N-acylated alanine methyl esters (NAMEs) from Roseovarius tolerans, structural analogs of quorum-sensing autoinducers, N-acylhomoserine lactones

Article abstract of DOI:10.1002/cbdv.201300210

The Roseobacter clade is one of the most important bacteria group living in the ocean. Liquid cultures of Roseovarius tolerans EL 164 were investigated for the production of autoinducers such as N-acylhomoserine lactones (AHLs) and other secondary metabolites. The XAD extracts were analyzed by GC/MS. Two AHLs, Z7-C14: 1-homoserine lactone (HSL) and C15: 1-HSL, were identified. Additionally, the extract contained five compounds with molecular-ion peaks at m/z 104, 145, and 158, thus exhibiting mass spectra similar to those of AHLs with corresponding peaks at m/z 102, 143, and 156. Isolation of the main compound by column chromatography, NMR analysis, dimethyl disulfide derivatization for the determination of the location of the CiC bond and finally synthesis of the compound with the proposed structure confirmed the compound to be (Z)-N-(hexadec-9-enoyl)alanine methyl ester. Four additional minor compounds were identified as C14: 0-, C15: 0-, C16: 0-, and C17: 1-N-acylated alanine methyl esters (NAMEs). All NAMEs have not been described from natural sources before. A BLASTp search showed the presence of AHL-producing luxI genes, but no homologous genes potentially responsible for the structurally closely related NAMEs were found. The involvement of the NAMEs in chemical communication processes of the bacteria is discussed. Copyright

Full text of DOI:10.1002/cbdv.201300210

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1559
N-Acylated Alanine Methyl Esters (NAMEs) from Roseovarius tolerans,
Structural Analogs of Quorum-Sensing Autoinducers, N-Acylhomoserine
Lactones
by Hilke Brunsa), Verena Thiela), Sonja Vogetb), Diana Patzeltc), Rolf Danielb), Irene Wagner-
Dçblerc), and Stefan Schulz*a)
a) Institute of Organic Chemistry, TU Braunschweig, Hagenring 30, DE-38106 Braunschweig
(phone: þ495313915271; e-mail: stefan.schulz@tu-bs.de)
b) Institute of Microbiology and Genetics, Georg-August University of Gçttingen, Grisebachstr. 8,
DE-37077 Gçttingen
c) Helmholtz Centre for Infection Research, Inhoffenstraße 7, DE-38124 Braunschweig
The Roseobacter clade is one of the most important bacteria group living in the ocean. Liquid
cultures of Roseovarius tolerans EL 164 were investigated for the production of autoinducers such as N-
acylhomoserine lactones (AHLs) and other secondary metabolites. The XAD extracts were analyzed by
GC/MS. Two AHLs, Z7-C14 :1-homoserine lactone (HSL) and C15 :1-HSL, were identified. Addition-
ally, the extract contained five compounds with molecular-ion peaks at m/z 104, 145, and 158, thus
exhibiting mass spectra similar to those of AHLs with corresponding peaks at m/z 102, 143, and 156.
Isolation of the main compound by column chromatography, NMR analysis, dimethyl disulfide
derivatization for the determination of the location of the C¼C bond and finally synthesis of the
compound with the proposed structure confirmed the compound to be (Z)-N-(hexadec-9-enoyl)alanine
methyl ester. Four additional minor compounds were identified as C14 :0-, C15 :0-, C16 :0-, and C17:1-
N-acylated alanine methyl esters (NAMEs). All NAMEs have not been described from natural sources
before. A BLASTp search showed the presence of AHL-producing luxI genes, but no homologous genes
potentially responsible for the structurally closely related NAMEs were found. The involvement of the
NAMEs in chemical communication processes of the bacteria is discussed.
Introduction. – Autoinducers are small diffusible signal molecules secreted by
bacteria in a cell-to-cell communication process called quorum sensing that is used to
regulate their gene expression in a cell density-dependent manner [1–3]. Several
widely different traits such as cell differentiation, expression of virulence factors,
flagella or biofilm formation are known to be quorum sensing regulated [4–7].
Bacteria of the Roseobacter clade constitute some of the most important bacteria in the
ocean [8][9]. In this clade quorum sensing regulates, e.g., antibiotic production in
Phaeobacter [10], flagella formation, type-IV secretion, and the mode of cell division in
Dinoroseobacter [11], inhibits biofilm formation and activates motility in Ruegeria [12],
and influences motility in Silicibacter [13]. The dominant class of autoinducers used in
Gram-negative bacteria, to which the Roseobacter clade belongs, are N-acylhomoserine
lactones (AHLs) [14]. Quorum sensing mediated by AHLs is one of the best-
understood bacterial systems at the molecular level [15]. Usually, the quorum-sensing
circuit consists of a LuxI/LuxR-type system, with LuxI being the synthase of the
autoinducer and LuxR its co-evolved receptor. After the synthesis of the autoinducer
ꢀ 2013 Verlag Helvetica Chimica Acta AG, Zꢁrich

1560
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
by LuxI, it freely diffuses in and out of the cell. Productive binding to the receptor
LuxR only occurs at a certain threshold concentration for which a high intracellular
ligand concentration and thus a high cell density is needed. The LuxRꢀautoinducer
complex then acts as a transcriptional regulator, binding to DNA and inducing
expression of downstream genes. In many but not all bacteria, the activated LuxR
complex at the same time also enhances the production of the autoinducer itself, thus
resulting in a positive feedback loop [16].
AHL Autoinducers share several structural characteristics. They are comprised of a
homoserine ring in l-configuration which is unsubstituted in the b- and g-position, and
N-acylated with an acyl chain derived from fatty-acid biosynthesis [5]. The chain length
varies from 4 to 18 C-atoms with C14, C16, and C18 dominating in bacteria of the
Roseobacter clade [17]. The acyl residue is most often unbranched and saturated,
mono- or diunsaturated. The monounsaturated chain normally carries a C¼C bond in
(Z)-configuration. Diunsaturated AHLs, e.g. N-[(2E,9Z)-hexadeca-2,9-dienoyl]homo-
serine lactone, have been identified in Jannaschia helgolandensis, Staleya guttiformis,
and Dinoroseobacter shibae [17]. AHLs with a rare methyl branching in iso position
have been identified in Aeromonas culicicola 3249T [18]. Additionally, modifications at
C(3) of the acyl chain are possible comprising a OH or C¼O function. For example, N-
(3-hydroxydecanoyl)homoserine lactone (3-OH-C10-AHL) has been identified in
Roseobacter gallaeciensis [17], Sinorhizobium meliloti [19], Pseudomonas fluorescens
[20], and Burkholderia pseudomallei [21]. The keto compound 3-oxo-C14-AHL was
identified in R. tolerans EL-78 and EL-90 [17], as well as in S. meliloti besides other 3-
oxo-AHLs [22]. In addition, AHLs carrying a p-coumaroyl residue derived from
environmental p-coumaric acid were identified in Rhodopseudomonas palustris,
Bradyrhizobium sp., and Silicibacter pomeroyi [23]. Usually, several AHLs are
produced by one strain in nanomolar concentrations or less. The pattern produced is
stable for a certain organism under identical cultivation conditions but different species
produce different patterns.
During our in-depth study on the variation of AHLs in various members of the
Roseobacter clade, we encountered unknown related metabolites. We report here the
identification of five N-acylated alanine methyl esters (NAMEs), the structures of
which show striking similarities to AHLs produced by R. tolerans. A XAD enrichment
method was used to extract these secondary metabolites from Roseovarius cultures
[24]. The compounds were identified by GC/MS analysis, isolation of the major
compound, and synthesis. A preliminary bioassay was performed to evaluate the
function of the compounds.
Results and Discussion. – Extraction. The secondary metabolites released by
liquid cultures of Roseovarius tolerans EL-164 were collected by extraction via
Amberlite XAD-16. The resin was added to freshly inoculated liquid cultures of R.
tolerans. This procedure removes bacterial metabolites from the aqueous phase of the
culture, thus likely enhancing their production if a concentration-dependent feedback
loop exists for the target compounds. The concentrated extracts were investigated by
GC/MS. Two AHLs, N-[(Z)-tetradec-9-enoyl]homoserine lactone (2, C14 :1-AHL)
and N-[(Z)-pentadec-9-enoyl]homoserine lactone (7), were readily identified (Fig. 1).

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1561
Fig. 1. Expansion of the total ion chromatogram (24–55 min) of the XAD extract of R. tolerans showing
the AHLs 2 and 7, as well as the NAMEs 1 and 3–6
Compound 2 has previously been detected in R. tolerans in addition to the
corresponding C14 :0-, C16 :0-, and C16 :1-AHLs [17].
Structure Elucidation. However, five compounds, 1 and 3–6, showed mass spectra
very similar to those of the AHLs. These mass spectra are characterized by fragment-ion
peaks at m/z 104, 145, and 158, compared to those at m/z 102, 143, and 156, in case of
AHLs (Fig. 2). These fragment ions can be explained by cleavage of the amide-bond,
and b- and g-cleavage as indicated in the figure. The amount of the major compound 1
was sufficient for isolation by column chromatography and investigation by NMR
spectroscopy. The NMR spectra showed an ester-bound Me group with a chemical shift
of d 3.75 (s), a CH H-atom next to a CO group with a signal at d 4.61 (quint., J¼7.2) and
a N-atom in a NꢀCHꢀMe spin system, suggesting an alanine methyl ester structure.
Moreover, the typical NMR signals of an acyl chain established that the amino group
was acylated by a long-chain fatty acid with a C¼C bond in (Z)-configuration.
By addition of dimethyl disulfide (DMDS) to the crude extract and subsequent
investigation by GC/MS, the location of the C¼C bond was elucidated [18][25][26].
The mass spectrum of the addition product is shown in Fig. 3. Cleavage preferentially
takes place between the methylsulfanyl groups and results in two characteristic ion
peaks at m/z 145 [C7H14SMe]þ and 288 [MeOCOCHCH3NHꢀCOC8H15SMe]þ ,
indicating the C¼C bond to be located at C(9). The latter fragment ion contains the
alanine methyl ester moiety which is lost to give rise to the fragment-ion peak at m/z
185. The structure proposal was supported by HR-MS data (Table), indicating the
molecular formula C20H37NO3.

1562
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Fig. 2. Mass spectra of a) (S)-N-[(Z)-hexadec-9-enoyl]alanine methyl ester (1), and b) N-[(Z)-tetradec-
9-enoyl]homoserine lactone (2)
To establish the structure, N-[(Z)-hexadec-9-enoyl]alanine methyl ester (1) was
synthesized by coupling palmitoleic acid with l-alanine methyl ester (Scheme 1).
Comparison of the spectral data of the natural product with those of the synthetic
sample confirmed the proposed structure. In addition, the GC retention indices (RIs)
of the natural and synthetic compound match (RInat ¼2479, RIsyn ¼2477).
The structures of the remaining four N-acylated alanine methyl esters (NAMEs),
3–6, were suggested based on the analysis of their mass spectra (Fig. 4). Compound 6
(C17:1-NAME) shows a mass spectrum almost identical to that of 1. Since the
molecular-ion Mþ, peak at m/z 353 is 14 u higher compared to 1, indicating an
additional CH2 group, 6 was proposed to be the higher homolog N-[(Z)-heptadec-9-
enoyl]alanine methyl ester. The DMDS derivative evidenced the location of the double
bond at C(9). Unsaturated AHLs show a characteristic loss of the homoserine moiety
([Mꢀ101]þ ) [17]. The unsaturated C16 :1- and C17:1-NAMEs exhibit peaks of
corresponding fragment ions, [Mꢀ103]þ , resulting from the loss of the alanine methyl
ester moieties at m/z 236 and 250, but in low intensities. The mass spectra of compounds

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1563
Fig. 3. Mass spectrum of the DMDS derivative of 1. The ion peaks at m/z 145 and 288 indicate the C¼C
bond at C(9).
Table. HR-MS Data of Natural N-Acylalanine Methyl Esters
Molecular ion
[MþH]þ or [MþNa]þ [m/z]
Acyl side chain
calc.
found
3
4
5
1
6
C18H35NNaOþ3
C19H37NNaOþ3
C20H40NOþ3
336.2509
350.2671
342.3008
340.2852
354.3008
336.2509
350.2666
342.3003
340.2845
354.3003
C14H27O
C15H29O
C16H31O
C16H29O
C17H31O
C20H38NOþ3
C21H40NOþ3
3–5 also exhibit the fragment-ion peaks at m/z 104, 145, and 158, typical for NAMEs,
albeit with different ion intensities compared to the unsaturated NAMEs. For the latter,
the base peak is at m/z 104, whereas it is at m/z 145 for compounds 3–5. Similarly, mass
spectra of AHLs show an increased intensity of the fragment-ion peak at m/z 102 for
unsaturated AHLs and at m/z 143 for saturated AHLs [17][27]. Taking into account
the molecular-ions peak at m/z 313, 327 and 341, which are small but significant,

1564
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Scheme 1. Synthesis of a) Unsaturated NAMEs 1 and 6, and b) Saturated NAMEs 3, 4, and 5. NaHMDS,
Sodium hexamenthyldisilazide (¼sodium bis(trimethylsilyl)amide; TPAP, tetrapropylammonium per-
ruthenate; NMO, N-methylmorpholine N-oxide.
compounds 3–5 were proposed to be N-(tetradecanoyl)-, N-(pentadecanoyl)-, and N-
(hexadecanoyl)alanine methyl esters, respectively.
Since acyl side chains with an uneven number of C-atoms are less abundant in
bacteria, a Me-branched side chain was also considered for these compounds. We have
previously developed a RI calculation model for AHLs and related compounds,
allowing determination of the branch position in long-chain aliphatic compounds
[18][28]. For an additional Me group in the chain, RI are by 30–75 units higher, while
an addition within the chain leads to an increment of 100 units. The RI of 3–5, 2292,
2395, and 2500, indicate a uniform, unbranched side chain, established also by

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1565
Fig. 4. Mass spectra of the NAMEs a) 3, b) 4, c) 5, and d) 6
comparison with the synthetic material (see below). All structural proposals were
supported by HR-ESI-MS data (Table).
Synthesis. For final verification, compounds 3–6 were synthesized as outlined in
Scheme 1. The synthesis of 6 started by conversion of 9-bromononan-1-ol (8) to the
corresponding Wittig salt (9-hydroxynonyl)triphenylphosphonium bromide (9) and
subsequent Wittig olefination with heptanal to give (Z)-heptadec-9-en-1-ol (10).
Oxidation with tetrapropylammonium perruthenate (TPAP) yielded (Z)-heptadec-9-
enoic acid (12). Conversion to the corresponding acid chloride with oxalyl chloride and
subsequent coupling with l-alanine methyl ester hydrochloride gave target compound
6 (C17:1-NAME). In analogy to the last reaction step, conversion of tetradecanoic,
pentadecanoic, and hexadecanoic acid into the corresponding acid chlorides and
coupling with l-alanine methyl ester hydrochloride furnished the NAMEs 3–5,
respectively. All synthetic material was identical to the corresponding natural
compounds, and the retention indices were in good agreement.
Assignment of Absolute Configuration. The absolute configuration of the alanine
methyl ester part was assigned by cleavage of the amide bond of the NAMEs in the
natural sample as well as in the synthetic 1, followed by derivatization of the amine
function with ethyl chloroformate (ClCOOEt; ECF) (Scheme 2). These derivatives

1566
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Scheme 2. Cleavage of the Amide Bond and ECF Derivatization for NAME 1. ECF, ClCOOEt.
were analyzed by GC on chiral phases (Fig. 5). Comparison by coinjection experiments
with a racemate of alanine methyl ester established the absolute configuration of the
natural NAMEs as exclusively l.
AHL Synthethase. A search with the Pfam domain for autoinducer synthetases
(PF00765) as well as BLASTp searches with the amino acid sequences of known AHL
synthetases of members of the Roseobacter clade resulted in one AHL synthetase
(ROT_01220). Examination of the surrounding neighborhood revealed the presence of
a LuxR-type regulator (ROT_01219), completing the two-component system. The
gene organization showed the highest synteny as well as high sequence homologies of
the AHL synthase to the closely related strains Roseovarius sp. 217 and Roseovarius sp.
TM1035 (Fig. 6). This AHL synthase seems to be responsible for the production of
both identified AHLs, 2 and 7. With 2 being the main and 7 the minor product, the
normal substrate for the AHL synthase seems to be the C14 :1 acyl chain, though
C15 :1 is also converted. There are known examples in which a single LuxI-type AHL
synthase is responsible for the production of several structurally similar AHLs [22].
N-Acylated Amino Acid Synthases (NASs). Other N-acylated amino acids are
structurally similar to the AHLs and NAMEs mentioned. In screenings for antibiotic
compounds in cosmid libraries of environmental DNA expressed in Escherichia coli by
Van Wagoner and Clardy, the most frequently encountered small molecules were N-
acylated amino acids [29]. So far, N-acylated amino acids of tyrosine [30], tryptophan,
arginine [31], and phenylalanine [32] have been described, each comprising both
saturated and monounsaturated fatty-acid residues. The chain length of the fatty-acid
residues of long-chain N-acyltyrosines ranges from C8 to C18, with C14 as the most
frequent residue. Fatty-acid residues observed with tryptophan have a chain length of
C16 and C18, whereas both arginine and phenylalanine residues vary from C14 to C16.
These N-acylated amino acids are produced by a family of N-acyl amino acid synthases
(NASs). NASs like FeeM share structural similarities with GCN5-related N-acyl
transferases (GNATs) including acyl homoserine lactone synthases [29]. Like AHL
synthases, NASs form an amide bond between an amino group and a fatty acid. In
addition, the fatty acid is acyl carrier protein-bound as used by AHL synthases rather
than coenzyme A-bound like used by most other GNAT like proteins. Based on these
similarities in structure and substrate usage between AHL synthases and GNATs, the

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1567
Fig. 5. Determination of the configuration of the alanine part of NAMEs of R. tolerans by chiral GC on a
ß-DEX 225 phase. a) NAME 1. b) Co-injection NAME 1 and racemate. c) Natural sample. d) Co-
injection natural sample and racemate. e) Racemate.
authors proposed also a functional link between the enzyme products with N-acyl
amino acids functioning as bacterial-signaling agents [29].
The search for a N-acylalanine synthase proved difficult. So far, the family of N-acyl
amino acid synthases (NASs) comprises synthases for the amino acids tyrosine,
tryptophan, and arginine, as well as phenylalanine [30–32]. The known NASs show
neither on the gene nor on the protein level any significant sequence similarity. Even in
a group of ten sequenced N-acyltyrosine synthases, no absolutely conserved residues in
the protein sequence were detected [33]. Thus, it is not surprising that a BLASTp
search against known NASs sequences gave no results.
Bioassays. Due to the striking similarity of NAMEs and AHLs, the autoinducing
activity of the NAMEs was tested in bioassays specific for long-chain AHLs. The
quorum-sensing as well as the quorum-quenching activity was tested using the reporter
strain P. putida pKRꢀC12 under standard conditions [34]. Even at a relatively high
concentration of 5 mm, no induction of the reporter strain was observed nor did the
NAMEs inhibit the induction of the reporter in the presence of structurally similar
AHLs. Since neither quorum-sensing nor quorum-quenching activity was observed in
these tests, NAMEs might have a different function for the bacterium. One could even

1568
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
Fig. 6. Gene organization in the neighborhood of the LuxI like synthetase (red arrow) in R. tolerans and
its homology to Roseovarius sp. 217 and Roseovarius sp. TM1035
speculate that, for the so far unknown function of NAMEs, it might be important that
there is no interference whatsoever with the AHL quorum-sensing system of the
bacterium. The inactivity in the AHL assay may be a prerequisite for proper function of
NAMEs, because the actual function of the NAMEs in the physiology of R. tolerans is
unknown.
Conclusions. – In summary, we have shown that Roseovarious tolerans EL-164
secretes new natural compounds, N-acylated alanine methyl esters during cultivation
alongside with the well-known signaling compounds of the Roseobacter clade, AHLs.
This distinguishes this bacterium from many other bacteria of this clade, which very
often can secrete AHLs. Their structural analogy suggests a role in bacterial
communication, but a specific function of associated genes has not been found yet.
The synthetic route developed allows investigation of this bacterium and its
communication system in more detail.
We thank the Deutsche Forschungsgemeinschaft for a grant from the collaborative research center
SFB/TRR 51.
Experimental Part
General. Chemicals were purchased from SigmaꢀAldrich Chemie GmbH (DE-Steinheim) or from
Acros Organics (Geel, Belgium), and used without further purification. Solvents were purified by
distillation and dried according to standard procedures. Moisture- and/or oxygen-sensitive reactions were
carried out under N2 in vacuum-heated flasks with dried solvents. TLC: 0.20 mm Macherey-Nagel silica
gel plates (Polygram SIL G/UV254). Column chromatography (CC): Merck silica gel 60 (0.040–
0.063 mm) using standard flash chromatographic methods. GC/MS: HP6890 GC system connected to a
HP5973 Mass Selective Detector fitted with a BPX-5 fused silica cap. column (25 mꢁ0.22 mm i.d.,
0.25 mm film, SGE Inc., Melbourne, Australia); conditions: inlet pressure; 97.0 kPa; He, 45.5 ml minꢀ1
;

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1569
injection volume; 1 ml; injector; 2508; transfer line; 3008; electron energy; 70 eV. The GC was
programmed as follows: 508 (5 min isothermic), increasing at 108 minꢀ1 to 3208, and operated in split
mode (35 :1); carrier gas (He); 1.2 ml minꢀ1. GC/MS Analyses of XAD extracts and of the synthesized
NAMEs: Agilent GC 7890A system connected to a 5975C mass-selective detector (Agilent) fitted with a
HP-5 MS fused silica cap. column (30 mꢁ0.25 mm i.d., 0.22 mm film; Hewlett-Packard, Wilmington,
USA), conditions: inlet pressure; 67.5 kPa; He, 24.2 ml minꢀ1; injection volume; 1 ml; injector; 2508;
transfer line; 3008; electron energy; 70 eV. The GC was programmed as follows: 508 (5 min isothermic),
increasing at 58 minꢀ1 to 3208, and operated in splitless mode; carrier gas (He): 1.2 ml minꢀ1. Chiral GC/
MS analyses: the same system but fitted with a bꢀDEX 225 column (30 mꢁ0.32 mm i.d.; 0.25 mm film,
SGE Analytical Science, Victoria, Australia), conditions: inlet pressure; 23.1 kPa; He, 25.0 ml minꢀ1
;
injection volume; 1 ml; injector; 2108; transfer line; 3008; electron energy; 70 eV. The GC was
programmed as follows: 308 (5 min isothermic), increasing at 108 minꢀ1 to 1108 and then with 258 minꢀ1
to 2108, and operated in splitless mode; carrier gas (He): 2.0 ml minꢀ1. Retention indices (RIs) were
determined from a homologous series of n-alkanes (C8 ꢀC33). UV Spectra: Varian Cary 100 Bio
spectrometer. IR Spectra: Bruker Tensor 27 ATR spectrometer. NMR Spectra: Bruker DRX-400
(400 MHz), AV III-400 (400 MHz), or AV II-600 (600 MHz) spectrometers; referenced to TMS (d
1
0.00 ppm) for H-NMR and CHCl3 (d 77.01 ppm) for 13C-NMR, chemical shifts are in ppm, coupling
constants J in Hz.
Strains, Culture Conditions, and Extraction. Roseovarius tolerans EL-164 was obtained from the
Deutsche Sammlung fꢀr Mikroorganismen und Zellkulturen (DSMZ). Precultures were routinely grown
on marine broth medium (MB, Carl Roth, (DE-Karlsruhe)) in Erlenmeyer flasks at 288 on a rotary
shaker at 150 rpm. Erlenmeyer flasks (500 ml) containing 100 ml of MB were inoculated with 2%
preculture, and 2% of Amberlite XAD-16 (SigmaꢀAldrich, Germany) was added. After 14 d of growth,
the resin was filtered off and extracted with 3ꢁ50 ml of CH2Cl2/H2O 3 :1 (v/v). The two phases were
separated, the org. phase was dried (MgSO4), and the solvent was evaporated under reduced pressure.
The extract was concentrated at 608 under N2 to a volume of ca. 500 ml. For HPLC analysis, 400 ml of an
extract was evaporated to dryness and redissolved in 300 ml of MeCN.
Sequencing and Phylogenetic Analysis. The genome of R. tolerans EL164 was sequenced using the
Illumina sequencing technology. The draft genome consists of 273 scaffolds with 3.15 Mbp. Open reading
frames (ORFs) were identified and annotated using the automatic pipeline from the Integrated Microbial
Genomes (IMG) Database [35]. In total, 3.243 genes could be identified of which 3.200 encode proteins.
Bioassays. The NAMEs 1 and 3–5 were tested according to the following procedure. The sensor
strain P. putida pKR-C12 [34] was inoculated from plates into preculture which was grown on LB
medium (20 ml with 20 mg/ml gentamycin) at 308 overnight. The next day fresh medium was added, and
the culture was grown on a shaking platform for 1–2 h until an OD620 value of 1.0 was reached. For the
test, LB medium (99 ml) and the NAME (1 ml of a 1 mm stock soln. in DMSO) were pipetted into 96-well
microtitre plates, and the sensor strain (100 ml) was added (end concentration of NAMEs 5 mm). Each of
the NAMEs was measured in triplicate. Microtitre plates were incubated at 308 and shaken. After 0, 0.5,
1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5.0, and 5.5 h, fluorescence was determined in a Victor 1420 Multilabel
Counter (PerkinꢀElmer) at an excitation wavelength l of 485 nm and a detection wavelength l of
535 nm. OD620 was also measured. DMSO was used as negative control, and synthetic acyl homoserine
lactones (3-oxo-C12 :0-, C14 :0-, C15 :0-, C16 :0-, and C16 :1-AHLs) were used as positive controls. Fold
induction of fluorescence was calculated by dividing the specific fluorescence (gfp535/OD620) of the test
sample by the specific fluorescence of the negative control. Assays were repeated if positive-control
values were below those previously determined. To test the quorum-quenching activity, each of the
NAME 1, 3, and 5 (5 mm) was incubated with its structural analog AHL C14 :0, C15 :0, and C16 :1, resp.
(5 mm).
Derivatizations and Cleavage. To 50 ml of the sample, 50 ml of DMDS and 5 ml of I2 soln. (5% in Et2O)
were added. The mixture was heated to 608 in a closed vial for 12 h. The soln. was hydrolyzed with sat.
Na2S2O3 soln. and extracted with AcOEt (2ꢁ100 ml). The combined org. layers were dried (MgSO4),
and the sample was concentrated at 608 under N2 to a volume of ca. 30 ml [25][26].
Cleavage of the Amide Bond. For the cleavage, 100 ml of an XAD extract were evaporated to dryness
and dissolved in 200 ml of MeOH containing 3% H2SO4. The mixture was heated to 658 in a closed vial for

1570
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
48 h. Then, a spatula of NaHCO3 was added for neutralization. The solvent was evaporated at 608 under
N2, and the residue was dissolved in 100 ml of CH2Cl2.
Ethyl Chloroformate (ClCOOEt) Derivatization. For the derivatization, 50 mg of sample were
diluted in 100 ml of H2O/EtOH/pyridine 60 :32 :8. Upon addition of 5 ml of ClCOOEt, the closed reaction
tube was shaken for 5 s. Then, 100 ml of CH2Cl2 containing 1% ClCOOEt were added. The org. phase was
removed with a syringe and dried (MgSO4) [36].
Syntheses. (S)-N-(Tetradecanoyl)alanine Methyl Ester (¼ Methyl N-Tetradecanoyl-l-alaninate; 3).
Representative Procedure for Amide Coupling. Under N2, tetradecanoic acid (0.86 ml, 2.5 mmol,
1.0 equiv.) was dissolved in CH2Cl2 (12 ml) and a drop of DMF was added [37][38]. Upon slow addition
of (COCl)2 (1.48 ml, 3.75 mmol, 1.5 equiv.) at 08, the mixture was stirred until no gas formation was
observed anymore. The solvent was removed under reduced pressure. K2CO3 (1.04 g, 7.50 mmol,
3 equiv.) was added to a soln. of l-alanine methyl ester hydrochloride (0.35 g, 2.5 mmol, 1.0 equiv.) in
H2O/CH2Cl2 1:1 (v/v) at 08, and the mixture was stirred for 5 min. The acid chloride was added dropwise
in CH2Cl2 (4 ml). After stirring for 3 h, the phases were separated, and the H2O phase was extracted with
CH2Cl2 (3ꢁ5 ml). The combined org. phases were dried with MgSO4, and the solvent was removed
under reduced pressure. The crude product was purified by flash chromatography (FC; SiO2; pentane/
AcOEt 2 :1) to afford pure 3 (0.45 g, 57%). White solid. Rf (pentane/AcOEt 2 :1): 0.40. RI: 2293.
1H-NMR (400 MHz, CDCl3): 6.09 (d, J ¼ 7.0, NH); 4.61 (quint., J ¼ 7.2, CH); 3.75 (s, MeO); 2.21 (t, J¼
7.6, COCH2); 1.63 (quint., J¼7.4, COCH2CH2); 1.40 (d, J ¼ 7.2, MeCH); 1.32–1.25 (m, 10 CH2); 0.88 (t,
J ¼ 6.9, MeCH2). 13C-NMR (100 MHz, CDCl3): 173.7 (NHC¼O); 172.6 (C¼O); 52.4 (MeO); 47.8 (CH);
36.5 (CH2C¼O); 31.9 (CH2); 29.6 (4 CH2); 29.4 (CH2); 29.3 (2 CH2); 29.2 (CH2); 25.2 (CH2); 22.6
(MeCH2); 18.5 (MeCH); 14.1 (MeCH2). EI-MS (70 eV): 313 (<1, Mþ ), 254 (6), 158 (6), 145 (54), 104
(15), 102 (8), 86 (5), 69 (5), 57 (10), 55 (14), 44 (100), 43 (16).
Compounds 4 and 5 were synthesized analogously with the respective acids.
(S)-N-(Pentadecanoyl)alanine Methyl Ester (¼ Methyl N-Pentadecanoyl-l-alaninate; 4). Yield:
67%. White solid. Rf (pentane/Et2O 2 :1): 0.11. RI: 2395. 1H-NMR (400 MHz, CDCl3): 6.06 (d, J ¼ 6.8,
NH); 4.62 (quint., J¼7.2, CH); 3.75 (s, MeO); 2.21 (t, J¼7.7, COCH2); 1.63 (quint., J¼7.4, COCH2CH2);
1.40 (d, J¼7.2, MeCH); 1.30–1.25 (m, 11 CH2); 0.88 (t, J¼6.9, MeCH2). 13C-NMR (100 MHz, CDCl3):
173.7 (NHC¼O); 172.7 (C¼O); 52.4 (MeO); 47.8 (CH); 36.5 (CH2C¼O); 31.9 (CH2); 29.7(CH2); 29.6
(4 CH2); 29.4 (CH2); 29.3 (2ꢁCH2); 29.2 (CH2); 25.6 (CH2); 22.7 (MeCH2); 18.6 (MeCH); 14.1
(MeCH2). EI-MS (70 eV): 327 (2, Mþ ), 268 (7), 158 (8), 145 (64), 104 (17), 102 (8), 86 (5), 69 (6), 57
(11), 55 (15), 44 (100), 43 (17).
(S)-N-(Hexadecanoyl)alanine Methyl Ester (¼ Methyl N-Hexadecanoyl-l-alaninate; 5). Yield: 65%.
White solid. Rf (pentane/Et2O 2 :1): 0.12. RI: 2500. 1H-NMR (400 MHz, CDCl3): 5.98 (d, J ¼ 6.8, NH);
4.61 (quint., J¼7.2, CH); 3.75 (s, MeO); 2.20 (t, J¼7.7, COCH2); 1.63 (quint., J¼7.4, COCH2CH2); 1.40
(d, J¼7.2, MeCH); 1.30–1.25 (m, 12 CH2); 0.88 (t, J¼6.8, MeCH2). 13C-NMR (100 MHz, CDCl3): 173.7
(NHC¼O); 172.6 (C¼O); 52.4 (MeO); 47.9 (CH); 36.6 (CH2C¼O); 31.9 (CH2); 29.7(3 CH2); 29.6
(3 CH2); 29.5 (CH2); 29.3 (2 CH2); 29.2 (CH2); 25.6 (CH2); 22.7 (MeCH2); 18.6 (MeCH); 14.1 (MeCH2).
EI-MS (70 eV): 341 (2, Mþ ), 282 (5), 158 (8), 145 (63), 104 (18), 102 (8), 86 (5), 69 (6), 57 (12), 55 (15),
44 (100), 43 (18).
(S)-N-[(Z)-Hexadec-9-enoyl]alanine Methyl Ester (¼ Methyl N-[(9Z)-Hexadec-9-enoyl]-l-alani-
nate; 1). Under N2, the acid 11 (0.10 g, 0.39 mmol, 1.0 equiv.) was dissolved in dry CH2Cl2 (5 ml), and a
drop of DMF was added. Upon slow addition of (COCl)2 (0.05 ml, 0.59 mmol, 1.5 equiv.) at 08, the
mixture was stirred until no gas formation was observed anymore. The solvent was removed under
reduced pressure. Et3N (0.16 ml, 1.18 mmol, 3 equiv.) was added to a soln. of l-alanine methyl ester
hydrochloride (0.08 g, 0.59 mmol, 1.5 equiv.) in CH2Cl2 (5 ml), and the mixture was stirred for 5 min. The
acid chloride was added dropwise in CH2Cl2 (5 ml) [37]. After stirring for 3 h, AcOEt (25 ml) was added,
and the org. phase was washed with sat. NaHCO3 soln. (2ꢁ10 ml) and H2O (2ꢁ10 ml). The phases were
separated, and the org. phase was dried (MgSO4). The removal of the solvent under reduced pressure
was followed by purification of the crude extract by FC (RP-18; MeCN/H2O 6 :1) to give 1 (64 mg, 46%).
1
Yellow oil. Rf (MeOH/H2O 2 :1): 0.32. RI: 2477. H-NMR (400 MHz, CDCl3): 6.04 (d, J ¼ 6.8, NH);
5.84–5.30 (m, HC¼CH); 4.61 (quint., J¼7.2, CH); 3.75 (s, MeO); 2.20 (t, J¼7.7, COCH2); 2.04–1.09 (m,
CH2CH¼CHCH2); 1.63 (quint., J¼7.3, COCH2CH2); 1.40 (d, J¼7.2, MeCH); 1.35–1.24 (m, 8 CH2);

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1571
0.88 (t, J¼6.9, MeCH2). 13C-NMR (100 MHz, CDCl3): 173.7 (NHC¼O); 172.6 (C¼O); 129.9 (HC¼CH);
129.7 (HC¼CH); 52.4 (MeO); 47.8 (CH); 36.5 (CH2C¼O); 31.7 (CH2); 29.7(2 CH2); 29.2 (2 CH2); 29.1
(CH2); 28.9 (CH2); 27.2 (H2CHC¼CHCH2); 27.1 (H2CHC¼CHCH2); 25.5 (CH2); 22.6 (MeCH2); 18.6
(MeCH); 14.1 (MeCH2). EI-MS (70 eV): 339 (5, Mþ ), 282 (3), 158 (8), 145 (13), 104 (100), 81 (8), 69
(10), 67 (12), 55 (23), 44 (62), 41 (15).
Synthesis of (9-Hydroxynonyl)(triphenyl)phosphonium Bromide (9). Under N2, a soln. of 9-
bromononan-1-ol (1.00 g, 4.48 mmol, 1.0 equiv.) and Ph3P (1.29 g, 4.93 mmol, 1.1 equiv.) in MeCN was
heated to reflux for 48 h [39]. The crude product was purified by FC by gradient elution (CH2Cl2,
CH2Cl2/MeOH 15 :1, CH2Cl2/MeOH 10 :1) to afford 9 in quantitative yield. Viscous yellow oil. Rf
(CH2Cl2/MeOH 15 :1): 0.16. 1H-NMR (400 MHz, CDCl3): 7.85–7.80 (m, 9 arom. H); 7.75–7.71 (m,
6 arom. H); 3.68–3.62 (m, CH2); 3.57 (t, J¼6.6, CH2OH); 2.73 (br. s, OH); 1.67–1.59 (m, 2 CH2); 1.49
(quint., J¼6.9, CH2); 1.28–1.21 (m, 4 CH2). 13C-NMR (100 MHz, CDCl3): 134.9 (d, J¼2.9, 3 Ar); 133.3
(d, J¼10.0, 6 Ar); 103.3 (d, J¼12.5, 6 Ar); 118.0 (d, J¼85.9, 3 Ar); 62.1 (CH2OH); 32.3 (CH2CH2OH);
30.0 (d, J¼15.6, CH2); 28.7 (d, J¼10.2, CH2); 28.5 (CH2); 25.3 (CH2); 22.7 (CH2); 22.3 (CH2), 22.2
(CH2). 31P-NMR (162 MHz, CDCl3): 24.66 (s, PPh3).
Synthesis of (9Z)-Heptadec-9-en-1-ol (10). In a round-bottom flask flushed with N2, the Wittig salt 9
was dissolved in dry THF (20 ml). NaHMDS was added dropwise at 08, and the mixture was stirred for
45 min at r.t. [40]. The bright orange soln. was cooled to ꢀ788, and octanal was added slowly. After
stirring for 1 h at ꢀ788, the mixture was allowed to warm to r.t. It was poured on ice-cooled pentane
(100 ml) and Ph3PO was filtered off. After removing two thirds of the solvent, the mixture was
precipitated on SiO2 and purified by flash chromatography (SiO2; pentane/AcOEt 10 :1) to give 10
(0.22 g, 63%). Colorless liquid. Rf (pentane/AcOEt 10 :1): 0.26. RI: 1975. 1H-NMR (400 MHz, CDCl3):
5.39–5.31 (m, HC¼CH); 3.64 (t, J¼6.6, CH2OH); 2.04–1.99 (m, H2CHC¼CHCH2); 1.60–1.53 (m,
CH2CH2OH); 1.37–1.22 (m, 10 CH2); 0.88 (t, J¼6.9, Me). 13C-NMR (100 MHz, CDCl3): 130.0
(HC¼CH); 129.8 (HC¼CH); 63.1 (CH2OH); 32.8 (CH2CH2OH); 31.9 (CH2); 29.8 (CH2); 29.7 (CH2);
29.5 (CH2); 29.4 (CH2); 29.3 (CH2); 29.2 (2 CH2); 27.2 (H2CHC¼CHCH2); 25.7 (CH2); 22.7 (MeCH2);
14.1 (MeCH2). EI-MS (70 eV): 254 (<1, Mþ ), 236 (14), 152 (5), 138 (10), 123 (20), 109 (36), 96 (70), 82
(83), 67 (80), 55 (100), 43 (55), 41 (91).
Synthesis of (9Z)-Heptadec-9-enoic Acid (12). To a soln. of 10 (0.18 g, 0.71 mmol, 1.0 equiv.), N-
methylmorpholine N-oxide (NMO; 0.83 g, 7.07 mmol, 10.0 equiv.), and H2O (0.12 ml, 7.07 mmol,
10 equiv.) in MeCN (25 ml), tetrapropylammonium perruthenate (TPAP; 0.03 g, 0.07 mmol, 0.1 equiv.)
was added, and the mixture was stirred over night [41]. After removal of the solvent under reduced
pressure, the crude mixture was purified by FC (SiO2; pentane/AcOEt 10 :1 with 1% AcOH) to give 12
1
(0.11 g, 0.24 mmol, 60%). Colorless oil. Rf (pentane/AcOEt 10 :1): 0.16. H-NMR (300 MHz, CDCl3):
5.40–5.29 (m, HC¼CH); 2.35 (t, J¼7.5, CH2OH); 2.04–1.98 (m, H2CHC¼CHCH2); 1.68–1.59 (m,
CH2CH2OH); 1.40–1.22 (m, 9 CH2); 0.88 (t, J¼6.8, CH3). 13C-NMR (75 MHz, CDCl3): 179.6 (COOH);
130.0 (HC¼CH);129.7 (HC¼CH); 33.9 (CH2COOH); 31.9 (CH2); 29.8 (CH2); 29.7 (CH2); 29.5 (CH2);
29.3 (CH2); 29.2 (CH2); 29.1 (2 CH2); 29.0 (CH2); 27.2 (H2CHC¼CHCH2); 27.1 (H2CHC¼CHCH2); 24.7
(CH2); 22.7 (CH2Me); 14.1 (Me).
Compound 6 was synthesized as described for 1.
Synthesis of (S)-N-[(Z)-Heptadec-9-enoyl)alanine Methyl Ester (¼ Methyl N-[(9Z)-heptadec-9-
enoyl]-l-alaninate; 6). Yield: 67%. Yellow oil. Rf (MeCN/H2O 6 :1): 0.48. RI: 2573. 1H-NMR (400 MHz,
CDCl3): 6.01 (d, J ¼ 6.8, NH); 5.38–5.30 (m, HC¼CH); 4.61 (quint., J¼7.2, CH); 3.75 (s, MeO); 2.20 (t,
J¼7.6, COCH2); 2.03–1.08 (m, CH2CH¼CHCH2); 1.67–1.60 (m, COCH2CH2); 1.40 (d, J¼7.2, MeCH);
1.34–1.25 (m, 9 CH2); 0.88 (t, J¼6.9, MeCH2). 13C-NMR (100 MHz, CDCl3): 173.7 (NHC¼O); 172.6
(C¼O); 130.0 (HC¼CH); 129.7 (HC¼CH); 52.4 (MeO); 47.9 (CH); 36.6 (CH2C¼O); 31.9 (CH2); 29.8
(CH2); 29.7 (CH2); 29.3 (CH2); 29.2 (3 CH2); 29.1 (CH2); 27.2 (H2CHC¼CHCH2);25.5 (CH2); 22.7
(MeCH2); 18.6 (MeCH); 14.1 (MeCH2). EI-MS (70 eV): 353 (5, Mþ ), 294 (3), 158 (8), 145 (14), 104
(100), 81 (8), 69 (9), 67 (11), 55 (22), 44 (60), 41 (14).

1572
CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
REFERENCES
[1] W. C. Fuqua, S. C. Winans, E. P. Greenberg, J. Bacteriol. 1994, 176, 269.
[2] B. L. Bassler, R. Losick, Cell 2006, 125, 237.
[3] C. Fuqua, M. R. Parsek, E. P. Greenberg, Annu. Rev. Genet. 2001, 35, 439.
[4] M. R. Parsek, E. P. Greenberg, Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 8789.
[5] S. R. Chhabra, B. Philipp, L. Eberl, M. Givskov, P. Williams, M. Camara, ꢂExtracellular
communication in bacteriaꢃ, in ꢂChemistry of Pheromones and Other Semiochemicals 2ꢃ, in ꢂTopics
in Current Chemistryꢃ Vol. 240, Ed. S. Schulz, Springer, Berlin, Heidelberg, 2005, p. 279.
[6] L. Passador, J. M. Cook, M. J. Gambello, L. Rust, B. H. Iglewski, Science 1993, 260, 1127.
[7] D. G. Davies, M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, E. P. Greenberg, Science
1998, 280, 295.
[8] T. Brinkhoff, H.-A. Giebel, M. Simon, Arch. Microbiol. 2008, 189, 531.
´
[9] A. Buchan, J. M. Gonzalez, M. A. Moran, Appl. Environ. Microbiol. 2005, 71, 5665.
[10] M. Berger, A. Neumann, S. Schulz, M. Simon, T. Brinkhoff, J. Bacteriol. 2011, 193, 6576.
[11] D. Patzelt, H. Wang, I. Buchholz, M. Rohde, L. Grçbe, S. Pradella, A. Neumann, S. Schulz, S.
Heyber, K. Mꢁnch, R. Mꢁnch, D. Jahn, I. Wagner-Dçbler, J. Tomasch, ISME J. 2013, doi: 10.1038/
ismej.2013.107.
[12] J. Zan, E. M. Cicirelli, N. M. Mohamed, H. Sibhatu, S. Kroll, O. Choi, C. L. Uhlson, C. L.
Wysoczinski, R. C. Murphy, M. E. A. Churchill, R. T. Hill, C. Fuqua, Mol. Microbiol. 2012, 85, 916.
[13] R. Belas, E. Horikawa, S. Aizawa, R. Suvanasuthi, J. Bacteriol. 2009, 191, 4502.
[14] J. S. Dickschat, Nat. Prod. Rep. 2010, 27, 343.
[15] G. D. Geske, J. C. OꢃNeill, H. E. Blackwell, Chem. Soc. Rev. 2008, 37, 1432.
[16] W. R. J. D. Galloway, J. T. Hodgkinson, S. D. Bowden, M. Welch, D. R. Spring, Chem. Rev. 2011, 111,
28.
[17] I. Wagner-Dçbler, V. Thiel, L. Eberl, M. Allgaier, A. Bodor, S. Meyer, S. Ebner, A. Henning, R.
Pukall, S. Schulz, ChemBioChem 2005, 6, 2195.
[18] V. Thiel, B. Kunze, P. Verma, I. Wagner-Dçbler, S. Schulz, ChemBioChem 2009, 10, 1861.
[19] M. Teplitski, A. Eberhard, M. R. Gronquist, M. Gao, J. B. Robinson, W. D. Bauer, Arch. Microbiol.
2003, 180, 494.
[20] P. D. Shaw, G. Ping, S. L. Daly, C. Cha, J. E. Cronan Jr., K. L. Rinehart, S. K. Farrand, Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 6036.
[21] R. L. Ulrich, D. Deshazer, H. B. Hines, J. A. Jeddeloh, Infect. Immun. 2004, 72, 6589.
´
[22] M. M. Marketon, M. R. Gronquist, A. Eberhard, J. E. Gonzalez, J. Bacteriol. 2002, 184, 5686.
[23] A. L. Schaefer, E. P. Greenberg, C. M. Oliver, Y. Oda, J. J. Huang, G. Bittan-Banin, C. M. Peres, S.
Schmidt, K. Juhaszova, J. R. Sufrin, C. S. Harwood, Nature 2008, 454, 595.
[24] B. Kunze, R. Jansen, F. Sasse, G. Hçfle, H. Reichenbach, J. Antibiot. 1998, 51, 1075.
[25] S. Schulz, C. Arsene, M. Tauber, J. N. McNeil, Phytochemistry 2000, 54, 325.
[26] B. A. Leonhardt, E. D. DeVilbiss, J. Chromatogr. 1985, 322, 484.
[27] M. Camara, M. Daykin, S. R. Chhabra, Methods Microbiol. 1998, 27, 319.
[28] S. Schulz, Lipids 2001, 36, 637.
[29] R. M. Van Wagoner, J. Clardy, Structure 2006, 14, 1425.
[30] S. F. Brady, J. Clardy, J. Am. Chem. Soc. 2000, 122, 12903.
[31] S. F. Brady, J. Clardy, Org. Lett. 2005, 7, 3613.
[32] J. W. Craig, M. A. Cherry, S. F. Brady, J. Bacteriol. 2011, 193, 5705.
[33] S. F. Brady, C. J. Chao, J. Clardy, Appl. Environ. Microbiol. 2004, 70, 6865.
[34] K. Riedel, M. Hentzer, O. G. Eisenberger, B. Huber, A. Steidle, H. Wu, N. Høiby, M. Givskov, S.
Molin, L. Eberl, Microbiology 2001, 147, 3249.
[35] V. M. Markowitz, I.-M. A. Chen, K. Palaniappan, K. Chu, E. Szeto, Y. Grechkin, A. Ratner, B.
Jacob, J. Huang, P. Williams, M. Huntemann, I. Anderson, K. Mavromatis, N. N. Ivanova, N. C.
Kyrpides , Nucl. Acids Res. 2012, 40, D115.
[36] P. Husek, J. Chromatogr. 1991, 552, 289.

CHEMISTRY & BIODIVERSITY – Vol. 10 (2013)
1573
´
[37] A. Avenoza, C. Cativiela, J. H. Busto, M. A. Fernandez-Recio, J. M. Peregrina, F. Rodrꢄguez,
Tetrahedron 2001, 57, 545.
[38] S. R. Angle, R. M. Henry, J. Org. Chem. 1998, 63, 7490.
[39] R. M. Waters, D. J. Voaden, J. D. Warthen Jr., Org. Prep. Proc. Int. 1978, 10, 5.
[40] H. J. Bestmann, W. Stransky, O. Vostrowsky, Chem. Ber. 1976, 109, 1694.
[41] A.-K. C. Schmidt, C. B. W. Stark, Org. Lett. 2011, 13, 4164.
Received June 28, 2013