Synthesis and preliminary biological characterization of new semisynthetic derivatives of ramoplanin

Article abstract of DOI:10.1021/jm070042z

Ramoplanin is a glycolipodepsipeptide antibiotic active against Gram-positive bacteria including vancomycin-resistant enterococci. Ramoplanin inhibits bacterial cell wall biosynthesis by a mechanism different from that of glycopeptides and hence does not show cross-resistance with these antibiotics. The systemic use of ramoplanin has been so far prevented because of its low local tolerability when injected intravenously. To overcome this problem, the fatty acid side chain of ramoplanin was selectively removed and replaced with a variety of different carboxylic acids. Many of the new ramoplanin derivatives showed antimicrobial activity similar to that of the natural precursor coupled with a significantly improved local tolerability. Among them the derivative in which the 2-methylphenylacetic acid has replaced the di-unsaturated fatty acid side chain (48) was selected as the most interesting compound and submitted to further in vitro and in vivo characterization studies.

Full text of DOI:10.1021/jm070042z

J. Med. Chem. 2007, 50, 3077-3085
3077
Synthesis and Preliminary Biological Characterization of New Semisynthetic Derivatives of
Ramoplanin
Romeo Ciabatti,* Sonia I. Maffioli, Gianbattista Panzone, Augusto Canavesi, Elena Michelucci, Paolo S. Tiseni,
Ettore Marzorati, Anna Checchia, Matteo Giannone, Daniela Jabes, Gabriella Romano`, Cristina Brunati,
Gianpaolo Candiani, and Franca Castiglione
Vicuron Pharmaceuticals, Gerenzano, Varese, Italy
ReceiVed January 12, 2007
Ramoplanin is a glycolipodepsipeptide antibiotic active against Gram-positive bacteria including vancomycin-
resistant enterococci. Ramoplanin inhibits bacterial cell wall biosynthesis by a mechanism different from
that of glycopeptides and hence does not show cross-resistance with these antibiotics. The systemic use of
ramoplanin has been so far prevented because of its low local tolerability when injected intravenously. To
overcome this problem, the fatty acid side chain of ramoplanin was selectively removed and replaced with
a variety of different carboxylic acids. Many of the new ramoplanin derivatives showed antimicrobial activity
similar to that of the natural precursor coupled with a significantly improved local tolerability. Among
them the derivative in which the 2-methylphenylacetic acid has replaced the di-unsaturated fatty acid side
chain (48) was selected as the most interesting compound and submitted to further in vitro and in vivo
characterization studies.
Introduction
Ramoplanin (see Figure 1) is a glycolipodepsipeptide anti-
biotic obtained from fermentation of Actinoplanes ATCC33076.
It is active against Gram-positive aerobic and anaerobic bacteria,
including VRE.a,1 Ramoplanin inhibits bacterial cell wall
biosynthesis by a mechanism different from those of other cell
wall synthesis inhibitors (â-lactams, glycopeptides such as
vancomycin and teicoplanin, and lipopeptides such as dapto-
mycin) and therefore does not show cross-resistance with them.
Its unique mechanism of action has been the subject of a number
of recent studies.2 Because of its potent antimicrobial activity,
ramoplanin could be an effective antibiotic for treating serious
Gram-positive infections. However, while demonstrating excel-
lent antimicrobial activity in the mouse septicemia infection
model,3 ramoplanin had poor local tolerability upon intravenous
injection and also caused hematuria, which would be unaccept-
able in clinical practice.
A ramoplanin derivative with the antimicrobial activity of
the natural product but without its tolerability issues could be
a potential agent for the parenteral treatment of serious infections
caused by multidrug-resistant Gram-positive bacteria. Initial
derivatization utilized the simplest reactions feasible on such a
complex molecule: (1) hydrogenation of the two double bonds
of the fatty acid side chain to produce tetrahydroramoplanin,4
(2) removal of the dimannosyl residue linked to the amino acid
at position 11 to produce the ramoplanin aglycon,5 and (3)
guanylation of the amino groups of the two ornithines to produce
the diarginine analogue.6 While these analogues had antimi-
* To whom correspondence should be addressed. Phone: +39
023546509. Fax: +39 029 647 4358. E-mail: Romeo.Ciabatti@
hotmail.it.
Figure 1. Structure of ramoplanin main component A2.
a Abbreviations: VRE, vancomycin-resistant enterococci; FmocOSu,
N-(9-fluorenylmethoxycarbonyloxy)succinimide; TEA, triethylamine;
DMF, dimethylformamide; TFA, trifluoroacetic acid; DCM, dichlo-
romethane; PhNCS, phenyl isothiocyanate; Fmoc, 9-fluorenylmethoxycar-
bonyloxy; Cbz, carbobenzyloxy; PTSA, p-toluenesulfonic acid; TBAF,
tetrabutylammonium fluoride; PyBOP, benzotriazol-1-yloxy-2-tris(pyrro-
lidin-1-yl)phosphonium hexafluorophosphate; DMSO, dimethylsulfoxide;
MIC, minimal inhibitory concentration; PBS, phosphate buffer solution;
OD, optical density.
crobial activity comparable to that of ramoplanin, tolerability
was not substantially improved. We then turned to modification
of the lipidic part of the molecule, attaching various carboxylic
acids to the N-terminal amino acid. We report on the chemical
strategy we followed to selectively remove the fatty acid side
chain of ramoplanin, the synthesis of new derivatives, and the
biological activity of these compounds.
10.1021/jm070042z CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/02/2007

3078 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Scheme 1. Chemical Deacylation of Ramoplanina
Ciabatti et al.
a Reagents and conditions: (a) FmocNOSu, TEA, DMF; (b) ozone, -78 °C, 3:1 methanol/DMF; (c) benzylamine, NaCNBH3, DMF; (d) PhNCS, 1:1
H2O/pyridine, then 1:1 TFA/DCM. Ramoplanin structure is schematized, with only the functional groups involved in the reactions illustrated. The ring
represents the peptidic backbone (sugars not shown). The primary amines belong to the ornithines in positions 4 and 10.
Scheme 2. Synthesis of 2Z,4E-7-Methyl-octa-2,4-dienoic Acid (Fatty Acid 5a of Ramoplanin A2)a
a Reagents and conditions: (a) NaH, THF at 0 °C; then BuLi, THF at -78 °C, i-PrCH2CHO; (b) NaBH4, MeOH, -30 °C; (c) toluene, PTSA, reflux; (d)
TBAF, THF, room temp.
Chemistry
resynthesized starting from synthon 4 (Scheme 3). Its biological
activity and physicochemical characteristics (MIC, hemolytic
Because of the chemical complexity of ramoplanin and the
presence of many amidic bonds, selective hydrolytic removal
of the fatty acid side chain appeared challenging. To overcome
this problem, we took advantage of the presence of the double
bonds on the fatty acid side chain. In particular, the transforma-
tion of the R,â-double bond in an aldehyde group and a
subsequent conversion of it into an amine moiety would have
given us access to the well-known Edman degradation, thus
allowing a selective and smooth removal of the original fatty
acid side chain. Following the above strategy, deacylation of
ramoplanin was thus accomplished by chemical methods (see
Scheme 1).
Additionally, in order to selectively derivatize the N-terminal
amine, the amino groups of ornithines 4 and 10 had to be
protected. Among the protective groups compatible with the
degradation scheme, Fmoc was found to be the most suitable.
Although Cbz could be used for deacylation, it was not suitable
for the subsequent synthesis of unsaturated derivatives because
of its hydrogenolytic cleavage. Fmoc groups were introduced
by reaction of ramoplanin with Fmoc succinimidyl activated
reagent in the presence of TEA. DiFmoc-ramoplanin 1 was
treated with ozone in a mixture of 1:3 DMF/methanol at
-78 °C, followed by addition of triphenylphosphine. The
resulting aldehyde 2 was reductively aminated with benzylamine
in the presence of NaCNBH3 to give compound 3, which has a
new N-terminal amino acid (N-Bn glycine). This amino acid
was easily removed by Edman degradation upon treatment with
phenylisothiocyanate in 1:1 pyridine/water followed by acidi-
fication with anhydrous trifluoroacetic acid. The overall yield
of the final key synthon, diFmoc-deacyl-ramoplanin 4, was 40%.
In order to verify that the degradation scheme did not affect
the chirality of the natural scaffold, ramoplanin A2 (5) was
1
activity, HPLC retention time, H NMR, 13C NMR, IR, [R]D)
were identical to those of ramoplanin.
The fatty acid side chain of ramoplanin A2,5a, was synthe-
sized following the procedure reported in Scheme 2. The R,â-
unsaturated lactone 63 was the key intermediate in the synthesis
of 5a because it yielded the desired cis configuration at the R,â
double bond.7 It was prepared by the condensation of 60 with
isovaleraldehyde, followed by the reduction of 61 to produce
the racemic dialcohol 62, which was then cyclized by refluxing
with acid.
Several di-unsaturated carboxylic acids (40a-45a), analogues
of the natural fatty acid chain of ramoplanin, were prepared
and condensed with 4. The di-unsaturated carboxylic acids were
synthesized starting from the corresponding aldehyde. The
synthesis was the same as that used to regenerate ramoplanin
A2 (Scheme 2). In order to obtain the isomers with 2E,4E double
bond geometry, the lactone was opened by basic treatment with
1 M NaOH for 2 h at room temperature, obtaining the more
stable trans,trans isomers. The structures of the di-unsaturated
acids are in Table 2.
Starting from the key intermediate diFmoc-deacyl-ramoplanin
4, more than 100 semisynthetic analogues were produced.8 To
facilitate screening a large number of carboxylic acids, up to
12 reactions, each one carried out with 10-20 mg of diFmoc-
deacyl-ramoplanin 4, were run simultaneously; the progress of
the reactions was followed by HPLC. DiFmoc-deacyl-ramopla-
nin 4 was condensed with carboxylic acids 5a-39a in dry DMF,
using PyBOP as the condensing agent (see Scheme 3). When
the amidation step was complete (1-12 h), the Fmoc protection
was removed in situ by adding 5% piperidine and allowing the
reaction to proceed for 30 min at room temperature. The pH

Semisynthetic DeriVatiVes of Ramoplanin
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 3079
Scheme 3. Synthesis of Ramoplanin Analoguesa
somewhat less hemolytic than the 2-cis,4-trans isomer (compare
ramoplanin A2 vs 40, 41 vs 42, and 43 vs 44). All of the di-
unsaturated analogues had antibacterial activity comparable to
that of ramoplanin. Comparison of aliphatic derivative 10 with
its isomer 46, both containing eight carbon atoms, showed an
effect of the geometry of the fatty acid moiety; while the two
compounds had similar antimicrobial activity, the hemolytic
activity of 46 was somewhat lower than that of its linear isomer
10.
Testing of the benzylic derivatives in powder form confirmed
their relatively low hemolytic activity. The reduced antimicrobial
activity of compound 32 was improved by addition of a methyl
group, and the hemolytic activity was lowered. Among the three
possible isomers (48, 49, and 33), the one with the methyl group
in position 2 (48) was slightly less hemolytic than the other
two. The same effect of the position of methyl groups was
observed in comparing two dimethyl compounds, 53 and 54.
Compound 54, with both methyl groups in the ortho position,
was less hemolytic than 53, while having comparable antibacte-
rial activity. Additionally, comparison of the benzoic compound
47, with its benzylic analogue 48, again demonstrated that the
benzylic compound had both more potent antimicrobial activity
and lower hemolytic activity than the benzoic analogue. The
lower hemolytic activity of the benzylic derivatives might be
due to their having different amphipathic properties, as described
for other molecules.10 However, the assessment of the above
hypothesis would deserve further investigation to completely
elucidate the role of the lipid side chain. Recent studies
demonstrated in fact that the lipid side chain of ramoplanin is
not essential for the binding of ramoplanin with the biological
target; nevertheless, it plays a key role in the biological activity
probably as it helps target ramoplanin to bacterial membranes,
thus positioning the antibiotic near its target, lipid II, which is
located on the outer surface of the bacterial membrane.11
In the benzylic series, increasing the number of carbon atoms
(50 and 55) did not affect antibacterial activity but led to an
increase in hemolytic activity. A trifluoromethyl substitution
(52) did not greatly affect antibacterial activity but resulted in
greater hemolysis, while the introduction of a nitro moiety (51)
decreased the antibacterial activity somewhat.
Among the new analogues, 45, 46, and 48 were selected for
in vivo tolerability testing (Table 3). Ramoplanin, administered
to rats at a dose of 10 mg/kg and at a drug concentration of 1
mg/mL, invariably caused red or dark urine (hematuria) within
24 h, while tails (injection sites) became dark or discolored 1-2
days postdose (data not shown). Lower regimens at 5 or 10
mg/kg both at a drug concentration of 0.5 mg/mL produced
variable results.
Analogues were solubilized in 5% glucose and administered
to rats by intravenous injection into tail vein. The primary goal
of these experiments was to evaluate the tolerability profile of
these compounds by visually examining the urine emitted by
rats within the first several hours after treatment.
Compound 48 was the analogue with the highest flebotoler-
ability profile in these experimental conditions, with no hema-
turia at a dose of 20 mg/kg and at a drug concentration of 8
mg/mL (heading “20, 8” in Table 3). After one or two treatments
with compound 48, urine samples were similar in color to those
of rats given 5% glucose and no macroscopic lesions, such as
dark tail, at injection sites or signs of suffering were observed.
The other two analogues tested in this experiment, i.e.,
compounds 45 and 46, had a tolerability profile lower than that
of compound 48. In fact, compound 45 caused hematuria at 10
mg/kg and 8 mg/mL, but hematuria was still present at a lesser
a Reagents and conditions: (a) RCOOH (5a-55a) PyBOP in dry DMF;
(b) 5% piperidine in DMF 30 min at room temp.
was adjusted to 6-7 with 1 N HCl. The reaction mixtures were
analyzed by LC-MS to verify the identity of the derivatives.
In each run, a previously characterized derivative was resyn-
thesized as an internal reference. The solutions were neutralized
and evaluated for antibacterial and hemolytic activity without
purification. Reduced hemolysis in vitro was the parameter
chosen to predict in vivo tolerability. Ramoplanin at 100 mg/L
produces 90-100% hemolysis. The new analogues were tested
at 90 and 180 mg/L. The uncondensed carboxylic acids were
also tested as controls. Biological activity data for a selection
of the most interesting derivatives (6s-39s)9 are in Table 1.
Results and Discussion
Among the aliphatic derivatives (6s-13s), compounds with
substituents containing eight or nine carbon atoms (10s and 11s)
had better antimicrobial activity than compounds with shorter
(6s-9s) or longer (12s and 13s) chains. All of the aliphatic
derivatives were less hemolytic than ramoplanin, although some
hemolytic activity was seen at 180 mg/L with the two most
hydrophobic derivatives (12s and 13s).
Among the benzoic acid derivatives (14s -27s) benzoic acid
14s had little hemolytic activity, but its antibacterial activity
was also greatly reduced. When the substituent was an aliphatic
chain (15s-18s), antimicrobial activity increased with the length
from C1 to C4 (15s-17s) but was greatly decreased with a C8
substituent (18s). The hemolytic activity of 16s-18s was higher
than that of 14s and 15s. When the substituent was an ether
moiety, there was a moderate effect on antibacterial activity,
but there was a great reduction in hemolytic activity (22s and
23s vs 16s and 17s). A double substitution (26s) did not improve
the antibacterial activity. The introduction of a chlorine group
(27s) reduced the antibacterial activity and increased hemolysis
relative to compound 23s. Compounds with polar substituents
(20s and 21s) had greatly decreased antimicrobial activity. Most
of the naphthyl derivatives (28s-31s) had good antimicrobial
activity and were somewhat less hemolytic than ramoplanin.
In the benzylic class (32s-39s), the benzylic derivative itself
(32s) had slightly better activity than the benzoic analogue 14s
against staphylococci. The effects of substituents on the
antimicrobial activity of the benzylic compounds were similar
to that of the benzoic derivatives. The hemolytic activity of the
benzylic derivatives was lower than that of all the other series.
We therefore prepared a greater number of compounds of this
class (Table 2 compounds 32, 33, 47-55). New derivatives 40-
55 and selected members of other classes (10, 31) were prepared
as powders to better evaluate their antibacterial and hemolytic
activities
Among the di-unsaturated analogues (40-45), as for the
saturated derivatives, the hemolytic activity decreased somewhat
with the number of carbon atoms in the chain (5, 41, 43, 45).
The comparison of pairs of analogues differing only in double
bond geometry showed the trans,trans conformation to be

3080 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Ciabatti et al.
Table 1. Biological Activities of Derivatives Tested from Using Parallel Synthesis Solutions
extent at the doses of 10 mg/kg and 4 mg/mL or 20 mg/kg and
4 mg/mL. Compound 46 also developed hematuria at 10 mg/
kg and 8 mg/mL but not at 10 mg/kg and 4 mg/mL or at 20
mg/kg and 4 mg/mL, indicating a decreased tolerability profile
compared with that of compound 48. The lower tolerability of
compounds 45 and 46 was also accompanied by an equally
lower local tolerability, since injection sites of tails after
treatments were macroscopically in poor condition.
In conclusion, these studies indicated that chemical modi-
fications of the natural molecule of ramoplanin, in order to
obtain analogues with a better flebotolerability profile, were
possible. In particular, they indicated that the modifications
play a crucial role in allowing us to administer to rats a higher
injectable dose than the natural molecule of ramoplanin
permitted and, at the same time, maintaining the same in vitro
antimicrobial profile. Because of its improved tolerability profile,

Semisynthetic DeriVatiVes of Ramoplanin
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 3081
Table 2. Biological Activity of Compounds Synthesized as Powder
Table 3. In Vivo Tolerability of Selected Ramoplanin Derivativesa
are the same as for (a) except for the following: instrument, Varian
9010; column, Merck Lichrocart 125-4 Lichrosphere 100 RP-8 (5
µm); linear gradient from 30% B to 40% B over 30 min and then
from 40% B to 80% B over 5 min. Details for (c) are the same as
for (a) except for the following: linear gradient from 15% B to
37% B over 20 min, from 37% B to 43% B over 5 min, and then
from 43% B to 58% B over 5 min. Details for (d) are the same as
for (a) except for the following: instrument, HP 1090. Details for
(e) are the same as for (a) except for the following: instrument,
Varian 9010; column, Merck Lichrocart 125-4 Lichrosphere 100
RP-8 (5 µm); linear gradient from 35% B to 70% B over 35 min.
Details for (f) are the same as for (e) except for the following:
linear gradient from 20% B to 40% B over 30 min and then from
40% B to 80% B over 5 min.
treatment
[dosage (mg/kg), drug concentration (mg/mL)]
compd
vehicle
10, 8
10, 4
20, 4
20, 8
45
46
48
3/3 rats A 5/6 rats P
3/3 rats A 3/3 rats P
3/3 rats A 3/3 rats A not tested
2/3 rats A 2/3 rats A not tested
3/3 rats A 3/3 rats A not tested
not tested
3/3 rats A
a P ) presence of hematuria, pink or red urine. A ) absence of hematuria,
yellow or colorless urine.
compound 48 was chosen for further in vitro and in vivo
characterization.
For NMR analysis, 1H chemical shifts are in ppm. NMR measure-
ments were carried out on a Bruker DRX 500 spectrometer operat-
ing at 500.13 MHz. Samples containing 5 mM of ramoplanin
derivative in D2O-DMSO-d6 (4:1) were utilized. For identification
Conclusion
These studies demonstrated that the tolerability profile of
ramoplanin can be improved by chemical modification while
maintaining antibacterial activity. Replacement of the original
fatty acid chain with different chemical residues has shown that
the antimicrobial activity and hemolytic effect can be modulated,
depending on the nature and the structure of the residue. In
addition, we have synthesized a number of new derivatives in
which the two activities were separated. Among them 48, in
which the 2-methylphenylacetic acid replaced the natural chain,
was selected as the most interesting compound and was
submitted to further in vitro and in vivo characterization studies.
1
and assignment of the spin systems, two-dimensional H COSY
and TOCSY spectra were recorded at 313 K. Amino acid residues
are numbered according to the literature,12 and only the peak reso-
nances differing from natural ramoplanin are reported in this table.
For MS analysis, the mass spectra were recorded on an LCQ
Advantage ThermoFinnigan instrument.
Chemistry. 4,10-diFmoc-Protected Ramoplanin (1). A solution
of ramoplanin dihydrochloride (110.6 g, 40 mmol) in DMF (500
mL) was maintained at 0 °C with stirring under nitrogen atmo-
sphere. To this solution Fmoc-ONSu (6.8 g, 20 mmol) and TEA
(5.8 mL, 41.2 mmol) were added, maintaining the mixture at
0-5 °C. After 5 min additional Fmoc-ONSu (6.8 g, 20 mmol) and
TEA (5.8 mL, 41.2 mmol) were added. After another 5 min, Fmoc-
ONSu (13.6 g, 40 mmol) was added. The mixture temperature was
allowed to rise to room temperature. The reaction was monitored
by HPLC analysis (a) (retention time of 25.6 min). After HPLC
control an additional 10.8 g of Fmoc-ONSu was necessary to
complete the reaction. After 30 min, acetic acid (20 mL) was added
Experimental Section
General Part. HPLC analysis details are as follows: (a)
instrument, Shimadzu SCL-6B; column, Merck Lichrocart 125-4
Lichrosphere 100 RP-18 (5 µm); flow, 1 mL/min; detector UV λ
) 270 nm; injection volume, 10 µL; phase A, 0.05 M HCOONH4;
phase B, MeCN; linear gradient from 35% B to 40% B over 15
min and then from 40% B to 70% B over 20 min. Details for (b)

3082 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Ciabatti et al.
Table 4. HPLC Retention Times and MS Data for Compounds 5-55a
and the reaction mixture was poured into ethyl acetate (1L) filtered
and dried. An amount of 133 g of a solid product was obtained.
The solid was washed with stirring in methanol/water (1:9), and
the mixture was adjusted to pH 4.5-5 with acetic acid. The solid
was filtered and dried at 35 °C under reduced pressure, obtaining
126.8 g of white solid. Yield 100%.
4,10-diFmoc-Protected Ramo-NHCOCHO (2). Into a solution
of 4,10-diFmoc protected ramoplanin (1) obtained in the previous
step (126 g) in methanol/DMF (3:1, 3 L) and cooled to -78 °C,
ozone was bubbled (170 mmol, at a flow rate of 100 L/h of oxygen
containing 5% of ozone) with stirring. The mixture was maintained
at -78 °C for 30 min. The reaction was monitored by HPLC
analysis (a) (retention time of 7.5 min). The excess of ozone was
eliminated by bubbling nitrogen into the solution. Triphenylphos-
phine was added (25 g), and the mixture was allowed to reach room
temperature. Methanol was evaporated under reduced pressure, and
the residual DMF solution was poured into ethyl acetate (8 L) with
stirring. The precipitate was filtered, washed with ethyl acetate (3
× 150 mL), and dried at room temperature, obtaining 132 g of a
solid that was used for the following step.
4,10-diFmoc Protected Ramoplanin-NHCOCH2NHCH2Ph
(3). To a solution of 4,10-diFmoc-protected ramoplanin-CHO (2)
(130 g, 45 mmol) and benzylamine hydrobromide (43 g, 224 mmol)
in anhydrous DMF (1 L), NaCNBH3 (4.23 g, 67 mmol) was added
with stirring at room temperature. The mixture was stirred for 2 h.
The reaction was monitored by HPLC analysis (a) (retention time
19.6 min). The solution was poured into water (10 L). The
precipitate was filtered and dried at 35 °C under reduced pressure,
obtaining 130 g of crude product. The crude product (107 g) was
dissolved at 35-40 °C in 1.5 L of a 1:1 acetonitrile/water mixture
at pH 2.5 (1 N HCl). To this solution, with stirring, silanized silica
gel was added (300 g). After 30 min, acetonitrile was evaporated
under reduced pressure and the water suspension was charged at
the top of a silanized silica gel column (diameter 7.5 cm, height
100 cm) previously equilibrated with water. Elution was with a
water/acetonitrile gradient from 85:15 to 1:1. Fractions containing
the product were collected, and the acetonitrile was evaporated
under reduced pressure. The precipitate was filtered, washed with
water (100 mL), and dried at 35 °C under reduced pressure,
obtaining 54.6 g of white solid. Yield was 42%, starting from
ramoplanin.
4,10-diFmoc Protected Ramoplanin-NH2 (4). To a solution of
4,10-diFmoc protected ramoplaninNHCOCH2NHCH2Ph (3) (17 g,
5.65 mmol) in 1:1 pyridine/water (340 mL), phenyl isothiocyanate
(0.74 mL, 6.35 mmol) was added while stirring at room temperature.
The reaction was monitored by HPLC analysis (a) (retention time
of 24.7 min). After 1 h the solvent was evaporated and the residue
was suspended in toluene (50 mL) and evaporated. This operation
was repeated twice. The solid was then suspended in dichlo-
romethane (100 mL) and added with TFA (100 mL). After 15 min
at 40 °C and HPLC control (a) (retention time of 9.5 min), the
mixture was evaporated under reduced pressure and the oil obtained
was triturated with diethyl ether (100 mL) and dried at 35-40 °C
under reduced pressure, obtaining 17 g of solid. The solid was
suspended in water, and the suspension was stirred at room
temperature for 2 h and filtered. The solid was dried at 35-40 °C
under reduced pressure, obtaining 15 g of white solid. Yield 91%.
Preparation of Carboxylic Acids (5a-55a) To Be Condensed
on 4,10-diFmoc-Protected Ramoplanin-NH2 (4). Many of the
carboxylic acids to be condensed with 4,10-diFmoc protected
ramoplanin-NH2 (4) are commercially available. Those carboxylic
acids not commercially available were prepared according to the
following procedure.
tR (min) of diFmoc
tR (min) of
lower isotope
compd
derivatives
final compd
molecular weight
5
6
7
8
9
34.2 (b), 25.6 (d)
15.5 (d)
18.6 (d)
21.3 (d)
33.7 (b)
24.7 (d)
26.3 (d)
27.5 (d)
34.9 (d)
17.1 (d)
20.9 (d)
21.7 (d)
25.3 (d)
31.6 (d)
21.8 (d)
10.2 (d)
17.3 (d)
32.4 (b)
25.2 (d)
26.8 (d)
29.6 (d)
27.6 (d)
26.8 (d)
22.1 (d)
20.9 (d)
22.6 (d)
33.7 (b)
30.2 (b)
21.6 (d)
22.0 (d)
11.8 (d)
9.2 (e)
10.9 (b)
2.8 (d)
4 (d)
2552
2483
2500
2514
2528
2542
2556
2570
2625
2520
2534
2548
2576
2632
2554
2536
2562
2564
2592
2606
2634
2664
2660
2570
2570
2584
2584
2534
2548
2568
2550
2564
2548
2550
2562
2552
2538
2538
2538
2524
2524
2542
2534
2548
2548
2562
2579
2602
2562
2562
2562
4.51 (d)
8.9 (b)
5.5 (d)
8.4 (d)
12.7 (d)
19.1 (d)
8.4 (d)
4.2 (d)
5.1 (d)
6.5 (d)
14.2 (d)
6.3 (d)
2.1 (d)
3.85 (d)
5.6 (b)
5.6 (d)
5.8 (d)
6.2 (d)
5.9 (d)
5.3 (d)
4.9 (d)
20.9 (d)
5 (d)
7.7 (b)
4.2 (b)
4.5 (d)
4.9 (d)
2.3 (d)
14.4 (f)
4.5 (d)
4.15 (d)
5 (d)
13.7 (b)
8.2 (b)
9.5 (b)
8 (b)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
21.4 (d)
20.7 (d)
22.5 (d)
34.2 (b)
33.8 (b)
33.7 (b)
33.6 (b)
33.9 (b)
33.2 (b)
25.0(a)
9.6 (b)
5.5 (b)
21.73 (c)
16.27 (c)
18.83 (c)
17.9 (f)
19.0 (f)
13.9 (f)
19.7 (f)
21.1 (f)
19.0 (f)
17.3-18.9 (f)
18.9(a)
21.7(a)
10.4 (e)
10.9 (e)
9.3 (e)
11.1 (e)
12.0 (e)
10.9 (e)
11.3-12.0 (e)
a Letters in parentheses refer to the HPLC conditions described in the
Experimental Section.
cyclohexane) was added dropwise to the reaction mixture, and
stirring was continued for a further 10 min. Isovaleraldehyde (16.5
mL, 154 mmol) was then added in one portion. After a further 10
min the mixture was poured into a HCl solution (50 mL of 37%
HCl in 400 mL of water). Ether was added, and the aqueous layer
was removed and extracted again with 2 × 40 mL of ether. The
ether extracts were combined, washed with a saturated brine
solution, dried over sodium sulfate, and filtered, and the solvent
was removed under reduced pressure. The oily residue was purified
by flash chromatography using 8:2 hexane/ethyl acetate as eluent,
1
obtaining 16.6 g of product (50%). H NMR (CDCl3, 500 MHz):
5-Hydroxy-7-methyl-3-oxooctanoic Acid Ethyl Ester (61).
Sodium hydride, as a 60% mineral oil dispersion (6.8 g, 169 mmol),
was weighed into a dry round bottomed flask, and dry tetrahydro-
furan (220 mL) was added. The flask was cooled in ice and flushed
with nitrogen. Ethyl acetoacetate (20 g, 154 mmol) was added
dropwise to the cooled and stirred slurry, and the mixture was stirred
for 10 min after the addition was complete. The solution was cooled
at -78 °C. A solution of n-butyllithium (85 mL of 2 M solution in
δ 0.95 (d, 6H), 1.19 (m, 1H), 1.28 (t, 3H), 1.51 (m, 1H), 1.80 (m,
1H), 2.65 (dd, 1H), 2.68 (s broad, 1H), 2.73 (dd, 1H), 3.47 (s, 2H),
4.21 (q, 2H).
3,5-Dihydroxy-7-methyloctanoic Acid Ethyl Ester (62). So-
dium borohydride (1.58 g, 41.6 mmol) was added to a stirred
solution of 61 (9 g, 41.6 mmol) in MeOH (100 mL) at -30 °C.
Stirring was continued at the same temperature for 2 h. Then a
saturated ammonium chloride solution was added and methanol

Semisynthetic DeriVatiVes of Ramoplanin
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 3083
Table 5. 1H Chemical Shifts (in ppm) of Selected Compoundsa
residue
22
31
32
45
48
54
side chain
1.37 (3H)
4.13 (2H)
7.57 (2H)
6.98 (2H)
3.99 (2Η)
7.95 (1H)
7.91 (1H)
7.87 (1H)
7.57 (1Η)
7.56 (1H)
7.47 (1H)
7.39 (1H)
4.55
1.83-2.24
3.95
1.16-1.22
1.33
3.55 (2Η)
7.3 (2H)
7.35 (3H)
0.98 (3H)
5.56 (1H)
6.49 (1H)
7.07 (1Η)
6.17 (1Η)
2.14 (2Η)
3.52 (2Η)
2.17 (3Η)
7.1 (1Η)
7.7 (1Η)
7.18 (1H)
7.21 (1H)
3.55 (2Η)
2.19 (6Η)
7.03 (2Η)
7.13 (2Η)
Asn (1)
CHR
CH2â
CHR
CH2â
CHγ
Me
4.71
1.88-1.22
4.18
1.4-1.44
1.4
0.68
4.34
1.46
4.7
1.79-2.21
4.06
1.42-1.46
1.44
0.62-0.74
4.25
4.7
1.66-2.18
4.285
1.503
1.503
0.765-0.8
4.36
4.55
1.68-2.18
4.07
1.39
1.45
0.67-0.76
4.195
4.75
1.6-2.2
4.13
1.46
1.46
0.72-0.79
4.18
Leu (15)
0.43-0.56
4.097
0.93
Ala (16)
CHR
Me
1.244
1.428
1.178
1.14
a NMR conditions are detailed in the Experimental Section.
was evaporated under reduced pressure. The mixture was extracted
with ethyl acetate, and the organic layer was dried over sodium
sulfate and concentrated to give the alcohol 62, which was used
without any further purification for the following step.
Following the same procedure but with different starting alde-
hydes, the following carboxylic acids were synthesized.
2E,4E-6-Methyl-2,4-heptadienoic acid (42a). Starting aldehyde
1
was isobutyraldehyde, giving 42a. H NMR (CDCl3, 500 MHz):
δ 1.07 (d, 6H), 2.43 (m, 1H), 5.8 (d, 1H, J ) 15.2 Hz), 6.15 (m,
6-Isobutyl-5,6-dihydro-2-pyrone (63). Toluene (100 mL) and
PTSA (800 mg) were added to (62), and the mixture was refluxed
for 2 h. The reaction was quenched with water, and the mixture
was extracted with ethyl acetate. The organic layer was dried over
sodium sulfate and evaporated to give 6 g of lactone (63), which
1H), 6.24 (m, 1H), 7.26 (m, 1H).
2E,4E-2,4-Octadienoic Acid (44a). Starting aldehyde was
1
butyraldehyde, giving 44a. H NMR (CDCl3, 500 MHz): δ 0.93
(t, 3H), 1.45 (m, 2H), 2.17 (m, 2H), 5.77 (d, 1H, J ) 15.3 Hz),
6.15-6.3 (m, 2H), 7.25 (m, 1H).
1
was used as such for the following step. H NMR (CDCl3, 500
4-Butoxybenzoic Acid (23a). Step 1. A mixture of 4-hydroxy-
benzaldehyde (8.2 mmol), butyl bromide (8.2 mmol), K2CO3 (8.2
mmol), and KI (8.2 mmol) in acetone (15 mL) was stirred at reflux
for 6 h. Acetone was evaporated, and the semisolid residue was
dissolved in water and extracted with ethyl acetate. The organic
layer was washed with 0.1 N NaOH and then with saturated brine,
dried over sodium sulfate, and evaporated under reduced pressure
to give a crude product, which was used as such for the following
step. Yield 100%.
Step 2. AgNO3 solution (4.6 M in water, 0.56 mL) was added
to the solution of the compound obtained according to step 1 (1.12
mmol) in ethanol (6.7 mL). KOH (5.6 mL of a 1 M solution in
water) was added, and the reaction mixture was stirred at room
temperature for 2 h. The solid was filtered, and the aqueous solution
was acidified with concentrated HCl and extracted with diethyl
ether. The organic phase was washed with water, dried over sodium
sulfate, and evaporated under reduced pressure to give the desired
4-butoxybenzoic acid 23a, which was used without further purifica-
tion.
MHz): δ 0.97 (d, 6H), 1.41 (m, 1H), 1.80 (m, 1H), 1.92 (m, 1H),
2.31 (m, 2H), 4.52 (m, 1H), 6.04 (m, 1H), 6.90 (m, 1H).
2Z,4E-7-Methylocta-2,4-dienoic Acid (5a). A mixture of the
lactone 63 (6 g) and n-Bu4NF‚3H2O (10 g) in THF (100 mL) was
stirred at room temperature for 3 h under nitrogen. Water was added,
and the mixture was extracted with ethyl acetate. The organic layer
was washed with saturated brine solution, dried over sodium sulfate,
and evaporated under reduced pressure to give 3.8 g of dienoic
acid 5a. Yield 60% from 61. 1H NMR (CDCl3, 500 MHz): δ 0.93
(d, 6H), 1.73 (m, 1H), 2.11 (m, 1H, J ) 7.4 Hz), 5.6 (d, 1H, J )
11.4), 6.12 (dt, 1H, J ) 15.2 and 7.4), 6.66 (dd, 1H, J1 ) J2 )
11.3), 7.33 (m, 1H, J ) 15.2 Hz). 13C NMR (CDCl3): δ 170.6,
147.0, 145.3, 127.4, 113.9, 41.7, 29.1, 22.2.
Following the same procedures but with different starting
aldehydes, the following carboxylic acids were synthesized.
2Z,4E-6-Methyl-2,4-heptandienoic Acid (41a). Starting alde-
hyde was isobutyraldehyde, giving 41a. 1H NMR (CDCl3, 500
MHz): δ 1.05 (d, 6H), 2.48 (m, 1H), 5.6 (d, 1H, J ) 11.31 Hz),
6.08 (m, 1H), 6.65 (dd, 1H), 7.3 (m, 1H).
Following the same procedure but using the appropriate benzoic
acid and alkyl bromide, the following carboxylic acids were
synthesized.
2Z,4E-2,4-Octadienoic Acid (43a). Starting aldehyde was
1
butyraldehyde, giving 43a. H NMR (CDCl3, 500 MHz): δ 0.98
4-Pentyloxybenzoic Acid (24a). 1H NMR (DMSO-d6, 500
MHz): δ 0.93 (t, 3H), 1.41 (m, 4H), 1.74 (m, 2H), 4.05 (t, 2H),
7.00 (d, 2H), 7.88 (d, 2H).
(t, 3H), 1.45 (m, 2H), 2.17 (m, 2H), 5.58 (d, 1H, J ) 11.33 Hz),
6.12 (m, 1H, J ) 15.2 Hz), 6.65 (dd, 1H), 7.34 (m, 1H).
2Z,4E-2,4-Heptadienoic Acid (45a). Starting aldehyde was
propionaldehyde, giving 45a. 1H NMR (CDCl3, 500 MHz): δ 1.07
(t, 3H), 2.25 (m, 2H), 5.59 (d, 1H, J ) 11.35 Hz), 6.17 (m, 1H),
6.66 (dd, 1H), 7.35 (m, 1H).
4-Heptyloxybenzoic Acid (25a). 1H NMR (DMSO-d6, 500
MHz): δ 0.88 (t, 3H), 1.3 (m, 6H), 1.74 (m, 2H), 4.05 (t, 2H),
7.00 (d, 2H), 7.88 (d, 2H).
3,4-Dibutoxy-benzoic acid (26a). 1H NMR (DMSO-d6, 500
MHz): δ 0.94 (t, 6H), 1.44 (m, 4H), 1.71 (m, 4H), 3.98 (t, 2H),
4.03 (t, 2H), 7.05 (d, 1H), 7.45 (s, 1H), 7.52 (d, 1H).
Compounds 5-55. To a solution of the 4,10-diFmoc protected
ramoplanin-NH2 4 (0.35 mmol), triethylamine (1.05 mmol), and
the suitable carboxylic acid (5a-55a) (0.525 mmol) in DMF (12.5
mL) was added PyBOP (0.52 mmol) with stirring at room
temperature. The reaction was monitored by HPLC analysis (see
Table 4). The mixture was allowed to react at room temperature,
and after 5 h, piperidine (0.6 mL) or alternatively 2,2,6,6-
tetramethylpiperidine (1.8 mL) was added to remove the protecting
group from the ornithine moieties. The reaction was continued at
room temperature and monitored by HPLC (see Table 4). After 30
2E,4E-7-Methyl-2,4-octadienoic Acid (5a). The trans,trans
isomer of 40a was synthesized following the same synthetic strategy
of compound 5a but changing the final step. A mixture of the
lactone 63 (2.143 mmol) and 30% NaOH (11 mL) was stirred at
reflux for 1 h. The mixture was acidified with 5 N HCl to pH 3
and then extracted with ethyl acetate. The organic layer was washed
with saturated brine, dried over sodium sulfate, and evaporated
under reduced pressure to give a crude product, which was purified
by flash chromatography (9:1 DCM/MeOH). An amount of 0.37 g
of the desired compound were obtained. Yield 60%. 1H NMR
(CDCl3, 500 MHz): δ 0.9 (d, 6H, Me2-CH), 1.76 (m, 1H, CHMe2),
2.1 (t, 2H, CH2CHdCH), 5.82 (d, 1H, J ) 15.36 Hz, CHCOOH),
6.12-6.3 (m, 2H), 7.29 (m, 1H).

3084 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Ciabatti et al.
min, diluted HCl was added (6.5 mL of a 1 M solution). The
resulting solutions can be tested for antimicrobial and hemolytic
activities. Alternatively, the product can be purified by preparative
HPLC and lyophilized. The derivatives were characterized by MS
spectrometry (see Table 4) and some of them by NMR spectroscopy
(see Table 5).
MICs were determined by broth microdilution method according
to the NCCLS procedure.13 Microorganisms were grown in cation-
adjusted Muller-Hinton broth, in Todd-Hewitt broth (streptococci
only), or in RPMI medium (Candida albicans only). Inocula were
5 × 105 CFU/mL. MICs were read after 24 h of incubation at
35 °C.
Hemolytic Activity. Hemolysis of erythrocytes is considered to
be an indicator of the local tolerability of ramoplanin analogues.
Hemolysis testing was performed according to the method suggested
in the literature. Initial experiments were carried out directly on
the solutions derived from the reaction of the amidation of 4,10-
diFmoc protected ramoplanin-NH2 4 as previously described. The
reaction solutions (resulting from the addition of 1 M hydrochloric
acid) were diluted at 180 mg/L by adding 0.1% peptone and 0.8%
NaCl (PBS). Additional experiments were carried out on the
powdered compound: the ramoplanin analogues were dissolved at
40.000 µg/mL in DMSO and then diluted 1:5 in 0.1% peptone and
0.8% NaCl (PBS). Whole-blood sample was obtained from the
abdominal aorta of rats and diluted 1:100 in PBS before the test.
PBS and 3% saponin in distilled water were the 0% and 100%
hemolysis controls, respectively. The reaction solutions or solubi-
lized powders to be tested were diluted 1:5 in triplicate into the
diluted blood and incubated in a water bath at 37 °C for 45 min.
The blood was then centrifuged at 2500-3000g for 10 min, and
0.1 mL of each supernatant was diluted in 0.9 mL of Drabkin’s
reagent. The OD of the samples was measured at 540 nm versus a
blank consisting of Drabkin’s reagent plus 0.1 mL of PBS. The
percent hemolysis was calculated as ∆x/∆t × 100, where ∆x )
mean OD540 (sample minus blank) and ∆t ) mean OD540 (minus
blank) of the positive control.
In Vivo Tolerability. Compounds were solubilized in 5%
glucose. Three to six rats for dosing of test compound were treated
by intravenous injection at 24 h intervals for 1-2 days. Urine
samples were examined for the presence of blood (hematuria), and
macroscopic observation of the injection sites and general behavior
were also recorded. Rats were killed 24 h after the last treatment.
Positive controls consisting of three rats given 5% glucose were
used. Our previous studies performed in our labs have demonstrated
that ramoplanin, administered to rats at a dose of 10 mg/kg and at
a drug concentration of 1 mg/mL, invariably caused red or dark
urine (hematuria) within 24 h while tails (injection sites) became
dark or discolored 1-2 days postdose (data not shown). Lower
regimens at 5 or 10 mg/kg, both at a drug concentration of 0.5
mg/mL, produced variable results. These in vivo ramoplanin studies
also demonstrated that the drug concentration (mg/mL) more than
the dosage (mg/kg) itself played a crucial role in causing hematuria.
Therefore, in order to find a derivative with enhanced flebotoler-
ability properties, it has been decided to start these in vivo studies
with an initial dose of 10 mg/kg at a drug concentration of 8 mg/
mL (10 mg/kg and 8 mg/mL). As second step, and in the case of
a positive result, other rats were immediately dosed with the same
test compound at 20 mg/kg at the same drug concentration (20 mg/
kg and 8 mg/mL), which was the highest dose-concentration tested.
In the case of a negative result at 20 mg/kg and 8 mg/mL, the
same dose but at a lower drug concentration (20 mg/kg and 4 mg/
mL) was assessed. Conversely, in the case of a negative result with
an initial dose of 10 mg/kg and 8 mg/mL, the same dose
administered at a lower concentration (10 mg/kg at a drug
concentration of 4 mg/mL; 10 mg/kg and 4 mg/mL) was tested. In
this case, whether a positive result was found, the corresponding
higher dose as mg/kg at the same drug concentration was tested.
With this experimental design, it was possible to verify different
conditions and, at the same time, to determine the influence of the
drug concentration versus the dose on the tolerability.
Acknowledgment. This work was partially supported by the
Italian Government throught grants to the Progetto ex art.11
L451/94 Rif.3932 “Ricerca, Caratterizzazione e Sviluppo di
Nuovi Antibiotici”. Part of this work was presented at 43rd
Annual ICAAC, Chicago, September 14-17, 2003.
References
(1) (a) Rolston, K. V. I.; Dholakia, N.; Ho, D. H.; LeBlanc, B.; Dvorak,
T.; Streeter, H. In-vitro activity of ramoplanin (a novel lipoglyco-
peptide), vancomycin and teicoplanin against Gram-positive clinical
isolates from cancer patients. J. Antimicrob. Chemother. 1996, 38,
265-269. (b) O’Hare, M. D.; Felmingham, D.; Gruneberg, R. N.
The in vitro activity of ramoplanin (A-16686/MDL 62,128), vanco-
mycin and teicoplanin against methicillin-susceptible and methicillin-
resistant staphylococcus spp. Drugs Exp. Clin. Res. 1988, 14, 617-
619. (c) Lawrence, T.; Rotstein, C.; Beam, T. R., Jr.; Gorzynski, E.
A.; Amsterdam, A. In vitro activities of ramoplanin, selected
glycopeptides, fluoroquinolones and other antibiotics against blood-
stream isolates of Gram-positive cocci. Antimicrob. Agents Chemoth-
er. 1993, 37, 896-900. (d) Johnson, C. C.; Taylor, S.; Pitsakis, P.;
May, P.; Levison, M. E. Bactericidal activity of ramoplanin against
antibiotic-resistant enterococci. Antimicrob. Agents Chemother. 1992,
36, 2342-2345. (e) Walker, S.; Chen, L.; Hu, Y.; Rew, Y.; Shin,
D.; Boger, D. L. Chemistry and biology of ramoplanin: a lipogly-
codepsipetide with potent antibiotic activity. Chem. ReV. 2005, 105,
449-475.
(2) (a) Lo, M.; Men, H.; Branstrom, A.; Helm, J.; Yao, N.; Goldman,
R.; Walker, S. A new mechanism of action proposed for ramoplanin.
J. Am. Chem. Soc. 2000, 122, 3540-3541. (b) Lo, M.; Helm, J.;
Sarngadharan, G.; Pelczer, I.; Walker, S. A new structure for the
substrate-binding antibiotic ramoplanin. J. Am. Chem. Soc. 2001, 123,
8640-8641. (c) Helm, J.; Chen, L.; Walker, S. Rethinking ramopla-
nin: the role of substrate binding in inhibition of peptidoglycan
biosynthesis J. Am. Chem. Soc. 2002, 124, 13970-13971. (d) Cudic,
P.; Kranz, J. K.; Behenna, D. C.; Kruger, R. G.; Tadesse, H.; Wand,
A. J.; Veklich, Y. I.; Weisel, J. W.; McCafferty, D. G. Complexation
of peptidoglycan intermediates by the lipoglycodepsipeptide antibiotic
ramoplanin: minimal structural requirements for intermolecular
complexation and fibril formation. Proc. Natl. Acad. Sci. U.S.A. 2002,
99, 7384-7389. (e) Chen, L.; Yuan, Y.; Helm, J. S.; Hu, Y.; Rew,
Y.; Shin, D.; Boger, D. L.; Walker, S. Dissecting ramoplanin:
mechanistic analysis of synthetic ramoplanin analogues as a guide
to the design of improved antibiotics. J. Am. Chem. Soc. 2004, 126,
7462-7463 (f) McCafferty, D. G.; Cudic, P.; Frankel, B. A.;
Barkallah, S.; Kruger, R. G.; Li, W. Chemistry and biology of the
ramoplanin family of peptide antibiotics. Biopolymers 2002, 66, 261-
284. (g) Jiang, W.; Wanner, J.; Lee, R. J.; Bounaud, P. Y.; Boger,
D. L. Total synthesis of the ramoplanin A2 and ramoplanose aglycon.
J. Am. Chem. Soc. 2002, 124, 5288-5290. (h) Jiang, W.; Wanner,
J.; Lee. R. J.; Bounaud, P. Y.; Boger, D. L. Total synthesis of the
ramoplanin A2 and ramoplanose aglycon. J. Am. Chem. Soc. 2003,
125, 1877-1887. (i) Hu, Y.; Helm, J.; Chen, L.; Ye, X.; Walker, S.
Ramoplanin inhibits bacterial transglycosilases by binding as a dimer
to lipid II. J. Am. Chem. Soc. 2003, 125, 8736-8737. (j) Rew, Y.;
Shin, D.; Hwang, I.; Boger, D. L. Total synthesis and examination
of three key analogues of ramoplanin: a lipoglycodepsipeptide with
potent antibiotic activity. J. Am. Chem. Soc. 2004, 126, 1041-1043.
(3) Pallanza, R.; Berti, M.; Scotti, R.; Randini, E.; Arioli, V. A-16686,
a new antibiotic from Actinoplanes II. Biological properties. J.
Antibiot. 1984, 37, 318-324.
(4) Ciabatti, R.; Cavalleri, B. Hydrogenated Derivatives of Antibiotic
A/16686. U.S. Patent 5108988, 1988.
(5) (a) Ciabatti, R.; Cavalleri, B. Aglycons of A/16686 Antibiotics.
European Patent 337203, 1989; Chem. Abstr. 1990, 112, 179893.
(b) Jiang, W.; Wanner, J.; Lee, R. J.; Bounaud, P.-Y.; Boger, D. L.
Total synthesis of the ramoplanin A2 and ramoplanose aglycon. J.
Am. Chem. Soc. 2003, 125, 1877. (c) Rew, Y.; Shin, D.; Hwang, I.;
Boger, D. L. Total synthesis and examination of three key analogs
of ramoplanin: a lipoglycodepsipeptide with potent antibiotic activity.
J. Am. Chem. Soc. 2004, 126, 1041. (d) Wanner, J.; Tang, D.;
McComas, C. C.; Crowley, B. M.; Jiang, W.; Moss, J.; Boger, D. L.
A new and improved method for deglycosidation of glycopeptide
antibiotics exemplified with vancomycin, ristocetin and ramoplanin.
Bioorg. Med. Chem. Lett. 2003, 13, 1169.
(6) Guanylated derivatives were synthesized in our laboratories, but the
results have not been published. (a) For the present study, the
derivatives were resynthesized as described in the following: Cudic,
P.; Behenna, D. C.; Kranz, J. K.; Kruger, R. G.; Wand, A. J.; Veklich,
Y. I.; Weisel, J. W.; McCafferty, D. G. Functional analysis of the
lipoglycodepsipeptide antibiotic ramoplanin. Chem. Biol. 2002, 9,
1-20. For other ornithines derivatives, see the following: (b) Helm,
J. S.; Chen, L.; Walker, S. Rethinking ramoplanin: the role of

Semisynthetic DeriVatiVes of Ramoplanin
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 3085
substrate binding in inhibition of peptidoglycan biosynthesis. J. Am.
Chem. Soc. 2002, 124, 13970. (c) Jiang, W.; Wanner, J.; Lee, R. J.;
Bounaud, P.-Y.; Boger, D. L. Total synthesis and examination of
three key analogs of ramoplanin: a lipoglycodepsipeptide with potent
antibiotic activity. J. Am. Chem. Soc. 2003, 125, 1877.
(10) (a) Kondejewski, L. H.; Jelokhani-Niaraki, M.; Farmer, S. W.; Lix,
B.; Kay, C. M.; Sykes, B. D.; Hancock, R. E. W.; Hodges, R. S.
Dissociation of antimicrobial and hemolytic activities in cyclic
peptides diastereomers by systematic alterations in amphipathicity.
J. Biol. Chem. 1999, 274, 13181-13192. (b) Yamada, K.; Shinoda,
S.-s.; Oku, H.; Komagoe, K.; Katsu, T.; Katakai, R. Synthesis of
low-hemolytic antimicrobial dehydropeptides based on Gramicidin
S. J. Med. Chem. 2006, 49, 7592-7595.
(11) (a) Walker, S.; Chen, L.; Hu, Y.; Rew, Y.; Shin, D.; Boger, D. L.
Chemistry and biology of ramoplanin: a lipoglycodepsipetide with
potent antibiotic activity. Chem. ReV. 2005, 105, 449-475. (b) Shin,
D.; Rew, Y.; Boger, D. L. Natural product synthesis special feature:
total synthesis and structure of the ramoplanin A1 and A3 aglycons:
two minor components of the ramoplanin complex. Proc. Natl. Acad.
Sci. U.S.A. 2004, 101, 11977-11979.
(12) (a) Ciabatti, R.; Kettenring, J. K.; Winters, G.; Tuan, G.; Zerilli, L.;
Cavalleri, B. Ramoplanin (A-16686), a new glycolipodepsipeptide
antibiotic. III. Structure elucidation. J. Antibiot. 1989, 42, 254-267.
(b) Kettenring, J. K.; Ciabatti, R.; Winters, G.; Tamborini, G.;
Cavalleri, B. Ramoplanin (A-16686), a new glycolipodepsipeptide
antibiotic. IV. Complete sequence determination by homonuclear 2D
NMR spectroscopy. J. Antibiot. 1989, 42, 268-275. (c) Kurz, M.;
Guba, W. 3D structure of ramoplanin: a potent inhibitor of bacterial
cell wall synthesis. Biochemistry 1996, 35, 12570-12575.
(7) (a) Huckin, S. N.; Weiler, L. Aldol reactions of the dianion of
â-ketoester. Tetrahedron Lett. 1971, 50, 4835-4838. (b) Masamune,
S.; Imperiali, B.; Garvey, D. S. Synthesis of ansamycins: the ansa
chain of rifamycin S. J. Am. Chem. Soc. 1982, 104, 5528-5531. (c)
Roush, W. R.; Spada, A. Enantio- and stereoselective synthesis of
the dihydroxyactadienoic acid fragments of the roridins and tricho-
verrins. Tetrahedron Lett. 1982, 23, 3773-3776. (d) Yoshida, T.;
Saito, S. A new method for synthesis of R,â-unsaturated δ-lactones
via Michael addition using methyl (phenylsulfinyl)acetate. Chem. Lett.
1982, 1587-1590. (e) Nakata, T.; Hata, N.; Oishi, T. A stereose-
lective synthesis of (2Z,4E)-dienoic acid involving masked functional
groups: nBu4NF induced ring-opening of R,â-unsaturated δ-lactone.
Heterocycles 1990, 30, 333-334. (f) Corey, E. J.; Schmidt, G. Studies
on the total synthesis of rifamycin. A method for the stereospecific
synthesis of the carbonyl-conjugated 1,3 dienic unit [C(15) to C(20)].
Tetrahedron Lett. 1979, 25, 2317-2320.
(8) (a) Ciabatti, R.; Maffioli, S. I.; Checchia, A.; Romano`, G.; Candiani,
G.; Panzone, G. A Process for the Production of Ramoplanin-like
Amide Derivatives. WO 2003076460, 2003; Chem.Abstr. 2003, 139,
261566. (b) Raju, B.; Ciabatti, R.; Maffioli, S. I.; Singh, U.; Romano`,
G.; Michelucci, E.; Tiseni, P. S.; Candiani, G.; Kim, B.; O’Dowd,
H. Ramoplanin Derivatives Possessing Antibacterial Activity. WO
2007001335, 2007.
(13) Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria
That Grow Aerobically; Approved Guideline; The National Com-
mittee for Clinical Laboratory Standards (NCCLS), January 2003.
(9) The superscript(s) is used when the derivatives were tested for
biological activity without isolation from the reaction solution.
JM070042Z