Structure-activity relationships and enzyme inhibition of pantothenamide-type pantothenate kinase inhibitors

Article abstract of DOI:10.1016/j.bmc.2005.09.021

A set of novel pantothenamide-type analogues of the known Staphylococcus aureus pantothenate kinase (SaPanK) inhibitors, N-pentyl, and N-heptylpantothenamide, was synthesized in three series. The first series of analogues (1-3) were designed as molecular probes of the PanK binding site to elucidate important structure-activity relationships (SAR). The second series of analogues (4-16) were designed using structural information obtained from the Escherichia coli PanK (EcPanK) structure by targeting the pantothenate binding site and the adjacent phenylalanine-lined lipophilic pocket. Insight into the antimicrobial effect of N-pentylpantothenamide (N5-Pan) through its conversion to the antimetabolite ethyldethia-CoA and further incorporation into an inactive acyl carrier protein analogue drove the development of the third series of analogues (17-25) to enhance this effect using substrate-like substitutions. Each of the analogues was screened for enzyme inhibition activity against a panel of pantothenate kinases consisting of EcPanK, Aspergillus nidulans (AnPanK), SaPanK, and the murine isoform (MmPanK1α). Series 1 demonstrated only modest inhibitory activity, but did reveal some important SAR findings including stereospecific binding. Series 2 demonstrated a much higher inhibition rate for the entire series and significant inhibition was seen with analogues containing alkyl substituents. Series 3 demonstrated the most preferential inhibition profile, with the highest inhibitory activity against the SaPanK and MmPanK1α. The MmPanK1α protein was inhibited by a broad spectrum of the compounds, whereas the E. coli enzyme showed greater selectivity. The overall activity data from these analogues suggest a complex and non-enzyme specific SAR for pantothenamide substrate/inhibitors of the different PanK enzymes.

Full text of DOI:10.1016/j.bmc.2005.09.021

Bioorganic & Medicinal Chemistry 14 (2006) 1007–1020
Structure–activity relationships and enzyme inhibition of
pantothenamide-type pantothenate kinase inhibitors
Kristopher G. Virga,a Yong-Mei Zhang,b Roberta Leonardi,b Robert A. Ivey,b
Kirk Hevener,a Hee-Won Park,b Suzanne Jackowski,b Charles O. Rockb and
Richard E. Leea,*
aDepartment of Pharmaceutical Sciences, University of Tennessee Health Science Center, 847 Monroe Ave, Rm327,
Memphis, TN 38163, USA
bDepartments of Infectious Diseases and Structural Biology, St. Jude ChildrenÕs Research Hospital, Memphis, TN 38105, USA
Received 2 July 2005; revised 7 September 2005; accepted 7 September 2005
Available online 5 October 2005
Abstract—A set of novel pantothenamide-type analogues of the known Staphylococcus aureus pantothenate kinase (SaPanK) inhib-
itors, N-pentyl, and N-heptylpantothenamide, was synthesized in three series. The first series of analogues (1–3) were designed as
molecular probes of the PanK binding site to elucidate important structure–activity relationships (SAR). The second series of ana-
logues (4–16) were designed using structural information obtained from the Escherichia coli PanK (EcPanK) structure by targeting
the pantothenate binding site and the adjacent phenylalanine-lined lipophilic pocket. Insight into the antimicrobial effect of N-pen-
tylpantothenamide (N5-Pan) through its conversion to the antimetabolite ethyldethia-CoA and further incorporation into an inac-
tive acyl carrier protein analogue drove the development of the third series of analogues (17–25) to enhance this effect using
substrate-like substitutions. Each of the analogues was screened for enzyme inhibition activity against a panel of pantothenate
kinases consisting of EcPanK, Aspergillus nidulans (AnPanK), SaPanK, and the murine isoform (MmPanK1a). Series 1 demonstrat-
ed only modest inhibitory activity, but did reveal some important SAR findings including stereospecific binding. Series 2 demon-
strated a much higher inhibition rate for the entire series and significant inhibition was seen with analogues containing alkyl
substituents. Series 3 demonstrated the most preferential inhibition profile, with the highest inhibitory activity against the SaPanK
and MmPanK1a. The MmPanK1a protein was inhibited by a broad spectrum of the compounds, whereas the E. coli enzyme showed
greater selectivity. The overall activity data from these analogues suggest a complex and non-enzyme specific SAR for pantothena-
mide substrate/inhibitors of the different PanK enzymes.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
genomics, and high-throughput screening techniques,
which have uncovered multitudes of potential targets.4
Coenzyme A (CoA) biosynthesis is one pathway that
has garnered much interest recently as a potential target
for antimicrobial therapy.5–7 CoA is the major acyl
group carrier in biology and is an essential cofactor in
intermediary metabolism.8 CoA is assembled in a
five-step pathway starting with pantothenic acid (vita-
min B5). Pantothenate kinase (PanK, the product of
the coaA gene) catalyzes the first and regulatory step
in the CoA biosynthetic pathway.8,9 A comparison of
the prokaryotic and eukaryotic PanK protein sequences
reveals a very low sequence homology, leading to the
hypothesis that selectivity might be possible in the
development of small molecule inhibitors.7,10
Increasing bacterial resistance to commonly used antibi-
otics is a great concern for the healthcare industry.1
Infections caused by resistant bacterial strains, such as
MRSA (methicillin-resistant Staphylococcus aureus),
VRE
(vancomycin-resistant
enterococcus),
and
MDRTB (multi-drug resistant tuberculosis), have led
to greatly increasing morbidity and mortality rates.2
This has led to an urgent need to develop new, more
potent antimicrobial agents with novel mechanisms of
action.3 These efforts have been greatly enhanced in
recent years with the advances in structural biology,
Keywords: Pantothenate kinase; Pantothenamide; Fatty acid synthase;
Structure based drug design; Coenzyme A.
*
The N-alkylpantothenamide class of antibacterials was
first reported in 1970 as active against Escherichia coli,13
Corresponding author. Tel.: +1 901 448 6018; fax: +1 901 448
6828; e-mail: relee@utmem.edu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.09.021

1008
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
Figure 1. Cross-sectional view with MOLCAD surface of the EcPanK enzyme binding pocket. Areas in brown are hydrophobic. Areas in blue are
hydrophilic. And areas in green are neutral. (a) Surface of the EcPanK binding site with pantothenate bound. The hydrophobic pocket (brown)
extends just beyond the acid functional group of the pantothenate. (b) N-Pentylpantothenamide docked into the active site of EcPanK. The 5-carbon
alkyl chain is predicted to extend into the hydrophobic pocket. There are two phenylalanine amino acid moieties in the pocket available for possible
ring stacking interaction with aromatic functional groups.
and their inhibitory activity attributed to their ability to
be phosphorylated by PanK and incorporated by down-
stream enzymes into inactive CoA analogues.14 Recent-
ly, these pantothenic acid analogues (4 and 5) were
reported to be potent inhibitors of the S. aureus PanK;7
however, subsequent research demonstrated that these
antimetabolites are substrates for the enzyme. Further-
more, the antibacterial action of N5-Pan and N7-Pan
is due to their incorporation into acyl carrier protein
(ACP) and the subsequent inhibition of fatty acid bio-
synthesis.6,15 Using the new binding information ob-
tained from the latest structure for the E. coli PanK
(Fig. 1A), N5-Pan was docked into the active site
(Fig. 1B) providing a basis for the design of new inhib-
itors/substrates for the enzyme.12 The aim of this work
was to build on this information to develop an SAR
model for PanK inhibitor/substrates, with the goal of
achieving selectivity and understanding the basis for
their antibacterial activity.
(H+) ion exchange column to convert the products to
the free-acid form. The final products were then convert-
ed to their Ca2+ salts. The synthesis of compound 3 and
its intermediates (A–C) was adapted from the procedure
of Freskos and is detailed under the compound proce-
dures (Scheme 2).18
The synthesis of the N-alkylpantothenamides (series 2
and 3) was adapted from the procedure of Choudhry
et al.7 Included in the series were the two known inhib-
itors 4 and 5. The analogues of series 2 (4–16) were de-
signed to probe the phenylalanine-lined hydrophobic
pocket of the enzyme, which extends just beyond the
acid moiety of the pantothenate binding site in Ec-
PanK. The aromatic amines (6–10 and 12) of series 2
were selected for potentially favorable ring stacking
interactions between two phenylalanine residues within
the hydrophobic pocket of the binding site (Table 1
and Fig. 1B). The analogues of series 3 (17–25) were
designed as antimetabolite compounds to resemble
the pantetheine moiety of the 4-phosphopantetheine
product which is produced further down in the CoA
pathway as well as the analogue ACP products demon-
strated by 4 and 5.6,14,15 The functional groups chosen
consisted of ethers, thioethers, and free hydroxyls and
were intended to mimic the biosynthetic intermediates
produced in the fatty acid synthase pathway. The reac-
tions were performed under inert conditions in discrete
reaction vessels. The free-acid form of pantothenate
used in each of the reaction schemes was obtained by
dissolving sodium pantothenate in methanol and subse-
quently passing through an Amberliteꢁ IR-120 ion ex-
change column. The pantothenate was then dried
2. Results and discussion
2.1. Chemistry
The synthesis of the two Ho-pantothenate (HoPan) iso-
mers (1 and 2) was adapted from Kopelevich et al.17 The
syntheses of 1 and 2 were performed by the reaction of
pantolactone (R or S) with 4-aminobutyric acid in the
presence of diethylamine (Scheme 1). The purification
of the reaction mixtures was performed by aqueous
work-up to remove unreacted pantolactone and ion ex-
change chromatography using an Amberliteꢁ IR-120
O
O
O
O
a
HO
HN
O
OH
H2N
+
OH
OH
R) or (S)
HO
(
1 (R)o r 2 (S)
Scheme 1. Synthesis of 1 and 2. Reagents and conditions: (a) triethylamine, methanol, 60 ꢂC, overnight.

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1009
O
O
O
O
O
O
O
O
HO
HN
O
OH
a
b
c
+
H2N
OH
OTf
N3
OH
N3
d
O
HO
HN
O
OH
H2N
3
Scheme 2. Synthesis of 3. Reagents and conditions: (a) trifluoro-methanesulfonic anhydride, 2,6-lutidine, DCM, ꢀ20 ꢂC, 1 h; (b) NaN3, DMF, 0 ꢂC,
2 h; (c) triethylamine, methanol, 60 ꢂC, overnight; (d) NH4OH, pyridine, PPh3, 6 h, ambient temperature.
Table 1. Percent of enzyme inhibition for series 1 analogues against the different PanK enzymes
HO
R'
O
HO
R'
O
R
R
2 only
Compound
R
R0
% Inhibition at 100 lM
EcPanK
AnPanK
MmPanK1a
SaPanK
O
O
O
1
2
3
OH
OH
NH2
29.4 0.1
80.6 0.97
2.7 0.9
78.7 6.0
HN
HN
HN
OH
OH
OH
12.5 4.3
11 1.6
0
0
0
54.6 1.7
13.1 1.6
0
under high heat and vacuum to complete dryness. The
N-alkylpantothenamides were then prepared by
reaction of the free-acid pantothenate with each of
the corresponding amines in the presence of diphenyl-
phosphoryl azide and triethylamine (Scheme 3). The
amines used to form the terminal amide were hand-
selected based on chemical structure and variability.
For compounds 4–16, each of the reaction mixtures
was purified by flash chromatography. For the purifica-
tion of the more polar compounds (17–25), each of the
reaction mixtures was first passed over a freshly acti-
vated Amberliteꢁ IR-400 basic ion exchange column
to scavenge the diphenylphosphoric acid byproduct,
as it eluted very closely with each of the desired prod-
ucts by TLC. Each of the compounds was then purified
by flash chromatography.
3. Biology
PanKs are found in all species, although their primary
structures have significant differences. We selected the
panel of PanKs for our analysis to cover the types of Pan-
Ks found in nature. The prototypical bacterial PanKs,
typified by the coaA gene product (EcPanK), are clearly
distinct from the eukaryotic enzymes typified by the Pan-
Ks from Aspergillus (AnPanK) and the mouse (MmPan-
K1a). The enzyme from S. aureus (SaPanK) is different
from both, although it is more closely related to the meta-
zoan enzymes than the E. coli enzyme. Each of the ana-
logues was screened for inhibitory activity against the
panel of PanK enzymes as determined by a decrease in
enzymatic activity demonstrated by each enzymeÕs ability
to convert pantothenate into 40-phosphopantothenate.
HO
HN
O
O
HO
HN
O
O
a,b
+
H2N R
HO
OH
HO
HN R
Scheme 3. Synthesis of 4–25. Reagents and conditions: (a) diphenylphosphoryl azide, DMF, 0 ꢂC; (b) triethylamine, 24 h.

1010
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
In the initial series of pantothenamide-type analogues
(1–3), Ho-pantothenate (HoPan), a previously reported
pantothenate analogue, was synthesized in both the pure
(R,1) and (S,2) isomeric forms as a starting point. The
HoPan analogue contains a c-aminobutyric acid func-
tionality in place of the b-alanine functionality of the
Table 2. Percent of enzyme inhibition for series 2 and 3 analogues against the different PanK enzymes
HO
HN
O
O
R
HO
Compound
R
% Inhibition at 100 lM
EcPanK
AnPanK
MmPanK1a
SaPanK
4
5
72.2 0.7
12.6 0.3
10.7 4.3
81.4 1.1
98.8 0.1
HN
76.4 0.01
89.6 0.6
98.2 0.1
HN
HN
6
7
8
57.8 0.6
59.5 0.3
35.7 0.8
40.7 1.3
80.9 0.6
89.5 2.0
HN
O
O
44.2 0.7
71.7 0.2
70.5 0.3
84.6 2.5
17.7 9.1
86.4 0.9
HN
HN
O
O
9
33.4 7.1
7.3 7.6
28.4 1.0
15.9 0.9
57.4 1.9
73.0 0.5
8.7 8.0
9.3 1.8
O
N
10
N
N
O
11
12
13
10.7 6.8
55.7 2.5
3.01 2.9
13.7 2.0
54.7 0.1
71.9 0.6
39.1 0.2
82.2 0.5
93.3 0.3
4.2 7.4
53.9 3.6
76.6 0.6
HN
HN
O
HN
HN
14
15
16
17
18
19
20
21
22
23
24
25
82.5 4.4
71.2 0.04
90.6 0.4
61.9 5.0
28.7 2.6
3.4 10.6
53.7 1.2
54.3 3.8
32.3 2.6
11.5 1.7
38.6 8.5
28.6 5.8
51.4 4.5
46.2 3.0
92.2 0.1
49.7 0.5
26.8 3.2
25.9 2.1
61.0 1.9
72.6 0.4
62.1 0.1
36.7 3.5
52.9 1.0
41.8 5.7
86.8 0.3
89.9 0.6
98.9 0.2
85.0 0.01
67.6 0.1
56.7 2.4
82.9 0.4
87.2 1.9
61.2 0.04
70.3 3.7
64.9 1.9
68.9 1.2
99.1 0.1
97.1 0.01
97.2 0.1
96.5 0.3
90.1 0.2
72.2 0.4
97.6 0.1
97.7 0.01
96.2 0.1
79.9 0.5
94.4 0.2
95.1 0.2
HN
HN
S
HN
HN
HN
O
OH
S
HN
HN
S
HN
HN
HN
O
OH
O
O
HN

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1011
native pantothenate. Enzyme inhibition data from the
two isomers revealed a distinct preference for the (R)
conformation over the (S), especially with respect to
the AnPanK and MmPanK1a enzymes (Table 1).
Importantly, the EcPanK was not inhibited by HoPan
to a significant degree. The importance of the C3
hydroxyl moiety was explored through substitution of
a primary amine (3) in place of the hydroxyl. PanK inhi-
bition was almost completely lost following this substi-
tution (Table 1). None of these compounds exhibited
an MIC against E. coli (>400 lM).
A second series of pantothenamide analogues were syn-
thesized to explore the structural diversity allowed in the
alkyl chain of the known active compounds 4 and 5
(Table 2). Series 2 (4–16) was designed based on the
pantothenamide scaffold of the two known inhibitors 4
and 5 using the recent structural information from
EcPanK.12 The remaining analogues (17–25) are also
pantothenamides, but were developed as antimetabolites
to more closely mimic the structure of CoA and enhance
the antibacterial effect seen with 4 and 5.7 The results
from the PanK assays showed that many of the com-
pounds demonstrate EcPanK inhibitory activity similar
to those of 4 and 5. However, the inhibitors were not
selective for the bacterial enzyme since the degree of en-
zyme inhibition was equal to or greater than that of
MmPanK1a. There were several analogues for which
the MmPanK1a enzyme showed the highest percent
inhibition (7, 12, 13, and 16) and a couple of instances
where it demonstrated the only significant inhibition
among the four enzymes (9–11). Most notably, the
MmPanK1a showed a greater selectivity for the aromat-
ic amines compared to the other enzymes. The SaPanK
showed an almost equally high percent inhibition across
the entire series. The similarity in the SAR between the
MmPanK1a and the SaPanK reflects the closer similar-
ity of SaPanK to metazoan enzymes than to
EcPanK.10,15 For compounds 13, 19, and 23, EcPanK
was the only enzyme not inhibited to a significant
degree, indicating differences in the structure of the
pantothenate binding site of the EcPanK compared to
the other enzymes. The binding promiscuity of this
group of dissimilar enzymes was disappointing with
respect to the goal of identifying selective inhibitors.
IC50 values were also obtained for the two known
inhibitors (4 and 5) against the EcPanK and MmPan-
K1a to demonstrate an inhibition pattern for each of
the series analogues consistent with single-site inhibition
(Fig. 2). As indicated by the IC50 plots, inhibition of the
enzymes by the two compounds occurs in a concentra-
tion-dependent manner.
Figure 2. Inhibition of EcPanK and MmPanK1a by compounds 4 and
5. Panel A, compounds 4 (s) and 5 (d) inhibited EcPanK with an IC50
value of about 70 lM. Panel B, compounds 4 (s) and 5 (d) inhibited
MmPanK1a with an IC50 value of about 20 lM.
even lower MIC than 4, which remained unchanged at
50 lM, indicating that 5 is a substrate for a TolC-depen-
dent pump. Analogues 6, 8, and 14 regained detectable
MIC values of 200 lM, suggesting that a TolC efflux
system(s) was partly responsible for conferring resis-
tance to the compounds. Only 21 from the second ana-
logue set in the series was active with a detectable MIC
of 50 lM. It was expected that 17 would show the great-
est antimicrobial activity in vitro of the compounds de-
signed as antimetabolites (17–25) as it demonstrated the
highest percent inhibition against the EcPanK enzyme.
Furthermore, once phosphorylated by PanK, it was
the most closely related structure to 40-phosphopant-
etheine. However, this compound was inactive in vivo.
Antimicrobial susceptibility testing of the second ana-
logue series of compounds against the wild-type E. coli
strain, UB1005, was conducted to explore the antibacte-
rial action of these inhibitors. The results from this assay
revealed that only 4 was active (Table 3). In a second
MIC assay, each of the analogues in the second two ser-
ies was screened against E. coli strain ANS1, which is
defective in TolC-dependent type I secretion of small
molecules and proteins (Table 3).16 Interestingly, 5 had
antibiotic activity in this TolC-negative screen to an
An MIC assay against S. aureus was also performed
(Table 3). For this assay, six of the more active com-
pounds from series 2 (4–6 and 14–16) and all of the ana-
logues from series 3 (17–25) were screened based on
their potent enzyme inhibition, including 4 and 5, for

1012
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
Table 3. MIC of pantothenamide analogues against E. coli and
S. aureus
tested showed any significant MIC activity. Only 6, 14,
and 15 had detectable S. aureus MIC values of 100,
100, and 200 lM, respectively. These results were unex-
pected as each of the analogues tested had significant
inhibitory activity against SaPanK. Many of the series
2 and 3 analogues demonstrated substantial enzyme
activity relative to 4 and 5. These results have deter-
mined that many of the compounds, which showed sig-
nificant activity in the enzyme inhibition assay, are
completely inactive against live cells.
Inhibitor
MIC
E. coli UB1005 (lM) E. coli ANS1 (lM) S. Aureus
4
5
50
>200
>200
>200
>200
>200
>200
>200
>200
>200
>200
>200
>200
ND
50
25
25 lM
78 nM
100 lM
ND
6
200
>200
200
7
8
ND
ND
9
>200
>200
>200
>200
>200
200
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
ND
ND
One explanation for the inability of the analogues to
inhibit bacterial growth was that they were not convert-
ed by the downstream enzymes to their respective CoA
analogues. Alternatively, these CoA analogues may
have not been substrates for AcpS, thus failing to be
incorporated into ACP. We tested the first point by
establishing an in vitro assay, followed by HPLC analy-
sis that couples the phosphorylation of the analogue by
EcPanK to its incorporation into CoA via CoaD and
CoaE, the enzymes that catalyze the last two steps in
CoA biosynthesis (Fig. 3, top panel). We subsequently
included AcpS and apo-ACP in the assay and we mon-
itored the conversion of apo-ACP to the corresponding
pantothenamide holo-ACPs. The reaction mixtures were
analyzed by SDS–PAGE on gels containing low concen-
trations of urea (Fig. 3, bottom panel) that are able to
resolve different ACP forms at alkaline pH.22 Impor-
tantly, all of the designed compounds were readily
metabolized by the enzymes in this system to form not
only inactive analogue CoA derivatives, but also
ND
ND
100 lM
200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200 lM
>200
>200
>200
>200
>200
>200
50
ND
ND
ND
ND
ND
ND
>200
>200
>200
>200
ND
ND
which MIC values had been previously reported.7 The
data from this assay validate the previously reported po-
tent activity of 4 and 5 against S. aureus (MIC 25 lM
and 78 nM, respectively). None of the other analogues
Figure 3. Conversion of compounds 4, 5 to CoA, and 17–25 analogues and in vitro incorporation into ACP. Top panel: example of HPLC analysis
of reaction mixtures containing one of the 17–25 compounds and either CoaADE (thick line) or CoaDE alone (control, thin line). Bottom panel:
SDS–PAGE analysis of the holo-ACP species formed in the presence of AcpS and the CoA analogues derived from compounds 4, 5, and 17–25.
These compounds were pre-incubated with CoaADE before adding apo-ACP and AcpS, as detailed in the Section 5.

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1013
inactive ACP. These data validate our design criteria for
producing compounds that would function in this path-
way. Thus, we conclude that the introduction of a polar
atom (O or S) into the alkyl chain must prevent the up-
take of the inhibitors into the cell.
amine substitution is likely the result of the imidazole
ring acting as an H-bond donator rather than an H-
bond acceptor. A possible explanation then for the inac-
tivity of the amine substituted (3) in place of the C3
hydroxyl is that the primary amine exists in the proton-
ated state and therefore, cannot accept the proton from
the H-bond donating imidazole. The C4 carbonyl and
N5 amide were maintained throughout each of the ser-
ies; however, they are both known to form H-bond
interactions with Tyr175 and Asn282, respectively.12
The C8 carboxyl was extended by one carbon in the case
of c-aminobutyric acid substitution (1 and 2). This
extension of the carbon chain is not detrimental to bind-
ing, but it is reasonable to assume that a two-carbon
spacer is optimal as revealed by binding interactions of
the C8 carboxyl in the EcPanK structure. Removal of
the C8 carboxyl is disfavored as indicated by the PanK
(ADP) Pan ternary complex and reduced inhibitory
activity of a number of analogues designed without the
C8 carboxyl functionality (data not shown). However,
substitution of the carboxyl with an amide functionality,
as in the pantothenamide analogues, is able to retain
binding affinity. It has been determined that the C8 car-
boxyl forms H-bond interactions with both Asn282 and
Tyr240.12 This explains how the substitution of the car-
boxyl functionality with that of an amide is acceptable,
as the amide is still capable of forming H-bond interac-
tions. Although, this substitution does appear to be lim-
ited to secondary amides. The binding of a tertiary
amide, compound 10, was severely diminished except
in the mammalian enzyme, perhaps indicating that an
H-bond donor is required in this position. The SAR of
the lipophilic pocket extending beyond the carboxylate
binding site of pantothenic acid has proven to be more
complex. The EcPanK demonstrated more favorable
binding to linear alkyl substituents that were not overly
bulky or sterically hindered (4, 5, 8, and 9). For the aro-
matic analogues (6–9 and 12) inhibition was slightly
weaker than the linear alkyl substituents, showing no
3.1. Structure–activity relationships developed for PanK
enzymes
The initial structure–activity relationship for PanK was
developed from the compounds in the first series of
pantothenamide-type analogues and corresponds to
the interactions involving EcPanK (Table 1 and
Fig. 4). There is some overlay between the SAR of all
of the PanK enzymes tested, but each has also demon-
strated unique interactions of their own. The SAR for
the EcPanK is the one detailed here. The primary alco-
hol on the C1 carbon, for which no modifications were
attempted, is the site of phosphorylation. It forms an
H-bond interaction with the side-chain carboxyl of
Asp127 as indicated by the PanK ADP Pan ternary
complex.12 This interaction indicates that a deprotona-
tion occurs to facilitate phosphoryl transfer. For the
C3 hydroxyl, collective data from each of the PanK en-
zymes suggest stereoselective binding for the (R) isomer
over the (S) for optimal binding. Binding was dimin-
ished for each of the enzymes with (S)-HoPan compared
to (R)-HoPan, especially for AnPanK and MmPanK1a.
Substitution at this position by a primary amine was
also disfavored. More conclusive data are still required
for the absolute SAR determination at this position
for EcPanK due to low-level binding of the enzyme
for each analogue in series 1. However, this SAR for Ec-
PanK is supported by the known interactions deter-
mined in the PanK ADP Pan ternary complex, where
the hydroxyl at this position was found to form an H-
bond interaction with the imidazole ring of His177.12
From the structure, it can also be rationalized that the
reduction in binding affinity seen with the primary
Carbonyl portion is
nonessential for binding.
-OH forms two H-bond
interactions. Amide substitution
is acceptable (non-tertiary).
Amide nitrogen forms a direct
H-bond with Asn 282 (may be
required for binding). Carbonyl
forms a weak H-bond to Tyr 175
coupled through water (may not
be required for binding).
Phosphorylation site
C2
Linear alkyl chains appear to be the
most favored. Polar substituents are
less favored. Phenyl rings were
C6
forms an H-bond interaction
that is required for phosphoryl
transfer.
HO
HN
O
N5
C3
C1
C4
C8
C7
tolerated. Activity decresed as the
number of ring substituents increased.
HO
O
OH
Two carbon spacer
is optimal for binding.
(R) isomer is the
preferred stereochemistry
for optimal binding.
Hydroxyl H-bond acceptor
required functionality.
Figure 4. Developed SAR of the resolved binding interactions for inhibitors of EcPanK.

1014
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
enhanced binding from interactions with the phenylala-
nine rich pocket. Binding was greater for the larger
1-methyl-3-phenyl propyl substituent (8) than for the
phenethyl (6). Methoxy-substituted phenethyl and
benzyl analogue (7, 9, and 12) activities decreased with
an increase in the number of methoxy substituents. Bind-
ing for this enzyme was also severely diminished for com-
pounds containing polar functionalities (11, 19, and 23).
activity are being converted into CoA and ACP antime-
tabolites. In fact, most of the analogues that were inac-
tive in the MIC assay were nonetheless substrates for the
pathway enzyme in vitro. Since the one of the strains
used in the test (ANS1) lacked the TolC-dependent ef-
flux pumps, we conclude that these compounds most
likely lack antimicrobial activity because they are unable
to efficiently enter the cell. This is a somewhat surprising
conclusion in light of the small difference in their struc-
tures. The inclusion of a single polar atom into the alkyl
chain is sufficient to render the compound inert as an
antimicrobial agent. This is difficult to reconcile with a
diffusion-mediated mechanism for transport into the cell
suggesting that the pantothenamides may be taken up
by a protein mediated process. Current work is focused
on understanding the mechanism for pantothenamide
uptake since it is clear that this barrier will be a critical
issue that needs to be addressed in the development of
additional pantothenamide antimetabolites.
Some secondary SAR for the other PanK enzymes has
been determined. The AnPanK demonstrated the lowest
rate of inhibition towards the series as a whole. Howev-
er, it did demonstrate a higher tolerance toward steric
hindrance (13) compared to EcPanK. It also demon-
strated the highest rate of inhibition toward the (R)-Ho-
Pan. No qualitative SAR was able to be obtained for the
AnPanK enzyme. The mammalian isoform, MmPan-
K1a, demonstrated the highest rate of inhibition toward
the series as a whole, as previously mentioned. It too did
not seem to be affected by the presence of bulky or ste-
rically hindered functionalities (7, 8, 13, and 16). Inter-
estingly, it was the only enzyme to demonstrate
significant inhibition in the presence of a tertiary amide
at N5 (10). The SaPanK also demonstrated a high rate
of inhibition. The highest percent inhibitions for this en-
zyme were seen with linear alkyl substituents (4, 5, 14,
20, and 21). Reasonable inhibition was also seen in the
presence of non-substituted aromatic functionalities as
well as hydroxyls (6, 8, 19, and 23). Binding for this en-
zyme was most severely diminished with the addition of
electron donating O-methoxy substituents placed on
aromatic rings and in the case of the ethyl morpholine
(7, 9, and 11). It is uncertain as to whether these effects
were a result of electron density, polarity, or the increase
in steric bulk.
5. Materials and methods
5.1. General
All the anhydrous solvents and starting materials were
purchased from Aldrich Chemical Company (Milwau-
kee, WI). All reagent grade solvents used for chroma-
tography were purchased from Fisher Scientific
(Sewanee, GA). Chemical reactions were monitored
by thin layer chromatography (TLC) on pre-coated
Merck 60 F254 silica gel plates and visualized by stain-
ing using a solution of ceric sulfate (1 g) and ammoni-
um molybdate (25 g) dissolved in 500 mL of 2 M
sulfuric acid. Product purification for compound 3
was performed by HPLC on a Gilson 215 HPLC (Mid-
dletown, WI) equipped with an Xterraꢁ Prep MS C18
5 lm, 19 · 100 mm column (Milford, MA); gradient
MeCN/H2O, 0–100% (25 min) with dual wavelength
detection at 215 and 254 nm. A Biotage Horizon chro-
matography system and silica gel cartridges were used
to purify reaction mixtures 4–25, Biotage Inc. (Lake
Forest, VA). Purity was ascertained by analytical
HPLC analysis for compounds 1–25 on a Shimadzu
LC20 AD equipped with a Phenomenex LunaTM (Tor-
rance, CA) 3 lm C18, 4.6 · 50 mm column; gradient
MeCN/H2O (0.01% TFA), 0–100% (10 min) with detec-
tion at 215 nm giving single peaks with retention times
reported below. Melting points were obtained with a
Thomas Scientific (Swedesboro, NJ) Uni-melt capillary
melting point apparatus and are uncorrected. Optical
rotations were obtained using a Rudolph Instruments
(Fairfield, NJ) DigiPol automatic polarimeter with sam-
4. Conclusion
This work reveals important SAR interactions within
the pantothenate binding site of PanK. The N-alkyl-
pantothenamide series of compounds has been expand-
ed and their activities against homologues of the PanK
enzyme from four species were determined. Most nota-
ble is that the absolute (R)-isomeric conformation of
the C3 hydroxyl is required for binding to all of the en-
zymes and substitutions at this position are not tolerat-
ed. The hydrogen bond donor/acceptor at the C8
carboxylate is also required for binding to all non-mam-
malian enzymes. Probing of the lipophilic pocket
beyond the pantothenate binding site has revealed
distinct SAR interactions for each of the four enzymes
tested. Most of the compounds in the series demonstrat-
ed significant activity against the different enzymes in an
inhibition assay and were subsequently demonstrated to
be substrates for the enzyme. The mammalian form of
PanK proved to be highly promiscuous, demonstrating
the highest degree of inhibition across the series as a
whole. These initial results also suggest that selectivity
will be a challenge in discovering potent inhibitors
directed solely to the bacterial enzyme. MIC assays
using two strains of E. coli as well as S. aureus support
the theory that the active compounds with antimicrobial
1
ples dissolved in MeOH at 10 mg/ml. All H and 13C
NMR spectra were recorded on a Bruker ARX-300
1
(300 and 75 MHz for H and 13C NMR, respectively)
1
or Varian INOVA-500 (500 and 125 MHz for H and
13C NMR, respectively) spectrometer. Chemical shifts
are reported in parts per million (d) relative to residual
solvent peak or internal standard (tetramethylsilane)
and coupling constants (J) are reported in hertz (Hz).
Mass spectra were recorded on a Bruker Esquire (Bille-
rica, MA) LC/MS using ESI.

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1015
5.2. General procedure for preparation of 1 and 2
(5 g, 17.7 mmol) was added dropwise. The reaction mix-
ture was stirred for 20 min at ꢀ78 ꢂC and then for 1 h
at room temperature. The reaction mixture was concen-
trated in vacuo and then extracted with Et2O partitioned
with 5% aqueous potassium hydrogen sulfate followed by
washing with saturated sodium bicarbonate and saturat-
ed brine. The solution was dried over sodium sulfate and
evaporated to complete dryness in vacuo to yield 3.85 g
(91% yield) of a clear viscous oil, which crystallized upon
cooling to form a white solid. Mp 29–30 ꢂC; [a]D ꢀ3.8 (c
In two separate 100 mL reaction vessels, 4-aminobutyric
acid (340 mg, 3.3 mmol) and diethylamine (418 lL,
4 mmol) were dissolved in 20 mL methanol. The mix-
tures were warmed to dissolve the 4-aminobutyric acid
and convert it to the diethylamine salt. Then to the first
reaction vessel (R)-pantolactone (390 mg, 3 mmol) was
added and to the second vessel (S)-pantolactone
(390 mg, 3 mmol) was added. The reactions were then
stirred overnight at 60 ꢂC. The reaction mixtures were
evaporated to dryness and dissolved in 20 mL water.
Each mixture was passed over an Amberliteꢁ IR-120
(H+) ion exchange column (2 · 10 cm) and eluted with
water until neutrality was obtained. These solutions
were then washed three times with dichloromethane to
remove any excess pantolactone. The extracted solutions
were then evaporated to dryness to yield the (R)-4-(2,4-
dihydroxy-3,3-dimethyl-butyramido)butyric acid (1)
and (S)-4-(2,4-dihydroxy-3,3-dimethyl-butyramido)bu-
tyric acid (2) products, respectively, as oils. The Ca2+
salts of each product were obtained by dissolving each
in 5 mL methanol with the addition of 0.6 equiv of cal-
cium hydroxide. The solutions were gently warmed to
40 ꢂC and filtered to remove any undissolved solid.
The filtered solutions were evaporated to dryness to
yield each product as a white amorphous solid.
1
1, MeOH; T = 25.5 ꢂC); H NMR (500 MHz, CDCl3) d
5.10 (s, 1H), 4.13 (dd, J = 9.3, 26.8 Hz, 2H), 1.34 (s,
3H), 1.25 (s, 3H); 13C NMR (75 MHz, CDCl3) d 167.9,
120.1, 85.2, 75.2, 40.0, 21.6, 19.0; MS (ESIꢀ) 261.1
(Mꢀ1); LC retention time 6.39 min.
5.6. (R)-3-Azido-4,4-dimethyl-dihydro-furan-2-one (B)
In a 100 mL dry round-bottomed flask, A (2.6 g,
10 mmol) was dissolved in dry DMF and cooled to
0 ꢂC followed by the addition of NaN3 (715 mg,
11 mmol). Under continuous stirring, the mixture was
allowed to warm up to room temperature and stirred
for an additional 2 h. The mixture was evaporated to
dryness in vacuo to remove excess DMF and purified
by flash chromatography using a linear gradient of
petroleum ether/ethyl acetate 30–90% to afford 540 mg
(35% yield) of B as a white crystalline solid. Mp 54–
1
5.3. (R)-4-(2,4-Dihydroxy-3,3-dimethyl-butyrylamino)-
butyric acid (1)
55 ꢂC; [a]D 103.3 (c 1, MeOH; T = 25.7 ꢂC); H NMR
(500 MHz, CDCl3) d 4.01 (dd, J = 9.0, 25.4 Hz, 3H),
1.24 (s, 3H), 1.10 (s, 3H); 13C NMR (75 MHz, CDCl3)
d 172.4, 76.3, 65.9, 40.3, 22.7, 19.5; MS (ESI+) 178.3
(M+23); LC retention time 5.68 min.
The synthesis was performed as detailed in the general
procedure to afford 406 mg (53% yield) of 1 as a white
powder. Mp 62–64 ꢂC; [a]D 33.9 (c 1, MeOH;
1
T = 27 ꢂC); H NMR (500 MHz, D2O) d 3.84 (s, 1H),
5.7. (R)-4-(2-Azido-4-hydroxy-3,3-dimethyl-butyryl-
amino)-butyric acid (C)
3.37 (d, J = 11.2 Hz, 1H), 3.25 (d, J = 11.2 Hz, 1H),
3.09 (t, J = 6.8 Hz, 2H), 2.07 (t, J = 7.8 Hz, 2H), 1.63
(qin, J = 7.3 Hz, 2H), 0.79 (s, 3H), 0.76 (s, 3H); 13C
NMR (75 MHz, D2O) d 182.3, 174.5, 75.5, 67.9, 38.3,
38.1, 34.3, 24.9, 20.1, 18.8; MS (ESIꢀ) 231.9 (Mꢀ1);
LC retention time 4.06 min.
In a 100 mL reaction vessel, 4-aminobutyric acid (365 mg,
3.5 mmol) and diethylamine (418 lL, 4 mmol) were dis-
solved in 20 mL of methanol. The mixture was warmed
to dissolve the 4-aminobutyric acid and to convert it to
the diethylamine salt. Intermediate B (0.5 g, 3.2 mmol)
was then added to the solution and the reaction mixture
was stirred overnight at 60 ꢂC. The reaction mixture was
evaporated to dryness and dissolved in 20 mL water.
The mixture was passed over an Amberliteꢁ IR-120
(H+) ion exchange column (2 · 10 cm) and eluted with
water until neutrality was obtained. The solution was then
extracted three times with dichloromethane to remove
any excess pantolactone and dried in vacuo to afford
400 mg (48% yield) C as a clear oil. [a]D 1.4 (c 1, MeOH;
5.4. (S)-4-(2,4-Dihydroxy-3,3-dimethyl-butyrylamino)-
butyric acid (2)
The synthesis was performed as detailed in the general
procedure to afford 389 mg (51% yield) of 2 as a white
powder. Mp 60–62 ꢂC; [a]D ꢀ31.8 (c 1, MeOH;
1
T = 26.2 ꢂC); H NMR (500 MHz, CH3OD) d 3.95 (s,
1H), 3.49 (d, J = 11.0 Hz, 1H), 3.43 (d, J = 11.0 Hz,
1H), 3.28 (t, J = 6.8 Hz, 2H), 2.26 (t, J = 7.1 Hz, 2H),
1.82 (qin, J = 6.8 Hz, 2H), 0.97 (s, 3H), 0.95 (s, 3H);
13C NMR (75 MHz, D2O) d 181.1, 174.5, 75.5, 67.9,
38.1 (2C), 38.1, 33.5, 24.6, 20.1, 18.8; MS (ESIꢀ) 231.8
(Mꢀ1); LC retention time 3.93 min.
1
T = 24.8 ꢂC); H NMR (500 MHz, CDCl3) d 6.88–6.98
(br s, 1H), 4.04 (s, 1H), 3.60 (t, J = 7.1 Hz, 2H), 3.40 (q,
J = 6.8 Hz, 2H), 2.44 (qin, J = 6.3 Hz, 2H), 1.99 (qin,
J = 7.1 Hz, 2H), 1.28 (s, 6H); 13C NMR (75 MHz, CDCl3)
d 175.9, 169.0, 68.5, 67.5, 38.9, 38.0, 30.9, 24.0, 20.1, 19.2;
MS (ESIꢀ) 256.9 (Mꢀ1); LC retention time 4.35 min.
5.5. (S)-Trifluoro-methanesulfonic acid 4,4-dimethyl-2-
oxo-tetrahydro-furan-3-yl ester (A)
5.8. (R)-4-(2-Amino-4-hydroxy-3,3-dimethyl-butyryl-
amino)-butyric acid (3)
In a 100 mL dry round-bottomed flask, (S)-pantolactone
(2.1 g, 16 mmol) was dissolved in 25 mL dry DCM and
1.63 mL pyridine (20 mmol) under inert conditions. The
reaction was cooled to ꢀ78 ꢂC and triflic anhydride
In a 25 mL reaction vessel, C (330 mg, 1.2 mmol) was
dissolved in a solution of concentrated NH4OH (4 mL)

1016
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
and pyridine (3.5 mL) followed by the addition of tri-
phenylphosphine (0.9 g, 3.4 mmol). The reaction mix-
ture was stirred for 6 h and evaporated to dryness in
vacuo. The crude material was purified by reverse-phase
HPLC with collection set at 215 nm using a linear gradi-
ent of 0.05% TFA in water to 100% acetonitrile. The
appropriate fractions were pooled and dried in vacuo
to afford 267 mg (62% yield) of 3 as a clear oil.
[a]D 21.8 (c 1, MeOH; T = 24.3 ꢂC); 1H NMR
(75 MHz, CDCl3) d 173.6, 170.9, 76.7, 70.2, 39.2,
38.73, 35.3, 34.8, 31.2, 28.9, 28.4, 26.4, 22.0, 20.6, 20.0,
13.5; MS (ESIꢀ) 315.2 (Mꢀ1); LC retention time
5.53 min.
5.12. 2,4-Dihydroxy-3,3-dimethyl N-(2-phenethylcarba-
moyl-ethyl)-butyramide (6)
The synthesis was performed as detailed in the general
procedure to afford 408 mg of product (63% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.54 (t,
J = 6.1 Hz, 1H), 7.28 (t, J = 7.6 Hz, 2H), 7.21 (t,
J = 7.6 Hz, 1H), 7.16 (d, J = 7.1 Hz, 2H), 6.58 (d,
J = 4.9 Hz, 1H), 4.87 (br s, 1H), 4.26 (br s, 1H), 3.97
(s, 1H), 3.54–3.36 (m, 6H), 2.78 (t, J = 7.3 Hz, 2H),
2.37 (t, J = 6.1 Hz, 2H), 0.95 (s, 3H), 0.90 (s, 3H); 13C
NMR (75 MHz, CDCl3) d 173.5, 171.1, 138.2, 128.1
(2), 128.1 (2), 126.0, 76.7, 70.2, 40.3, 38.8, 35.2, 35.0,
34.8, 20.6, 20.0; MS (ESIꢀ) 321.1 (Mꢀ1); LC retention
time 4.92 min.
(500 MHz, CH3OD)
d
3.75 (s, 1H), 3.53 (d,
J = 11.0 Hz, 1H), 3.48 (d, J = 10.7 Hz, 1H), 3.32 (t,
J = 6.8 Hz, 2H), 2.39 (t, J = 7.3 Hz, 2H), 1.85 (qin,
J = 7.1 Hz, 2H), 1.08 (s, 3H), 1.04 (s, 3H); 13C NMR
(75 MHz, D2O) d 181.5, 167.4, 68.8, 59.5, 38.8, 36.0,
34.0, 24.5, 20.8, 18.9; MS (ESI+) 233.1 (M+1); LC reten-
tion time 3.99 min.
5.9. General procedure for preparation of 4–25
To the free-acid form of pantothenate (6.8 mmol for 4
and 5, 2 mmol for 6–16, and 8 mmol for 17–25) dis-
solved in 10 mL dry DMF in discrete reaction vessels
under inert conditions were added diphenylphosphoryl
azide (DPPA, 1.5 equiv for 4–16 and 1.25 equiv for
17–25) and 1 of each of the 22 amines (1.5 equiv for
4–16 and 1.25 equiv for 17–25). The reaction mixtures
were then cooled to 0 ꢂC and triethylamine (1.5 equiv
for 4–16 and 1.25 equiv for 17–25) was added to each.
The reactions were stirred at 0 ꢂC for 2 h, followed by
stirring at room temperature for 12 h. The reaction vol-
umes were reduced under vacuum at 70 ꢂC to remove
dimethylformamide. Each of the crude mixtures was
purified by flash chromatography as previously detailed.
5.13. 2,4-Dihydroxy-3,3-dimethyl N-{2-[2-(3,4-dime-
thoxy-phenyl)-ethylcarbamoyl]-ethyl}-butyramide (7)
The synthesis was performed as detailed in the general
procedure to afford 422 mg of product (55% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.46 (t,
J = 6.1 Hz, 1H), 6.80 (d, J = 8.6 Hz, 1H), 6.71 (d,
J = 6.1 Hz, 2H), 6.23 (t, J = 5.6 Hz, 1H), 4.45 (br d,
J = 4.6 Hz, 1H), 3.97 (br d, J = 4.2 Hz, 1H), 3.90 (br s,
1H), 3.86 (d, J = 4.4 Hz, 6H), 3.58–3.42 (m, 6H), 2.74
(t, J = 7.1 Hz, 2H), 2.39 (t, J = 6.1 Hz, 2H), 0.99 (s,
3H), 0.91 (s, 3H); 13C NMR (75 MHz, CDCl3) d
173.3, 170.9, 148.5, 147.3, 130.7, 120.2, 111.6, 111.0,
76.9, 70.3, 55.4 (2), 40.3, 38.8, 35.2, 34.7, 34.6, 20.8,
19.9; MS (ESIꢀ) 381.1 (Mꢀ1); LC retention time
4.71 min.
5.10. 2,4-Dihydroxy-3,3-dimethyl N-(2-pentylcarbamoyl-
ethyl)-butyramide (4)
The synthesis was performed as detailed in the general
procedure to afford 1.18 g of product (60% yield) as a
viscous oil, which crystallized upon cooling. Mp 72–
5.14. 2,4-Dihydroxy-3,3-dimethyl N-[2-(1-methyl-3-phen-
yl-propylcarbamoyl)-ethyl]-butyramide (8)
74 ꢂC; 1H NMR (500 MHz, CDCl3)
d 7.54 (t,
J = 6.3 Hz, 1H), 6.45 (t, J = 5.5 Hz, 1H), 4.75 (d,
J = 3.3 Hz, 1H), 4.16 (br s, 1H), 3.98 (d, J = 2.6 Hz,
1H), 3.6–3.4 (m, 4H), 3.19 (q, J = 6.2 Hz, 2H), 2.43 (t,
J = 5.9 Hz, 2H), 1.47 (qin, J = 7.3 Hz, 2H), 1.36–1.23
(m, 4H), 0.97 (s, 3H), 0.91 (s, 3H), 0.88 (t, J = 7.0 Hz,
3H); 13C NMR (75 MHz, CDCl3) d 173.6, 170.9, 76.9,
70.3, 39.2, 38.8, 35.3, 34.8, 28.9, 28.5, 21.8, 20.8, 19.9,
13.4; MS (ESIꢀ) 287.2 (Mꢀ1); LC retention time
4.87 min.
The synthesis was performed as detailed in the general
procedure to afford 471 mg of product (67% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.51 (t,
J = 4.6 Hz, 1H), 7.27 (t, J = 7.6 Hz, 2H), 7.20–7.13 (m,
3H), 6.12 (t, J = 8.3 Hz, 1H), 4.59 (t, J = 5.1 Hz, 1H),
4.10–3.94 (m, 3H), 3.60–3.40 (m, 4H), 2.62 (t,
J = 8.3 Hz, 2H), 2.42–2.30 (m, 2H), 1.81–1.68 (m, 2H),
1.14 (d, J = 6.6 Hz, 3H), 0.97 (s, 3H), 0.91 (s, 3H); 13C
NMR (75 MHz, CDCl3) d 173.4, 170.2, 141.1, 127.9
(2), 127.7 (2), 125.4, 76.9, 70.3, 44.8, 38.8, 37.8, 35.4,
34.8, 31.9, 20.8, 20.3, 20.0; MS (ESIꢀ) 349.2 (Mꢀ1);
LC retention time 5.30 min.
5.11. 2,4-Dihydroxy-3,3-dimethyl N-(2-heptylcarbamoyl-
ethyl)-butyramide (5)
The synthesis was performed as detailed in the general
procedure to afford 1.54 g of product (71% yield) as a
viscous oil, which crystallized upon cooling. Mp 74–
5.15. 2,4-Dihydroxy-3,3-dimethyl N-[2-(3,4,5-trimethoxy-
benzylcarbamoyl)-ethyl]-butyramide (9)
76 ꢂC; 1H NMR (500 MHz, CDCl3)
d
7.50 (t,
The synthesis was performed as detailed in the general
procedure to afford 480 mg of product (60% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.54 (t,
J = 6.1 Hz, 1H), 7.12 (t, J = 5.4 Hz, 1H), 6.47 (s, 2H),
4.87 (d, J = 4.2 Hz, 1H), 4.27 (d, J = 5.8 Hz, 3H), 3.94
(d, J = 4.4 Hz, 1H), 3.81 (s, 6H), 3.78 (s, 3H),
J = 5.8 Hz, 1H), 6.21 (br s, 1H), 4.47 (br s, 1H), 4.00
(d, J = 5.1 Hz, 1H), 3.91 (br s, 1H), 3.62–3.42 (m, 4H),
3.20 (q, J = 6.5 Hz, 2H), 2.43 (t, J = 5.8 Hz, 2H), 1.48
(t, J = 6.9 Hz, 2H), 1.34–1.22 (m, 4H), 0.99 (s, 3H),
0.92 (s, 3H), 0.88 (t, J = 7.2 Hz, 3H); 13C NMR

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1017
3.58–3.44 (m, 2H), 3.40 (q, J = 8.6 Hz, 2H), 2.44 (t,
J = 6.1 Hz, 2H), 0.92 (s, 3H), 0.86 (s, 3H); 13C NMR
(75 MHz, CDCl3) d 173.5, 170.9, 152.7 (2), 136.5,
133.4, 104.5 (2), 76.6, 70.0, 60.2, 55.5 (2), 43.3, 38.6,
35.1, 34.7, 20.5, 19.9; MS (ESIꢀ) 397.1 (Mꢀ1); LC
retention time 4.62 min.
(dq, J = 11.2, 5.4 Hz, 2H), 2.60–2.50 (m, 1H), 2.44 (t,
J = 6.1 Hz, 2H), 2.44–2.36 (m, 1H), 1.94 (qin,
J = 2.9 Hz, 1H), 1.86–1.74 (m, 2H), 1.50 (ddd,
J = 13.9, 6.1, 2.2 Hz, 1H), 1.22 (s, 3H), 1.08 (d,
J = 7.3 Hz, 3H), 1.03 (s, 3H), 1.00 (s, 3H), 0.92 (s,
3H), 0.86 (d, J = 9.8 Hz, 1H); 13C NMR (75 MHz,
CDCl3) d 173.5, 170.2, 76.9, 70.3, 47.5, 47.2, 45.2,
41.0, 38.8, 37.9, 36.4, 35.5, 34.8, 34.5, 27.4, 22.8, 20.8,
20.2, 19.9; MS (ESIꢀ) 353.2 (Mꢀ1); LC retention time
5.63 min.
5.16. 2,4-Dihydroxy-3,3-dimethyl N-[3-(4-benzyl-pipera-
zin-1-yl)-3-oxo-propyl]-butyramide (10)
The synthesis was performed as detailed in the general
procedure to afford 402 mg of product (53% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.42 (t,
J = 5.9 Hz, 1H), 7.35–7.25 (m, 5H), 4.01 (br s, 1H),
3.99 (s, 1H), 3.66–3.52 (m, 5H), 3.52 (s, 2H), 3.47 (s,
2H), 3.43 (s, 2H), 2.54 (t, J = 5.6 Hz, 2H), 2.47–2.36
(m, 4H), 1.00 (s, 3H), 0.90 (s, 3H); 13C NMR
(75 MHz, CDCl3) d 172.9, 169.3, 136.7, 128.6 (2),
127.8 (2), 126.8, 77.0, 70.4, 62.2, 52.3, 52.0, 44.9, 41.2,
38.9, 34.3, 32.1, 20.9, 19.8; MS (ESIꢀ) 376.3 (Mꢀ1);
LC retention time 4.98 min.
5.20. 2,4-Dihydroxy-3,3-dimethyl N-(2-propylcarbamoyl-
ethyl)-butyramide (14)
The synthesis was performed as detailed in the general
procedure to afford 453 mg of product (87% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.62 (t,
J = 6.4 Hz, 1H), 6.66 (t, J = 5.6 Hz, 1H), 4.03 (s, 1H),
3.64–3.46 (m, 4H), 3.21 (q, J = 6.6, 2H), 2.48 (t,
J = 6.4, 2H), 1.55 (sextet, J = 7.3 Hz, 2H), 1.01 (s, 3H),
0.96 (s, 3H), 0.95 (s, 3H); 13C NMR (75 MHz, CDCl3)
d 173.5, 171.0, 76.8, 70.2, 40.9, 38.8, 35.3, 34.8, 22.1,
20.7, 19.9, 10.8; MS (ESIꢀ) 259.1 (Mꢀ1); LC retention
time 4.26 min.
5.17. 2,4-Dihydroxy-3,3-dimethyl N-[2-(2-morpholin-4-
yl-ethylcarbamoyl)-ethyl]-butyramide (11)
The synthesis was performed as detailed in the general
procedure to afford 386 mg of product (58% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.60 (t,
J = 5.9 Hz, 1H), 6.83 (s, 1H), 3.99 (s, 1H), 3.72 (t,
J = 4.4 Hz, 4H), 3.54 (q, J = 5.9 Hz, 2H), 3.48 (d,
J = 11.2 Hz, 1H), 3.42 (d, J = 11.0 Hz, 1H), 3.35 (qin,
J = 6.6 Hz, 2H), 2.56–2.41 (m, 8H), 0.96 (s, 3H), 0.92
(s, 3H); 13C NMR (75 MHz, CDCl3) d 173.6, 171.2,
76.3, 69.8, 66.0 (2), 56.7, 52.8 (2), 38.7, 35.3, 35.2,
34.8, 20.5, 20.1; MS (ESIꢀ) 330.2 (Mꢀ1); LC retention
time 4.01 min.
5.21. 2,4-Dihydroxy-3,3-dimethyl N-(2-isobutylcarba-
moyl-ethyl)-butyramide (15)
The synthesis was performed as detailed in the general
procedure to afford 423 mg of product (77% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.65 (t,
J = 6.1 Hz, 1H), 7.04 (t, J = 5.9 Hz, 1H), 5.23 (br s,
1H), 4.61 (br s, 1H), 3.98 (s, 1H), 3.6–3.36 (m, 4H),
3.01 (t, J = 6.4 Hz, 2H), 2.45 (t, J = 6.1 Hz, 2H), 1.74
(septet, J = 6.6 Hz, 1H), 0.94 (s, 3H), 0.90 (s, 3H), 0.88
(d, J = 6.6 Hz, 6H); 13C NMR (75 MHz, CDCl3) d
173.6, 171.1, 76.5, 70.0, 46.5, 38.7, 35.2, 34.9, 27.8,
20.9, 20.0, 19.6 (2); MS (ESIꢀ) 273.2 (Mꢀ1); LC reten-
tion time 4.52 min.
5.18. 2,4-Dihydroxy-3,3-dimethyl N-[2-(4-methoxy-ben-
zylcarbamoyl)-ethyl]-3,3-dimethyl-butyramide (12)
The synthesis was performed as detailed in the general
procedure to afford 496 mg of product (73% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.53 (t,
J = 6.1 Hz, 1H), 7.14 (d, J = 8.6 Hz, 2H), 6.94 (t,
J = 5.4 Hz, 1H), 6.82 (d, J = 8.6 Hz, 2H), 4.82 (d,
J = 2.9 Hz, 1H), 4.27 (d, J = 5.6 Hz, 2H), 4.23, (br s,
1H), 3.93 (d, J = 3.4 Hz, 1H), 3.76 (s, 3H), 3.49 (qin,
J = 6.1 Hz, 2H), 3.42 (d, J = 11.0 Hz, 1H), 3.37 (d,
J = 11.2 Hz, 1H), 2.41 (t, J = 6.1 Hz, 2H), 0.92 (s, 3H),
0.87 (s, 3H); 13C NMR (75 MHz, CDCl3) d 173.6,
170.9, 158.4, 129.6, 128.5 (2), 113.5 (2), 76.6, 70.0,
54.7, 42.4, 38.7, 35.1, 34.8, 20.5, 20.0; MS (ESIꢀ) 337.1
(Mꢀ1); LC retention time 4.79 min.
5.22. 2,4-Dihydroxy-3,3-dimethyl N-[2-(3,7-dimethyl-
octa-2,6-dienylcarbamoyl)-ethyl]-butyramide (16)
The synthesis was performed as detailed in the general
procedure to afford 326 mg of product (46% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.54 (t,
J = 6.1 Hz, 1H), 6.09 (t, J = 4.6 Hz, 1H), 5.21 (t,
J = 7.1 Hz, 1H), 5.12 (t, J = 6.8 Hz, 1H), 4.05 (s, 1H),
3.87 (t, J = 6.1 Hz, 2H), 3.68–3.49 (m, 4H), 2.48 (t,
J = 6.1 Hz, 2H), 2.16–2.02 (m, 4H), 1.73 (s, 3H), 1.65
(s, 3H), 1.05 (s, 3H), 0.97 (s, 3H); 13C NMR (75 MHz,
CDCl3) d 173.1, 170.7, 139.7, 131.3, 123.2, 118.9, 77.0,
70.4, 38.9, 38.8, 37.1, 35.2, 34.8, 25.9, 25.1, 20.9, 19.9,
17.1, 15.7; MS (ESIꢀ) 353.2 (Mꢀ1); LC retention time
5.75 min.
5.19. 2,4-Dihydroxy-3,3-dimethyl N-[2-(2,6,6-trimethyl-
bicyclo[3.1.1]hept-3-ylcarbamoyl)-ethyl]-butyramide (13)
5.23. 2,4-Dihydroxy-3,3-dimethyl N-[2-(2-methylsulfanyl-
ethylcarbamoyl)-ethyl]-butyramide (17)
The synthesis was performed as detailed in the general
procedure to afford 491 mg of product (69% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.55 (t,
J = 6.1 Hz, 1H), 6.23 (d, J = 8.3 Hz, 1H), 4.59 (br s,
1H), 4.24 (qin, J = 7.9 Hz, 1H), 4.04 (br s, 1H), 4.01
(d, J = 4.9 Hz, 1H), 3.56 (qin, J = 6.1 Hz, 2H), 3.48
The synthesis was performed as detailed in the general
procedure to afford 394 mg of product (17% yield) as a
viscous oil, which crystallized upon cooling. Mp 61–
63 ꢂC; 1H NMR (500 MHz, CDCl3)
d 7.48 (t,

1018
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
J = 6.1 Hz, 1H), 6.58 (t, J = 5.6 Hz, 1H), 4.45 (br s, 1H),
3.99 (s, 1H), 3.89 (br s, 1H), 3.62–3.52 (m, 2H), 3.52–
3.38 (m, 4H), 2.63 (dt, J = 1.0, 6.6 Hz, 2H), 2.46 (t,
J = 6.1 Hz, 2H), 2.11 (s, 3H), 0.99 (s, 3H), 0.92 (s,
3H); 13C NMR (75 MHz, CDCl3) d 173.2, 171.0, 77.0,
70.3, 38.8, 37.3, 35.3, 34.7, 33.2, 20.9, 19.9, 14.4; MS
(ESIꢀ) 291.0 (Mꢀ1); LC retention time 4.28 min.
CDCl3) d 173.6, 171.1, 76.6, 70.1, 38.7, 38.1, 35.3,
34.8, 31.0, 28.0, 20.6, 20.0, 14.9; MS (ESI+) 329.1
(M+23); LC retention time 4.45 min.
5.28. 2,4-Dihydroxy-3,3-dimethyl N-[2-(3-ethoxy-pro-
pylcarbamoyl)-ethyl]-butyramide (22)
The synthesis was performed as detailed in the general
procedure to afford 1.21 g of product (48% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.50 (t,
J = 6.1 Hz, 1H), 6.67 (t, J = 4.9 Hz, 1H), 4.52 (d,
J = 5.4 Hz, 1H), 4.00 (d, J = 5.4 Hz, 1H), 3.96 (t,
J = 5.9 Hz, 1H), 3.61–3.43 (m, 8H), 3.34 (qin,
J = 5.7 Hz, 2H), 2.41 (t, J = 6.1 Hz, 2H), 1.77 (qin,
J = 5.9 Hz, 2H), 1.20 (t, J = 7.1 Hz, 3H), 1.00 (s, 3H),
0.92 (s, 3H); 13C NMR (75 MHz, CDCl3) d 173.2,
170.8, 76.9, 70.3, 68.9, 65.9, 38.8, 37.7, 35.2, 34.8, 28.4,
20.9, 19.9, 14.6; MS (ESI+) 327.2 (M+23); LC retention
time 4.33 min.
5.24. 2,4-Dihydroxy-N-[2-(2-methoxy-ethylcarbamoyl)-
ethyl]-3,3-dimethyl-butyramide (18)
The synthesis was performed as detailed in the general
procedure to afford 197 mg of product (9% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.67 (t,
J = 6.1 Hz, 1H), 7.28 (t, J = 5.4 Hz, 1H), 5.26 (d,
J = 5.1 Hz, 1H), 4.63 (t, J = 5.6 Hz, 1H), 3.98 (d,
J = 5.4 Hz, 1H), 3.57–3.36 (m, 8H), 3.34 (s, 3H), 2.46
(t, J = 6.4 Hz, 2H), 0.94 (s, 3H), 0.91 (s, 3H); 13C
NMR (75 MHz, CDCl3) d 173.6, 171.3, 76.3, 70.4,
69.8, 58.0, 38.7, 38.6, 35.1, 34.8, 20.5, 20.0; MS (ESIꢀ)
275.0 (Mꢀ1); LC retention time 3.98 min.
5.29. 2,4-Dihydroxy-N-[2-(3-hydroxy-propylcarbamoyl)-
ethyl]-3,3-dimethyl-butyramide (23)
5.25. 2,4-Dihydroxy-N-[2-(2-hydroxy-ethylcarbamoyl)-
ethyl]-3,3-dimethyl-butyramide (19)
The synthesis was performed as detailed in the general
procedure to afford 660 mg of product (25% yield) as a
viscous oil. 1H NMR (500 MHz, CD3OD) d 3.91 (s,
1H), 3.61 (t, J = 6.4 Hz, 2H), 3.56–3.44 (m, 2H), 3.49
(d, J = 10.7 Hz, 1H), 3.41 (d, J = 11.0 Hz, 1H), 3.28 (t,
J = 6.8 Hz, 2H), 2.45 (t, J = 6.8 Hz, 2H), 1.73 (qin,
J = 6.8 Hz, 2H), 0.94 (s, 6H); 13C NMR (75 MHz,
CD3OD) d 174.1, 171.9, 75.4, 68.5, 58.5, 38.4, 35.5,
34.5, 34.5, 31.2, 19.4, 19.0; MS (ESIꢀ) 275.0 (Mꢀ1);
LC retention time 3.80 min.
The synthesis was performed as detailed in the gener-
al procedure to afford 710 mg of product (33% yield)
1
as a viscous oil. H NMR (500 MHz, CD3OD) d 3.91
(s, 1H), 3.61 (t, J = 5.6 Hz, 2H), 3.56–3.38 (m, 4H),
3.31 (t, J = 5.9 Hz, 2H), 2.46 (t, J = 6.6 Hz, 2H),
0.94 (s, 6H); 13C NMR (75 MHz, CD3OD) d 174.1,
172.1, 75.4, 68.4, 59.6, 41.0, 38.4, 34.6, 34.5, 19.4,
19.0; MS (ESIꢀ) 261.0 (Mꢀ1); LC retention time
3.71 min.
5.30. 2,4-Dihydroxy-N-[2-(3-methoxy-propylcarbamoyl)-
ethyl]-3,3-dimethyl-butyramide (24)
5.26. 2,4-Dihydroxy-3,3-dimethyl N-[2-(2-ethylsulfanyl-
ethylcarbamoyl)-ethyl]-butyramide (20)
The synthesis was performed as detailed in the general
procedure to afford 1.35 g of product (57% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.45 (t,
J = 5.9 Hz, 1H), 6.45 (br s, 1H), 4.25 (d, J = 5.6 Hz,
1H), 3.99 (d, J = 5.4 Hz, 1H), 3.73 (t, J = 5.9 Hz, 1H),
3.60–3.53 (m, 2H), 3.51–3.44 (m, 4H), 3.40–3.28 (m,
5H), 2.42 (t, J = 6.1 Hz, 2H), 1.77 (qin, J = 6.1 Hz,
2H), 1.01 (s, 3H), 0.93 (s, 3H); 13C NMR (75 MHz,
CDCl3) d 173.3, 170.9, 76.9, 70.9, 70.3, 58.2, 38.8,
37.4, 35.3, 34.8, 28.5, 20.8, 19.9; MS (ESIꢀ) 289.0
(Mꢀ1); LC retention time 4.11 min.
The synthesis was performed as detailed in the general
procedure to afford 709 mg of product (28% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.47 (t,
J = 6.1 Hz, 1H), 6.56 (t, J = 5.6 Hz, 1H), 4.41 (d,
J = 5.4 Hz, 1H), 4.00 (d, J = 5.4 Hz, 1H), 3.86 (t,
J = 5.9 Hz, 1H), 3.62–3.53 (m, 2H), 3.52–3.37 (m, 4H),
2.67 (dt, J = 1.7, 6.8 Hz, 2H), 2.55 (q, J = 7.3 Hz, 2H),
2.46 (t, J = 6.1 Hz, 2H), 1.26 (t, J = 7.3 Hz, 3H), 0.99
(s, 3H), 0.92 (s, 3H); 13C NMR (75 MHz, CDCl3) d
173.3, 171.0, 76.9, 70.3, 38.8, 38.0, 35.3, 34.8, 30.6,
25.0, 20.8, 19.9, 14.2; MS (ESIꢀ) 305.0 (Mꢀ1); LC
retention time 4.53 min.
5.31. 2,4-Dihydroxy-3,3-dimethyl N-[2-(2-ethoxy-ethyl-
carbamoyl)-ethyl]-butyramide (25)
5.27. 2,4-Dihydroxy-3,3-dimethyl N-[2-(3-methylsulfanyl-
propylcarbamoyl)-ethyl]-butyramide (21)
The synthesis was performed as detailed in the general
procedure to afford 1.26 g of product (53% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.52 (t,
J = 6.1 Hz, 1H), 6.64 (t, J = 5.1 Hz, 1H), 4.74 (d,
J = 5.4 Hz, 1H), 4.10 (t, J = 5.6 Hz, 1H), 3.99 (d,
J = 5.4 Hz, 1H), 3.60–3.37 (m, 10H), 2.45 (q,
J = 6.1 Hz, 2H), 1.20 (t, J = 7.1 Hz, 3H), 0.98 (s, 3H),
0.92 (s, 3H); 13C NMR (75 MHz, CDCl3) d 173.4,
171.2, 76.8, 70.1, 68.2, 65.9, 38.8, 38.7, 35.2, 34.8, 20.9,
19.9, 14.5; MS (ESIꢀ) 289.1 (Mꢀ1); LC retention time
4.18 min.
The synthesis was performed as detailed in the general
procedure to afford 863 mg of product (34% yield) as a
viscous oil. 1H NMR (500 MHz, CDCl3) d 7.62 (t,
J = 6.1 Hz, 1H), 7.08 (t, J = 5.6 Hz, 1H), 5.07 (d,
J = 5.1 Hz, 1H), 4.46 (t, J = 5.4 Hz, 1H), 3.99 (d,
J = 4.9 Hz, 1H), 3.60–3.39 (m, 4H), 3.31 (q,
J = 6.8 Hz, 2H), 2.52 (t, J = 7.1 Hz, 2H), 2.45 (t,
J = 6.4 Hz, 2H), 2.09 (s, 3H), 1.80 (qin, J = 7.1 Hz,
2H), 0.96 (s, 3H), 0.92 (s, 3H); 13C NMR (75 MHz,

K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
1019
5.32. Assay of PanK
centration) to reaction mixtures containing Tris–HCl
(50 mM, pH 7.5), ATP (5 mM), MgCl2 (5 mM), DTT
(2 mM), and EcPanK + CoaDE (0.6 lg each) of in a to-
tal volume of 40 lL. Control reactions were set up with-
out EcPanK. Reaction mixtures were incubated at 37 ꢂC
for 30 min and then stopped by placing the tubes in boil-
ing water for 5 min. Precipitated proteins were pelleted
by centrifugation (14,000 rpm, 5 min) and the superna-
tants were filtered through 0.2 lm Spin-X centrifuge
tube filters (Costar) before being injected (20 lL) onto
the HPLC column. Products were eluted at 0.8 mL/
min with K2HPO4 (100 mM, pH 7.0 at 4 ꢂC) and an
increasing concentration of MeOH from 5% to 60% in
16.6 min, followed by isocratic elution with 60% MeOH
from t = 16.6 to 21.8 min. UV–vis products were detect-
ed at 254 nm. ATP, CoA, and dephospho-CoA stan-
dards were eluted at 3.83, 7.81, and 10.56 min,
respectively. Single products were detected in complete
reaction mixtures containing compounds 17 and 19–25
with retention times of 8.93, 5.89, 10.46, 10.08, 9.78,
6.61, 8.54, and 8.71 min, respectively. Analysis of the
reaction mixture containing 18 revealed the formation
of a major product with a retention time of 7.46 min
and of a secondary product eluted at 10.03 min. Omis-
sion of CoaE from this reaction yielded only the product
at 10.03 min and allowed its identification as the dephos-
pho-CoA analogue derived from 21. N5- and N7-CoA
were eluted at 12.67 and 16.86 min, respectively. No
product was detected in the control reaction mixtures
lacking EcPanK. The formation of CoA analogues de-
rived from compounds 4, 5, and 17–25 was monitored
by HPLC at 4 ꢂC on an AKTA Explorer 10S system
(Amersham Biosciences) equipped with a micromolar
RPC C2/C18 ST 4.6/100 column (Amersham
Biosciences).
The standard PanK assays contained D-[1-14C]panto-
thenate (45 lM; specific activity 55 mCi/mmol), ATP
(100 lM), MgCl2 (10 mM), Tris–HCl (0.1 M, pH 7.5),
and PanK in a total volume of 40 lL.19,20 The com-
pounds were dissolved in DMSO and included in the
assay at 100 lM. Control assays included the equal vol-
ume of DMSO solvent. A panel of PanK enzymes was
used to test the compounds: purified E. coli PanK
(100 ng), purified S. aureus PanK (20 ng), and purified
A. nidulans PanK (200 ng) and mPanK1a (40 lg cellu-
lar lysate). The mixture was incubated at 37 ꢂC for
10 min. The reaction was stopped by adding 4 lL of
10% (v/v) acetic acid to the mix. Then 40 lL of the mix-
ture was deposited onto a Whatmann DE81 ion ex-
change filter disk that was washed in three changes of
1% acetic acid in 95% ethanol (25 mL/disk; 20 min
wash) to remove unreacted pantothenate. 40-phospho-
pantothenate was quantitated by counting the dried
disk in 3 mL of scintillation solution. A second assay
was used to determine the phosphorylation of the pan-
tothenate analogues by PanK. The reaction mix con-
tained pantothenate or pantothenate analogue,
[c-32P]ATP (0.25 mM; specific activity 1 Ci/mmol),
MgCl2 (10 mM), Tris–HCl (0.1 M, pH 7.5), and PanK.
After incubation at 37 ꢂC for 20 min, the reaction was
stopped by adding 4 lL of 0.5 M EDTA. A 10 lL ali-
quot of the reaction mixture was spotted onto an acti-
vated silica gel H plate, which was developed with
butanol/acetate/water (5:2:4, v/v/v). The dried plate
was exposed to a storage phosphor screen, and the
phosphorylated product was quantitated using Image-
Quant 5.2 software (Molecular Dynamics).
5.33. Determination of the MIC
The in vitro incorporation of the CoA analogues into
ACP was assayed by pre-incubating compounds 17–25
(1 mM final concentration) or DMSO (negative control)
with EcPanK + CoaDE (0.6 lg each) in a 40 lL reaction
mixture containing Tris–HCl (50 mM, pH 7.5), ATP
(5 mM), MgCl2 (5 mM), and DTT (2.5 mM). Reaction
mixtures were incubated for 2 h at 37 ꢂC and then AcpS
(3.5 lg) and apo-ACP (60 lg) were added to each mix-
ture and the concentration of MgCl2 was adjusted to
55 mM in a total volume of 60 lL. Reactions were incu-
bated at 37 ꢂC for a further 30 min and then analyzed by
electrophoresis on 13% polyacrylamide gels containing
0.5 M urea.22 Compounds 4 and 5 were included as po-
sitive controls.
The MICs of the test compounds against E. coli strain
ANSI and S. aureus strain RN4220 were determined
by a broth microdilution method. ANS1 or RN4220
was grown to mid-log phase in 1% tryptone broth
and then diluted 30,000-fold in the same medium. A
10 lL aliquot of the diluted cell suspension (3000–
5000 colony forming units) was used to inoculate each
well of a 96-well plate (U-bottom with low evaporation
lid) containing 100 lL of tryptone broth with the indi-
cated concentration of inhibitors. The plate was incu-
bated at 37 ꢂC for 20 h before being read with a
FusionTM Universal Microplate Analyzer (Packard,
Canada) at 600 nm. The absorbance was normalized
to the DMSO solvent treated control which was con-
sidered as 100%.
References and notes
5.34. In vitro conversion of compounds 17–25 to CoA
analogues and incorporation into ACP
1. Goldmann, D. A.; Weinstein, R. A.; Wenzel, R. P.;
Tablan, O. C.; Duma, R. J.; Gaynes, R. P.; Schlosser, J.;
Martone, W. J. J. Am. Med. Assoc. 1996, 275, 1544.
2. Levy, S. B.; Marshall, B. Nat. Med. 2004, 10, S122.
3. Rachakonda, S.; Cartee, L. Curr. Med. Chem. 2004,
11775.
4. Schmid, M. B. Antibiot. Dev. Res. 2001, 197.
5. Gerdes, S. Y.; Scholle, M. D.; DÕSouza, M.; Bernal, A.;
Baev, M. V.; Farrell, M.; Kurnasov, O. V.; Daugherty, M.
D.; Mseeh, F.; Polanuyer, B. M.; Campbell, J. W.;
E. coli ACP, holo-ACP synthase (AcpS), and EcPanK
were purified as previously described.11,15,21 E. coli
CoaD and CoaE were expressed as C-terminal and N-
terminal His-tagged proteins, respectively, and purified
in a single step by nickel-affinity chromatography. Stock
solutions (10 or 20 mM) of compounds 4, 5, and 17–25
were prepared in DMSO and added (0.5 mM final con-

1020
K. G. Virga et al. / Bioorg. Med. Chem. 14 (2006) 1007–1020
Anantha, S.; Shatalin, K. Y.; Chowdhury, S. A.; Fonstein,
M. Y.; Osterman, A. L. J. Bacteriol. 2002, 184, 4555.
6. Zhang, Y. M.; Frank, M. W.; Virga, K. G.; Lee, R. E.;
Rock, C. O.; Jackowski, S. J. Biol. Chem. 2004, 279,
50969.
12. Ivey, R. A.; Zhang, Y. M.; Virga, K. G.; Hevener, K.; Lee,
R. E.; Rock, C. O.; Jackowski, S.; Park, H. W. J. Biol.
Chem. 2004, 279, 35622.
13. Clifton, G.; Bryant, S. R.; Skinner, C. G. Arch. Biochem.
Biophys. 1970, 137, 523.
7. Choudhry, A. E.; Mandichak, T. L.; Broskey, J. P.; Egolf,
R. W.; Kinsland, C.; Begley, T. P.; Seefeld, M. A.; Ku, T.
W.; Brown, J. R.; Zalacain, M.; Ratnam, K. Antimicrob.
Agents Chemother. 2003, 47, 2051.
14. Strauss, E.; Begley, T. P. J. Biol. Chem. 2002, 277, 48205.
15. Leonardi, R.; Chohnan, S.; Zhang, Y. M.; Virga, K. G.;
Lee, R. E.; Rock, C. O.; Jackowski, S. J. Biol. Chem. 2005,
280, 3314.
8. Jackowski, S. Biosynthesis of Pantothenic acid and
Coenzyme A. In Escherichia coli and Salmonella typhimu-
rium; Neidhardt, F. C., Curtiss, R., Gross, C. A.,
Ingraham, J. L., Lin, E. C., Low, K. B., Mgasanik, B.,
Reznikoff, W., Riley, M., Schaechter, M., Umbarger, H.
E., Eds.; American Society of Microbiology: Washington
DC, 1996.
9. Rock, C. O.; Calder, R. B.; Karim, M. A.; Jackowski, S.
J. Biol. Chem. 2000, 275, 1377.
10. Calder, R. B.; Williams, R. S.; Ramaswamy, G.; Rock, C.
O.; Campbell, E.; Unkles, S. E.; Kinghorn, J. R.;
Jackowski, S. J. Biol. Chem. 1999, 274, 2014.
16. Jackowski, S.; Zhang, Y. M.; Price, A. C.; White, S. W.;
Rock, C. O. Antimicrob. Agents Chemother. 2002, 46,
1246.
17. Kopelevich, V. M.; Evdokimova, G. S.; Marieva, T. D.;
Shmuilovich, L. M. Pharm. Chem. J. 1971, 5, 534.
18. Freskos, J. N. Synth. Commun. 1994, 24, 557.
19. Rock, C. O.; Karim, M. A.; Zhang, Y. M.; Jackowski, S.
Gene 2002, 291, 35.
20. Rock, C. O.; Park, H. W.; Jackowski, S. J. Bacteriol. 2003,
185, 3410.
21. Rock, C. O.; Cronan, J. E. Anal. Biochem. 1980, 102,
362.
11. Yun, M.; Park, C. G.; Kim, J. Y.; Rock, C. O.; Jackowski,
S.; Park, H. W. J. Biol. Chem. 2000, 275, 28093.
22. Heath, R. J.; Rock, C. O. J. Biol. Chem. 1995, 270,
26538.