A plant-like biosynthesis of benzoyl-CoA in the marine bacterium 'Streptomyces maritimus'

Article abstract of DOI:10.1016/S0040-4020(00)00765-1

The first plant-like biosynthesis of benzoic acid in a prokaryote, the marine actinomycete 'Streptomyces maritimus', has been characterized. Feeding experiments with ²H- and ¹³C-labeled intermediates revealed that phenylalanine is metabolized to benzoyl-coenzyme A (CoA) by means of a phenylalanine ammonia lyase and subsequent β-oxidation of cinnamoyl-CoA. Benzoyl-CoA serves as the rare starter unit of a type II polyketide synthase producing the structurally novel bacteriostatic polyketides enterocin and the wailupemycins. The results from the feeding study are consistent with the proposed biosynthetic model deduced from the cloned enterocin biosynthesis gene cluster. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00765-1

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 9115±9120
A Plant-like Biosynthesis of Benzoyl-CoA in the Marine
Bacterium `Streptomyces maritimus'
Christian Hertwecka and Bradley S. Moorea,b,
*
aDepartment of Chemistry, University of Washington, Box 351700, Seattle, WA 98195-1700, USA
bDivision of Medicinal Chemistry, College of Pharmacy, University of Arizona, Tucson, AZ 85721-0207, USA
Dedicated to Professor Paul J. Scheuer, B. S. M.'s scienti®c grandfather, in celebration of his 50-year tenure at the University of Hawaii and his inspiration
to the ®eld of marine natural products
Received 12 May 2000; accepted 25 June 2000
AbstractÐThe ®rst plant-like biosynthesis of benzoic acid in a prokaryote, the marine actinomycete `Streptomyces maritimus', has been
characterized. Feeding experiments with 2H- and 13C-labeled intermediates revealed that phenylalanine is metabolized to benzoyl-coenzyme
A (CoA) by means of a phenylalanine ammonia lyase and subsequent b-oxidation of cinnamoyl-CoA. Benzoyl-CoA serves as the rare starter
unit of a type II polyketide synthase producing the structurally novel bacteriostatic polyketides enterocin and the wailupemycins. The results
from the feeding study are consistent with the proposed biosynthetic model deduced from the cloned enterocin biosynthesis gene cluster.
q 2000 Elsevier Science Ltd. All rights reserved.
Introduction
benzoyl-CoA.6 Other rare examples of bacterial natural
products possessing a probable benzoate-derived residue
include thiangazole7 and benzoyl a-l-rhamno-pyranoside.8
In eukaryotic systems, benzoic acid is conversely a common
metabolite and is a component of many important natural
products, including salicylic acid,9 cocaine,10 taxol,11 and
the zaragozic acids.12
The bacteriostatic agents enterocin (vulgamycin) (2) and
wailupemycins A±D (3±6) are members of a series of struc-
turally diverse polyketides produced by the marine bac-
terium `Streptomyces maritimus' (Scheme 1).1±3 Classical
feeding experiments demonstrated that enterocin is derived
from benzoic acid (1), seven acetate units, and the methyl
group of methionine and undergoes a novel Favorskii-like
carbon rearrangement.4 We recently cloned, sequenced, and
heterologously expressed the enterocin biosynthesis gene
cluster (enc) and demonstrated that the 21.3 kb cluster
contains an iterative type II polyketide synthase (PKS)
and is extremely versatile in encoding the biosynthesis of
the diverse set of polyketides.3,5 Feeding studies with [ring-
d5]benzoic acid to the transformant S. lividans K4-114/
pJP15F11-1 that carries the enc cluster established the
breadth of polyketide biosynthesis by this single gene set
as nine uncharacterized structural analogs were additionally
produced.3
In spite of its wide occurrence in plants and fungi, the
biosynthesis of benzoic acid has not been ®rmly estab-
lished.9 Two pathways have been characterized which
similarly involve conversion of phenylalanine (7) to
cinnamic acid (8) by phenylalanine ammonia lyase (PAL)
(Scheme 2). The routes diverge at this intermediate, one
involving a b-oxidation pathway (route a) and the second
a retro-aldol path through benzaldehyde (route b). The only
known bacterial benzoate pathway is anaerobic and
involves transamination of phenylalanine to phenylpyruvate
followed by two successive a-oxidative decarboxylations
(route c).13 Recently, a direct conversion of shikimic acid
to benzoic acid was postulated in a streptomycete (route d).8
Benzoic acid is a rare PKS starter unit that has been impli-
cated in only one other bacterial polyketide biosynthetic
pathway. The myxobacterial antifungal macrolide soraphen
A is derived from a modular type I PKS that is primed with
Analysis of the enc gene cluster revealed seven open
reading frames (ORFs) arranged on four transcripts
neighboring the minimal enc PKS that are putatively
involved in the biosynthesis and attachment of the
benzoyl-CoA starter unit.5 The nucleotide sequence of
these ORFs is very suggestive for a eukaryotic-like benzoic
acid biosynthetic pathway in this prokaryote. Feeding
experiments with labeled intermediates were conducted in
`S. maritimus' and veri®ed that the enterocin benzoic acid
starter unit is biosynthesized via the plant-like pathway.
Keywords: benzoic acid; biosynthesis; phenylalanine ammonia lyase; poly-
ketide synthase; Streptomyces.
* Corresponding author. Division of Medicinal Chemistry, College of
Pharmacy, University of Arizona, Tucson, AZ 85721-0207, USA. Tel.:
11-520-626-6931; fax: 11-520-626-2466;
e-mail: moore@pharmacy.arizona.edu
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00765-1

9116
C. Hertweck, B. S. Moore / Tetrahedron 56 (2000) 9115±9120
Scheme 1. Structures of enterocin (2) and wailupemycins A±D (3±6) from `Streptomyces maritimus' and their biosynthesis from benzoyl-CoA (1a).
Scheme 2. Biosynthetic routes to benzoyl-CoA in plants and fungi (paths a and b), in anaerobic bacteria (path c) and from shikimic acid (path d).

C. Hertweck, B. S. Moore / Tetrahedron 56 (2000) 9115±9120
9117
Scheme 3. Reagents: (i) DBU, LiCl, (EtO)2P(O) 13CH213CO2Et, MeCN, rt; (ii) 1 M NaOH, rt; (iii) CrCl2, LiI, Br13CH213CO2Et, THF, rt 2558C; (iv) PDC,
DCM, rt. The asterisks denote 13C atoms.
Results and Discussion
yielded the pure free acids 8±10 without side reactions. All
labeled compounds exhibited a very high (.98%) degree of
isotopic labeling.
Although analysis of the enc gene cluster suggested a plant-
like benzoic acid biosynthetic pathway,5 it did not unequi-
vocally differentiate between the b-oxidative and the retro-
aldol routes. Furthermore, while benzoyl-CoA is the
accepted enterocin PKS starter unit, b-ketophenylpropio-
nyl-CoA 10a may be an alternative starter unit that requires
one fewer malonyl-CoA condensations. Thus, doubly
labeled phenylpropanoid biosynthetic intermediates were
synthesized with label in both the phenyl ring and the propyl
side chain to determine the metabolic fate of the precursor in
enterocin biosynthesis.
Feeding studies
The intermediacy of phenylalanine (7) in the `S. maritimus'
benzoic acid biosynthetic pathway was initially con®rmed
in a feeding experiment with [ring-d5]phenylalanine. A 24%
incorporation was measured by mass spectrometry on the
enriched enterocin (Table 1). The MS fragmentation pattern
of the pentadeuterated enterocin was consistent with the
label con®ned to the benzyl ring. This high incorporation
was consistent with that of enterocin derived from
[d5]benzoic acid which in a parallel culture was likewise
enriched at 31%.
Synthesis of precursors
Synthesis of the doubly labeled putative phenylpropionyl
intermediates 8±10 was straightforward and highly ef®cient
as outlined in Scheme 3. Horner±Emmons reaction of
commercially available [ring-d5]benzaldehyde and triethyl
[13C2]phosphonoacetate using DBU/LiCl as base14 provided
the ethyl (E)-cinnamate 8b in high yield. GC and NMR
analyses corroborated that the cinnamic ester was free
from the (Z)-isomer and displayed a high degree of isotope
labeling. Labeled ethyl 3-hydroxy-3-phenylpropionate 9b
was conveniently prepared as the racemate from [ring-
d5]benzaldehyde and ethyl [1,2-13C2]bromoacetate by a
novel chromium Reformatsky reaction.15 While the classic
zinc variant of this reaction is known to be sluggish and
unreliable on a small scale, the chromium variation offered
an excellent yield (78%) even on a microscale (0.5 mmol).
Subsequent PDC oxidation16 of the b-hydroxy ester almost
quantitatively furnished the corresponding b-keto propio-
nate 10b under mild conditions. Basic hydrolysis of all
three ethyl esters with 1 M NaOH at ambient temperature
Similarly, high incorporation of the phenylpropanoids 8±10
(10±36%) was detected by MS analysis (Table 1), yet only
after loss of the 13C2-label in the side chain. No intact incor-
poration was detected, clearly indicating that the propionate
side chain is cleaved before incorporation of the benzoyl unit.
Thus, benzoyl-CoA (1a), and not b-ketopropionyl-CoA
10a, is the enc starter unit. Relative to cinnamate, approxi-
mately half of racemic 3-hydroxypropionate 9 was incorpo-
rated, leading to the assumption that only one enantiomer is
accepted by the dehydrogenase. In fact, similar observations
have been made during biosynthetic studies of cocaine,
where a dramatic preference for the R-enantiomer was
exhibited.10 Whether or not a stereospeci®c dehydrogenase
is involved in the `S. maritimus' benzoic acid pathway, as
well as its stereospeci®city, will be addressed at a later time
by asymmetric synthesis and feeding of labeled (R)- and
(S)- hydroxypropionates. Furthermore, the relative low
incorporation of the b-ketoacid 10 is probably due to its
instability in culture, which results in spontaneous
decarboxylation. Its incorporation, nonetheless, provides
strong support for the plant-like b-oxidative pathway. The
bacterial a-oxidative pathway (route c) does not appear to
be operative in `S. maritimus', as in addition to the genetic
analysis of the cluster, the a-oxidative intermediate phenyl-
acetate (11) was not incorporated into the benzoate-derived
moiety of enterocin (Table 1).
Table 1. Relative incorporation of labeled precursors into enterocin (2)
Precursora
% Enrichment
in 2
[ring-2H5]Benzoic acid (1)
31
[ring-2H5]Phenylalanine (7)
(2E)-[ring-2H5,1,2-13C2]Cinnamic acid (8)
[ring-2H5,1,2-13C2]-3-Hydroxy-3-phenylpropionic
acid (9)
24
36b
16b
[ring-2H5,1,2-13C2]-3-Oxo-3-phenylpropionic acid
(10)
10b
These data are consistent with the genetic analysis of the enc
biosynthetic gene set. Conversion of l-phenylalanine to
trans-cinnamic acid by the putative encP gene product
PAL constitutes the initial reaction in the pathway (Scheme
4). This is only the second report of a PAL catalyzed reac-
tion in a bacterium17 and the ®rst associated with a bacterial
biosynthesis of benzoic acid. Activation by the putative
[ring-2H5]Phenylacetic acid (11)
n.d.c
a 40 mmol precursor/100 mL `S. maritimus' culture.
b Enrichment based on the incorporation of the d5-benzyl group with loss
of the side chain 13C2-label.
c Not detectable.

9118
C. Hertweck, B. S. Moore / Tetrahedron 56 (2000) 9115±9120
Scheme 4. Proposed biosynthesis of benzoyl-CoA in `S. maritimus' and arrangement of the putative biosynthetic genes in the enc gene cluster (GenBank
accession number AF254925). EncH: acyl-CoA ligase; EncI: Enoyl-CoA hydratase; EncJ: b-ketothiolase; EncABC: minimal PKS; EncL: aryl transferase;
EncM: oxygenase; EncN: aryl-CoA ligase; EncP: phenylalanine ammonia lyase (PAL).
acyl-CoA ligase EncH to cinnamoyl-CoA (8a) followed by
a b-oxidative pathway involving hydration (putatively cata-
lyzed by EncI), dehydrogenation (putatively catalyzed by
EncM), and thiolysis (putatively catalyzed by EncJ) yields
the benzoyl-CoA starter unit. Thiolysis may alternatively
generate the free acid, which would then require conversion
to the CoA thioester 1a by the putative EncN aryl-CoA
ligase. Correlation of the biochemical reactions to the
putative functional assignments of the enc gene products
will be resolved at the biochemical and genetic levels
with recombinant proteins.
glassware using standard gas tight syringes, cannulas and
septa. Solvents and reagents were dried prior to use accord-
ing to standard procedures. For column chromatographic
separations 230±400 mesh silica gel (Aldrich) was used.
NMR spectra were recorded on a Bruker AM-500 spec-
trometer. Chemical shifts are given in ppm and are adjusted
to the TMS scale by reference to the solvent signal.
Coupling constants (J) are given in Hertz (Hz). GC-MS
was carried out on a Hewlett±Packard 5890 GC connected
to a 5971A Mass Spectrometer. LC-MS analyses were
performed using a Hewlett±Packard HP1100 LC pump
with a photo diode-array detector linked to a Bruker HP
Esquire Ion Trap Mass Spectrometer operating in the nega-
tive ion mode. A Beckman Ultrasphere C18 column
(4.6 mm£15 cm) was used at a ¯ow rate of 1 mL/min
with a linear solvent gradient of 5% MeOH in 0.15%
TFA/H2O to 100% MeOH over a period of 45 min.
In conclusion, the ®rst plant-like benzoate pathway in a
bacterium has been characterized in `S. maritimus'. Feeding
experiments with labeled intermediates support the
biochemical pathway suggested by the sequence analysis
of the enc gene cluster. Due to the clustering of the benzoate
biosynthesis genes in this prokaryote, we regard
`S. maritimus' as an excellent model to fully characterize
the eukaryotic-like benzoic acid biosynthetic pathway at the
enzymatic and genetic levels.
Materials
Triethyl [13C2]phosphonoacetate, chromium (II) chloride
(anhydrous), and LiI (anhydrous) were purchased from
Aldrich Co., [ring-2H5]benzoic acid, [ring-2H5]phenyl-
alanine, and ethyl [13C2]bromoacetate were obtained from
Cambridge Isotope Labs, and [ring-2H5]benzaldehyde and
[ring-2H5]phenylacetic acid were acquired from CDN
Isotopes.
Experimental
General methods
All reactions were carried out under argon in ¯ame-dried

C. Hertweck, B. S. Moore / Tetrahedron 56 (2000) 9115±9120
9119
Fermentation
(9). To a gray-green suspension of anhydrous CrCl2 (251
mg, 2.05 mmol) and anhydrous LiI (11 mg, 0.08 mmol) in
dry THF (3 mL) were added [ring-2H5]benzaldehyde (75 ml,
0.74 mmol) and ethyl [13C2]bromoacetate (91 ml, 0.82
mmol). The mixture was stirred at ambient temperature
for 15 min and then warmed to 558C for 1 h. The maroon
suspension was cooled to ambient temperature, quenched
with brine (3 mL) and vigorously stirred for 15 min. The
organic layer was separated, and the aqueous phase was
extracted with Et2O (3£5 mL). The combined organic
extracts were washed with H2O, dried (Na2SO4), and
concentrated under reduced pressure. The residue was
puri®ed on silica with 2:1 hexane/Et2O and gave the ethyl
ester 9b (116 mg) as a colorless oil in 78% yield. EIMS [IP
70 eV; m/z (% rel int)] 201 (M1z, 31), 172 (4), 156 (6), 126
(12), 112 (100), 110 (62), 90 (21), 84 (62), 72 (10), 62 (19),
52 (11). Ester 9b (40 mg, 0.2 mmol) was subsequently
hydrolyzed as described for 8b to the colorless oil 9
(28 mg) in 81% yield.19 1H NMR (CDCl3) d 2.78 (2H,
dddd, Jˆ129, 17.8, 6.2, 3.7 Hz, H-2), 5.15 (1H, m, H-3),
(no aromatic H detectable).20
`S. maritimus' spores from a plate culture were inoculated
into 100 mL of A1 medium in a 500 mL Erlenmeyer ¯ask.
The medium contained 10 g soluble potato starch, 4 g yeast
extract, 2 g peptone, and 28 g Instant Oceanw per liter H2O
and buffered at pH 8 with 10 mL of 1 M Tris buffer. The
inoculated culture was incubated at 288C for 2 days on a
rotary shaker at 200 rpm. The inoculum (10 mL) was then
transferred to 90 mL of A1 and incubated at 288C for 5 days
with shaking at 200 rpm.
Feeding experiments with labeled precursors
Single doses of precursors were administered to the fermen-
tation at the time of inoculation, and the cultures were
harvested 5 days later. Precursors were added as sterile
solutions (40 mmol in 100 ml DMSO) in the amounts
indicated per 100 mL culture: [ring-2H5]phenylalanine
(7.9 mg), (20E)-[ring-2H5,1,2-13C2]cinnamic acid (7.3 mg),
[ring-2H5,1,2-13C2]-3-hyroxy-3-phenylpropionic acid (8.1
mg), [ring-2H5,1,2-13C2]-3-oxo-3-phenylpropionic acid
(8.0 mg), [ring-2H5]phenylacetic acid (5.6 mg), and [ring-
2H5]benzoic acid (5.1 mg). The cultures were centrifuged
at 6000£g for 8 min, and the supernatants were extracted
with EtOAc (3£50 mL). The crude extracts were dried
(Na2SO4) and evaporated in vacuo. The residue was
dissolved in 0.5 mL MeOH and analyzed by LC-ESMS
(5 ml in 500 ml MeOH). Enterocin (2): ESMS (m/z) 443
(M2H), 425 (M2H3O), 399, 381, 355, 337, 275, 257.
Enterocin-d5: ESMS (m/z) 448 (M2H), 430 (M2H3O),
404, 386, 360, 342, 280, 262.
[ring-2H5,1,2-13C2]-3-Oxo-3-phenylpropionic acid (10).
b-Hydroxyester 9b (50 mg, 0.25 mmol) was added to an
ice-cold suspension of pyridinium chlorochromate
(0.4 mmol) in CH2Cl2 (1.5 mL) and stirred at ambient
temperature for 15 h. The suspension was ®ltered, the
®ltrate dried in vacuo and the residue chromatographed on
silica with 3:1 hexane/ Et2O providing 45 mg ethyl ester 10b
as a colorless oil in 91% yield.16 EIMS [IP 70 eV; m/z (% rel
int)]: 200 (M1z11, 35), 154 (9) 138 (17), 126 (24), 110
(100), 82 (41), 54 (17), 45 (57). Ester 10b (40 mg,
0.2 mmol) was subsequently hydrolyzed as described for
8b to give the free acid 10 (29 mg, 84% yield) as a colorless
oil.21 1H NMR (CDCl3) d (mixture of 76% keto and 24%
enol tautomers22) 3.98 (1.52H, dd, Jˆ131, 18 Hz, H-2),
5.69 (0.24H, d, Jˆ176 Hz, enol H-2), (no aromatic H
detectable).23
Synthesis of labeled precursors
(2E)-[ring-2H5,1,2-13C2]Cinnamic acid (8). To a supension
of LiCl (32.2 mg, 0.6 mmol) in dry MeCN (1.0 mL) at 08C
were subsequently added triethyl [13C2]phosphonoacetate
(120 ml, 0.6 mmol), 1,8-diazabicyclo-[5.4.0]-undec-7-ene
(75 ml, 0.5 mmol) and [ring-2H5]benzaldehyde (51 ml,
0.5 mmol). After stirring for 10 min at ambient temperature,
H2O (1 mL) was added and immediately the pH was
adjusted to 7.0 by quick titration with dilute HCl. The
aqueous phase was extracted with Et2O (3£5 mL) and the
combined organic phases were washed with brine and dried
over MgSO4. After evaporation of the solvent in vacuo, the
residue was puri®ed on silica with 1:1 pentane/Et2O, provid-
ing 80.2 mg ethyl ester 8b as a colorless oil in 87% yield.
EIMS [IP 70 eV; m/z (% rel int)] 183 (M1z, 28), 155 (17),
138 (100), 109 (42), 81 (15), 54 (7). Ester 8b (0.3 mmol,
55 mg) was added to 1 M aqueous NaOH (2 mL, excess)
and the emulsion was stirred overnight at ambient tempera-
ture, at which time no ester was detectable by TLC analysis.
The reaction mixture was extracted with Et2O (2£2 mL),
and the ethereal solution was discarded. The aqueous
phase was acidi®ed to pH 1 with 1N HCl and extracted
with Et2O (5£3 mL). The combined organic extracts were
dried (Na2SO4) and concentrated to afford 38 mg (81%) of
pure acid 8. 1H NMR (CDCl3) d 6.41 (1H, ddd, Jˆ163, 16,
2.8 Hz, H-2), 7.75 (1H, ddd, Jˆ16, 6.75, 2.6 Hz, H-3), (no
aromatic H detectable).18
Acknowledgements
We thank B. S. Davidson (Utah State U) for the strain
`S. maritimus' and M. Sadilek (UW) for LC-MS assistance.
This research was generously supported by the National
Oceanic and Atmospheric Administration, US Department
of Commerce, through the National and Washington Sea
Grant programs (NA76RG01149, project no. R/B-28) and
by the NIH (AI47818). C. H. wants to thank the Alexander
von Humboldt foundation for support by a Feodor Lynen
postdoctoral fellowship.
References
1. Sitachitta, N.; Gadepalli, M.; Davidson, B. S. Tetrahedron
1996, 52, 8073.
2. Bernan, V. S.; Piel, J.; Davidson, B. S.; Moore, B. S. Submitted
for publication; GenBank accession number AF233338.
3. Piel, J.; Hoang, K.; Moore, B. S. J. Am. Chem. Soc. 2000, 122,
5415.
4. Seto, H.; Sato, T.; Urano, S.; Uzawa, J.; Yonehara, H.
Tetrahedron Lett. 1976, 4367.
[ring-2H5,1,2-13C2]-3-Hydroxy-3-phenylpropionic acid

9120
C. Hertweck, B. S. Moore / Tetrahedron 56 (2000) 9115±9120
5. Piel, J.; Shipley, P. R.; Hertweck, C.; Hunt, D. M.; Newman,
M. S.; Moore, B. S. Submitted for publication.
1998, 251, 907; Rhee, S. K.; Fuchs, G. Eur. J. Biochem. 1999, 262,
507.
6. Schupp, T.; Toupet, C.; Cluzel, B.; Neff, S.; Hill, S.; Beck, J. J.;
Ligon, J. M. J. Bacteriol. 1995, 177, 3673; Hill, A. M.; Harris, J. P.;
Siskos, A. P. J. Chem. Soc., Chem. Commun. 1998, 2361.
7. Jansen, R.; Kunze, B.; Reichenbach, H.; Jurkiewicz, E.;
14. Trost, B. M.; Metz, P.; Hane, J. T. Tetrahedron Lett. 1986, 27,
5691; Coe, J. W.; Roush, W. R. J. Org. Chem. 1989, 54, 915.
15. Wessjohann, L.; Gabriel, T. J. Org. Chem. 1997, 62, 3772.
16. Corey, E. J., Schmidt, G. Tetrahedron Lett. 1979, 399;
Fadnavis, N. W.; Vadivel, S. K.; Sharfuddin, M.; Bhalerao, U. T.
Tetrahedron: Asymmetry 1997, 8, 4003.
È
Hunsmann, G.; Ho¯e, G. Liebigs Ann. Chem. 1992, 357; Kunze,
B.; Jansen, R.; Pridzun, L.; Jurkiewicz, E.; Hunsmann, G.; Ho¯e,
È
G.; Reichenbach, H. J. Antibiot. 1993, 46, 1752.
8. Grond, S.; Langer, H.-J.; Henne, P.; Sattler, I.; Thiericke, R.;
17. Bezanson, G. S.; Desaty, D.; Emes, A. V.; Vining, L. C. Can.
J. Microbiol. 1970, 16, 147.
È
Grabley, S.; Zahner, H.; Zeeck, A. Eur. J. Org. Chem. 2000,
929.
18. Freeman, F.; Chang, L. Y.; Kappos, J. C.; Sumarta, L. J. Org.
Chem. 1987, 52, 1460.
9. Ribnicky, D. M.; Shulaev, V.; Raskin, I. Plant Physiol. 1998,
118, 565; Jarvis, A. P.; Schaaf, O.; Oldham, N. J. Planta (in press).
10. Bjorklund, J. A.; Leete, E. Phytochemistry 1992, 31, 3883.
11. Rohr, J. Angew. Chem., Int. Ed. 1997, 36, 2190.
12. Bergstrom, J. D.; Dufresne, C.; Bills, G. F.; Nallinomstead,
M.; Byrne, K. Annu. Rev. Microbiol. 1995, 49, 607.
13. Schneider, S.; Mohamed, M. E.; Fuchs, G. Arch. Microbiol.
1997, 168, 310; Schneider, S.; Fuchs, G. Arch. Microbiol. 1998,
169, 509; Hirsch, W.; Schagger, H.; Fuchs, G. Eur. J. Biochem.
19. Boaz, N. W. J. Org. Chem. 1992, 57, 4289.
20. Fleming, I.; Winter, S. B. D. J. Chem. Soc., Perkin Trans. 1
1998, 2687.
21. Ohta, S.; Tsujimura, A.; Okamoto, M. Chem. Pharm. Bull.
1981, 29, 2762.
22. Rosenfeld, S. M.; Cotell, C. M.; Smith, J. C. J. Chem. Soc.,
Chem. Commun. 1985, 402.
23. Rathke, M. W.; Nowak, M. A. Synth. Commun. 1985, 15,
1039.