Acylation of novobiocin by carboxylic acid anhydrides: Preparation and characterization of semi-synthetic novenamines

Article abstract of DOI:10.1016/j.tet.2012.01.042

The acylation of novobiocin by carboxylic acid anhydrides leading to preparation of three families of semi-synthetic acylated novenamine analogues is reported. ESI-MS was used to monitor the reaction progress and enabled the isolation of intermediate compounds that provided insights into the sequence of acylation steps during the reaction. The chemoselectivity of the reaction depends on the carboxylic acid anhydride. The potential of the acylated novenamine analogues as leads for the development of antibacterial agents is discussed.

Full text of DOI:10.1016/j.tet.2012.01.042

Tetrahedron 68 (2012) 2306e2312
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Acylation of novobiocin by carboxylic acid anhydrides: preparation
and characterization of semi-synthetic novenamines
*
Yifat Berkov-Zrihen, Roi Rutenberg, Micha Fridman
School of Chemistry, Raymond and Beverly Sackler Faculty of Exact Sciences, Tel Aviv University, Ramat Aviv, Tel Aviv 69978, Israel
a r t i c l e i n f o
a b s t r a c t
Article history:
The acylation of novobiocin by carboxylic acid anhydrides leading to preparation of three families of
semi-synthetic acylated novenamine analogues is reported. ESI-MS was used to monitor the reaction
progress and enabled the isolation of intermediate compounds that provided insights into the sequence
of acylation steps during the reaction. The chemoselectivity of the reaction depends on the carboxylic
acid anhydride. The potential of the acylated novenamine analogues as leads for the development of
antibacterial agents is discussed.
Received 26 October 2011
Received in revised form 25 December 2011
Accepted 16 January 2012
Available online 21 January 2012
Keywords:
Novenamine
Ó 2012 Elsevier Ltd. All rights reserved.
Noviose
Novobiocin
Aminocoumarins
Semi-synthetic aminocoumarins
1. Introduction
Aminocoumarins and their analogues also inhibit the activity of
other enzymes including heat shock protein 90 (Hsp90),3e5 HIV-1
Discovered over six decades ago, members of the amino-
coumarin family of antibiotics, including novobiocin, coumermycin
A1 and clorobiocin (Fig. 1), competitively inhibit the ATPase
activity of the bacterial type II topoisomerase DNA gyrase.1,2
RNase H6 and also suppress HIV-1 integrase.7 Aminocoumarins
also suppress sister chromatid exchange (SCE) induced by TNFa 8
.
Target specificities and biological activities vary dramatically
among different aminocoumarin analogues,9 and there is a need for
Fig. 1. Natural aminocoumarin antibiotic structures.
* Corresponding author. E-mail address: mfridman@post.tau.ac.il (M. Fridman).
0040-4020/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2012.01.042

Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
2307
methods for the preparation of new aminocoumarin derivatives
with novel activities and target specificities. Chemoenzymatic,10,11
chemical synthesis12e14 and synthetic biology15 methods for the
preparation of aminocoumarin derivatives have been reported.
However, the highly functionalized chemical structure of amino-
coumarins makes it challenging to chemo-selectively derivatize
these molecules and develop semi-synthetic methodologies for the
preparation of their analogues.
One chemoselective semi-synthetic transformation was de-
veloped by Hinman and co-workers,16 who reported the conversion
of novobiocin to novobiocin methyl oxazole analogue 1 (Scheme 1a)
using an excess of acetic anhydride in heated pyridine.
different families of novobiocin analogues can be prepared
depending on the choice of carboxylic acid anhydride.
2. Results and discussion
To study the acylation of novobiocin, an excess (12 equiv) of the
arbitrarily chosen isobutyric anhydride and dry pyridine (12 equiv)
were added to novobiocin in DMF as solvent under inert conditions.
The reaction was kept at ambient temperature for 2 h and then
gradually heated to 70 ꢀC over 8 h. Negative electrospray ionization
mass spectrometry (ESI-MS) reaction monitoring indicated that
initially, the peak corresponding to the molecular weight of novo-
biocin ([MꢁH]ꢁ, m/z 611.7) disappeared while that of the mono-
isobutyryl novobiocin appeared ([MꢁH]ꢁ, m/z 681.2). The mono-
isobutyryl novobiocin rapidly disappeared concurrently with the
appearaꢁnce of a peak corresponding to di-isobutyryl novobiocin
([MꢁH] , m/z 752.0). Upon gradual heating of the reaction to 70 ꢀC,
peaks corresponding to the molecular weight of di-isobutyryl
novenamine ([MꢁH]ꢁ, m/z 563.8), to the molecular weight of tri-
acylated novenamine ([MꢁH]ꢁ, m/z 633.6) and to the molecular
weight of 3-dimethylallyl-4-O-butyryl-benzoic acid ([MꢁH]ꢁ, m/z
275.4) were observed. After additional stirring at 70 ꢀC, the tri-
acylated novenamine peak disappeared and only the di-isobutyryl
novenamine and 3-dimethylallyl-4-O-butyryl-benzoic acid peaks
remained. Similar ESI-MS reaction progress profiles were observed
when other anhydrides were used in the reaction. The exception
was pivalic anhydride. Initially, as in reactions with other anhy-
drides, the molecular weights corresponding to the conversion of
novobiocin to mono- and di-pivaloyl novobiocin were detected
([MꢁH]ꢁ, m/z 695.8 and [MꢁH]ꢁ, m/z 779.9, respectively). How-
ever, once the temperature was elevated, only the molecular
weight corresponding to mono-pivaloyl novenamine ([MꢁH]ꢁ, m/z
507.6) was detected; no di- or tri-pivaloyl novenamine was ob-
served. By performing the reaction several times and terminating
the reaction after each step, mono- and di-acylated novobiocin
intermediate compounds were isolated by reverse-phase HPLC, and
their structures were determined by 1H NMR spectroscopy.
Although the intermediate acylation products had limited sta-
bility, in the case of pivalic and isobutyric anhydrides mono- and di-
acylated novobiocin intermediate compounds were both separable
by HPLC and were stable enough for 1H NMR characterization. Two
different mono-pivaloyl novobiocin intermediate compounds were
isolated. In one, acylation had occurred at the phenol position of the
3-dimethylallyl-4-hydroxybenzoyl segment of novobiocin (com-
pound 4, Scheme 2a). The second was the enol ester of novobiocin
(compound 5).
Scheme 1. Semi-synthetic aminocoumarin analogues. (a) Ac2O (excess), wet pyridine,
110 ꢀC, 4 h. (b) (i) RCOCl or (RCO)2O (excess), wet pyridine,
D. (ii) PTSA, methanol.
Based on this chemical transformation, Godfrey and co-workers
developed a three step synthesis of enamide analogues of cou-
mermycin A1 (Scheme 1b).17 In the initial step, 2,20-O,O-di-tetra-
hydropyranyl protected compound 2 was prepared from cou-
mermycin A1 to avoid acylation on the C-2 axial alcohol of the
noviose carbohydrate unit. Addition of excess of acyl chlorides or
carboxylic acid anhydrides to 2 in heated wet pyridine followed by
the acidic removal of the tetrahydropyranyl groups resulted in
moderate yields of enamide structures, such as 3 (Scheme 1b).
Unlike the reaction with novobiocin, treatment of coumermycin A1
under these conditions did not result in oxazole products. Two
major structural features differentiate novobiocin from cou-
mermycin A1 (Fig. 1): whereas in novobiocin the C-3 alcohol of the
noviose sugar segment is substituted by a carbamoyl group, the
noviose C-3 alcohol in coumermycin A1 is substituted by a 5-
methyl-2-carboxy-pyrrole ester. In novobiocin, a single noven-
amine unit is attached to a 3-dimethylallyl-4-hydroxybenzoyl
moiety through an enamide bond; in coumermycin A1 a 3-
methylpyrrole-2,4-bis carboxylic acid unit connects two
novenamine-like segments through enamide bonds. These two
structural differences affect the chemical reactivity and stability of
the two aminocoumarins.
The identities of the intermediates were determined by 1H NMR
spectroscopy (Fig. 2). A down-field 1H NMR shift of the aromatic
protons of the 3-dimethylallyl-4-O-pivaloyl-benzoyl segment (H-2,
H-5 and H-6) of 4 when compared to the chemical shifts of the
aromatic protons of the 3-dimethylallyl-4-hydroxybenzoyl seg-
ment of novobiocin (Fig. 2a) confirmed the position of the pivaloyl
group in compound 4 (Fig. 2b). For compound 5, a change in the 1H
NMR chemical shifts of the coumarin aromatic protons (H-40 and H-
50) and no change in the chemical shift of the protons (H-2, H-5 and
H-6) on the 3-dimethylallyl-4-hydroxybenzoyl segment was ob-
served (Fig. 2c). In addition, an NOE effect between the methyl
groups of the pivaloyl ester and H-40 was observed by 1D-NOE
experiments, confirming the structure of enol-ester 5.
When isobutyric anhydride was used, two mono-isobutyryl
novobiocin intermediates were isolated. Unlike the two mono-
pivaloyl novobiocin intermediates 4 and 5, in the reaction with
isobutyric anhydride the intermediates were identified as the iso-
butyryl enol-ester 6 and the asymmetric enimide 7 (Scheme 2b).
1D-NOE experiments confirmed the proximity between the iso-
propyl protons of the isobutyryl enamide of 7 and the aromatic
In this manuscript we describe the semi-synthetic preparation
of acylated novenamines starting from novobiocin. We discuss the
different steps of the reaction mechanism and demonstrate how

2308
Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
Scheme 2. Mono- and di-acylated novobiocin intermediates. (a) Pivalic anhydride (8 equiv), pyridine (8 equiv), DMF, room temperature. (b) Isobutyric anhydride (8 equiv), pyridine
(8 equiv), DMF, room temperature.
protons of the 3-dimethylallyl-4-hydroxybenzoyl segment (H-2
and H-6). For the asymmetric enimide 7, duplicated 1H NMR signals
were observed for H-2, H-5 and H-6 that appeared in a 1:1 ratio,
indicating a population of two different conformers (Fig. 2d). In the
case of pivalic anhydride, di-pivaloyl novobiocin product 8 (Scheme
2a) was found to be acylated on the 3-dimethylallyl-4-
hydroxybenzoyl to give the 3-dimethylallyl-4-O-pivaloyl-benzoyl
segment and on the enol position to form the enol-pivaloyl ester.
The di-isobutyryl novobiocin product 9 (Scheme 2b) showed 1H
NMR characteristics similar to these of 8. Similar to the di-pivaloyl
novobiocin 8, di-isobutyryl novobiocin 9 was acylated at the 3-
dimethylallyl-4-O-isobutyryl-benzoyl segment and on the enol
position.
Based on the intermediate compounds structures and on the
ESI-MS reaction progress profile we suggest that as a result of the
second acylation and upon heating, the reaction proceeds via an
asymmetric enimide that is formed as a result of the migration of
the acyl group from the enol-ester of the di-acylated novobiocin
to the enamide nitrogen. The isolated intermediate asymmetric
enimide 7 (Scheme 2b), which is a precursor of the di-acylated
novobiocin intermediate 9, supports this mechanistic suggestion.
Finally, pyridine-mediated cleavage of the 3-dimethylallyl-4-O-
acyl-benzoyl segment of the asymmetric enimide intermediate of
novobiocin yielded the enamide product. Previous studies suggest
that the acylation reaction of coumermycin A1 (3) may proceed via
a similar mechanism.18,19
The role of the nitrogen base was studied by using DMF as
a solvent and the non-nucleophilic di-isopropyl ethylamine as the
base. Under these conditions, the di-acylated novobiocin was
readily formed; however, the acylated novenamine final products
were not detected even after prolonged heating. When pyridine or
Fig. 2. 1H NMR of the aromatic region of reaction intermediates.
imidazole (equimolar to the anhydride) was added, gradual

Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
2309
formation of acylated novenamine products was observed by ESI-
MS. This experiment indicates that a nucleophilic nitrogen is re-
quired to facilitate the 3-dimethylallyl-4-acylbenzoate unit cleav-
age. Although the first acylation on novobiocin varied among the
phenol, enol and enamide nitrogen, in both isolated di-acylated
novobiocin intermediates the phenol position was acylated.
Moreover, ESI-MS monitoring of all of the tested anhydrides in-
dicated that no cleavage of the 3-dimethylallyl-4-hydroxybenzoyl
segment occurred before the formation of the di-acylated novobi-
ocin was completed. This can be explained by the fact that the
electron withdrawing effect caused by the acylation of the phenol
position of the 3-dimethylallyl-4-hydroxybenzamide increases the
electrophilicity of the benzamide carbonyl and facilitates the
pyridine-mediated cleavage of this segment.
When performed in pyridine using DMF as solvent and pyridine
in an equimolar ratio to the carboxylic acid anhydride yields of the
products isolated upon HPLC purification of the reaction mixtures
were best. In general, with fewer than 8 equiv of anhydride, re-
action yields were lower and the sequence did not proceed beyond
the mono-acylated novobiocin. We synthesized a series of di-
acylated novenamines (compounds 10e20, Table 1) and isolated
them in moderate to good isolated yields (27e83%). 1H and 1D-
TOCSY NMR analysis confirmed that compounds 10e15 were ac-
ylated on the C-2 alcohol of the noviose carbohydrate segment and
on the enamine of the novenamine unit to form the corresponding
enamide. In the case of pivalic anhydride, the final product was
identified as mono-pivaloyl enamide 16 (56% isolated yield). This
result was observed only when pivalic anhydride with the hindered
anhydrides, a dynamic mixture of di-acylated novenamine prod-
ucts containing the symmetric enimides 19 and 20 and the enol-
ester enamides 19a and 20a were obtained (32% isolated yield of
the dynamic mixture 19/19a and 33% isolated yield for the dynamic
mixture 20/20a, Table 1).
1H NMR spectra of the same mixture at different temperatures
and times indicated different ratios of the symmetric enimides and
the corresponding enol-ester enamides were present. Although
reactions were performed using an excess of anhydride and at an
elevated temperature, which should enable acylation of every nu-
cleophilic group on the molecule with no selectivity, chemo-
selectivity was observed when aromatic carboxylic acid anhydrides
were used. As with pivalic anhydride, no noviose C-2 alcohol es-
terification of the reaction products was observed for these anhy-
drides. Hence, with these anhydrides no protecting group
manipulation was required to prevent acylation from taking place
at the noviose C-2 alcohol.
As aminocoumarins and their analogues bind to several differ-
ent proteins, the semi-synthetic novenamine analogues 10e20
have a broad range of potential biological activities.1e7 Since ami-
nocoumarins were initially identified as antibacterial agents, we
tested several analogues representing the three subfamilies of the
semi-synthetic acylated novenamines 10e20 in bacterial growth
assays. We chose two Gram positive bacteria and two Gram nega-
tive bacteria for this analysis (Table 2) including Arthrobacter NRRL
B-3652, a species that expresses an amidase that cleaves the
enamide bond of novobiocin and clorobiocin and demonstrates
high resistance to novobiocin. The di-isobutyric and di-crotonyl
novenamines representing the first family of acylated noven-
amine did not inhibit growth of the tested Gram positive and Gram
negative bacteria. When mixtures of aromatic carboxylic acid an-
hydride analogues 19/19a and 20/20a were tested, modest in-
hibition of Gram positive cell growth and no inhibition of Gram
negative bacterial growth were observed. The mono-pivaloyl
enamide 16, representing the third subfamily of the semi-
synthetic acylated novenamine analogues, was significantly more
active against both strains of Gram positive bacteria yet had no
activity against the tested Gram negative strains. This suggests that
mono-acylated enamide novenamine analogues may serve as
promising leads for the development of antibiotics with activity
against Gram positive bacteria.
tert-butyl group
hydride with a less hindered tertiary carbon
a
to the carbonyl was used; when isobutyric an-
to the carbonyl was
a
used, the C-2 alcohol of the noviose was acylated. When aromatic
acid anhydrides were used, several unexpected results were ob-
served. When the reaction was performed using benzoic anhydride,
the final product, obtained in 83% isolated yield, was identified as
the symmetric enimide novenamine analogue 17; a similar result
was obtained when p-methoxy benzoic anhydride was used (the
products was the symmetric enimide novenamine analogue 18,
which was obtained in 45% isolated yield). However, when 4-fluoro
benzoic anhydride or 4-methylbenzoic anhydride was used, a sin-
gle HPLC peak of di-acylated novenamine product was isolated. 1H
NMR characterization indicated that for these two aromatic acid
Table 1
Acylated novenamine analogues
Compound
Anhydride
R1
R2
R3
R4
Yield
10
11
12
13
14
15
16
17
18
19
19a
20
20a
Isobutyric
Acetic
Isovaleric
4-n-Chlorobutyric
Crotonic
Propionic
Pivalic
Benzoic
p-Methoxy benzoic
4-Fluoro-benzoic
4-Fluoro-benzoic
4-Methylbenzoic
4-Methylbenzoic
Isobutyryl
Acetyl
Isovaleryl
4-n-Chlorobutyryl
Crotonyl
Propionyl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Isobutyryl
Acetyl
Isovaleryl
4-n-Chlorobutyryl
Crotonyl
Propionyl
Pivaloyl
Benzoyl
p-Methoxy benzoyl
4-Fluorobenzoyl
4-Fluorobenzoyl
4-Methylbenzoyl
4-Methylbenzoyl
H
H
H
H
H
H
H
H
80%
43%
42%
43%
27%
35%
56%
83%
45%
32%a
Benzoyl
p-Methoxy benzoyl
4-Fluorobenzoyl
H
4-Methylbenzoyl
H
H
H
4-Fluorobenzoyl
H
4-Methylbenzoyl
33%a
a
Yield of dynamic mixtures obtained for 4-fluoro-benzoic anhydride (19/19a) and for the 4-methylbenzoic anhydride (20/20a).

2310
Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
Table 2
MIC values (
(250 mmꢂ21.20 mm). All reagents were purchased from com-
m
g/mL) of acylated novenamine analogues
Bacterial strainsa
mercial sources and used without further purification. 4-n-Chlor-
obutyric anhydride was synthesized according to a previously
reported procedure.20
A
B
C
D
Novobiocin
10
14
16
19/19a
20/20a
<0.125
>256
>256
32
128
128
0.5
128
>256
>256
>256
>256
>256
>256
4.1.1. General novobiocin acylation reaction demonstrated for ana-
logue 10. Novobiocin sodium salt (200 mg, 0.31 mmol) was dis-
solved in DMF (3 mL). Isobutyric anhydride (0.41 mL, 2.52 mmol)
and pyridine (0.2 mL, 2.52 mmol) were added, and the solution was
stirred at ambient temperature. Monitoring of the reaction by ESI-
MS indicated the disappearance of the starting material ([MꢁH]ꢁ,
m/z 611.7) and the formation of the di-acylated novobiocin
([MꢁH]ꢁ, m/z 752.0). After heating the reaction to 70 ꢀC for an
additional 8 h, full conversion to enamide 10 ([MꢁH]ꢁ, m/z 563.8)
was detected. Upon complete conversion, the crude was evapo-
rated under vacuum, re-dissolved in a minimal volume of CH3CN
and purified by reverse-phase HPLC at a flow rate of 15.0 mL/min.
HPLC solvent A was H2O and solvent B was CH3CN. The elution
gradient was 95% A for 4 min followed by 95e0% A over 20 min and
100% B for 3 min. The product was detected at 210 and 254 nm
using a double-channel UV detector (product retention time-
¼25.5 min). Fractions containing the pure product were concen-
trated under vacuum, dissolved in H2O and freeze dried to yield
enamide 10 as a white solid (142 mg, 80%), mp 115e125 ꢀC. 1H NMR
>256
>256
32
>256
256
>256
>256
>256
>256
>256
a
MIC values were determined for the Gram positive strains (A) Staphylococcus
epidermidis ATCC12228 and (B) bacillus subtilis ATCC6633 and for the Gram nega-
tive strains (C) Escherichia coli BL21 (DE3) and (D) Arthrobacter NRRL B-3652.
Strains were tested using the double dilution method. Experiments were
performed in triplicate.
3. Conclusions
In this study we describe acylation conditions of novobiocin for
the preparation of several families of acylated novenamine ana-
logues. ESI-MS proved an essential reaction monitoring method that
enabled us to isolate reaction intermediates and gain insight intothe
steps of the acylation sequence. By using different anhydrides, we
demonstrated the preparation of different families of semi-
synthetic aminocoumarins in a single synthetic step from com-
mercially available novobiocin. When alkyl substituted anhydrides
were used, di-acylated novenamines (compounds 10e15) were
formed. In these derivatives, the 3-dimethylallyl-4-acylbenzoate
segment of novobiocin was replaced by an anhydride acyl group,
and the axial C-2 alcohol of the noviose carbohydrate segment was
converted to the corresponding ester. When a highly hindered an-
(400 MHz, MeOD):
d
1.23e1.29 (15H, m, H-600, COCH(CH3)2), 1.42
(3H, s, H-700), 2.33 (3H, s, H-100), 2.70 (1H, h, J 7.0 Hz, COCH(CH3)2),
2.88 (1H, h, J 6.8 Hz, COCH(CH3)2), 3.52 (1H, d, J 8.9 Hz, H-400), 3.60
(3H, s, H-800), 5.45e5.48 (2H, m, H-200, H-300), 5.64 (1H, d, J 2.4 Hz, H-
100), 7.22 (1H, d, J 8.9 Hz, H-50), 7.79 (1H, d, J 8.8 Hz, H-40). 13C NMR
(125.7 MHz, MeOD) d 9.5, 20.0, 20.3, 20.7 (2C), 24.3, 29.6, 36.1, 37.0,
hydride with a quaternary carbon
a to the carbonyl was used, as in
62.7, 71.6, 72.5, 80.8, 83.9, 98.0,104.4,112.5,113.4,115.8,124.1,152.5,
158.7, 159.4, 159.5, 163.7, 178.3, 181.8. Positive HRESIMS, m/z calcd
587.2217 for C27H36N2O11Na, found 587.2222 [MþNa]þ.
the case of pivalic anhydride, only the mono-acylated enamide
(compound 16) was formed and no noviose carbohydrate axial C-2
alcohol acylation took place even though an excess of anhydride was
used. When aromatic acid anhydrides were used, a third family of
novobiocin semi-synthetic analogues was formed (compounds
18e20). For some of the aromatic acid anhydrides, stable symmetric
enimide derivatives could be isolated. For others, dynamic mixtures
of the enol-esters and the corresponding symmetric enimide ana-
logues were isolated. These acylated novenamines may possess
a variety of biological activities. Here we used standard antibacterial
growth assays to demonstrate that mono-acylated enamide de-
rivatives, such as compound 16 may serve as leads for the de-
velopment of novenamine-based antibacterial agents.
4.1.2. Di-acetyl novenamine (compound 11). Novobiocin sodium
salt (200 mg, 0.31 mmol), DMF (3 mL), acetic anhydride (0.24 mL,
2.52 mmol), pyridine (0.2 mL, 2.52 mmol). HPLC product retention
time 20.7 min. Product yield (69 mg, 0.14 mmol, 43%), mp
80e90 ꢀC. 1H NMR (500 MHz, MeOD): 1.21 (3H, s, H-600), 1.37 (3H,
d
s, H-700), 2.16 (3H, s, H-100), 2.20 (3H, s, COCH3), 2.25 (3H, s, COCH3),
3.49 (1H, d, J 8.8 Hz, H-400), 3.53 (3H, s, H-800), 5.42e5.45 (2H, m, H-
200, H-300), 5.62 (1H, s, H-100), 7.15 (1H, d, J 8.9 Hz, H-50), 7.69 (1H, d, J
8.9 Hz, H-40). 13C NMR (125.7 MHz, MeOD):
d 9.4, 21.6, 23.5, 24.2,
29.6, 62.7, 71.4, 72.8, 80.8, 83.7, 97.9, 104.1, 112.5, 113.0, 115.8, 124.0,
125.4, 159.0, 159.4, 159.5, 163.6, 172.4, 175.1. Positive HRESIMS m/z
calcd 531.1591 for C23H28N2O11Na, found 531.1589 [MþNa]þ.
4. Experimental section
4.1. General
4.1.3. Di-isovaleryl novenamine (compound 12). Novobiocin sodium
salt (300 mg, 0.47 mmol), DMF (3 mL), isovaleric anhydride
(0.75 mL, 3.78 mmol), pyridine (0.3 mL, 3.78 mmol). HPLC retention
time 26.9 min. Product yield (117 mg, 0.19 mmol, 42%), mp
NMR spectra were recorded on 200, 400 or 500 MHz in-
struments for 1H NMR and 100.6 MHz or 125.7 MHz for 13C NMR.
Chemical shifts are reported in parts per million (ppm). 1H NMR
spectra were calibrated to CD3OD (3.34 ppm), or DMSO (2.54 ppm).
13C NMR spectra were calibrated to CD3OD (49.86 ppm) or DMSO
(40.45 ppm). Multiplicities are reported using the following ab-
breviations: s¼singlet, d¼doublet, dd¼doublet of doublet,
t¼triplet, q¼quartet, h¼heptet and m¼multiplet. Coupling con-
stants (J) are given in Hertz. In addition to 1H and 13C NMR spec-
trometry, the exact structures of the reported compounds were
determined using sets of 1D-TOCSY and 1D-NOE experiments. IR
spectra were recorded on a Bruker FT-IR spectrometer TENSOR 27
using an ATR detector (see Supplementary data). Unless otherwise
mentioned, all reactions were conducted under an argon atmo-
sphere using anhydrous solvents. Reactions were monitored using
an LC-ESI mass detector. Compounds were purified on a preparative
75e85 ꢀC. 1H NMR (400 MHz, MeOD):
d 1.00e1.06 (12H, m,
COCH2CH(CH3)2), 1.22 (3H, s, H-600), 1.41 (3H, s, H-700), 2.10e2.20
(2H, m, COCH2CH(CH3)2), 2.29 (3H, s, H-100), 2.29e2.32 (2H, m,
COCH2CH(CH3)2), 2.38 (2H, d, J 5.8 Hz, COCH2CH(CH3)2), 3.53 (1H, d,
J 9.0 Hz, H-400), 3.60 (3H, s, H-800), 5.44e5.49 (2H, m, H-200, H-300),
5.65 (1H, d, J 2.4 Hz, H-100), 7.22 (1H, d, J 9.0 Hz. H-50), 7.80 (1H, d, J
8.9 Hz, H-40). 13C NMR (125.7 MHz, MeOD):
d 9.5, 23.6 (4C), 24.2,
27.8, 28.4, 29.7, 45.0, 46.5, 62.7, 71.6, 72.7, 80.9, 83.9, 98.1, 104.2,
112.5, 113.5, 115.9, 124.1, 152.6, 159.2, 159.4, 159.5, 163.7, 174.3, 177.4.
Positive HRESIMS m/z calcd 615.2530 for C29H40N2O11Na, found
615.2527 [MþNa]þ.
4.1.4. Di-4-n-chlorobutyryl novenamine (compound 13). Novobiocin
sodium salt (200 mg, 0.31 mmol) DMF (3 mL), 4-chloro-n-butyric
anhydride (0.92 g, 4.09 mmol), pyridine (0.32 mL, 4.09 mmol).
HPLC system using a Phenomenex Luna Axia 5-mm C-18 column

Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
2311
HPLC retention time 24.9 min. Product yield (85 mg, 0.13 mmol,
3.45e3.48 (1H, m, H-400), 3.48 (3H, s, H-800), 4.04 (1H, m, H-200), 5.16
(1H, dd, J1 3.1 Hz, J2 9.7 Hz, H-300), 5.41 (1H, d, J 2.4 Hz, H-100), 6.89
(1H, d, J 8.8 Hz, H-50), 7.35e7.40 (4H, m, benzoyl), 7.43e7.45 (2H, m,
benzoyl), 7.56 (1H, d, J 8.6 Hz, H-40), 7.75e7.78 (4H, m, benzoyl). 13C
43%), mp 80e90 ꢀC. 1H NMR (200 MHz, MeOD): 1.22 (3H, s, H-600),
d
1.41 (3H, s, H-700), 2.09e2.27 (4H, m, COCH2CH2CH2Cl), 2.32 (3H, s,
H-100), 2.66 (2H, t, J 7.2 Hz, COCH2CH2CH2Cl), 2.72 (2H, t, J 7.0 Hz,
COCH2CH2CH2Cl), 3.51 (1H, d, J 8.9 Hz, H-400), 3.60 (3H, s, H-800), 3.70
NMR (100.6 MHz, MeOD):
d 9.6, 24.0, 29.6, 62.1, 70.0, 71.6, 78.9,
(2H, t,
J
6.4 Hz, COCH2CH2CH2Cl), 3.71 (2H, t,
J
6.4 Hz,
82.2, 99.6, 102.9, 109.1, 113.1, 118.5, 123.8, 128.9 (8C), 132.1 (2C),
137.8 (2C), 153.2, 156.8, 157.5 (2C), 162.7, 170.6, 174.7. Positive
HRESIMS m/z calcd 655.1904 for C33H32N2O11Na, found 655.1907
[MþNa]þ.
COCH2CH2CH2Cl), 5.44e5.52 (2H, m, H-200, H-300), 5.68 (1H, d, J
2.3 Hz, H-100), 7.21 (1H, d, J 9.0 Hz, H-50), 7.78 (1H, d, J 8.9 Hz, H-40).
13C NMR (125.7 MHz, MeOD):
d 9.5, 24.2, 29.6, 29.8, 30.4, 32.9, 34.6,
45.6, 45.9, 62.8, 71.5, 73.0, 80.9, 83.8, 98.0, 104.1, 112.6, 113.1, 115.9,
124.1, 152.6, 159.3, 159.4, 159.5, 163.7, 174.1, 176.8. Positive HRESIMS
m/z calcd 655.1437 for C27H34Cl2N2O11Na, found 655.1435
[MþNa]þ.
4.1.9. Di-4-methoxybenzoyl novenamine (compound 18). Novobiocin
sodium salt (300 mg, 0.47 mmol), DMF (3 mL), 4-methoxybenzoic
anhydride (1.08 g, 3.78 mmol), pyridine (0.3 mL, 3.78 mmol). HPLC
retention time 21.0 min. Product yield (147 mg, 0.21 mmol, 45%),
4.1.5. Di-crotonyl novenamine (compound 14). Novobiocin sodium
salt (200 mg, 0.31 mmol), DMF (3 mL), crotonic anhydride (0.6 mL,
4.09 mmol), pyridine (0.32 mL, 4.09 mmol). HPLC retention time
19.7 min. Product yield (47 mg, 0.08 mmol, 27%), mp 90e100 ꢀC. 1H
mp 110e120 ꢀC. 1H NMR (400 MHz, MeOD): 1.17 (3H, s, H-600),
d
1.35 (3H, s, H-700), 2.26 (3H, s, H-100), 3.56 (3H, s, H-800), 3.57 (1H, d,
J 9.7 Hz, H-400), 3.69 (6H, s, OMe of 4-methoxybenzoyl), 4.20 (1H, t,
J 2.9 Hz, H-200), 5.33 (1H, dd, J1 3.1 Hz, J2 9.7 Hz, H-300), 5.53 (1H, d, J
2.6 Hz, H-100), 6.88 (2H, d, J 8.9 Hz, 4-methoxybenzoyl), 6.90 (2H, d,
J 8.9 Hz, 4-methoxybenzoyl), 7.07 (1H, d, J 8.9 Hz, H-50), 7.73 (1H,
d, J 8.8 Hz, H-40), 7.83 (2H, d, J 8.9 Hz, 4-methoxybenzoyl), 7.86
(2H, d, J 8.9 Hz, 4-methoxybenzoyl). 13C NMR (100.6 MHz, MeOD):
NMR (400 MHz, DMSO):
d
1.15 (3H, s, H-600), 1.39 (3H, s, H-700), 1.91
(dd, 3H, J1 1.3 Hz, J2 6.3 Hz, COCHCHCH3), 1.95 (dd, 3H, J1 1.5 Hz, J2
6.9 Hz, COCHCHCH3), 2.23 (3H, s, H-100), 3.47 (1H, d, J 9.0 Hz, H-400),
3.52 (3H, s, H-800), 5.34e5.39 (2H, m, H-200, H-300), 5.73 (1H, d, J
2.9 Hz, H-100), 6.02 (1H, dd, J1 1.5 Hz, J2 15.7 Hz, COCHCHCH3), 6.44
(1H, dd, J1 1.8 Hz, J2 15.4 Hz, COCHCHCH3), 6.84e6.92 (1H, m,
COCHCHCH3), 7.01e7.10 (1H, m, COCHCHCH3), 7.19 (1H, d, J 9.0 Hz,
d
9.6, 24.1, 29.8, 56.7 (2C), 62.7, 71.8, 73.9, 80.6, 83.6, 100.7, 106.8,
111.1, 115.1 (4C), 115.4, 118.8, 124.7, 130.7 (2C), 132.7 (2C), 133.3
(2C), 154.5, 159.1 (2C), 160.0, 164.8, 164.9, 166.6, 174.1, 176.8. Posi-
tive HRESIMS m/z calcd 715.2115 for C35H36N2O13Na, found
715.2119 [MþNa]þ.
H-50), 7.75 (1H, d, J 8.9 Hz, H-40). 13C NMR (125.7 MHz, DMSO):
d 9.4,
18.9, 19.0, 24.2, 29.1, 62.1, 69.2, 70.9, 79.5, 82.2, 96.3, 102.3, 111.3,
112.7, 114.0, 112.9, 123.2, 125.6, 142.6, 148.1, 151.3, 156.9, 157.5, 158.9,
161.3, 165.7, 167.1. Positive HRESIMS m/z calcd 583.1904 for
C27H32N2O11Na, found 583.1905 [MþNa]þ.
4.1.10. Di-4-fluorobenzoyl novenamine (dynamic mixture of com-
pounds 19 and 19a). Novobiocin sodium salt (75 mg, 0.12 mmol),
DMF (1 mL), 4-fluorobenzoyl anhydride (250 mg, 0.95 mmol),
4.1.6. Di-propionyl novenamine (compound 15). Novobiocin sodium
salt (300 mg, 0.47 mmol), DMF (3 mL), propionic anhydride
(0.48 mL, 3.78 mmol), pyridine (0.3 mL, 3.78 mmol). HPLC retention
time 23.3 min. Product yield (88 mg, 0.16 mmol, 35%), mp
pyridine (77
m
L, 0.95 mmol). HPLC retention time 19.7 min.
Product yield (25 mg, 0.04 mmol, 32%), mp 145e155 ꢀC. 1H NMR
(400 MHz, MeOD) of compound 19:
d
1.04 (3H, s, H-600), 1.25 (3H,
s, H-700), 2.10 (3H, s, H-100), 3.46 (3H, s, H-800), 3.47 (1H, m, H-400),
4.0 (1H, t, J 2.8 Hz, H-200), 5.13 (1H, dd, J1 3.3 Hz, J2 9.7 Hz, H-300),
5.43 (1H, d, J 2.5 Hz, H-100), 6.92 (1H, d, J 8.8 Hz, H-50), 7.15e7.26
(4H, m, aromatic), 7.77e7.82 (4H, m, aromatic), 7.58 (1H, d, J
120e130 ꢀC. 1H NMR (400 MHz, MeOD):
d
1.19e1.27 (3H, m, H-600),
1.19e1.27 (6H, m, COCH2CH3), 1.41 (3H, s, H-700), 2.27 (2H, q, J 9.3 Hz,
COCH2CH3), 2.33 (3H, s, H-100), 2.50 (2H, q, J 9.4 Hz, COCH2CH3),
3.52 (1H, d, J 8.9 Hz, H-400), 3.59 (3H. s, H-800), 5.45e5.50 (2H, m, H-
200, H-300), 5.66 (1H, d, J 2.2 Hz, H-100), 7.23 (1H, d, J 9.0 Hz, H-50), 7.80
8.3 Hz, H-40). 1H NMR (400 MHz, MeOD) of compound 19a:
d 1.05
(3H, s, H-600), 1.27 (3H, s, H-700), 2.30 (3H, s, H-100), 3.47(3H, s, H-
800), 3.47 (1H, m, H-400), 4.1 (1H, t, J 2.8 Hz, H-200), 5.13 (1H, dd, J1
3.4 Hz, J2 9.9 Hz, H-300), 5.43 (1H, d, J 2.6 Hz, H-100),7.15e7.26 (1H,
m, H-50), 7.28 (2H, t, J 8.9 Hz, aromatic), 7.42 (2H, t, J 8.8 Hz, ar-
omatic), 7.57 (1H, d, J 8.3 Hz, H-40), 7.87 (1H, d, J 8.8 Hz, aromatic),
7.88 (1H, d, J 8.8 Hz, aromatic), 8.18 (1H, d, J 8.9 Hz, aromatic), 8.19
(1H, d, J 8.9 Hz, aromatic). 13C NMR of the dynamic mixture of
(1H, d, J 8.9 Hz. H-40). 13C NMR (125.7 MHz, MeOD):
d 9.5, 10.1, 10.9,
21.2, 29.1, 29.6, 30.7, 62.7, 71.5, 72.7, 80.9, 83.9, 98.1, 104.0, 112.5,
113.5, 115.8, 124.1, 152.6, 159.4, 159.5, 159.7, 163.9, 175.7, 178.6.
Positive HRESIMS m/z calcd 559.1904 for C25H32N2O11Na, found
559.1906 [MþNa]þ.
4.1.7. Mono-pivalyl novenamine (compound 16). Novobiocin so-
dium salt (200 mg, 0.31 mmol), DMF (2 mL), pivalic anhydride
(0.51 mL, 2.52 mmol), pyridine (0.2 mL, 2.57 mmol). HPLC retention
time 23.1 min. Product yield (90 mg, 0.18 mmol, 56%), mp
compounds 19 and 19a (125.7 MHz, MeOD): d 10.8, 10.9, 25.4,
30.8, 30.9, 63.6, 63.7, 71.2, 71.4, 73.1, 73.2, 80.7, 81.0, 83.3, 83.4,
100.6, 100.7, 105.2, 111.3, 112.2, 113.7, 115.0, 115.2, 116.4, 117.6, 117.7,
117.8, 117.9, 118.1, 118.3, 118.6, 118.9, 119.2, 124.6, 125.4, 126.3,
131.9, 133.0, 133.1, 133.2, 133.3, 135.0, 135.9, 136.0, 152.8, 154.4,
156.7, 158.8, 159.2, 160.2, 162.0, 164.0, 165.5, 167.5, 167.8, 171.9,
175.3. Negative HRESIMS m/z calcd 667.1736 for C33H29F2N2O11,
found 667.1736 [MꢁH]ꢁ.
150e160 ꢀC. 1H NMR (400 MHz, MeOD):
d
1.17 (3H, s, H-600), 1.37
(3H, s, H-700), 1.38 (9H, s, CO(CH3)3), 2.31 (3H, s, H-100), 3.58 (3H, s,
H-800), 3.62 (1H, d, J 10.0 Hz, H-400), 4.26 (1H, t, J 2.4 Hz, H-200), 5.35
(1H, dd, J1 2.8 Hz, J2 9.8 Hz, H-300), 5.61 (1H, d, J 2.1 Hz, H-100), 7.23
(1H, d, J 8.9 Hz, H-50), 7.78 (1H, d, J 8.9 Hz, H-40). 13C NMR
(100.6 MHz, MeOD):
d
9.9, 24.1, 28.6 (3C), 29.9, 41.3, 62.7, 71.7, 73.8,
4.1.11. Di-4-methylbenzoyl novenamine (dynamic mixture of com-
pounds 20 and 20a). Novobiocin sodium salt (78 mg, 0.12 mmol),
DMF (1 mL), 4-methylbenzoic anhydride (250 mg, 0.98 mmol),
80.8, 83.5, 100.9, 103.9, 112.4, 113.3, 115.7, 124.1, 152.5, 159.3, 159.7,
160.0, 164.2, 182.7. Positive HRESIMS m/z calcd 531.1955 for
C24H32N2O10Na, found 531.1954 [MþNa]þ.
pyridine (77
yield (27 mg, 0.04 mmol, 33%), mp 110e120 ꢀC. 1H NMR for com-
pound 20 (500 MHz, MeOD):
1.21 (3H, s, H-600), 1.25 (3H, s, H-700),
mL, 0.95 mmol). HPLC retention time 20.5 min. Product
4.1.8. Di-benzoyl novenamine (compound 17). Novobiocin sodium
salt (300 mg, 0.47 mmol), DMF (3 mL), benzoic anhydride (0.85 mg,
3.78 mmol), pyridine (0.3 mL, 3.78 mmol). HPLC retention time
18.1 min. This compound was obtained as a hygroscopic white
solid. Product yield (248 mg, 0.39 mmol, 83%). 1H NMR (400 MHz,
d
2.11 (3H, s, H-100), 2.33 (6H, s, methyl on 4-methylbenzoyl), 3.46
(3H, s, H-800), 3.46e3.48 (1H, m, H-400), 4.20 (1H, t, J 2.5 Hz, H-200),
5.14 (1H, dd, J1 3.1 Hz, J2 9.8 Hz, H-300), 5.45 (1H, d, J 2.2 Hz, H-100),
6.96 (1H, d, J 8.9 Hz, H-50), 7.17e7.19 (4H, m, aromatic), 7.62e7.65
(5H, m, H-40, aromatic). 1H NMR for compound 20a (500 MHz,
DMSO):
d
1.07 (3H, s, H-600), 1.27 (3H, s, H-700), 2.11 (3H, s, H-100),

2312
Y. Berkov-Zrihen et al. / Tetrahedron 68 (2012) 2306e2312
MeOD):
d
1.06 (3H, s, H-600), 1.27 (3H, s, H-700), 2.30 (3H, s, H-100),
References and notes
2.33 (3H, s, methyl on 4-methylbenzoyl), 2.39 (3H, s, methyl on
4-methylbenzoyl), 3.48(3H, s, H-800), 3.44e3.51 (1H, m, H-400), 4.1
(1H, t, J 2.2 Hz, H-200), 5.18 (1H, dd, J1 3.0 Hz, J2 9.8 Hz, H-300), 5.58
(1H, d, J 2.3 Hz, H-100), 7.22 (1H, d, J 9.5 Hz, H-50), 7.25 (2H, t, J 8.1 Hz,
aromatic), 7.39 (2H, t, J 8.1 Hz, aromatic), 7.51 (1H, d, J 8.9 Hz, H-40),
7.72 (2H, d, J 8.0 Hz, aromatic), 8.0 (2H, d, J 8.1 Hz, aromatic). 13C
NMR of the dynamic mixture of compounds 20 and 20a
1. Maxwell, A. Mol. Microbiol. 1993, 9, 681e686.
2. Heide, L. Nat. Prod. Rep. 2009, 26, 1241e1250.
3. Yu, X. M.; Shen, G.; Neckers, L.; Blake, H.; Holzbeierlein, J.; Cronk, B.; Blagg, B. S.
J. Am. Chem. Soc. 2005, 127, 12778e12779.
4. Donnelly, A. C.; Mays, J. R.; Burlison, J. A.; Nelson, J. T.; Vielhauer, G.; Holz-
beierlein, J.; Blagg, B. S. J. J. Org. Chem. 2008, 73, 8901e8920.
5. Yun, B. G.; Huang, W.; Leach, N.; Hartson, S. D.; Matts, R. L. Biochemistry 2004,
43, 8217e8229.
6. Althaus, I. W.; Franks, K. M.; Langley, K. B.; Kezdy, F. J.; Peterson, T.; Buxer, S. E.;
Decker, D. E.; Dolak, L. A.; Ulrich, R. G.; Reusser, F. Cell. Mol. Life Sci. 1996, 52,
329e335.
7. Chiang, C. C.; Mouscadet, J. F.; Tsai, H. J.; Liu, C. T.; Hsu, L. Y. Chem. Pharm. Bull.
2007, 55, 1740e1743.
8. Lazutka, J. R.; Rudaitiene, S. Mutat. Res., Environ. Mutagen. Relat. Subj. 1992, 268,
217e221.
(100.6 MHz, MeOD): d 9.4, 9.5, 22.1, 22.3, 22.4, 23.8, 29.4, 29.6, 62.2,
69.7, 69.73, 71.5, 71.6, 79.1, 79.4, 81.8, 81.9, 99.4 (2C), 110.9, 112.2,
114.8, 123.0, 123.8, 125.6, 128.8, 129.3, 129.7, 130.14, 130.9, 134.4,
142.6, 143.4, 146.8, 151.3, 152.9, 155.3, 157.5, 158.6, 160.5, 162.5,
163.4, 166.9, 174.4. Negative HRESIMS m/z calcd 659.2241 for
C35H35N2O11, found 659.2244 [MꢁH]ꢁ.
9. Burlison, J. A.; Avila, C.; Vielhauer, G.; Lubbers, D. J.; Holzbeierlein, J.; Blagg, B. S.
J. J. Org. Chem. 2008, 73, 2130e2137.
10. Meyers, C. L. F.; Oberthuer, M.; Heide, L.; Kahne, D.; Walsh, C. T. Biochemistry
2004, 43, 15022e15036.
11. Fridman, M.; Balibar, C. J.; Lupoli, T.; Kahne, D.; Walsh, C. T.; Garneau-Tsodikova,
Acknowledgements
S. Biochemistry 2007, 46, 8462e8471.
12. Burlison, J. A.; Neckers, L.; Smith, A. B.; Maxwell, A.; Blagg, B. S. J. J. Am. Chem.
Soc. 2006, 128, 15529e15536.
13. Shen, G. Bioorg. Med. Chem. Lett. 2004, 14, 5903e5906.
14. Le Bras, G.; Radanyi, C.; Peyrat, J. F.; Brion, J. D.; Alami, M.; Marsaud, V.; Stella,
B.; Renoir, J. M. J. Med. Chem. 2007, 50, 6189e6200.
We would like to thank the Tel Aviv University Start-up Fund for
the financial support of this study. We would like to thank Prof.
Shmuel Carmeli from the School of Chemistry at Tel Aviv University
for his advice on compound characterization.
15. Heide, L. Biotechnol. Adv. 2009, 27, 1006e1014.
16. Hinman, J. W.; Caron, E. L.; Hoeksema, H. J. Am. Chem. Soc. 1957, 79, 3789e3800.
17. Keil, J. G.; Godfrey, J. C.; Cron, M. J.; Hooper, I. R.; De Nettleton, K. E. P.; Schmitz,
H. Pure Appl. Chem. 1971, 28, 571e601.
18. Keil, J. G.; Hooper, I. R.; Cron, M. J.; Farbig, O. B.; Nettleton, D. E. J. Antibiot. 1968,
21, 551e566.
19. Raiford, L. C.; LeRosen, A. L. J. Am. Chem. Soc. 1945, 67, 2163e2165.
20. Sakai, T.; Iwata, K.; Utaka, M.; Takeda, A. Bull. Chem. Soc. Jpn. 1987, 60,
1161e1162.
Supplementary data
1H, 13C NMR and FT-IR spectra of acylated novenamine ana-
logues 10e20. Supplementary data related to this article can be
found online at doi:10.1016/j.tet.2012.01.042.