
[8] L. Tang, S. Shah, L. Chung, J. Carney, L. Katz, C. Khosla, B. Julien, Science
2000, 287, 640–642.
[9] Y. Chai, S. Shan, K. J. Weissman, S. Hu, Y. Zhang, R. M?ller, Chem. Biol.
2012, 19, 361–371.
[10] S. C. Wenzel, F. Gross, Y. Zhang, J. Fu, F. A. Stewart, R. M?ller, Chem. Biol.
2005, 12, 349–356.
[11] F. Gross, M. W. Ring, O. Perlova, J. Fu, S. Schneider, K. Gerth, S. Kuhl-
mann, A. F. Stewart, Y. Zhang, R. M?ller, Chem. Biol. 2006, 13, 1253–
1264.
[12] Y. Zhang, J. P. P. Muyrers, G. Testa, A. F. Stewart, Nat. Biotechnol. 2000,
18, 1314–1317.
[13] Y. Zhang, J. P. Muyrers, J. Rientjes, A. F. Stewart, BMC Mol. Biol. 2003, 4,
1.
[14] Y. Zhang, F. Buchholz, J. P. Muyrers, F. A. Stewart, Nat. Genet. 1998, 20,
123–128.
[15] J. Fu, S. C. Wenzel, O. Perlova, J. Wang, F. Gross, Z. Tang, Y. Yin, A. F.
Stewart, R. M?ller, Y. Zhang, Nucleic Acids Res. 2008, 36, e113.
[16] T. M. Binz, S. C. Wenzel, H. J. Schnell, A. Bechthold, R. M?ller, ChemBio-
Chem 2008, 9, 447–454.
[17] M. Wolpert, L. Heide, B. Kammerer, B. Gust, ChemBioChem 2008, 9, 603–
612.
[18] O. Perlova, J. Fu, S. Kuhlmann, D. Krug, F. Stewart, Y. Zhang, R. M?ller,
Appl. Environ. Microbiol. 2006, 72, 7485–7494.
[19] J. Fu, X. Bian, S. Hu, H. Wang, F. Huang, P. M. Seibert, A. Plaza, L. Xia, R.
Muller, A. F. Stewart, Y. Zhang, Nat. Biotechnol. 2012, 30, 440–446.
[20] J. Clerc, M. Groll, D. J. Illich, A. S. Bachmann, R. Huber, B. Schellenberg,
R. Dudler, M. Kaiser, Proc. Natl. Acad. Sci. USA 2009, 106, 6507–6512.
[21] M. Groll, B. Schellenberg, A. S. Bachmann, C. R. Archer, R. Huber, T. K.
Powell, S. Lindow, M. Kaiser, R. Dudler, Nature 2008, 452, 755–758.
[22] C. R. Archer, D. L. T. Koomoa, E. M. Mitsunaga, J. Clerc, M. Shimizu, M.
Kaiser, B. Schellenberg, R. Dudler, A. S. Bachmann, Biochem. Pharmacol.
2010, 80, 170–178.
[23] J. Clerc, B. I. Florea, M. Kraus, M. Groll, R. Huber, A. S. Bachmann, R.
Dudler, C. Driessen, H. S. Overkleeft, M. Kaiser, ChemBioChem 2009, 10,
2638–2643.
[24] C. S. Coleman, J. P. Rocetes, D. J. Park, C. J. Wallick, B. J. Warn-Cramer, K.
Michel, R. Dudler, A. S. Bachmann, Cell Proliferation 2006, 39, 599–609.
[25] M. Oka, Y. Nishiyama, S. Ohta, H. Kamei, M. Konishi, T. Miyaki, T. Oki, H.
Kawaguchi, J. Antibiot. 1988, 41, 1331–1337.
[26] U. W?spi, D. Blanc, T. Winkler, P. Ruedi, R. Dudler, Mol. Plant-Microbe In-
teract. 1998, 11, 727–733.
[27] U. W?spi, P. Hassa, A. A. Staempfli, L. P. Molleyres, T. Winkler, R. Dudler,
Microbiol. Res. 1999, 154, 89–93.
[28] H. Amrein, S. Makart, J. Granado, R. Shakya, J. Schneider-Pokorny, R.
Dudler, Mol. Plant-Microbe Interact. 2004, 17, 90–97.
[29] H. J. Imker, C. T. Walsh, W. M. Wuest, J. Am. Chem. Soc. 2009, 131,
18263–18265.
[30] C. Ramel, M. Tobler, M. Meyer, L. Bigler, M. O. Ebert, B. Schellenberg, R.
Dudler, BMC Biochem. 2009, 10, 26.
[31] J. P. P. Muyrers, Y. Zhang, G. Testa, A. F. Stewart, Nucleic Acids Res. 1999,
27, 1555–1557.
[32] D. G. Yu, H. M. Ellis, E. C. Lee, N. A. Jenkins, N. G. Copeland, D. L. Court,
Proc. Natl. Acad. Sci. USA 2000, 97, 5978–5983.
[33] W. M. Wuest, D. Krahn, M. Kaiser, C. T. Walsh, Org. Lett. 2011, 13, 4518–
4521.
[34] B. D. Bennett, E. H. Kimball, M. Gao, R. Osterhout, S. J. Van Dien, J. D. Ra-
binowitz, Nat. Chem. Biol. 2009, 5, 593–599.
[35] J. Sambrook, D. W. Russell in Molecular Cloning: A Laboratory Manual,
Cold Spring Harbor Laboratory Press, New York, 2001, Section A2.2.
[36] G. Miksch, F. Bettenworth, K. Friehs, E. Flaschel, Appl. Microbiol. Biotech-
nol. 2005, 69, 312–320.
[37] P. R. Jensen, K. Hammer, Appl. Environ. Microbiol. 1998, 64, 82–87.
[38] J. Fu, M. Teucher, K. Anastassiadis, W. Skarnes, A. F. Stewart, Methods En-
zymol. 2010, 477, 125–144.
[39] N. Gaitatzis, A. Hans, R. M?ller, S. Beyer, J. Biochem. 2001, 129, 119–124.
[40] B. Silakowski, H. U. Schairer, H. Ehret, B. Kunze, S. Weinig, G. Nordsiek, P.
Brandt, H. Blçcker, G. Hçfle, S. Beyer, R. M?ller, J. Biol. Chem. 1999, 274,
37391–37399.
[41] W. W. Metcalf, W. H. Jiang, B. L. Wanner, Gene 1994, 138, 1–7.
Received: May 12, 2012
Published online on July 31, 2012
1952
www.chembiochem.org
? 2012 Wiley-VCH Verlag GmbH&Co. KGaA, Weinheim ChemBioChem 2012, 13, 1946–1952
Y. Zhang and R. M?ller et al.