A click chemistry approach to pleuromutilin derivatives, part 2: Conjugates with acyclic nucleosides and their ribosomal binding and antibacterial activity

Article abstract of DOI:10.1021/jm201266b

Pleuromutilin is an antibiotic that binds to bacterial ribosomes and thereby inhibit protein synthesis. A new series of semisynthetic pleuromutilin derivatives were synthesized by a click chemistry strategy. Pleuromutilin was conjugated by different linke

Full text of DOI:10.1021/jm201266b

Article
pubs.acs.org/jmc
A Click Chemistry Approach to Pleuromutilin Derivatives, Part 2:
Conjugates with Acyclic Nucleosides and Their Ribosomal Binding
and Antibacterial Activity
Ida Dreier,†,§ Surender Kumar,†,§,‡ Helle Søndergaard,†,§ Maria Louise Rasmussen,†,§
Lykke Haastrup Hansen,†,∥ Nanna Holmgaard List,§ Jacob Kongsted,§ Birte Vester,*,†,∥
and Poul Nielsen*,†,§
†Nucleic Acid Center, §Department of Physics, Chemistry and Pharmacy, and ∥Department of Biochemistry and Molecular Biology,
University of Southern Denmark, 5230 Odense M, Denmark
S
* Supporting Information
ABSTRACT: Pleuromutilin is an antibiotic that binds to
bacterial ribosomes and thereby inhibit protein synthesis. A
new series of semisynthetic pleuromutilin derivatives were
synthesized by a click chemistry strategy. Pleuromutilin was
conjugated by different linkers to a nucleobase, nucleoside, or
phenyl group, as a side-chain extension at the C22 position of
pleuromutilin. The linkers were designed on the basis of the
best linker from our first series of pleuromutilin derivatives
following either conformational restriction or an isosteric
methylene to oxygen exchange. The binding of the new
compounds to the Escherichia coli ribosome was investigated by
molecular modeling and chemical footprinting of nucleotide U2506, and it was found that all the derivatives bind to the specific
site and most of them better than pleuromutilin itself. The effect of the side-chain extension was also explored by chemical
footprinting of nucleotide U2585, and the results showed that all the compounds interact with this position to varying degrees.
Derivatives with a conformational restriction of the linker generally had a higher affinity than derivatives with an isosteric
exchange of one of the carbons in the linker with a hydrophilic oxygen. A growth inhibition assay with three different bacterial
strains showed significant activity of several of the new compounds.
side-chain extension at the C14 position can assume distinct
conformations in the binding pocket.4 The detailed interactions
of 23S rRNA and four pleuromutilin derivatives have been
determined by X-ray structures of antibiotic−ribosome
complexes.5−7
INTRODUCTION
■
Few new antibiotics reach the market, and at the same time, the
problem with bacterial resistance to many antibiotics is
increasing. Therefore, there is a need for developing new
antibiotics. Natural products or semisynthetic derivatives of
natural products have so far been the most successful way for
developing new antibiotics. The natural compound pleuro-
mutilin1 1 (Figure 1) has a modest antibacterial activity and
inhibits bacterial protein synthesis by binding to the ribosomes.
Modification of the C14 position of pleuromutilin has led to
some more active compounds, such as retapamulin 2, tiamulin
3, and valnemulin 4 (Figure 1). Retapamulin is the first
antibiotic in the pleuromutilin class to be approved for human
use; it is used in the treatment of topical skin infections.2
Tiamulin and valnemulin are used in veterinary medicine for
pigs and poultry. The interaction of tiamulin and valnemulin
with the bacterial ribosome has been investigated and it was
shown that the two drugs bind to domain V of 23S rRNA in the
peptidyltransferase center (PTC), thereby inhibiting peptide
bond formation by hindering a correct location of the tRNA.3
Chemical footprinting of various pleuromutilin derivatives
suggests a similar mode of binding of the tricyclic mutilin
core of all pleuromutilin derivatives in the PTC but that the
Previous work in our group has led to the synthesis and
analysis of binding of 19 semisynthetic pleuromutilin
derivatives.8 The derivatives were synthesized by use of a
parallel click chemistry strategy, where the C22 hydroxy-group
of pleuromutilin was substituted by an azide group and then
used in a standard Cu(I)-catalyzed alkyne−azide [3 + 2]
cycloaddition (CuAAC reaction)9,10 with a series of 19 terminal
alkynes. The alkynes were linked to different nucleobases,
nucleosides, or a phenyl group with a variation of the linker
length. The binding affinities of the 19 derivatives to Escherichia
coli ribosomes were evaluated by chemical footprinting. This
method shows how well the derivatives protect accessible Us in
23S rRNA against reaction with CMCT [N1-cyclohexyl-N3-(2-
morpholinoethyl)carbodiimide p-toluenesulfonate]. Previous
footprinting on pleuromutilin derivatives shows two clear
Received: September 22, 2011
Published: January 26, 2012
© 2012 American Chemical Society
2067
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
For the present study, we decided to elaborate on 5 by
varying the three-carbon linker, as well as to increase our
knowledge of the binding site by synthesis of other analogues in
combination with further molecular docking studies. Binding to
the E. coli ribosome is investigated by chemical footprinting,
and the potential biological activity is tested by determining the
inhibition on growth of three different bacteria.
RESULTS
■
Chemical Synthesis. A new series of pleuromutilin
derivatives was synthesized based on the design used in our
former study, by the click chemistry approach.8 The derivative
with the higest affinity from our previous series, 5, is used as a
lead compound. The aim was first of all to make conforma-
tional restrictions of the three-carbon linker between the
triazole ring and the nucleobase, as well as to make an isosteric
exchange of one of the carbons in the linker with oxygen in
order to increase the hydrophilicity of the linker. Thus
pleuromutilin 1 was converted to the azide 68,11 and reacted
with a series of alkynes 7−16 by the CuAAC reaction9,10 to give
a series of 10 new pleuromutilin triazole derivatives 17−25 and
27 (Scheme 1). The compounds were chosen on the basis of
lead compound 5, the previous study, and ease of synthesis.
The compounds 17, 19, and 20 were made as conformationally
restricted analogues of 5, and 18 was made in order to compare
the effect of two different nucleobases with the same restricted
linker. Compounds 22−24 were made as more hydrophilic
analogues of 5 and its corresponding thymine and phenyl
counterparts included in our first study.8 The compounds 21
and 25 were made to follow up on the conformational
restriction of a uracil analogue of 5 as well as on extension of
the linker with an oxygen atom. The unsubstituted triazole
compound 27 was included as a reference compound to
investigate the general effect of side-chain extensions at the
triazole. All CuAAC reactions proceeded in medium to good
yields to give the pure target compounds after simple
chromatographic purification. Finally, also a 1,5-disubstituted
1,2,3-triazole isomer of the lead derivative 5 was synthesized,
that is, 28 (Scheme 2). To give this regioisomer, a ruthenium
catalyst12 was applied in the cycloaddition reaction between the
azide derivative of pleuromutilin 6 and the terminal alkyne N9-
(pentyn-5-yl)adenine8 29.
The alkyne building blocks 7−15 used in the click reaction
were synthesized in a few steps from simple nucleobase or
nucleoside precursors. The alkynylated nucleobases with a
conformational restriction in the linker, 7 and 8, were
synthesized by Mitsunobu reactions with 2-penten-4-yn-1-ol
and N6-bis-Boc-adenine 3013 or N3-benzoylthymine 31,14
respectively (Scheme 3), followed by hydrolysis with either
hydrochloric acid or sodium methoxide in methanol to remove
the protecting groups. The alkyne building blocks 9 and 10,
which contain a more bulky aromatic but still conformationally
restricted linker, were also synthesized from 30 (Scheme 4).
The protected nucleobase was used in a copper-mediated
Chan−Lam−Evans-modified Ullmann condensation with
either (3-iodophenyl)boronic acid or (4-iodophenyl)boronic
acid according to a known procedure15 to give compounds 32
and 33.15 These were used in a Sonogashira coupling with
trimethylsilylacetylene to introduce the triple bond and provide
compounds 34 and 35. This was followed by a basic
deprotection of the TMS- and Boc-protecting groups in one
step to give compounds 9 and 10 (Scheme 4). Nucleoside 11
was synthesized from the TBS-protected 2′-C-allyl-2′-deoxyur-
Figure 1. Chemical structure of pleuromutilin and derivatives thereof.
footprints at U2506 and U2585.4 The protection at U2506 is
believed to correlate to binding affinity, while the effect at
U2585 tells whether the extension from the mutilin core is
placed in the neighborhood of U2585. All the derivatives were
found to bind in the PTC, and 11 showed stronger affinity for
the PTC than pleuromutilin itself. Compound 5 (Figure 1) was
the best binding derivative from the series.8 A common feature
for the four best derivatives was a three-carbon linker between
the triazole and the nucleobase. Another trend was that adenine
is favored over thymine or the simple phenyl group.
Footprinting at nucleotide U2585 demonstrated that all 19
derivatives also interact with this position, a position that is
unaffected by pleuromutilin itself. The two best derivatives
from the previous series, that is, 5 and a derivative with a
phenyl instead of adenine, were investigated by molecular
docking. This modeling suggested that the mutilin core binds
to the PTC in the same way as for tiamulin. It also suggested
that the side chain is extended into the peptide exit tunnel and
that a stacking interaction is taking place between the adenine/
phenyl group and U2586 or A2062. Finally, a stacking
interaction between the triazole and nucleobase U2585 was
indicated.
2068
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
Scheme 1. Overview of Click Chemistry Approach and Chemical Structure of the 10 Derivatives of Pleuromutilin and Applied
a
Terminal Alkynes
a
Reagents and conditions: (a) (i) TsCl, DMAP, CH2Cl2; (ii) NaN3, acetone, H2O (52%, two steps).8,11 (b) Sodium ascorbate, CuSO4, t-BuOH,
H2O or THF, H2O; 17 (29%), 18 (45%), 19 (64%), 20 (31%), 21 (62%), 22 (53%), 23 (65%), 24 (71%), 25 (76%). (c) TBAF, THF, 27 (58%, two
steps).
idine derivative 36,16 which was protected at the nucleobase
with a benzoyl group to give compound 37 (Scheme 5). This
was followed by an oxidative cleavage to give the aldehyde 38.
The aldehyde was then converted to an alkyne by use of the
Ohira−Bestmann reagent17,18 and debenzoylated to give 39,
followed by desilylation to give nucleoside 11. The two alkynes
12 and 13, containing a three-atom linker with an oxygen, were
was first deprotected with BCl3 in dichloromethane (DCM),
which gave a mixture of two products, 3-(hydroxymethyl)-1-[2-
(propargyloxy)ethyl]thymine and the target 1-[2-
(propargyloxy)ethyl]thymine. The mixture was treated with
NaOH in THF and water to give the pure alkyne 15.
Compound 27 was synthesized from the cycloaddition between
the azide derivative 6 and trimethylsilylacetylene to give
compound 26, followed by a desilylation with TBAF to give
27 (Scheme 1).
Chemical Footprinting of Pleuromutilin Derivatives in
PTC. The affinities of the 11 new derivatives (17−25, 27, and
28) as well as for the lead derivative from the previous series 5
and the derivatives tiamulin 3 and valnemulin 4 for the binding
site in the PTC of ribosomes from E. coli were examined by
chemical footprinting at nucleotide U2506 as described
synthesized by Vorbruggen coupling to bis(propargyloxy)-
̈
methane19 involving either the N6-formamidine-protected
adenine 4020 or unprotected thymine 41 (Scheme 6). The
benzyl propargyl ether 14 was synthesized according to a
known procedure from benzylbromide and propargyl alcohol.21
Alkyne 15 was synthesized from 3-(benzyloxymethyl)-1-(2-
hydroxyethyl)thymine 42,14,22 which was alkylated with
propargyl bromide to give 43 (Scheme 7). Compound 43
2069
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
a
a
Scheme 2. Synthesis of 1,5-Disubstituted Isomer of 5
Scheme 4. Synthesis of Alkynes 9 and 10
a
Reagents and conditions: (a) CpRuCl(COD), toluene, DMF (25%).
a
Scheme 3. Synthesis of Alkynes 7 and 8
a
Reagents and conditions: (a) (3-Iodophenyl)boronic acid or (4-
iodophenyl)boronic acid, Cu(OAc)2, DMF, Et3N, 3 Å MS; 32 (42%),
33 (51%). (b) TMS-acetylene, Pd(PPh3)2Cl2, CuI, DMF, Et3N; 34
(65%), 35 (86%). (c) KOH, MeOH; 9 (39%), 10 (83%)
side chains can move to accommodate the variable sizes and
chemical groups of the drugs.6 Ribosomes were incubated with
and without the drugs and exposed to CMCT that reacts with
accessible uridines (including U2506 in 23S rRNA). The
difference in reaction with and without drug can then be
monitored by primer extension with reverse transcriptase that
stops at CMCT-modified Us. The extension product is
visualized by gel electrophoresis and the resulting bands are
quantified by image analysis to obtain an assessment of the
protection against CMCT afforded by the drug. The results are
shown in Table 1 for two concentrations of the drug and are
ordered by protection at U2506 at 5 μM concentrations, with
1.00 representing no protection (no binding) and 0
representing total protection. All the derivatives were seen to
bind in the PTC and eight of them apparently bound better
than pleuromutilin itself. It should be noted that the
footprinting method is qualitatively good and very specific
but not very quantitatively precise. The best derivative 19
showed as good an affinity to the PTC as the two drugs
valnemulin 4 and tiamulin 3. In general, the conformationally
restricted derivatives 17−20 show a higher affinity for the PTC
than the derivatives with an isosteric exchange of one of the
carbons in the linker with a more hydrophilic oxygen 22−25.
An exception is compound 21, which has a conformationally
restricted linker but displays one of the lowest affinities for the
PTC. Again, the trend from the first series8 of increased binding
affinity for adenine derivatives as compared to thymine or
phenyl derivatives is observed since 17 binds more strongly
a
Reagents and conditions: (a) (i) 2-Penten-4-yn-1-ol, DIAD, PPh3,
THF; (ii) 3 M HCl, MeOH (84% estimated, two steps). (b) (i) 2-
Penten-4-yn-1-ol, DEAD, PPh3, toluene, DCM; (ii) NaOMe, MeOH
(17%, two steps).
previously.8 The rationale for using this protection as an affinity
measure came from previous footprinting experiments4,8 and
from the fact that crystal structures have revealed that the
position of the tricyclic mutilin core moiety is fixed in a tight
binding pocket including position U2506 in 23S rRNA.5−7 In
contrast, the RNA moieties interacting with the pleuromutilin
2070
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
a
a
Scheme 5. Synthesis of Nucleoside 11
Scheme 6. Synthesis of Alkynes 12 and 13
a
Reagents and conditions: (a) BzCl, pyridine (94%). (b) (i) OsO4,
NMO, acetone, H2O; (ii) NaIO4, dioxane, H2O (49%). (c) (i)
Dimethyl-2-oxopropylphosphonate, p-TsN3, K2CO3, CH3OH,
CH3CN; (ii) methanolic ammonia (46%). (d) TBAF, THF (57%).
a
Reagents and conditions: (a) (i) Bis(propargyloxy)methane, BSA,
TMS triflate, ClCH2CH2Cl, CH3CN; (ii) NH3(aq) (14%, two steps).
(b) Bis(propargyloxy)methane, BSA, TMS triflate, ClCH2CH2Cl,
CH3CN (27%).
than 18 and 22 binds more strongly than 23 and 24. The “1,5-
isomer” of 5, that is, compound 28, belongs to the strongest
binding analogues like the conformationally restricted ana-
logues 17 and 19. The derivative with the lowest affinity for the
PTC is the reference compound 27, which lacks a side-chain
extension at the triazole. Probably, the side-chain extension at
the triazole provides better binding affinity to the PTC by
making more interactions with the rRNA.
a
Scheme 7. Synthesis of Alkyne 15
Besides U2506, only U2585 has shown significant changes in
accessibility to CMCT upon binding of pleuromutilin
derivatives.4,8 The accessibility of nucleotide U2585 in 23S
rRNA, which is positioned relatively close to the binding site of
the tricyclic core of pleuromutilin, was therefore also
investigated. This protection is not believed to reflect the
strength of antibiotic binding (as for position U2506) but
merely reflects a closeness of the drug side chain to this
position.4,8 This might thus provide information about whether
the side chains of the derivatives are positioned in the same
orientation in the PTC. The results in Table 1 show that all the
novel derivatives protect U2585 to some extent and all of them
provided higher protection than pleuromutilin and (except 27)
tiamulin. In contrast, only derivative 19 gave better protection
at nucleotide U2585 than valnemulin.
a
Reagents and conditions: (a) NaH, propargyl bromide, THF (82%).
(b) (i) BCl3, DCM; (ii) NaOH, THF, H2O (13%, two steps).
Docking Studies. Three of the pleuromutilin derivatives
with good binding affinities (5, 19 and 28), as well as the
derivative showing the smallest binding affinity (27), were
chosen for a molecular docking investigation. A PTC ribosome
model was constructed that consists of all residues within 30 Å
from the PTC binding site. Due to the lack of crystal structures
of E. coli in complex with pleuromutilin derivatives, this model
was based on the X-ray structure of Deinococcus radiodurans in
complex with tiamulin.5 The docking experiments were
performed with Glide;23−26 details for the setup of the docking
experiments are given in the Experimental Section. Redocking
of tiamulin 3 into this PTC model placed the drug as in the X-
ray structure. The docking results for the four drugs revealed a
similar binding pattern as documented for tiamulin with the
2071
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
Table 1. Footprinting and MIC Values for Pleuromutilin and Derivatives Thereof
b
degree of CMCT mod. relative to positive control
MIC μg/mL
a
a
a
a
U2506, 5 μM
U2506, 50 μM
U2585, 5 μM
U2585, 50 μM
B. subtilis 168
L. innocua
E. coli AS19
Val 4
19
0.05
0.06
0.07
0.09
0.10
0.13
0.18
0.18
0.21
0.23
0.28
0.28
0.30
0.31
0.48
0.04
0.06
0.03
0.04
0.05
0.05
0.08
0.06
0.06
0.07
0.08
0.08
0.08
0.09
0.18
0.15
0.10
0.54
0.19
0.29
0.23
0.40
0.39
0.37
0.37
0.89
0.43
0.50
0.46
0.66
0.16
0.07
0.50
0.16
0.16
0.17
0.22
0.21
0.22
0.22
0.80
0.24
0.41
0.26
0.42
2
8
1
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
>32
<0.125
1
Tia 3
28
>32
>32
16
0.5−1
16−32
8
17
5
>32
>32
>32
>32
>32
32
8−16
>32
20
18
≥32
>32
22
23
>32
Pleuro 1
25
1−2
≥32
>32
>32
>32
>32
>32
24
21
>32
27
16−32
a
Selected positions in E. coli 23S rRNA affected by CMCT modification; U2506 at the binding pocket of the mutilin core and U2585 near the
variations in pleuromutilin conjugations. The numbers represent the relative CMCT accesibility in the presence of the drug, in 5 or 50 μM
concentrations, compared to in the absence of the drug. For example, 0.10 means that the intensity of the band on the gel in the presence of the drug
b
is only 10% of the intensity of the corresponding band from samples without the drug. Minimal inhibitory concentration (MIC) of the derivatives
determined in a plate assay with three different bacteria, E. coli AS19, Bacillus subtilis 168, and Listeria innocua.
Figure 2. Results of the docking of 5 (blue), 19 (purple), 27 (green), and 28 (yellow) into the PTC model binding site. Residue numbers shown are
according to E. coli 23S RNA numbering.
2072
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
pleuromutilin cores placed close to the U2506 position as
presented in Figure 2, which shows a superposition of the four
docked drugs. Based on the predicted binding affinities
(docking scores), the drugs were found to bind in the sequence
28 > 5 > 19 ≫ 27, which is in agreement with the footprinting
results. Only very small deviations (0.2 kcal/mol) were
observed in the binding affinities between the drugs 28, 5,
and 19; that is, these three drugs are found to bind with similar
strength. In contrast, the scores of the docking procedure
clearly distinguish between drugs with high/low binding
affinities according to the footprinting results. As observed
previously,8 the results of these docking experiments give
indications for a stacking interaction between the triazoles of
the drugs and U2585. This stacking interaction seems more
pronounced for the more flexible drugs (5, 27, and 28) than for
the very restricted derivative 19 (Figure 2). However, 19
demonstrates the strongest overall binding from the foot-
printing, and docking of natural pleuromutilin resulted in
almost the same docking score as for 27, indicating that this
stacking interaction between the triazole and U2585 is of minor
importance for the overall binding of the drugs. An interaction
between the side-chain adenines of 5, 19, and 28 with A2062
appears to be unlikely on the basis of the docking, and only
compound 19 seems partly oriented toward a stacking
interaction. However, A2062 is known to be very flexible and
the docking might not model this flexibility very well.
Potential Growth Inhibition of Bacteria by Pleuro-
mutilin Derivatives. Although binding affinity for the target is
an important parameter for the efficacy of a potential drug, the
ultimate criterion is activity against bacterial cells, and the
ability to perform in the cells is vital. As a first preliminary step
to investigate the biological relevance of the 11 derivatives of
pleuromutilin, the minimal inhibitory concentration (MIC)
values were determined for three different bacterial strains, a
drug-hypersensitive E. coli AS19 (Gram-negative), Bacillus
subtilis 168 (Gram-positive), and Listeria innocua (Gram-
positive). These are all nonpathogenic strains but if a
compound does not have any effect on any of these strains,
it is not likely to be useful as a general antibiotic, although the
possibility exists that it could be effective against specific strains.
The MICs are listed in Table 1, and due to the availability of
the synthesized drugs, only relatively low concentrations were
tested. L. innocua seems very resistant to pleuromutilins in
general, as only valnemulin inhibited growth at the concen-
tration tested. For B. subtilis 168, valnemulin was also the most
potent compound, while compound 19 was second best,
followed by 17 and then pleuromutilin. E. coli AS19 was also
most sensitive to valnemulin, followed by tiamulin, 19,
pleuromutilin, 17, 5, and then 27 and 28. Some of the new
compounds are thus comparable in potency to tiamulin, which
is the most successful of the commercial pleuromutilin products
if measured by usage. For further drug development, promising
candidates should be tested against pathogenic bacteria such as
Staphylococcus and Streptococcus.
with a conformationally restricted linker demonstrate better
binding affinities to the PTC than the derivatives with oxygen
in the linker. Compound 19 is best in both the footprinting and
MIC studies, which is also in good agreement with the present
docking results. Also, 17 and 28 show a slight improvement in
binding affinity as compared to 5, and this might reflect a small
change in the positioning of the adenine in the binding site.
The docking studies indicate that the pleuromutilin core is
located close to the U2506 position and that the triazole moiety
stacks with the position U2585. However, this stacking
interaction seems to be of minor importance for the overall
binding of the derivatives, since 27 demonstrates a low affinity
and since the stacking according to the docking study seems
less pronounced for the best binder 19 as compared to other
derivatives. Nevertheless, the triazole might have an important
role in directing the side chain toward a suitable position.
The relatively low variation in affinity observed is not
surprising due to the fact that the designs are based on a
reasonably good binder in the form of the pleuromutilin core.
Nevertheless, the variations in chemical structure may be
important when it comes to in vivo function. Compound 19 is
the best derivative in both our assays (Table 1) and is hereby
the superior candidate for further studies.
CONCLUSION
■
A series of 11 new pleuromutilin derivatives has been efficiently
synthesized by a parallel click chemistry strategy. The alkyne
building blocks were synthesized in relatively few steps, and in
combination with the click chemistry strategy, this gives easy
access to new active derivatives of pleuromutilin. The chemical
footprinting showed that all the new derivatives bind in the
PTC in the ribosome and seven of these bind better than
pleuromutilin itself. The docking reflected the footprinting
results, indicating the potential of this method in future design.
Derivative 19 showed binding affinity to the PTC at a level of
the known drugs valnemulin and tiamulin and is also the most
potent compound in the new series on the basis of MIC
experiments. The present close interplay between a straightfor-
ward and reliable docking procedure and a convenient
semisynthetic strategy opens the door for fast development of
new potent antibiotics based on pleuromutilin.
EXPERIMENTAL SECTION
■
General. All reactions with anhydrous solvents were performed
under an atmosphere of either nitrogen or argon. Column
chromatography was performed either as flash column chromatog-
raphy or as standard column chromatography carried out on glass
columns with silica gel 60 (particle size 0.040−0.063 mm.). Microwave
heated reactions were preformed with an Emrys Creator. NMR spectra
were recorded at 300 or 400 MHz for 1H NMR and at 75 or 101 MHz
for 13C NMR. The chemical shift values (δ) are reported in parts per
million (ppm) relative to tetramethylsilane as internal standard. Two-
dimensional (2D) spectra were used to assign NMR spectra, and
standard numbering of pleuromutilin and nucleosides was applied.
Electrospray ionization mass spectra (ESI-MS) were recorded in
positive-ion mode.
DISCUSSION
■
General Procedure for Synthesizing Pleuromutilin Deriva-
tives. The pleuromutilin azide 6 (0.07−0.20 mmol), the alkyne
derivative (0.07−0.21 mmol), sodium ascorbate (0.01−0.03 mmol),
and CuSO4·5H2O (0.01−0.04 mmol) were dissolved in t-BuOH/H2O
(2−3 mL, 1:1 v/v) in a microwave vial. The vial was sealed and heated
with magnetic stirring in the microwave reactor at 110 °C for 30 min.
The reaction mixture was concentrated under reduced pressure, and
the residue was purified by column chromatography (0−10% CH3OH
The results presented herein are in line with the outcome from
the first series8 that nucleobases conjugated to the C22 position
of pleuromutilin through a triazole linkage are well adopted
within the PTC binding pocket. Derivatives conjugated to the
nucleobase adenine give a better binding affinity than the
derivatives conjugated to thymine, which again is superior to a
simple phenyl group. Another general trend is that compounds
2073
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
(10% CH3OH in CH2Cl2). 1H NMR (CDCl3) δ 8.44 [1H, s, H2(A)],
8.21 (1H, s, Ph), 8.18 [1H, s, H8(A)], 7.99 [1H, s, H5(triazole)], 7.89
(1H, d, J = 7.8 Hz, Ph), 7.75 (1H, d, J = 8.9 Hz, Ph), 7,63 (1H, m, Ph),
6.42 (1H, dd, Jcis = 11.3 Hz, Jtrans = 17.4 Hz, H19), 5.98 [2H, br s,
NH2(A)], 5.84 (1H, d, J = 8.5 Hz, H14), 5.34 (1H, d, Jcis = 11.3 Hz,
H20), 5.24−5.09 (3H, m, H20 and H22), 3.40−3,34 (1H, m, H11),
2.33−2.08 (5H, m, H2, H4, H10, and H13), 1.79−1.10 (9H, m, H1,
H6, H7, H8, H13, and 11-OH), 1.38 (3H, s, H15), 1.19 (3H, s, H18),
0.89 (3H, d, J = 7.0 Hz, H17), 0.74 (3H, d, J = 7.0 Hz, H16). 13C
NMR (CDCl3) δ 216.7 (C3), 165.1 (C21), 156.0 [C6(A)], 153.9
[C2(A)], 150.1 [C4(A)], 147.2 [C4(triazole)], 139.6 [C8(A)], 138.7
(C19), 135.6, 132.3, 130.6, 125.5, 123.3 (Ph), 121.7 [C5(triazole)],
120.7 (Ph), 120.2 [C5(A)], 117.7 (C20), 74.7 (C11), 71.3 (C14), 58.1
(C4), 51.8 (C22), 45.5 (C9), 44.8 (C13), 44.1 (C12), 42.0 (C5), 36.7
(C6), 36.2 (C10), 34.5 (C2), 30.4 (C8), 26.9 (C7), 26.6 (C18), 24.9
(C1), 17.0 (C16), 14.8 (C15), 11.6 (C17). HRMS (ESI) m/z
661.3211 [M + Na]+, C35H42N8O4Na calcd 661.3221.
Synthesis of 22-{4-[4-(Adenin-9-yl)phenyl]-1,2,3-triazol-1-yl}-22-
deoxypleuromutilin (20). The general procedure was applied with
compound 6 (68 mg. 0.17 mmol), N9-(4-ethynylphenyl)adenine 10
(31 mg, 0.13 mmol), sodium ascorbate (3.4 mg, 17 μmol), and
CuSO4·5H2O (4.2 mg, 17 μmol) in t-BuOH/H2O (2 mL, 1:1 v/v).
Yield 26 mg (31%); Rf 0.3 (10% CH3OH in CH2Cl2). 1H NMR
(CDCl3) δ 8.45 [1H, s, H2(A)], 8.14 [1H, s, H8(A)], 8.06−8.04 (2H,
m, Ph), 7.94 [1H, s, H5(triazole)], 7.82−7.80 (2H, m, Ph), 6.44 (1H,
dd, Jcis = 10.9 Hz, Jtrans = 17.4 Hz, H19), 5.85 (1H, d, J = 8.4 Hz, H14),
5.62 [2H, br s, NH2(A)], 5.36 (1H, dd, Jgem = 1.5 Hz, Jcis = 10.9 Hz,
H20), 5.26−5.09 (3H, m, H20 and H22) 3.37 (1H, dd, J = 6.4 Hz, J =
10.4 Hz, H11), 2.36−2.09 (5H, m, H2, H4, H10, and H13), 1.80−1.11
(9H, m, H1, H6, H7, H8, H13, and 11-OH), 1.38 (3H, s, H15), 1.20
(3H, s, H18), 0.89 (3H, d, J = 6.8 Hz, H17), 0.75 (3H, d, J = 7.2 Hz,
H16). 13C NMR (CDCl3) δ 216.8 (C3), 165.2 (C21), 155.9 [C6(A)],
154.0 [C2(A)], 150.2 [C4(A)], 147.4 [C4(triazole)], 139.6 [C8(A)],
138.8 (C19), 134.8, 130.5, 127.4, 124.0 (Ph), 121.4 [C5(triazole)],
120.4 [C5(A)], 117.9 (C20), 74.7 (C11), 71.4 (C14), 58.2 (C4), 51.9
(C22), 45.6 (C9), 44.9 (C13), 44.2 (C12), 42.1 (C5), 36.7 (C6), 36.3
(C10), 34.6 (C2), 30.5 (C8), 27.0 (C7), 26.6 (C18), 25.0 (C1), 17.1
(C16), 14.8 (C15), 11.7 (C17). HRMS (ESI) m/z 639.3398 [M +
H]+, C35H43N8O4 calcd 639.3401.
in CH2Cl2) to give the product as a white foam. Purities ≥95% were
secured by HPLC.
Synthesis of 22-{4-[(E)-3-(Adenin-9-yl)propen-1-yl]-1,2,3-triazol-
1-yl}-22-deoxypleuromutilin (17). The general procedure was applied
with compound 6 (74 mg. 0.18 mmol), N9-[(E)-pent-2-en-4-yn-1-
yl]adenine 7 (30 mg, 0.15 mmol), sodium ascorbate (3.8 mg, 19
μmol), and CuSO4·5H2O (5.1 mg, 20 μmol) in t-BuOH/H2O (2 mL,
1
1:1 v/v). Yield 26 mg (29%); Rf 0.4 (10% CH3OH in CH2Cl2). H
NMR (CDCl3) δ 8.39 [1H, s, H2(A)], 7.87 [1H, s, H8(A)], 7.57 [1H,
s, H5(triazole)], 6.66 (1H, dt, J = 5.8, 16.0 Hz, N9-CH2CH), 6.53
(1H, d, J = 16.0 Hz, N9-CH2CHCH), 6.39 (1H, dd, Jcis = 11.1 Hz, Jtrans
= 17.4 Hz, H19), 5.81 (1H, d, J = 8.5 Hz, H14), 5.75 [2H, s, NH2(A)],
5.32 (1H, d, Jcis = 11.1 Hz, H20), 5.20 (1H, d, Jtrans = 17.4 Hz, H20),
5.11−4.97 (4H, m, H22, N9-CH2), 3.39−3.32 (1H, m, Hz, H11),
2.30−2.06 (5H, m, H2, H4, H10, and H13), 1.78−1.08 (9H, m, H1,
H6, H7, H8, H13, and 11-OH), 1.35 (3H, s, H15), 1.17 (3H, s, H18),
0.88 (3H, d, J = 7.0 Hz, H17), 0.70 (3H, d, J = 7.0 Hz, H16). 13C
NMR (CDCl3) δ 216.7 (C3), 165.1 (C21), 155.6 [C6(A)], 153.4
[C2(A)], 150.2 [C4(A)], 145.0 [C4(triazole)], 140.4 [C8(A)], 138.7
(C19), 125.7 (N9-CH2CH), 122.2, 122.1 [C5(triazole), N9-
CH2CHCH) 119.7 [C5(A)], 117.8 (C20), 74.7 (C11), 71.2 (C14),
58.1 (C4), 51.6 (C22), 45.5 (C9), 45.2 (N9-CH2), 44.8 (C13), 44.1
(C12), 42.0 (C5), 36.6 (C6), 36.2 (C10), 34.5 (C2), 30.4 (C8), 26.9
(C7), 26.5 (C18), 24.9 (C1), 16.9 (C16), 14.8 (C15), 11.6 (C17).
HRMS (ESI) m/z 625.3221 [M + Na]+, C32H42N8O4Na calcd
625.3221.
Synthesis of 22-(4-{[2R-(Uracil-1-yl)-4S-hydroxy-5R-
hydroxymethyl]tetrahydrofuran-3R-ylmethyl}-1,2,3-triazol-1-yl)-22-
deoxypleutomutilin (21). The general procedure was applied with
compound 6 (64 mg. 0.16 mmol), nucleoside 11 (35 mg, 0.13 mmol),
sodium ascorbate (3.3 mg, 17 μmol), and CuSO4·5H2O (4.3 mg, 17
μmol) in t-BuOH/H2O (2 mL, 1:1 v/v) was applied. Yield 55 mg
Synthesis of 22-{4-[(E)-3-(Thymin-1-yl)propen-1-yl]-1,2,3-triazol-
1-yl}-22-deoxypleuromutilin (18). The general procedure was slightly
modified with compound 6 (54 mg, 0.13 mmol), N1-[(E)-pent-2-en-
4-ynyl]thymine 8 (33 mg, 0.17 mmol), sodium ascorbate (5.3 mg, 27
μmol), and CuSO4·5H2O (4.5 mg, 18 μmol) in t-BuOH/H2O (3.0
mL, 1:1 v/v) and conventional heating at 80 °C for 21.5 h. Yield 35
1
(62%); Rf 0.5 (10% CH3OH in CH2Cl2). H NMR (DMSO-d6) δ
1
11.22 [1H, br s, NH(U)], 7.81 [1H, d, J = 8.1 Hz, H6(U)], 7.78 [1H,
s, H5(triazole)], 6.12 (1H, dd, Jcis = 11.2 Hz, Jtrans = 17.8 Hz, H19),
5.99 (1H, d, J = 9.1 Hz, H1′), 5.62 [1H, d, J = 8.1 Hz, H5(U)], 5.56
(1H, d, J = 8.3 Hz, H14), 5.39 (1H, s, H3′-OH) 5.29−5.03 (5H, m,
H22, H20, and H5′-OH), 4.53 (1H, d, J = 6.0 Hz, H11-OH), 4.12−
4.09 (1H, m, H3′), 3.90 (1H, t, J = 3.7 Hz, H4′) 3.54 (2H, d, J = 3.0
Hz, H5′), 3.42 (1H, t, J = 6.0 Hz, H11), 2.99−2.91 (1H, m, H2′-CH2),
2.62−2.53 (2H, m, H2′, H2′-CH2), 2.40 (1H, s, H4), 2.23−1.99 (4H,
m, H2, H10 and H13), 1.67−1.00 (8H, m, H1, H6, H7, H8, and H13),
1.24 (3H, s, H15), 1.07 (3H, s, H18), 0.81 (3H, d, J = 7.0 Hz, H17),
0.62 (3H, d, J = 7.0 Hz, H16). 13C NMR (DMSO-d6) δ 216.9 (C3),
165.6 (C21), 162.9 [C4(U)], 150.8 [C2(U)], 144.4 [C4(triazole)],
140.7, 140.4 [C6(U), C19], 123.8 [C5(triazole)], 115.4 (C20), 102.2
[C5(U)], 87.4 (C4′), 87.0 (C1′), 72.5 (C11), 71.8 (C3′), 70.4 (C14),
61.8 (C5′), 57.1 (C4), 50.9 (C22), 47.6 (C2′), 44.9 (C9), 44.1 (C12),
43.3 (C13), 41.5 (C5), 36.4 (C6), 36.2 (C10), 33.9 (C2), 30.0 (C8),
28.5 (C18), 26.5 (C7), 24.4 (C1), 19.7 (C2′-CH2), 16.1 (C16), 14.2
(C15), 11.5 (C17). HRMS (ESI) m/z 692.3258 [M + Na]+,
C34H47N5O9Na calcd 692.3266.
mg (45%); Rf 0.5 (10% CH3OH in CH2Cl2). H NMR (CDCl3) δ
9.19 [1H, s, NH(T)], 7.64 [1H, s, H5(triazole)], 7.05 [1H, s, H6(T)],
6.62 (1H, d, J = 16.0 Hz, N1-CH2CHCH), 6.49−6.36 (2H, m, N1-
CH2CH, H19), 5.81 (1H, d, J = 8.5 Hz, H14), 5.32 (1H, d, Jcis = 12.0
Hz, H20), 5.21 (1H, d, Jtrans = 17.4 Hz, H20), 5.11, 5.03 (2H, AB, J =
17.6 Hz, H22), 4.49 (2H, d, J = 6.1 Hz, N1-CH2), 3.37 (1H, dd, J =
6.5, 10.3 Hz, H11), 2.31−2.07 (5H, m, H2, H4, H10, and H13), 1.93
[3H, s, CH3(T)], 1.79−1.09 (9H, m, H1, H6, H7, H8, H13, and 11-
OH), 1.36 (3H, s, H15), 1.18 (3H, s, H18), 0.88 (3H, d, J = 6.9 Hz,
H17), 0.71 (3H, d, J = 7.0 Hz, H16). 13C NMR (CDCl3) δ 216.8
(C3), 165.1, 164.3 [C21, C4(T)], 150.9 [C2(T)], 145.1 [C4-
(triazole)], 139.8 [C6(T)], 138.7 (C19), 125.5 (N1-CH2CH), 122.6,
122.2 [N1-CH2CHCH, C5(triazole)], 117.7 (C20), 111.3 [C5(T)],
74.6 (C11), 71.2 (C14), 58.1 (C4), 51.6 (C22), 49.3 (N1-CH2), 45.5
(C9), 44.8 (C13), 44.1 (C12), 41.9 (C5), 36.6 (C6), 36.1 (C10), 34.5
(C2), 30.4 (C8), 26.9 (C7), 26.5 (C18), 24.9 (C1), 16.9 (C16), 14.7
(C15), 12.5 [CH3(T)], 11.6 (C17). HRMS (ESI) m/z 616.3114 [M +
Na]+, C32H43N5O6Na calcd 616.3106.
Synthesis of 22-{4-[3-(Adenin-9-yl)phenyl]-1,2,3-triazol-1-yl}-22-
deoxypleuromutilin (19). The general procedure was slightly modified
with compound 6 (28 mg, 0.07 mmol), N9-(3-ethynylphenyl)adenine
9 (16 mg, 0.07 mmol), sodium ascorbate (1.5 mg, 7.6 μmol), and
CuSO4·5H2O (2.1 mg, 8.4 μmol) in t-BuOH/H2O (2.5 mL, 1:1 v/v)
and conventional heating at 80 °C for 18 h. Yield 20 mg (64%); Rf 0.2
Synthesis of 22-[4-(Adenin-9-ylmethoxymethyl)-1,2,3-triazol-1-
yl]-22-deoxypleuromutilin (22). The general procedure was slightly
modified with compound 6 (61 mg, 0.15 mmol), compound 12 (38
mg, 0.19 mmol), sodium ascorbate (4.0 mg, 20 μmol), and
CuSO4·5H2O (10 mg, 40 μmol) in t-BuOH/H2O (3.0 mL, 1:1 v/v)
2074
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
and conventional heating at 80 °C for 20.5 h. Yield 49 mg (53%); Rf
H19), 5.81 (1H, d, J = 8.5 Hz, H14), 5.32 (1H, dd, Jgem = 1.4 Hz, Jcis =
1
0.1 (10% CH3OH in CH2Cl2). H NMR (CDCl3) δ 8.37 [1H, s,
11.1 Hz, H20), 5.21 (1H, dd, Jgem = 1.4 Hz, Jtrans = 17.4 Hz, H20),
5.12, 5.05 (2H, AB, JAB = 17.6 Hz, H22), 4.66 (2H, s, N1-
CH2CH2OCH2), 3.89 (2H, t, J = 4.9 Hz, N1-CH2CH2), 3.76 (2H, t, J
= 4.9 Hz, N1-CH2) 3.37 (1H, dd, J = 6.5, 10.4 Hz, H11), 2.31−2.06
(5H, m, H2, H4, H10, and H13), 1.89 [3H, d, J = 1.2 Hz, CH3(T)],
1.79−1.09 (9H, m, H1, H6, H7, H8, H13, and 11-OH), 1.35 (3H, s,
H15), 1.18 (3H, s, H18), 0.88 (3H, d, J = 7.0 Hz, H17), 0.70 (3H, d, J
= 7.1 Hz, H16). 13C NMR (CDCl3) δ 216.8 (C3), 165.2, 164.4 [C21,
C4(T)], 151.0 [C2(T)], 144.9 [C4(triazole)], 141.8 [C6(T)], 138.7
(C19), 124.0 [C5(triazole)], 117.7 (C20), 110.0 [C5(T)], 74.6 (C11),
71.2 (C14), 68.2 (N1-CH2), 64.3 (N1-CH2CH2OCH2), 58.0 (C4),
51.6 (C22), 48.3 (N1-CH2CH2), 45.5 (C9), 44.7 (C13), 44.1 (C12),
41.9 (C5), 36.6 (C6), 36.1 (C10), 34.5 (C2), 30.4 (C8), 26.9 (C7),
26.5 (C18), 24.9 (C1), 16.9 (C16), 14.7 (C15), 12.3 [CH3(T)], 11.6
(C17). HRMS (ESI) m/z 634.3189 [M + Na]+, C32H45N5O7Na calcd
634.3212.
H2(A)], 8.21 [1H, s H8(A)], 7.82 [1H, s, H5(triazole)], 6.37 (1H, dd,
Jcis = 11.2 Hz, Jtrans = 17.2 Hz, H19), 6.15 [2H, br s, NH2(A)], 5.78
(1H, d, J = 7.6 Hz, H14), 5.67 (2H, s, N9-CH2O), 5.30−5.07 (4H, m,
H20 and H22), 4.74 (2H, s, N9-CH2OCH2), 3.40−3.39 (1H, m,
H11), 2.25−2.05 (5H, m, H2, H4, H10, and H13), 1.76−1.07 (9H, m
H1, H6, H7, H8, H13, and 11-OH), 1.32 (3H, s, H15), 1.15 (3H, s,
H18), 0.87 (3H, d, J = 6.4 Hz, H17), 0.67 (3H, d, J = 6.8 Hz, H16).
13C NMR (CDCl3) δ 217.0 (C3), 165.2 (C21), 161.0 [C6(A)], 153.8
[C2(A)], 151.8 [C4(A)], 145.3 [C8(A)], 142.7 [C4(triazole)], 138.9
(C19), 125.4 [C5(triazole)], 117.6 (C20), 111.4 [C5(A)], 74.9 (N9-
CH2O), 74.6 (C11), 71.4 (C14), 60.8 (N9-CH2OCH2), 58.1 (C4),
51.8 (C22), 45.6 (C9), 44.8 (C13), 44.2 (C12), 42.0 (C5), 36.7 (C6),
36.2 (C10), 34.6 (C2), 30.5 (C8), 26.9 (C7), 26.7 (C18), 24.9 (C1),
17.0 (C16), 14.6 (C15), 11.7 (C17). HRMS (ESI) m/z 629.3168 [M
+ Na]+, C31H42N8O5Na calcd 629.3171.
Synthesis of 22-(1,2,3-Triazol-1-yl)-22-deoxypleuromutilin (27).
Azide 6 (50 mg. 0.12 mmol), sodium ascorbate (4.6 mg, 23 μmol),
and CuSO4·5H2O (5.0 mg, 20 μmol) were dissolved in t-BuOH/H2O
(4 mL, 1:1 v/v), and the reaction mixture was degassed for 10 min
with argon. (Trimethylsilyl)acetylene 16 (0.026 mL, 0.19 mmol) was
added and the reaction mixture was heated at 85 °C for 21 h and then
concentrated under reduced pressure. The residue was purified by
column chromatography (0−1.5% CH3OH in CH2Cl2), which gave a
mixture of compounds 26 and 27. The mixture was dissolved in
anhydrous THF (2 mL) and the solution was degassed for 10 min
with argon followed by addition of TBAF (0.05 mL, 0.05 mmol, 1 M
in THF). The reaction mixture was stirred at RT for 18 h, followed by
partitioning in CH2Cl2/10% NaHCO3 (10 mL, 2:1 v/v). The aqueous
phase was extracted with CH2Cl2 (2 × 5 mL). The organic phases
were collected and concentrated under reduced pressure. The residue
was purified by column chromatography (0−1.5% CH3OH in
CH2Cl2). The product was isolated as a white foam: yield 31 mg,
58%; Rf 0.2 (5% CH3OH in CH2Cl2). 1H NMR (CDCl3) δ 7.76 [1H,
s, CH(triazole)], 7.67 [1H, s, CH(triazole)], 6.41 (1H, dd, Jcis = 11.0
Hz, Jtrans = 17.4 Hz, H19), 5.82 (1H, d, J = 8.5 Hz, H14), 5.33 (1H, dd,
Synthesis of 22-[4-(Thymin-1-ylmethoxymethyl)-1,2,3-triazol-1-
yl]-22-deoxypleuromutilin (23). The general procedure was applied
with compound 6 (80 mg, 0.20 mmol), 1-(propargyloxymethyl)-
thymine 13 (30 mg, 0.15 mmol), sodium ascorbate (3.4 mg, 17 μmol),
and CuSO4·5H2O (5.1 mg, 20 μmol) in t-BuOH/H2O (2 mL, 1:1 v/
1
v). Yield: 60 mg, 65%; Rf 0.6 (10% CH3OH in CH2Cl2). H NMR
(CDCl3) δ 9.73 [1H, br s, NH(T)], 7.92 [1H, s, H5(triazole)], 7.22
[1H, d, J = 1.3 Hz, H6(T)], 6.40 (1H, dd, Jcis = 11.1 Hz, Jtrans = 17.4
Hz, H19), 5.81 (1H, d, J = 8.5 Hz, H14), 5.32 (1H, dd, Jgem = 1.4 Hz,
Jcis = 11.1 Hz, H20), 5.22−5.08 (5H, m, H20, N1-CH2O, H22), 4.74
(2H, s, N1-CH2OCH2), 3.36 (1H, dd, J = 6.5, 10.3 Hz, H11), 2.29−
2.02 (5H, m, H2, H4, H10, and H13), 1.94 [3H, d, J = 1.3 Hz,
CH3(T)], 1.78−1.08 (9H, m, H1, H6, H7, H8, H13, and 11-OH),
1.32 (3H, s, H15), 1.16 (3H, s, H18), 0.87 (3H, d, J = 7.0 Hz, H17),
0.70 (3H, d, J = 7.0 Hz, H16). 13C NMR (CDCl3) δ 216.8 (C3),
165.2, 164.1 [C21, C4(T)], 152.0 [C2(T)], 143.5 [C4(triazole)],
139.4 [C6(T)], 138.8 (C19), 125.5 [C5(triazole)], 117.6 (C20), 112.2
[C5(T)], 75.1 (N1-CH2O), 74.6 (C11), 71.0 (C14), 61.5 (N1-
CH2OCH2), 58.1 (C4), 51.6 (C22), 45.5 (C9), 44.7 (C13), 44.1
(C12), 41.9 (C5), 36.6 (C6), 36.1 (C10), 34.5 (C2), 30.4 (C8), 26.9
(C7), 26.5 (C18), 24.9 (C1), 16.9 (C16), 14.7 (C15), 12.4 [CH3(T)],
11.6 (C17). HRMS (ESI) m/z 620.3052 [M + Na]+, C31H43N5O7Na
calcd 620.3055.
Synthesis of 22-[4-(Benzyloxymethyl)-1,2,3-triazol-1-yl]-22-deox-
ypleuromutilin (24). The general procedure was slightly modified with
compound 6 (56 mg, 0.14 mmol), benzyl propargyl ether 14 (31 mg,
0.21 mmol), sodium ascorbate (5.9 mg, 30 μmol), and CuSO4·5H2O
(4.5 mg, 18 μmol) in THF/H2O (3.5 mL, 1:1 v/v), stirring at RT for
20 h, and flash column chromatography (0−100% EtOAc in CH2Cl2).
Yield 58 mg, 71%; Rf 0.2 (10% EtOAc in CH2Cl2). 1H NMR (CDCl3)
δ 7.65 [1H, s, H5(triazole)], 7.36−7.27 (H5, m, Ph), 6.41(1H, dd, Jcis
= 11.0 Hz, Jtrans = 17.4 Hz, H19), 5.82 (1H, d, J = 8.5 Hz, H14), 5.34
(1H, d, Jcis = 11.0 Hz, H20), 5.21 (1H, d, Jtrans = 17.4 Hz, H20), 5.10,
5.03 (2H, AB, J = 17.5 Hz, H22), 4.72 (2H, s, PhCH2O), 4.61 (2H, s,
PhCH2OCH2), 3.35 (1H, dd, J = 6.6, 10.6 Hz, H11), 2.30−2.05 (5H,
m, H2, H4, H10, and H13), 1.79−1.09 (9H, m, H1, H6, H7, H8, H13,
and 11-OH), 1.35 (3H, s, H15), 1.18 (3H, s, H18), 0.87 (3H, d, J =
7.0 Hz, H17), 0.71 (3H, d, J = 7.0 Hz, H16). 13C NMR (CDCl3) δ
216.7 (C3), 165.2 (C21), 146.0 [C4(triazole)], 138.7 (C19), 137.9,
128.6, 128.0, 127.9 (Ph), 123.9 [C5(triazole)], 117.8 (C20), 74.7
(C11), 72.6 (PhCH2OCH2), 71.1 (C14), 63.7 (PhCH2OCH2), 58.1
(C4), 51.7 (C22), 45.5 (C9), 44.8 (C13), 44.1 (C12), 42.0 (C5), 36.7
(C6), 36.2 (C10), 34.5 (C2), 30.5 (C8), 26.9 (C7), 26.5 (C18), 25.0
(C1), 17.0 (C16), 14.8 (C15), 11.6 (C17). HRMS (ESI) m/z
572.3094 [M + Na]+, C32H43N3O5Na calcd 572.3096.
Jgem = 0.9 Hz, Jcis = 11.0 Hz, H20), 5.21 (1H, dd, Jgem = 0.9 Hz, Jtrans
=
17.4 Hz, H20), 5.14, 5.08 (2H, AB, J = 17.5 Hz, H22), 3.36 (1H, dd, J
= 6.6, 10.3 Hz, H11), 2.30−2.06 (5H, m, H2, H4, H10, and H13),
1.79−1.09 (9H, m, H1, H6, H7, H8, H13, and 11-OH), 1.34 (3H, s,
H15), 1.18 (3H, s, H18), 0.88 (3H, d, J = 7.0 Hz, H17), 0.71 (3H, d, J
= 7.1 Hz, H16). 13C NMR (CDCl3) δ 216.7 (C3), 165.2 (C21), 138.7
(C19), 134.4 [CH(triazole)], 124.9 [CH(triazole)], 117.7 (C20), 74.6
(C11), 71.1 (C14), 58.1 (C4), 51.5 (C22), 45.5 (C9), 44.8 (C13),
44.1 (C12), 41.9 (C5), 36.6 (C6), 36.2 (C10), 34.5 (C2), 30.4 (C8),
26.9 (C7), 26.5 (C18), 24.9 (C1), 16.9 (C16), 14.7 (C15), 11.6
(C17). HRMS (ESI) m/z 452.2521 [M + Na]+, C24H35N3O4Na calcd
452.2520.
Synthesis of 22-{5-[3-(Adenin-9-yl)propyl]-1,2,3-triazol-1-yl}-22-
deoxypleuromutilin (28). A solution of Cp*RuCl(COD) (15 mg, 39
μmol) in toluene (1 mL) was degassed for 5 min with argon. N9-
(Pent-4-yn-1-yl)adenine8 (49 mg, 0.24 mmol) and a solution of
compound 6 (124 mg, 0.31 mmol) in toluene (2 mL), were added.
Five drops of DMF was added and the reaction mixture was stirred in
the microwave reactor at 110 °C for 15 min. The reaction mixture was
concentrated under reduced pressure, and the residue was absorbed
onto Celite and purified by column chromatography (0−10% CH3OH
in CH2Cl2) to give the product as a white foam: yield 36 mg, 25%; Rf
1
0.3 (10% CH3OH in CH2Cl2). H NMR (CDCl3) δ 8.34 [1H, s,
H2(A)], 7.78 [1H, s, H8(A)], 7.58 [1H, s, H4(triazole)], 6.36 (1H,
dd, Jcis = 11.0 Hz, Jtrans = 17.4 Hz, H19), 6.15 [2H, br s, NH2(A)], 5.76
(1H, d, J = 8.4 Hz, H14), 5.25 (1H, dd, Jgem = 1.4 Hz, Jcis = 11.0 Hz,
H20), 5.16 (1H, dd, Jgem = 1.4 Hz, Jtrans = 17.4 Hz, H20), 5.07, 4.92
(2H, AB, J = 17.7 Hz, H22), 4.27 (2H, t, J = 6.9 Hz, N9-CH2), 3.39−
3.36 (1H, m, H11), 2.64 (2H, t, J = 7.9 Hz, N9-CH2CH2CH2), 2.45−
1.94 (8H, m, N9-CH2CH2, H2, H4, H10, H13, and 11-OH), 1.77−
1.08 (8H, m, H1, H6, H7, H8, and H13), 1.35 (3H, s, H15), 1.16 (3H,
s, H18), 0.88 (3H, d, J = 7.0 Hz, H17), 0.64 (3H, d, J = 7.0 Hz, H16).
13C NMR (CDCl3) δ 216.7 (C3), 165.3 (C21), 155.8 [C6(A)], 153.2
Synthesis of 22-[4-(2-Thymin-1-yl)ethoxymethyl)-1,2,3-triazol-1-
yl]-22-deoxypleuromutilin (25). The general procedure was applied
with compound 6 (52 mg. 0.13 mmol), 1-[2-(propargyloxy)ethyl]-
thymine 15 (25 mg, 0.12 mmol), sodium ascorbate (2.8 mg, 14 μmol)
and CuSO4·5H2O (4.8 mg, 19 μmol) in t-BuOH/H2O (2 mL, 1:1 v/
1
v). Yield 56 mg, 76%; Rf 0.5 (10% CH3OH in CH2Cl2). H NMR
(CDCl3) δ 9.14 (1H, br s, NH), 7.61 [1H, s, H5(triazole)], 7.10 [1H,
d, J = 1.2 Hz, H6(T)], 6.41 (1H, dd, Jcis = 11.1 Hz, Jtrans = 17.4 Hz,
2075
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
[C2(A)], 150.2 [C4(A)], 140.2 [C8(A)], 138.9 (C19), 136.5
[C5(triazole)], 132.3 [C4(triazole)], 119.7 [C5(A)], 117.5 (C20),
74.6 (C11), 71.2 (C14), 58.0 (C4), 49.6 (C22), 45.5 (C9), 44.9
(C13), 44.1 (C12), 43.1 (N9-CH2), 41.9 (C5), 36.5 (C6), 36.1 (C10),
34.5 (C2), 30.4 (C8), 28.2 (N9-CH2CH2), 26.9 (C7), 26.7 (C18),
24.9 (C1), 20.5 (N9-CH2CH2CH2), 16.8 (C16), 14.7 (C15), 11.6
(C17). HRMS (ESI) m/z 627.3365 [M + Na]+, C32H44N8O4Na calcd
627.3378.
Footprinting Experiments. E. coli strain MRE600 was used to
prepare ribosomes for the chemical footprinting experiments the same
way as described previously.8 The cells were grown in LB medium and
lysed, and then ribosomes were isolated on sucrose gradients.
Antibiotic binding and chemical modification, as well as primer
extension and gel electrophoresis, were performed as described
previously,8 except that a Cy5-labeled primer (in a higher
concentration) was used instead of a 5′-32P-labeled deoxyoligonucleo-
tide primer. The gel bands were visualized and the intensities of the
modifications were assessed by use of a Typhoon trio laser scanner.
Experiments were performed at least two times, and the numbers in
Table 1 are an average of protections against CMCT modification
calculated from samples in the presence of drug relative to control
samples without drug, and the background value from unmodified
samples was subtracted..
Present Address
‡Department of Chemistry, University College, Kurukshetra
University, Kurukshetra 136119, India.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge The Danish Council for Independ-
ent Research | Technology and Production Sciences (FTP) and
The Danish National Science Foundation. J.K. also thanks the
Danish Natural Science Research Council/The Danish
Councils for Independent Research and the Lundbeck
Foundation for financial support.
ABBREVIATIONS:
■
PTC peptidyl transferase center; TBS tert-butyldimethylsilyl) ;
TBAF tetrabutylammonium fluoride; DEAD diethyl azodicar-
boxylate; DIAD diisopropyl azodicarboxylate; BSA bis-
(trimethylsilyl)acetamide; CMCT N1-cyclohexyl-N3-(2-
morpholino)ethylcarbodiimide p-toluenesulfonate
Drug Susceptibility Testing. The following strains and media
were used: E. coli AS19 in LB medium, B. subtilis 168 in 25% BHI/75%
LB/0.05% Tween, and L. innocua in 50% BHI/50% LB. Drug
susceptibility testing was performed in a microtiter plate format, where
OD values were measured at 450 nm with a microtiter plate reader
(Victor 3, Perkin-Elmer). Medium was inoculated with single colonies
and incubated overnight. The cultures were diluted to OD450 = 0.01,
and 100 μL of diluted culture was mixed with 100 μL of drug solutions
in a series with 2-fold concentration steps. The tested concentration
ranges were 0.125−32 μg/mL. MIC was defined as the drug
concentration at which the growth of the cultures was completely
inhibited after 24 h for E. coli AS19 and L. innocua and after 12 h for B.
subtilis 168 with incubation at 37 °C.
REFERENCES
■
(1) Kavanagh, F.; Hervey, A.; Robbins, W. J. Antibiotic substances
from Basidiomycetes: VIII. Pleurotus multilus (Fr.) Sacc. and Pleurotus
passeckerianus Pilat. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 570−574.
(2) Daum, R. S.; Kar, S.; Kirkpatrich, P. Retapamulin. Nat. Rev. Drug.
Discovery 2007, 6, 865−866.
(3) Poulsen, S. M.; Karlsson, M.; Johansson, L. B.; Vester, B. The
pleuromutilin drugs tiamulin and valnemulin bind to the RNA at the
peptidyl transferase centre on the ribosome. Mol. Microbiol. 2001, 41,
1091−1099.
(4) Long, K. S.; Hansen, L. H.; Jakobsen, L.; Vester, B. Interaction of
pleuromutilin derivatives with the ribosomal peptidyl transferase
center. Antimicrob. Agents Chemother. 2006, 50, 1458−1462.
(5) Schlunzen, F.; Pyetan, E.; Fucini, P.; Yonath, A.; Harms, J. M.
Inhibition of peptide bond formation by pleuromutilins: the structure
of the 50S ribosomal subunit from Deinococcus radiodurans in complex
with tiamulin. Mol. Microbiol. 2004, 54, 1287−1294.
(6) Davidovich, C.; Bashan, A.; Auerbach-Nevo, T.; Yaggie, R. D.;
Gontarek, R. R.; Yonath, A. Induced-fit tightens pleuromutilins
binding to ribosomes and remote interactions enable their selectivity.
Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 4291−4296.
(7) Gurel, G.; Blaha, G.; Moore, P. B.; Steitz, T. A. U2504 determines
the species specificity of the A-site cleft antibiotics: the structures of
tiamulin, homoharringtonine, and bruceantin bound to the ribosome.
J. Mol. Biol. 2009, 389, 146−156.
(8) Lolk, L.; Pohlsgaard, J.; Jepsen, A. S.; Hansen, L. H.; Nielsen, H.;
Steffansen, S. I.; Sparving, L.; Nielsen, A. B.; Vester, B.; Nielsen, P. A
click chemistry approach to pleuromutilin conjugates with nucleosides
or acyclic nucleoside derivatives and their binding to the bacterial
ribosome. J. Med. Chem. 2008, 51, 4957−4967.
(9) Tornøe, C. W.; Christensen, C.; Meldal, M. Peptidotriazoles on
solid phase: [1,2,3]-triazoles by regiospecific copper(I)-catalyzed 1,3-
dipolar cycloadditions of terminal alkynes to azides. J. Org. Chem.
2002, 67, 3057−3064.
Docking Procedures. The docking calculations were performed
by use of the Schrodinger 2010 program package. A PTC model was
̈
constructed by truncating the D. radiodurans ribosome structure5
(PDB entry 1XBP) to contain all residues within a spherical cut of 30
Å from the PTC binding site. This ribosome model was prepared by
use of the protein preparation wizard in the Schrodinger software. As a
̈
control, a redocking of tiamulin into the PTC model placed the drug
in the binding pocket identical to its position in the X-ray structure. All
docking experiments were based on the Glide module23−26 in the
Schrodinger 2010 program package in the XP (extra precision) mode.
̈
Flexibility was modeled by softening of the van der Waals radii with a
scaling factor of 0.8 for the nonpolar parts of the receptor atoms and
0.9 for the ligand atoms. In both cases the charge cutoff was set to 0.2.
Ranking of the docked ligands was based on the GlideScore (XP
GScore).
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures that are not covered in the main text,
for compounds 7−13, 15, 32, 34, 35, 37−39, and 43, and
1
selected NMR H NMR and 13C NMR spectra as well as
(10) Rostovtsev, V. V.; Green, L. G.; Fokin, V. V.; Sharpless, K. B. A
stepwise Huisgen cycloaddition process: copper(I)-catalyzed regiose-
lective “ligation” of azides and terminal alkynes. Angew. Chem., Int. Ed.
2002, 41, 2596−2599.
(11) Riedl, K. Studies on pleuromutilin and some of its derivatives. J.
Antibiot. 1976, 29, 132−139.
(12) Boren, B. C.; Narayan, S.; Rasmussen, L. K.; Zhang, L.; Zhao,
H.; Lin, Z.; Jia, G.; Fokin, V. V. Ruthenium-catalyzed azide-alkyne
cycloaddition: scope and mechanism. J. Am. Chem. Soc. 2008, 130,
8923−8930.
HPLC profiles for the pleuromutilin conjugates. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*(P.N.) Phone: +45 65502565, fax: +45 66158780, e-mail:
pon@ifk.sdu.dk; (B.V.) Phone: +45 65502406, fax: +45
65502467, e‑mail: b.vester@bmb.sdu.dk.
2076
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077

Journal of Medicinal Chemistry
Article
(13) Dey, S.; Garner, P. Synthesis of tert-butoxycarbonyl (Boc)-
protected purines. J. Org. Chem. 2000, 65, 7697−7699.
(14) Ludek, O. R.; Meier, C. Influence of the N3-Protection Group
on N1- vs. O2-Alkylation in the Mitsunobu Reaction. Eur. J. Org. Chem.
2006, 941−946.
(15) Jacobsen, M. F.; Knudsen, M. M.; Gothelf, K. V. Efficient N-
arylation and N-alkenylation of the five DNA/RNA nucleobases. J.
Org. Chem. 2006, 71, 9183−9190.
(16) Albæk, N.; Petersen, M.; Nielsen, P. Analogues of a locked
nucleic acid with three-carbon 2′,4′-linkages: synthesis by ring-closing
metathesis and influence on nucleic acid duplex stability and structure.
J. Org. Chem. 2006, 71, 7731−7740.
(17) Ohira, S. Methanolysis of dimethyl (1-diazo-2-oxopropyl)-
phosphonate: generation of dimethyl (diazomethyl)phosphonate and
reaction with carbonyl compounds. Synth. Commun. 1989, 19, 561−
564.
(18) Muller, S.; Liepold, B.; Roth, G. J.; Bestmann, H. J. An improved
̈
one-pot procedure for the synthesis of alkynes from aldehydes. Synlett
1996, 532−522.
(19) El-Brollosy, N. R.; Jørgensen, P. T.; Dahan, B.; Boel, A. M.;
Pedersen, E. B.; Nielsen, C. Synthesis of novel N-1 (allyloxymethyl)
analogues of 6-benzyl-1-(ethoxymethyl)-5-isopropyluracil (MKC-442,
emivirine) with improved activity against HIV-1 and its mutants. J.
Med. Chem. 2002, 45, 5721−5726.
(20) Kotian, P. L.; Kumar, V. S.; Lin, T. H.; El-Kattan, Y.; Ghosh, A.;
Wu, M.; Cheng, X.; Bantia, S.; Babu, Y. S.; Chand, P. An efficient
synthesis of acyclic N7- and N9-adenine nucleosides via alkylation with
secondary carbon electrophiles to introduce versatile functional groups
at the C-1 position of acyclic moiety. Nucleosides Nucleotides Nucleic
Acids 2006, 25, 121−140.
(21) Inamoto, K.; Yamamoto, A.; Ohsawa, K.; Hiroya, K.; Sakamoto,
T. Highly regioselective palladium-catalyzed annulation reactions of
heteroatom-substituted allenes for synthesis of condensed hetero-
cycles. Chem. Pharm. Bull. 2005, 53, 1502−1507.
(22) Shatila, R. S.; Bouhadir, K. H. Two simple protocols for the
preparation of diallylaminoethyl-substituted nucleic bases: a compar-
ison. Tetrahedron Lett. 2006, 47, 1767−1770.
(23) Glide, version 5.7. Schrodinger, LLC, New York, 2011.
̈
(24) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.;
Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.;
Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new
approach for rapid, accurate docking and scoring. 1. Method and
assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739−1749.
(25) Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.;
Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: a new approach for rapid,
accurate docking and scoring. 2. Enrichment factors in database
screening. J. Med. Chem. 2004, 47, 1750−1759.
(26) Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.;
Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T.
Extra precision Glide: docking and scoring incorporating a model of
hydrophobic enclosure for protein−ligand complexes. J. Med. Chem.
2006, 49, 6177−6196.
2077
dx.doi.org/10.1021/jm201266b | J. Med. Chem. 2012, 55, 2067−2077