Development of a method for the parallel synthesis and purification of N-substituted pantothenamides, known inhibitors of coenzyme A biosynthesis and utilization

Article abstract of DOI:10.1039/b811086g

N-Substituted pantothenamides are a class of pantothenic acid analogues which have been shown to act as inhibitors of coenzyme A biosynthesis and utilization, especially by blocking fatty acid metabolism through formation of inactive acyl carrier proteins. To fully explore the chemical diversity and inhibitory potential of these analogues we have developed a simple method for the parallel synthesis and purification of any number of pantothenamides from a single precursor, and subsequently evaluated a small library of these compounds as inhibitors of bacterial growth to demonstrate the potential and utility of the method.

Full text of DOI:10.1039/b811086g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Development of a method for the parallel synthesis and purification of
N-substituted pantothenamides, known inhibitors of coenzyme A
biosynthesis and utilization†
Marianne van Wyk and Erick Strauss*
Received 30th June 2008, Accepted 29th August 2008
First published as an Advance Article on the web 17th October 2008
DOI: 10.1039/b811086g
N-Substituted pantothenamides are a class of pantothenic acid analogues which have been shown to
act as inhibitors of coenzyme A biosynthesis and utilization, especially by blocking fatty acid
metabolism through formation of inactive acyl carrier proteins. To fully explore the chemical diversity
and inhibitory potential of these analogues we have developed a simple method for the parallel
synthesis and purification of any number of pantothenamides from a single precursor, and subsequently
evaluated a small library of these compounds as inhibitors of bacterial growth to demonstrate the
potential and utility of the method.
Introduction
Coenzyme A (CoA) is a ubiquitous and essential cofactor that is
mainly involved in carrying and activating acyl groups, especially
those involved in fatty acid metabolism. Due to its central
metabolic role, and the differences in how some of the processes
that rely on CoA operate in humans and bacteria, both CoA
biosynthesis and utilization have long been regarded as promising
targets for antimicrobial drug development.1 The search for
inhibitors of these processes have mainly focused on compounds
that are analogues of pantothenic acid (vitamin B5), the natural
precursor of CoA. Such compounds could have a multi-facetted
effect: they could potentially interfere with the cellular uptake of
exogenous pantothenic acid, or inhibit one or more of the five CoA
biosynthetic enzymes to reduce the levels of available intracellular
CoA, or be transformed by the biosynthetic enzymes into a CoA
analogue that could subsequently inhibit CoA-utilizing enzymes.
The N-substituted pantothenamides are a class of such pan-
tothenic acid analogue inhibitors that have received special
attention in several recent studies. Although these compounds
have been known as antibacterials for some time,2 it has only
recently been shown that they mainly exert their inhibitory effects
by being transformed in vivo by the CoA biosynthetic enzymes
CoaADE into the corresponding CoA analogues.3 Since these
CoA analogues do not have the essential sulfhydryl group that
confers on CoA its ability to act as an acyl carrier, they inhibit
fatty acid metabolism by transferring an inactive prosthetic group
to the acyl carrier proteins (ACPs) which are central to these
processes (Scheme 1).4,5
Scheme 1 N-Substituted pantothenamides (such as N-pentyl pan-
tothenamide shown above) act as antimetabolites by forming coenzyme
A analogues which do not contain the essential terminal thiol group of
natural CoA. These analogues subsequently transfer inactive prosthetic
groups to the acyl carrier proteins that are involved in fatty acid
metabolism, thereby blocking these processes.
or CoaA) inhibitors.6,7 As the first and committed step of CoA
biosynthesis, inhibition of PanK activity presents an obvious
strategy whereby CoA biosynthesis can be inhibited as an
alternative approach for antimicrobial development. However,
no pantothenamide characterized to date has been shown to
negatively affect a PanK enzyme’s ability to perform its ATP-
dependent phosphorylation reaction. Instead, these compounds
all seem to act as alternative (or competitive) substrates of these
enzymes which transform them into 4¢-phosphopantothenamides,
thereby reducing their effective concentration in the cell.
Pantothenamides have also been considered promising lead
compounds for the identification of pantothenate kinase (PanK,
Department of Biochemistry, Stellenbosch University, Matieland 7602, South
Africa. E-mail: estrauss@sun.ac.za; Fax: +27 21 808 5863; Tel: +27 21 808
5866
† Electronic supplementary information (ESI) available: Detailed synthetic
protocols and characterization data of thioester precursors and all
N-substituted pantothenamides. See DOI: 10.1039/b811086g
One reason why no pantothenamide that can inhibit PanK-
mediated phosphorylation has been identified as yet may be due
to the low-throughput manner in which they are synthesized. Each
compound is usually prepared individually by activation and sub-
sequent amidation of pantothenic acid’s carboxyl group, followed
4348 | Org. Biomol. Chem., 2008, 6, 4348–4355
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
by purification by column chromatography.3,4,6 Although such a
synthetic protocol is relatively simple to execute, the preparation
of a large number of different N-substituted pantothenamides—a
requirement for the effective high-throughput screening of these
compounds as potential PanK inhibitors—would be a time-
consuming process.
Results and discussion
Reaction conditions for aminolysis
In the process of developing a system that will allow for the
simultaneous preparation of various types of pantothenamides
we set out to determine optimized conditions for the type of
aminolysis reaction shown in Scheme 2. While we previously used
similar aminolysis reactions to produce several CoA analogues
from an S-phenyl thioester, the reaction design in that case was
hampered by several unique limitations imposed by the chemical
nature of the precursor and the resulting products.8 For example,
to ensure the solubility of the CoA derivatives the aminolysis
reaction had to be performed in aqueous solutions, which required
the use of large (up to ten equivalents) excesses of amine, as well
as long (more than six hours) reaction times. Moreover, since
various amines are insoluble in water this protocol precluded their
use or required the addition of organic solvents to the reaction
mixtures to increase their solubility. Performing the aminolysis
in an aqueous solution also increased the rate of hydrolysis of
the thioester precursor, and in some cases it became a significant
competing side reaction. Since pantothenic acid thioesters as well
as pantothenamides are soluble in organic solvents we decided
to investigate whether performing the aminolysis reaction in polar
solvents would address these issues. We first attempted the reaction
using methanol as solvent, but excluded it since transesterification
competed with the aminolysis reaction and produced methyl
pantothenate as side product (data not shown). We subsequently
selected acetonitrile as an alternative solvent since it has a low
boiling point, allows for good solubility of reagents and was ideal
for purification.
The rate of aminolysis of S-phenyl thiopantothenate 1b in
acetonitrile was subsequently tested in the presence of three
different amines: pentylamine as an example of a primary amine,
sec-butylamine as a typical secondary amine and benzylamine
as an example of an amine with a bulky side group (Scheme 3).
Tertiary amines, such as tert-butyl amine, have previously been
shown not to be reactive and were not tested. Each reaction
was performed with five equivalents of amine at 30 ◦C and
was monitored for a decrease in the concentration of S-phenyl
thiopantothenate 1b. This was done by removing an aliquot
of the reaction mixture at regular intervals and analyzing it
by HPLC. The results are shown in Fig. 1. The data obtained
for each aminolysis reaction were subsequently fit by regression
analysis to the rate equation for a second-order reaction. The rate
We therefore set out to develop a simple method for the
parallel synthesis of a large number of pantothenamides for
use in such inhibitor screens. We decided to base our method
on the known reactivity of activated thioesters towards amines,
since we have successfully exploited similar aminolysis chemistry
to prepare different CoA analogues from so-called pre-CoA
thioesters (Scheme 2).8 This strategy would involve the preparation
of S-phenyl thiopantothenate 1b (n = 2) as a single activated
pantothenic acid precursor; this compound can subsequently
be treated with a range of structurally diverse amines to allow
the simultaneous preparation of a wide array of N-substituted
pantothenamide products. To increase the structural diversity
of the resulting pool of pantothenamides even further we also
looked into the utilization of S-phenyl thio-a-pantothenate 1a
(n = 1) and S-phenyl thiohomopantothenate 1c (n = 3) as similarly
activated precursors. These thioesters either have one methylene
group removed from or added to the b-alanine moiety of natural
pantothenate derivatives. Finally, to ensure the practical utility of
such a method in the preparation of pure compounds for screening
purposes we also set out to devise a purification protocol for the
removal of all excess reagents and/or unreacted starting materials
as well as any potential side products that may result from the
aminolysis reaction. Here we describe the details of such a method,
and demonstrate that it can be employed to prepare a structurally
diverse library of N-substituted pantothenamides. Subsequent
biological evaluation of these compounds identified two previously
unknown pantothenamides as inhibitors of bacterial growth,
demonstrating that this method may enable us to identify the first
inhibitors of PanK activity.
constants (kobs) and the half-lifes (t1 ) of the various reactions were
2
subsequently determined from the reaction curves obtained in this
manner. These results are summarized in Table 1.
Scheme 3 Reaction conditions used to evaluate the rate of aminolysis of
S-phenyl thiopantothenate 1b with different amines.
Scheme 2 The parallel synthesis and purification of a library of
N-substituted pantothenamides is possible by using S-phenyl thioesters
1 (with n number of methylene groups between the two amide groups) as
precursors in aminolysis reactions with m number of amines. The resulting
library would have an n¥m number of members.
The kinetic analysis shows that the aminolysis of 1b proceeds
the fastest in the presence of a sterically unhindered primary
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4348–4355 | 4349
©

View Article Online
Table 1 Kinetic data for the aminolysis of S-phenyl thiopantothenate 1b
the excess amine. To enable the simultaneous removal of any
potentially unreacted thioester 1,4-diaminobutane was added to
the aminolysis reaction after three hours. This would result in the
formation of N-(4-amino-butyl) pantothenamide, a compound
which has a primary amine that would also be retained by a cation
exchange resin. We first used strongly acidic resins in this method
as such resins should be most effective in the removal of the excess
amine. However, the low pH of these resins (pH < 2) was found to
promote degradation of the pantothenamides to pantolactone and
a b-alanylamine which contaminated the final product and made
its purification difficult. We therefore chose to use a weakly acidic
cation exchange resin which has a higher pH (pH ~ 4) and that did
not cause any detectable degradation by lactonization, while still
removing all the excess amine. Finally, the thiophenol produced
during aminolysis as well as the reaction solvent were removed by
evaporation to give the pure pantothenamides as products.
in the presence of various amines
1
t /min
2
Amine
k
obs/mM-1 min-1
sec-butylamine
benzylamine
pentylamine
0.00380 0.0003
0.0125 0.0009
1.14 1.07
12.6
3.65
0.0421
One significant limitation to this first purification method is
that it cannot be applied to aminolysis reactions with amine
reactants that contain an additional amine functionality (such as
a tertiary amine) in its substructure as pantothenamides formed
in such cases would also be retained by cation exchange resins.
To address this problem we developed a silica chromatography-
based purification method for such pantothenamides (Scheme 4,
Method B). In this protocol, reaction mixtures were purified using
two sequential silica chromatography steps: the first removes the
unreacted thioester and thiophenol from the mixture, while the
second separates the excess amine from the final product. Pure
pantothenamides are obtained after evaporation of the solvent.
To demonstrate that these two purification methods can be
successfully applied as described above, aminolysis reactions of 1b
with pentylamine and 4-(2-aminoethyl)morpholine (the examples
shown in Scheme 4) were purified by following methods A and
B respectively. The resulting N-pentyl pantothenamide 4b and
N-(3-morpholin-4-yl-propyl) pantothenamide 45b purified in this
Fig. 1 Evaluation of the rate of aminolysis of S-phenyl thiopantothenate
1b at 30 ◦C in acetonitrile. The data points show the decrease in the
concentration of 1b over time in the presence of five equivalents of
sec-butylamine (ꢀ), benzylamine (᭢) and pentylamine (ꢀ) respectively.
A control reaction (●) contained no amine. The curves (dashed lines)
show the data fitted to the rate equation for a second-order reaction.
amine such as pentylamine, and that the reaction is essentially
complete within five minutes (reaction half-life of ~2.5 seconds).
The aminolysis with benzylamine, a primary but bulky amine
is nearly a 100-fold slower, while the reaction with a secondary
amine like sec-butylamine is nearly three-fold slower still. While
the analysis indicates that the rate of an aminolysis reaction is
clearly dependant on the structure of the amine, it also shows that
all the reactions were more than 90% complete after two hours.
Since a parallel synthesis of pantothenamides would entail the
use of a large variety of structurally diverse amines we therefore
decided to perform the aminolysis reactions for three hours at
30 ◦C to ensure the complete conversion of the thioester precursor
to pantothenamide.
1
manner were characterized by H and 13C NMR. Their purity
was also assessed by LC-MS analysis (Fig. 2) which indicated that
the only identifiable impurity in both cases was a small amount
of diphenyl phosphoric acid (the broad peak at 7.2 minutes).
This contaminant is a by-product of the diphenylphosphoryl
azide (DPPA)-mediated synthesis of the S-phenyl thioesters 1a
and 1b which was found to be present in some batches of these
compounds even after flash column chromatography. However, we
found that it can be excluded through careful aqueous work-up of
the coupling reaction mixtures prior to purification. Subsequent
aminolysis reactions performed with thioester precursors purified
in this manner and similarly analyzed by LC-MS did not show
any contaminants of the aminolysis products.
Purification of pantothenamide products
With optimized aminolysis conditions in hand we set out to
develop a simple purification protocol that would also be amenable
for use in a parallel synthetic method. Such a protocol would
mainly have to focus on the removal of the excess amine used in
the aminolysis—and the thiophenol produced as a by-product of
it—as the most important contaminants of the reaction. However,
it may also be possible that with certain sterically hindered amines
the aminolysis may not be complete after three hours, which
additionally would leave some unreacted thioester in the reaction
mixture. We therefore devised a purification protocol that would
address the removal of all these components in step-wise fashion
as shown in Scheme 4.
Parallel synthesis
The aminolysis reaction and the purification protocol were
subsequently combined to show that these methods can be used to
produce a large number of structurally and functionally diverse N-
substituted pantothenamides in parallel. The general procedure,
carried out in 96-well 2 mL deep-well plates, is outlined in
Scheme 5. The aminolysis reactions were performed first by
incubating the thioester 1a, 1b or 1c in the presence of five
equivalents amine for a period of three hours at 30 ◦C. For
purifications by method A (Scheme 4) the reaction mixtures
Two purification methods were developed. In the first and
more general method (Method A), we explored the use of
cation exchange chromatography with acidic resins to remove
4350 | Org. Biomol. Chem., 2008, 6, 4348–4355
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Scheme 4 Protocols used for the purification of N-substituted pantothenamides prepared by parallel aminolysis reactions of activated thioesters.
Method A: Aminolysis reactions with amines that do not contain other amino functionalities are purified using cation exchange chromatography with the
weakly acidic resin Amberlite IRC-86 to remove excess amine and unreacted thioester (after treatment of the latter with 1,4-diaminobutane), as exemplified
by the purification of pantothenamide 4b. Method B: If the amine used in the aminolysis reactions contains other amine functional groups (especially ter-
tiary amines) the resulting pantothenamide has to be purified by means of a two-step silica purification, as in the case of the morpholine-containing 45b.
contact time with the resin to completely remove the excess amine,
as well as any N-(4-amino-butyl) pantothenamide formed from
potentially unreacted thioester. The resin was thoroughly washed
with aqueous acetonitrile and the solvent as well as the remaining
thiophenol were removed by evaporation. Reaction mixtures
containing pantothenamides with tertiary amine functionalities
were purified by method B (Scheme 4) by transferring them to
filter plates loaded with silica gel equilibrated with acetonitrile.
The microcolumns were subsequently eluted with the help of a
vacuum manifold. Due to the limited volume of commercially
available 96-well filter plates two separate silica chromatography
steps were required to remove all impurities (see Experimental
section for details).
Table 2 shows the yields obtained for the aminolysis of thioesters
1a–c with 48 structurally diverse amines to give N-substituted
a-pantothenamides 2a–49a, N-substituted pantothenamides 2b–
49b and N-substituted homopantothenamides 2c–49c respectively
after purification. The pantothenamides 2–42 were purified by
method A using cation exchange chromatography, while pan-
tothenamides 43–49 were purified on silica gel by method B. The
results show that in total 142 different pantothenamides were
prepared and purified in parallel by the procedure outlined in
Scheme 5, and that for the types of amines chosen the purified
yields are generally in excess of 75% and in many cases above 90%.
Interestingly there seems to be little correlation between the yields
obtained for the three pantothenamides formed from the same
amine. For example, while no product was recovered from the
Fig. 2 HPLC analysis of two pantothenamides formed by aminolysis
of 1b and purified by the two different methods shown in Scheme 4.
N-pentyl pantothenamide 4b (panel A) was purified using cation exchange
resin-based method A while N-(3-morpholin-4-yl-propyl) pantothenamide
45b (panel B) was purified via silica chromatography-based method B. The
only observable impurity in both cases is a small amount of diphenyl
phosphoric acid, represented by the broad peak at 7.2 minutes (see text for
details).
were subsequently treated with 1,4-diaminobutane to remove
any unreacted thioester and then transferred to a 96-well filter
plate preloaded with weakly acidic cation exchange resin. The
plates were allowed to elute under gravity to ensure enough
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4348–4355 | 4351
©

View Article Online
and was available in our lab. We evaluated the potency of each
pantothenamide by comparing the growth of E. coli K12 in
1% tryptone growth medium to its growth in the same medium
with 50 mM of pantothenamide added. The growth inhibition
values for each of the prepared compounds are provided in
the Electronic Supplementary Information†. However, as shown
in Table 3 two pantothenamides (in addition to the known
inhibitor N-pentyl pantothenamide 4b) were identified that also
caused complete growth inhibition (less than 5% growth relative
to the controls) at this concentration. These compounds are
currently being investigated to determine their specific mechanism
of inhibition, i.e. whether like N-pentyl pantothenamide they
are mainly inhibitors of CoA utilization, or if they perhaps
inhibit one of the CoA biosynthetic enzymes such as PanK. This
result demonstrates that the parallel synthetic method for the
preparation of pantothenamides may successfully be applied in
the identification of new members of this class of pantothenic acid
analogues as potential new antimicrobials.
Conclusion
We have developed and successfully executed a general method
for the parallel synthesis of a diverse array of N-substituted
pantothenamide analogues from activated thioesters. This method
is based on determined optimal aminolysis reaction times for
most amines, regardless of structure. The devised protocol also
addresses several purification obstacles, and has been successfully
applied to the parallel purification of pantothenamides containing
diverse functional groups as amide substituents. Biological evalua-
tion of a small compound library prepared by this method allowed
for the identification of two previously unknown pantothenamides
as in vivo inhibitors of E. coli. This demonstrates its potential in
the discovery of new pantothenic acid analogues as inhibitors of
CoA biosynthesis and/or utilization.
Scheme 5 Protocol for the parallel synthesis of N-substituted pan-
tothenamides in 96-well format. Aminolysis reactions are performed in
deep-well plates, while purifications are carried out in 96-well filter plates
which act as microcolumns for chromatography. The final products are
obtained after evaporation of the volatiles from the column eluates.
aminolysis reactions of S-phenyl pantothenate thioesters 1b and
1c with tert-butylamine (used as a supposed negative control reac-
tion) the same reaction with S-phenyl a-thiopantothenate 1a gave
the corresponding pantothenamide 8a in 49% yield. However, in
general the a-pantothenamides and homopantothenamides were
obtained in lower yields than the N-substituted pantothenamides
formed from 1b. This suggests that the structure of the thioester
also affects the yield of the reaction, and that this phenomenon
should be taken into account when such a parallel synthetic
preparation of pantothenamides is performed.
Experimental
Material and methods
All chemicals, media and resins were purchased from Sigma-
Aldrich (Aldrich, Sigma or Fluka), Merck Chemicals or Acros
Organics and were of the highest available purity. All solvents used
for reactions and purification were CHROMASOLV HPLC grade
solvents form Sigma-Aldrich. Aminolysis reactions were done in
2 mL polypropylene 96-well deep-well plates from NUNC. Fil-
trations for purification purposes were done with AcroPrep 96-
well filter plates from Pall Life Sciences. Individual polypropylene
cluster tubes (Corning) were used in yield determination. Solvent
evaporation from deep-well plates was done on a Labconco
Centrivap concentrator. Inhibition studies were performed in
Greiner Bio-one Cellstar flat-bottomed 96-well suspension culture
plates. All ESI-MS and LC-MS analyses were also done at the
Central Analytical Facility (CAF) at Stellenbosch University using
a Waters 2690 Separations Module with a Waters 996 Photodiode
Array Detector for LC separations, followed by mass analysis on a
Waters Micromass Quattro mass spectrometer. All NMR analyses
were performed on Varian INOVA instruments (300 MHz and
400 MHz), also at CAF.
Evaluation of N-substituted pantothenamides as bacterial
growth inhibitors
To demonstrate the application and utility of the library of
prepared pantothenamides the purified compounds were tested as
inhibitors of Escherichia coli cell growth. Previous studies have
shown that N-substituted pantothenamides have antimicrobial
activities with minimal inhibitory concentration (MIC) values
ranging between 2–200 mM for different strains of E. coli.4,6 In
particular, the known growth inhibitor N-pentyl pantothenamide
4b showed inhibition of the E. coli wild-type strain UB1005 with
a MIC of 50 mM.6 This value can therefore be taken as the current
benchmark for pantothenamide inhibition of natural bacterial
strains. Consequently we decided to evaluate the inhibitory
potential of the pantothenamide library relative to this value to
determine if any of the compounds had a similar or improved
potency.
Inhibition was tested against the K12 strain of E. coli as it has
been sustained as a lab strain with minimal genetic manipulation9
4352 | Org. Biomol. Chem., 2008, 6, 4348–4355
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 2 Aminolysis of phenyl thioesters 1a, 1b and 1c with a variety of amines (2m–49m) to form the corresponding N-substituted a-pantothenamides (2a
to 49a), N-substituted pantothenamides (2b to 49b) and N-substituted homopantothenamides (2c to 49c). Yields were determined by weight determination
of each purified compound, followed by their characterization by 1H NMR
% Yields
% Yields
Entry
R-group
a (n = 1)
b (n = 2)
c (n = 3)
Entry
26
R-group
a (n = 1)
b (n = 2)
c (n = 3)
2
3
84
98
78
86
94
82
85
96
75
27
87
94
94
4
5
85
90
95
94
76
74
28
29
96
91
98
87
89
88
6
7
86
86
91
94
74
75
30
31
81
84
98
88
68
74
8
49
82
88
NR
97
NR
65
32
33
34
80
88
91
94
90
96
75
75
71
9
10
95
70
11
90
94
78
35
78
97
82
12
13
14
83
88
94
95
91
92
75
78
75
36
37
38
91
~90a
78
85
~90a
92
74
~61a
81
15
16
65
80
99
94
73
77
39
40
81
48
95
87
80
71
17
18
77
83
78
91
72
73
41
42
80
74
93
84
75
43
19
20
86
86
94
94
78
82
43
44
87
89
79
87
83
76
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4348–4355 | 4353
©

View Article Online
Table 2 (Contd.)
% Yields
% Yields
Entry
R-group
a (n = 1)
b (n = 2)
c (n = 3)
Entry
R-group
a (n = 1)
b (n = 2)
c (n = 3)
21
88
90
84
45
85
75
65
22
23
85
81
92
93
87
74
46
47
80
75
70
78
72
66
24
25
79
86
98
97
78
78
48
49
81
55
77
97
67
78
a Yield not accurate due to impurities present in amine starting material (based on 1H NMR analysis).
Table 3 N-substituted pantothenamides that showed inhibition of E. coli
K12 grown in 1% tryptone at a concentration of 50 mM
concentration of the S-phenyl thiopantothenate 1b remaining in
each sample was determined by integration of the corresponding
peak area, followed by correlation of this area to a standard curve
of known concentrations of 1b prepared as described below. All
samples were analyzed in triplicate.
Compound Name
Bacterial Growth (%)a
4b
14a
16a
N-pentyl pantothenamide
-2
2
5
1
N-cyclopentyl a-pantothenamide -2
Thioester standard curve. A standard curve of known S-
phenyl thiopantothenate concentrations was obtained by HPLC
analysis of thioester samples with concentrations ranging between
0 and 25 mM. From each of these samples 50 mL aliquots were
removed which were treated and analyzed exactly as described
above. Samples were injected in triplicate, and the standard curve
obtained by integration of the resulting peak areas.
N-cyclopropylmethyl
a-pantothenamide
-3
a Inhibition was evaluated by determining cell densities (OD600) of cultures
grown in the presence of inhibitor relative to control samples with no
inhibitor added. The reported values are the averages of four separately
tested samples, with the associated standard deviation.
Data fitting and analysis. The resulting data were analyzed
by plotting the determined concentrations of 1b against time,
followed by nonlinear regression analysis of the data using a
second-order rate eqn (1)
Aminolysis reaction conditions
Sample analysis. The rates of aminolysis were determined
by adding 20 mM S-phenyl thiopantothenate 1b to a solution
of 100 mM amine (either pentylamine, benzylamine or sec-
butylamine) in acetonitrile in a total volume of 1 mL. A negative
control reaction contained no amine. Periodically 50 mL aliquots
were removed from the solution and immediately loaded onto
30 mg (dry weight) of Amberlite IRC-86 ion exchange resin
prewashed with water. The resin was washed with 250 mL 40%
aqueous acetonitrile, and the eluent was concentrated until all the
solvent was removed. The residue was redissolved in 500 mL 50%
acetonitrile and 15 mL of each sample was analyzed by HPLC on a
Hewlett-Packard series 1100 HPLC system with dual wavelength
detection at 215 and 254 nm by using a Gemini 5 m C18 (110A), 2.0 ¥
30 mm column equilibrated with 100% solution A (0.01% TFA in
MilliQ water) and 0% solution B (acetonitrile). The column was
eluted with a linear gradient increasing to 100% B over 10 minutes,
followed by a isocratic elution with 100% solution B for a further
5 minutes. A flow rate of 0.35 mL min-1 was maintained. The
[A]0
f(x) =
(1)
1+ [A]0 kobs
x
The reaction rate constants (kobs) were obtained directly in this
manner (as one of the equation parameters) while the half-life
of each reaction was calculated from the rate constants. Data
analysis and plotting were performed using SigmaPlot 9.0 (Systat
software).
Parallel synthesis and purification
General procedure. The parallel aminolysis of S-phenyl
thioesters 1a, 1b and 1c with amines 2m–49m were performed
using 48 different amines (in fivefold excess) in 100% acetonitrile as
solvent. Solutions of the amines (100 mM final) in acetonitrile were
loaded into a 2 mL 96-well deep-well plate, followed by addition of
4354 | Org. Biomol. Chem., 2008, 6, 4348–4355
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
the desired thioester (20 mM final) also dissolved in acetonitrile.
The resulting solutions had a final volume of 1 mL. The deep-well
plate was capped and incubated at 30 ◦C while shaking vigorously
for 3 hours.
5 minutes. A flow rate of 0.35 mL min-1 was maintained. In
all cases the compounds gave single peaks at 215 nm in these
analyses. The optical purity of pantothenamides 4b and 45b was
assessed by performing 1H NMR analysis in the presence of
increasing amounts of tris-[3-(trifluoromethylhydroxymethylene)-
(+)-camphorato]europium (Eu(tfc)3). All other N-substituted
pantothenamides were characterized by H NMR. The detailed
characterization data are provided in the Electronic Supplemen-
tary Information (ESI)†.
Purification method A. After 3 hours the reaction mixtures
of compounds 2a–42a, 2b–42b and 2c–42c were treated with 1,4-
diaminobutane (100 mL; 550 mM). The resulting solutions were
incubated with shaking at 30 ◦C for another 60 minutes before
being loaded onto pre-washed Amberlite IRC-86 weak cation
exchange resin (300 mg dry weight per well) contained in a 1 mL
AcroPrep 96-well filter plate. The resin was eluted under gravity,
and was washed twice with 300 mL 40% aqueous acetonitrile. The
combined eluates were dried overnight on a Centrivap centrifugal
concentrator under reduced pressure. The resin was subsequently
washed with another 1200 mL 40% aqueous acetonitrile and the
eluate added to the dried products. The resulting solutions were
transferred to individual pre-weighed 96-well cluster tubes and
dried for another 72 hours by centrifugal concentration under
reduced pressure to remove all the solvent and thiophenol present
in the mixture. The cluster tubes containing the dried products
were subsequently weighed individually to determine the purified
yield of each compound.
1
Growth of E. coli K12 with N-substituted pantothenamides
Inhibition assays were performed by preparing a starter culture
of E. coli K12 in 1% tryptone containing four separate colonies
grown on LB agar plates. The starter culture was grown to mid-
log phase and then diluted 30 000-fold in the same medium. A
10 mL aliquot of the diluted cell suspension was used to inoculate
each well of a 96-well flat-bottomed plate containing 100 mL of
1% tryptone broth supplemented with a specific N-substituted
pantothenamide in final concentration of 50 mM. The plates were
incubated at 37 ◦C for 20 hours before the cell densities were
measured by reading the absorbance in each well at 600 nm. The
extent of growth in each well was determined by normalizing the
OD600 values relative to those of the negative control (containing
3% acetonitrile instead of pantothenamide), which were taken as
100% growth. Each compound was tested in quadruplicate.
Purification method B. The reaction mixtures of compounds
43a–49a, 43b–49b and 43c–49c were transferred to 1 mL AcroPrep
96-well filter plates pre-loaded with silica gel (300 mg dry weight
per well) that had been equilibrated with 100% acetonitrile. The
microcolumns were subsequently eluted by using a 96-well plate
vacuum manifold. The silica was then washed with 500 mL
acetonitrile, followed by elution of the product as well as some
remaining amine from the silica with ~3.6 mL methanol. The
eluates were captured in a new 96-well plate and were dried
overnight by centrifugal concentration under reduced pressure.
The resulting residues were resuspended in 150 mL methanol before
being loaded onto a second 1 mL AcroPrep 96-well filter plate
pre-loaded with silica gel (300 mg dry weight per well) that had
been equilibrated with 100% methanol. The products were eluted
with 2 mL methanol for 44a–c, 45a–c, 48a–c as well as 49a–c.
Compounds 43a–c, 46a–c and 47a–c were only fully removed from
the silica after elution with 6 mL of methanol. The eluates were
collected in individual pre-weighed 96-well cluster tubes, followed
by removal of the solvent by centrifugal concentration over
48 hours. The yields of the purified compounds were determined by
individual weighing of the cluster tubes containing the products.
Acknowledgements
The authors gratefully acknowledge the technical assistance
of J. Albert Abrie during the initial stages of this project. This
work was financially supported by a grant from the National
Research Foundation of South Africa (FA2005040600033), and
by Stellenbosch University.
References
1 C. Spry, K. Kirk and K. J. Saliba, FEMS Microbiol. Rev., 2008, 32, 56;
Y.-M. Zhang, S. W. White and C. O. Rock, J. Biol. Chem., 2006, 281,
17541.
2 G. Clifton, S. R. Bryant and C. G. Skinner, Arch. Biochem. Biophys.,
1970, 137, 523.
3 E. Strauss and T. P. Begley, J. Biol. Chem., 2002, 277, 48205.
4 Y.-M. Zhang, M. W. Frank, K. G. Virga, R. E. Lee, C. O. Rock and S.
Jackowski, J. Biol. Chem., 2004, 279, 50969.
5 A. C. Mercer and M. D. Burkart, Nat. Prod. Rep., 2007, 24, 750; R.
Leonardi, S. Chohnan, Y.-M. Zhang, K. G. Virga, R. E. Lee, C. O. Rock
and S. Jackowski, J. Biol. Chem., 2005, 280, 3314.
6 K. G. Virga, Y.-M. Zhang, R. Leonardi, R. A. Ivey, K. Hevener, H.-W.
Park, S. Jackowski, C. O. Rock and R. E. Lee, Bioorg. Med. Chem., 2006,
14, 1007.
7 A. E. Choudhry, T. L. Mandichak, J. P. Broskey, R. W. Egolf, C.
Kinsland, T. P. Begley, M. A. Seefeld, T. W. Ku, J. R. Brown, M. Zalacain
and K. Ratnam, Antimicrob. Agents Chemother., 2003, 47, 2051.
8 M. van Wyk and E. Strauss, Chem. Commun., 2007, 398.
9 F. R. Blattner, G. Plunkett, III, C. A. Bloch, N. T. Perna, V. Burland,
M. Riley, J. Collado-Vides, J. D. Glasner, C. K. Rode, G. F. Mayhew, J.
Gregor, N. W. Davis, H. A. Kirkpatrick, M. A. Goeden, D. J. Rose, B.
Mau and Y. Shao, Science, 1997, 277, 1453.
Characterization of N-substituted pantothenamides. Pan-
tothenamides 4a, 4b, 4c, 45a, 45b and 45c were fully characterized
by 1H and 13C NMR to demonstrate the efficacy and utility
of the two purification methods. The purity of these com-
pounds were further verified by LC-MS analysis on a X-bridge
C18 (110A) 3.5 um, 2.1 ¥ 50 mm column. The column was
equilibrated with 100% solution A (0.01% TFA in MilliQ water)
and 0% solution B (acetonitrile). The products eluted with a
linear gradient increasing to 100% B over 10 minutes, followed
by a isocratic elution with 100% solution B for a further
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 4348–4355 | 4355
©