Stereochemical elucidation and total synthesis of dihydropacidamycin D, a semisynthetic pacidamycin

Article abstract of DOI:10.1021/ja003292c

Hydrogenation of the C(4′) exocyclic olefin of the pacidamycins has been shown to produce a series of semisynthetic compounds, the dihydropacidamycins, with antimicrobial activity similar to that of the natural products. Elucidation of stereochemistry in

Full text of DOI:10.1021/ja003292c

870
J. Am. Chem. Soc. 2001, 123, 870-874
Stereochemical Elucidation and Total Synthesis of
Dihydropacidamycin D, a Semisynthetic Pacidamycin
Constantine G. Boojamra,* Re´my C. Lemoine, Julie C. Lee, Roger Le´ger,† Karin A. Stein,‡
Nicole G. Vernier, Angela Magon, Olga Lomovskaya, Patrick K. Martin,
Suzanne Chamberland,§ May D. Lee, Scott J. Hecker, and Ving J. Lee
Contribution from Microcide Pharmaceuticals, Inc., 850 Maude AVenue,
Mountain View, California 94043
ReceiVed September 6, 2000
Abstract: Hydrogenation of the C(4′) exocyclic olefin of the pacidamycins has been shown to produce a
series of semisynthetic compounds, the dihydropacidamycins, with antimicrobial activity similar to that of the
natural products. Elucidation of stereochemistry in the pacidamycins has been completed through a campaign
of natural product degradation experiments in combination with the total synthesis of the lowest-molecular
weight dihydropacidamycin, dihydropacidamycin D. The stereochemical identities of the tryptophan and two
alanine residues contained in pacidamycin D have been shown to be of the natural (S) configuration, and the
unique 3-methylamino-2-aminobutyric acid contained in this series of antibiotics has been shown to be of the
(2S,3S) configuration. Finally, the stereochemistry obtained by hydrogenation of the C(4′)-C(5′) exocyclic
olefin has been shown to be (R) at the C(4′) nucleoside site.
Introduction
samycins18 (also known as uridyl-peptide antibiotics, or UPAs,
see Table 1), presented themselves as attractive starting points
for the development of a new class of antibacterial drugs for
three reasons. First, both the pacidamycins and mureidomycins
have demonstrated promising bioavailability and mureidomycin
C has shown in vivo efficacy.11,15 Second, these antibiotics have
an unexploited mode of action; the mureidomycins have been
shown to inhibit translocase (transferase or MraY),19-21 an
essential enzyme in peptidoglycan biosynthesis22,23 in most
Gram-positive and Gram-negative organisms and mycobacteria.
Translocase is not the target of any agents in current clinical
use. Consequently, strains of Pseudomonas aeruginosa that are
resistant to â-lactam and fluoroquinolone antibiotics remain
sensitive to mureidomycin.24 Third, unlike other translocase
inhibitors such as tunicamycin,21 which also inhibits the
formation of the pentasaccharide core found in mammalian
N-linked glycoproteins, the pacidamycins and mureidomycins
are specific for their bacterial targets. Consistent with this,
mureidomycins A and C have been shown to have low
cytotoxicity.25
The increasing occurrence of multiresistant bacterial infections
in both nosocomial and community settings has prompted the
need for new antibiotic therapies.1-7 Strategies to this end have
included modification of existing classes of antibiotics, use of
agents to potentiate existing antibiotics against resistant organ-
isms, and the identification of new classes of antibiotics that
have unexploited modes of action and thus are not cross-resistant
with old classes.8,9 Three structurally related families of natural
products, the pacidamycins,10-12 mureidomycins,13-17 and nap-
* To whom correspondence should be addressed.
† Current address: ConjuChem Inc., 225 President-Kennedy Ave., Suite
3950, Montre´al, Que´bec, Canada H2X 3Y8.
‡ Current address: Roche Bioscience, 2401 Hillview Ave., Palo Alto,
CA 94304.
§ Current address: Universite´ de Sherbrooke, 2500 Boul. Universite´,
Sherbrooke, Que´bec, Canada J1K 2R1.
(1) Brown, P. D.; Lerner, S. A. Lancet 1998, 352, 1295-1302.
(2) Spach, D. H.; Black, D. Ann. Allergy, Asthma, Immunol. 1998, 81,
293-303.
(3) Bugg, T. D.; Walsh, C. T. Nat. Prod. Rep. 1992, 9, 199-215.
(4) Gold, H. S.; Moellering, R. C., Jr. N. Engl. J. Med. 1996, 335, 1445-
1453.
(5) Cohen, M. L. Ann. Emerg. Med. 1994, 24, 454-456.
(6) Jacoby, G. A. Annu. ReV. Med. 1996, 47, 169-179.
(7) Nikaido, H. Science 1994, 264, 382-388.
(8) Renau, T. E.; Hecker, S. J.; Lee, V. J. Antimicrobial Potentiation
Approaches: Targets and Inhibitors. In Annual Reports in Medicinal
Chemistry; Academic Press: New York, 1998; Vol. 33, Chapter 12.
(9) Chu, D. T.; Plattner, J. J.; Katz, L. J. Med. Chem. 1996, 39, 3853-
74.
(10) Chen, R. H.; Buko, A. M.; Whittern, D. N.; McAlpine, J. B. J.
Antibiot. (Tokyo) 1989, 42, 512-20.
(11) Fernandes, P. B.; Swanson, R. N.; Hardy, D. J.; Hanson, C. W.;
Coen, L.; Rasmussen, R. R.; Chen, R. H. J. Antibiot. (Tokyo) 1989, 42,
521-6.
(12) Karwowski, J. P.; Jackson, M.; Theriault, R. J.; Chen, R. H.; Barlow,
G. J.; Maus, M. L. J. Antibiot. (Tokyo) 1989, 42, 506-11.
(13) Isono, F.; Sakaida, Y.; Takahashi, S.; Kinoshita, T.; Nakamura, T.;
Inukai, M. J. Antibiot. (Tokyo) 1993, 46, 1203-7.
(14) Isono, F.; Inukai, M.; Takahashi, S.; Haneishi, T.; Kinoshita, T.;
Kuwano, H. J. Antibiot. (Tokyo) 1989, 42, 667-73.
(15) Isono, F.; Katayama, T.; Inukai, M.; Haneishi, T. J. Antibiot. (Tokyo)
1989, 42, 674-9.
(16) Isono, K. J. Antibiot. (Tokyo) 1988, 41, 1711-39.
(17) Inukai, M.; Isono, F.; Takahashi, S.; Enokita, R.; Sakaida, Y.;
Haneishi, T. J. Antibiot. (Tokyo) 1989, 42, 662-6.
(18) Chatterjee, S.; Nadkarni, S. R.; Vijayakumar, E. K.; Patel, M. V.;
Ganguli, B. N.; Fehlhaber, H. W.; Vertesy, L. J. Antibiot. (Tokyo) 1994,
47, 595-8.
(19) Isono, F.; Inukai, M. Antimicrob. Agents Chemother. 1991, 35, 234-
6.
(20) Brandish, P. E.; Burnham, M. K.; Lonsdale, J. T.; Southgate, R.;
Inukai, M.; Bugg, T. D. J. Biol. Chem. 1996, 271, 7609-14.
(21) Brandish, P. E.; Kimura, K. I.; Inukai, M.; Southgate, R.; Lonsdale,
J. T.; Bugg, T. D. Antimicrob. Agents Chemother. 1996, 40, 1640-4.
(22) Pless, D. D.; Neuhaus, F. C. J. Biol. Chem. 1973, 248, 1568-76.
(23) Struve, W. G.; Sinha, R. K.; Neuhaus, F. C. Biochemistry 1966, 5,
82-93.
(24) Isono, F.; Kodama, K.; Inukai, M. Antimicrob. Agents Chemother.
1992, 36, 1024-7.
10.1021/ja003292c CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/11/2001

Synthesis of Dihydropacidamycin D
J. Am. Chem. Soc., Vol. 123, No. 5, 2001 871
Table 1. Representative Pacidamycins and Mureidomycins
Figure 1. Hydrogenation of pacidamycin D.
R1
R2
R3
R4
R5
Mura A14
Mur B14
H
H
CH2(m-HOPh) (CH2)2SCH3 m-(HOPh)
uracil
Results and Discussion
CH2(m-HOPh) (CH2)2SCH3 m-(HOPh) dihydrouracil
uracil
Mur C14 glycyl CH2(m-HOPh) (CH2)2SCH3 m-(HOPh)
Hydrogenation of Pacidamycins. Early in the course of our
work, it became apparent that the unusual C(4′)-exocyclic olefin
would prove problematic from the point of view of chemical
stability and synthetic accessibility. If hydrogenation of the 4′-
exocyclic olefin could be shown to produce active compounds,
then dihydropacidamycins would be superior to pacidamycins
as medicinal targets. Hydrogenation of pacidamycins 1, 4, and
D was performed in DMF over 10% Pd/C at 1 atm (Figure 1).
Typically, at least 24 h were required for the complete consump-
tion of starting material. The product corresponding to the
reduced 4′-exocyclic olefin (dihydropacidamycin) formed in all
cases, along with an additional product having both the 4′-exo-
cyclic olefin and the uracil double bond reduced (tetrahydro-
pacidamycin). Further hydrogenation of pacidamycin D gave
tetrahydropacidamycin D, uncontaminated with any dihydro
compound. All of these reduced pacidamycins remained com-
parably active to the parent compounds against wild-type P.
aeruginosa. These results justified a program directed toward
the total synthesis of dihydropacidamycin-like compounds.
Since dihydropacidamycin D (2) was the least structurally
complex of the semisynthetic dihydropacidamycins, it was
chosen as our target for total synthesis. Before embarking on a
total synthesis of 2, it was first necessary to determine the
stereochemistry at several chiral centers in the molecule. The
assumption was made that the configuration of each corre-
sponding chiral center was constant throughout the pacidamycins
and mureidomycins; degradative studies were performed on
pacidamycins 1, 4, and 5 since they were available in larger
amounts than pacidamycin D. The C-terminal amino acid residue
of pacidamycin 1 could be removed employing carboxypeptidase
A, indicating that this residue was likely of the natural (S)
configuration. Further, the product of this enzymatic hydrolysis
was subjected to coupling with both (S)- and (R)-phenylalanine
isocyanates. The semisynthetic compounds resulting from the
coupling with the (S) amino acid isocyanate produced a
compound with anti-pseudomonal activity, while the (R)-de-
rived material was inactive.37 Therefore, the C-terminal amino
acid was assigned the natural (S) configuration. Our objective
was to establish the remaining centers through both analysis
of degraded pacidamycins and the comparison of totally
synthetic dihydropacidamycin D with semisynthetic dihydro-
pacidamycin D.
Mur D14 glycyl CH2(m-HOPh) (CH2)2SCH3 m-(HOPh) dihydrouracil
Pac 110 alanyl CH2(m-HOPh)
Pac 210 alanyl CH2(m-HOPh)
Pac 310 alanyl CH2(m-HOPh)
Me
Me
Me
Me
Me
Me
3-indolyl
Ph
m-(HOPh)
3-indolyl
Ph
uracil
uracil
uracil
uracil
uracil
uracil
Pac 410
Pac 510
Pac D36
H
H
H
CH2(m-HOPh)
CH2(m-HOPh)
Me
3-indolyl
a Mureidomycin ) Mur; Pacidamycin ) Pac
Interestingly, these antibiotics possess selective activity
against P. aeruginosa,24 an organism that is among the most
refractory to current therapies. While a selective anti-pseudomo-
nas agent is attractive from an academic standpoint, a limited
spectrum is a liability in the clinic where broad-spectrum empiric
therapy is often necessary. Members of the mureidomycin family
have been shown to be potent inhibitors of not only the
translocase of Escherichia coli20 but also of Staphylococcus
aureus,26 suggesting that a broad-spectrum agent based on the
UPA structure is possible. The pacidamycins and mureidomy-
cins suffer from the additional limitation that P. aeruginosa
develops resistance at a high frequency (10-5-10-6).11,15 Due
to the limited spectrum and frequency of resistance, efforts by
pharmaceutical companies in this area have been sparse.27-35
We hypothesized that the problems of resistance and spectrum
were a function of uptake (or lack thereof) into bacterial cells
and that modulation of the physical properties of this class of
antibiotic might allow for improved uptake, thus achieving a
broader-spectrum agent of reduced resistance frequency. To this
end, we required a synthetic protocol that provided access to
diverse UPA analogues. Before a program to synthesize diverse
UPAs could be executed, the stereochemistry at each of the
asymmetric centers needed to be established. This publication
describes the stereochemical elucidation of the UPAs and total
synthesis of a semisynthetic derivative of a recently isolated
pacidamycin, pacidamycin D.36
(25) Inukai, M.; Isono, F.; Takatsuki, A. Antimicrob. Agents Chemother.
1993, 37, 980-3.
(26) Malouin, F. 1997, unpublished results, Microcide Pharmaceuticals,
Inc.
(27) Godfrey, O. W.; Johnson, R. D.; Kastner, R. E. U.S. Patent 4,180,-
564, December 25, 1979.
(28) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. Eur. Patent Appl. 0 253 413 A3, May 20, 1987.
(29) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. Eur. Patent Specification 0 317 292 B1, November 16, 1988.
(30) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. U.S. Patent 5,213,974, May 25, 1993.
(31) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. U.S. Patent 5,039,663, August 13, 1991.
(32) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. Eur. Patent Appl. 0 517 345 A1, November 16, 1988.
(33) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. U.S. Patent 5,041,423, August 20, 1991.
(34) Haneishi, T.; Inukai, M.; Shimizu, K.; Isono, F.; Sakaida, Y.;
Kinoshita, T. Eur. Patent 0 317 292, 1993.
Isolation and Stereochemical Determination of 3-Meth-
ylamino-2-aminobutyric Acid. A sample of pacidamycin 5
(obtained by fermentation)12,36 was hydrolyzed according to a
modified literature procedure reported for the mureidomycins.14
The 3-methylamino-2-aminobutyric acid (DABA) was isolated
by size-exclusion chromatography. A literature report regarding
the related 3-amino-2-aminobutyric acid suggested that in an
(36) Fronko, R. M.; Lee, J. C.; Galazzo, J. G.; Lee, M. D. J. Antibiot.
2000, in press.
(35) Nadkarni, S. R.; Chatterjee, S.; Patel, M. V.; Desikan, K. R.; Ganguli,
B. N.; Fehlhaber, H. W.; Rupp, R. H. Eur. Pat. Appl. 0 487 756, 1992.
(37) Le´ger, R.; Lee, V. J.; Hecker, S. H. unpublished results, Microcide
Pharmaceuticals, Inc.

872 J. Am. Chem. Soc., Vol. 123, No. 5, 2001
Boojamra et al.
Figure 2. 1H NMR of 3-methylamino-2-aminobutyric acid obtained from the natural products.
Scheme 1
acidic medium, the ammonium moieties would adopt an anti
orientation to minimize electrostatic repulsion.38 Accordingly,
a 1H NMR spectrum was obtained in 2 N DCl in D2O, revealing
two isomeric diaminobutyrates. JH2-H3 for the major isomer was
4.4 Hz. The minor component, which is likely a diastereomer
of the major component, had JH2-H3 ) 2.8 Hz. The major isomer
was surmised to be that present in the pacidamycins since
resubjection of the mixture to the hydrolysis conditions caused
equilibration of the two components to a ∼1:1 ratio. As an anti
orientation of amino groups in the (2S,3S) or (2R,3R) leads to
an anti orientation of hydrogens, expected to have a larger
coupling constant, a tentative sterochemical assignment was
made that the diaminobutyrate of the pacidamycins was either
(2S,3S) or (2R,3R) (see Figure 2). This relative stereochemical
assignment was confirmed by addition of a totally synthetic
sample (vide infra, synthesis) of (2S,3S)-DABA to an NMR
sample of the DABA isolated from pacidamycin degradation;
enhancement of the major component was observed.
The absolute configuration of this residue was determined
by HPLC, employing a chiral mobile phase.39 Pacidamycin 4
was hydrolyzed, and the resultant mixture was shown to contain
(2S,3S)-, but not (2R,3R)-DABA by comparison with authentic
synthetic samples of both (vide infra). This assignment was
further corroborated by the observations that totally synthetic
dihydropacidamycins containing the (2S,3S)-DABA had anti-
pseudomonal activity, while only inactive compounds were
obtained from the (2R,3R)-DABA.
Stereospecific Synthesis of (2S,3S)- and (2R,3R)-3-Meth-
ylamino-2-aminobutyric Acid. Both enantiomers of azetidinone
7 were prepared according to literature procedure starting from
the appropriate enantiomer of Boc-protected threonine.40,41 The
azetidinone ring was hydrolyzed using NaOH, providing
compound 8, and the N-methyl group was introduced by
reductive amination with formaldehyde (see Scheme 1), afford-
ing 9. The benzyloxy group was removed by hydrogenolysis
to provide compound 10. A sample of 10 was further depro-
tected by removal of the Boc group, and the resulting 3-meth-
ylamino-2-aminobutyric acid demonstrated an identical 1H NMR
spectrum when mixed together with the sample isolated from
the natural product, thus confirming our initial relative stereo-
chemical assignment. At no point in the reaction sequence was
there any evidence (TLC or NMR) for the production of
diasteromeric products resulting from epimerization of the
R-carbon.
Synthesis of 5′-Amino-3′,5′-dideoxyuridine Derivatives 18
and 26. Since the stereochemistry at C(4′) of dihydropacida-
mycin D was unknown, we set out to synthesize both com-
pounds 18 and 26 for incorporation into totally synthetic
dihydropacidamycins. While this contribution was in prepara-
tion, an alternate synthetic method for preparation of compound
18 appeared in the literature.42 Our synthesis of 18 began with
12, obtained according to literature procedure (see Scheme
2).43,44 The free hydroxyl group was activated for radical
reduction using thiocarbonyldiimidazole. Reduction was per-
formed employing azacyclohexylcarbonitrile and tributyltin
hydride as reducing agent, providing compound 13.45 The 5′-
silyl ether was then selectively removed using 80% acetic acid
in H2O to afford compound 14. Compound 14 was treated with
p-tosyl chloride, and the resulting tosylate 15 was converted to
azide 16. Compound 16 afforded the desired nucleoside 18 upon
deprotection with TBAF and reduction with 1,3-propanedithiol.
The azide was reduced only after the silyl ether had been
removed since it was found that reduction of the azide in the
presence of the TBS-protected alcohol led to a product that
decomposed on standing. We chose propanedithiol as our
reductant since use of hydrogenation led to the desired product
(38) Hausmann, W. K.; Borders, D. B.; Lancaster, J. E. J. Antibiot.
(Tokyo) 1969, 22, 207-10.
(42) Bender, D. M.; Hennings, D. D.; Williams, R. M. Synthesis 2000,
399-402.
(43) Ogilvie, K. K.; Beaucage, S. L.; Schifman, A. L.; Theriault, N. Y.;
Sadana, K. L. Can. J. Chem. 1978, 56, 2768-2780.
(44) Hakimelahi, C. H.; Proba, Z. A.; Ogilvie, K. K. Tetrahedron Lett.
1981, 22, 5243.
(39) Weinstein, S.; Engel, M. H.; Hare, P. E. Anal. Biochem. 1982, 121,
370-377.
(40) Floyd, D. M.; Fritz, A. W.; Pluscec, J.; Weaver, E. R.; Cimarusti,
C. M. J. Org. Chem. 1982, 47, 5160-5167.
(41) Miller, M. J.; Mattingly, P. G.; Morrison, M. A.; Kerwin, J. F., Jr.
J. Am. Chem. Soc. 1980, 102, 7026-7032.
(45) Velazquez, S.; M.-J., C. Tetrahedron 1992, 48, 1683-1694.

Synthesis of Dihydropacidamycin D
J. Am. Chem. Soc., Vol. 123, No. 5, 2001 873
Scheme 2
Scheme 3
Scheme 4
contaminated with its dihydronucleoside analogue. Furthermore,
all byproducts of this reduction were methylene chloride soluble,
while the product remained in the aqueous phase during
extraction.
The synthesis of 26 began with 19, also prepared according
to literature procedure.46 Correction of the stereochemistry at
C(4′) was addressed concomitantly with deoxygenation at the
3′ site in an initial base-mediated elimination step (see Scheme
3), modified from the method of Nagpal and Horwitz;47 the
optimized sequence involved trapping the resulting crude free
alcohol as its tert-butyldimethylsilyl ether 20. The desired
stereochemistry, (R), at the C(4′) center was installed by a
hydrogenation directed to the â face of the molecule by the
bulky silyl ether-protecting group. The desired stereoisomer was
obtained in a 9:1 ratio relative to the undesired, which was
removed by recrystallization of 21. Brief treatment with 9/1
TFA:CH2Cl2 was found to be optimal for selectively depro-
tecting the carboxylic acid over the silylated alcohol; the free
carboxylic acid resulting from these deprotection conditions was
subjected to a two-step activation-reduction sequence, affording
alcohol 22. Comparison of isomers 22 and 14 provided un-
equivocal proof that we had installed the desired (R) stereo-
chemistry at C(4′) since 22 and 14 were spectroscopically
distinct, and 14 was known to be of the (S) configuration at
C(4′). Nitrogen functionality was introduced by tosylation of
22 followed by azide displacement, yielding compound 24.
The silyl ether was removed under standard conditions, and
finally the azide was reduced using neat 1,3-propanedithiol,
affording analytically pure compound 26, after an extractive
purification.
Total Synthesis of Dihydropacidamycin D (2). Given our
earlier stereochemical deductions, only three asymmetic centers
remained to be assigned: the R2 and R3 amino acid residues
(see Table 1), and C(4′) of the ribose. Fortunately, both
stereoisomers of the two amino acid components were com-
mercially available with the appropriate protecting groups, and
both C(4′) stereoisomers (18 and 26) were available via the
synthetic methods described above. Thus, we set out to pre-
pare all eight (23) possible isomers and to determine the
stereochemistry of dihydropacidamycin D through compari-
son of biological activity and physical properties of these
isomers with semisynthetic 2. Our synthetic protocol is exempli-
fied with L-amino acids and nucleoside 26 since these are the
components that provided a totally synthetic product with
physical and biological properties identical to that of semisyn-
thetic 2.
The urea portion of the molecule was prepared according to
the method of Majer and Randad.48 H-Ala-OBn and H-Trp-
(46) Corey, E. J.; Samuelsson, B. J. J. Org. Chem. 1984, 49, 4735.
(47) Nagpal, K. L.; Horwitz, J. P. J. Org. Chem. 1971, 36, 3743-3745.

874 J. Am. Chem. Soc., Vol. 123, No. 5, 2001
Boojamra et al.
Scheme 5
Table 2. Minimum Inhibitory Concentrations (MICs) for Key
Compounds
played significant activity against wild-type P. aeruginosa,
equivalent to that measured for the semisynthetic compound 2
(see Table 2).
minimum inhibitory
concentrations (µg/mL)49
strain
description
wild-type
Paca 4 Pac D
2
34
Conclusions
P. aeruginosa
32
64
64 32
In conclusion, the identity of the stereocenters of the
pacidamycins has been elucidated; all proteinogenic amino acids
in the pacidamycins are of the natural, (S) configuration, and
the DABA residue is of the (2S,3S) configuration. Hydrogena-
tion of the pacidamycins results in biologically active com-
pounds containing the (R) configuration at C(4′). A straight-
forward method is reported for the total synthesis of these
C(4′)-C(5′) saturated pacidamycins. This allows synthesis of
structurally varied pacidamycins based on adaptation of the
methodology reported herein to include other amino acid
residues, thereby facilitating the search for a broad-spectrum
agent of this type. The biological activities of a diverse synthetic
set of dihydropacidamycins will be reported in due course.
PAM102050
P. aeruginosa PAM1020 oprM::Hg51
4
4
8
4
PAM115450
OtBu were condensed with triphosgene, and the resulting
unsymmetrical urea 27 was debenzylated by hydrogenation at
1 atm. The liberated carboxylic acid was converted to the
pentafluorophenyl ester 29 under standard conditions (see
Scheme 4).
Boc-protected DABA (10) was coupled to the pentafluo-
rophenyl ester of Fmoc-L-alanine (Scheme 5). The dipeptide
30 was activated as its pentafluorophenyl ester and coupled with
aminonucleoside 26, providing compound 31. The C-terminal
portion of the molecule was appended by removal of the Boc
group followed by coupling with 29, affording 32. Deprotection
and subsequent HPLC purification provided 34, a compound
whose 1H NMR spectra and HPLC retention time were identical
to those of compound 2. The remaining seven stereoisomers
were synthesized by similar methods and their biological
activities measured. As final proof that 34 was indeed the
stereoisomer corresponding to semisynthetic 2, only 34 dis-
Acknowledgment. We gratefully acknowledge the assistance
of James Kanter in the preparation of nucleoside 18. We also
gratefully acknowledge Professor Jonathan A. Ellman and Derek
Cogan for useful discussions and technical assistance. Finally
we acknowledge Dr. Johanne Blais, Cynthia Dinh, and Monica
Hoang for microbiology assistance.
Supporting Information Available: All experimental pro-
cedures and compound characterization; reproductions of the
1H and 13C NMR spectra of totally synthetic dihydropacida-
mycin D (34) (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
(48) Majer, P.; Randad, R. S. J. Org. Chem. 1994, 59, 1937-1938.
(49) MICs were calculated by the broth microdilution method in
accordance with NCCLS standards: NCCLS 1997, 17(2).
(50) Lomovskaya, O.; Lee, A.; Hoshino, K.; Hiroko, I.; Mistry, A.;
Warren, M. S.; Boyer, E.; Chamberland, S.; Lee, V. J. Antimicrob. Agents
Chemother. 1999, 43, 1340-1346.
(51) Disruption of oprM results in inactivation of the MexAB-OprM
efflux pump.
JA003292C