Báo cáo khoa học: Deciphering the key residues in Plasmodium falciparum b-ketoacyl acyl carrier protein reductase responsible for interactions with Plasmodium falciparum acyl carrier protein pptx

Deciphering the key residues in Plasmodium falciparum
b-ketoacyl acyl carrier protein reductase responsible
for interactions with Plasmodium falciparum acyl
carrier protein
Krishanpal Karmodiya, Rahul Modak, Nirakar Sahoo, Syed Sajad and Namita Surolia
Molecular Biology and Genetics Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur, Bangalore, India
The human malaria-causing parasite Plasmodium falci-
parum harbors the type II fatty acid synthase (FAS)
[1,2], which is essential for its sustenance and survival.
In contrast to the multifunctional FAS enzyme in the
type I pathway operating in humans [3], the type II
FAS system has discrete enzymes for each step of the
pathway. Type II FAS in P. falciparum is one of the
pathways specific to its ‘plastid’ and has been validated
as a unique target for developing new antimalarials
[4–8].
During the elongation cycle of type II FAS, the
growing acyl chain, i.e. butyryl–acyl carrier protein
(ACP), is elongated successively in each round by two
carbon units by the action of four enzymes acting con-
secutively. First, b-ketoacyl ACP synthase (either FabB
or FabF) elongates the acyl-ACP of the C
n
acyl chain
Keywords
fatty acid synthase; fluorescence; malaria;
protein–protein interactions; surface
plasmon resonance
Correspondence
N. Surolia, Molecular Biology and Genetics
Unit, Jawaharlal Nehru Centre for Advanced

Scientific Research, Jakkur, Bangalore
560064, India
Fax: +91 80 22082766
Tel: +91 80 22082820-21
E-mail:
(Received 14 June 2008, revised 23 July
2008, accepted 25 July 2008)
doi:10.1111/j.1742-4658.2008.06608.x
The type II fatty acid synthase (FAS) pathway of Plasmodium falciparum is
a validated unique target for developing novel antimalarials, due to its
intrinsic differences from the type I pathway operating in humans.
b-Ketoacyl acyl carrier protein (ACP) reductase (FabG) performs the
NADPH-dependent reduction of b-ketoacyl-ACP to b-hydroxyacyl-ACP,
the first reductive step in the elongation cycle of fatty acid biosynthesis. In
this article, we report intensive studies on the direct interactions of Plasmo-
dium FabG and Plasmodium ACP in solution, in the presence and absence
of its cofactor, NADPH, by monitoring the change in intrinsic fluorescence
of P. falciparum FabG (PfFabG) and by surface plasmon resonance. To
address the issue of the importance of the residues involved in strong, spe-
cific and stoichiometric binding of PfFabG to P. falciparum ACP (PfACP),
we mutated Arg187, Arg190 and Arg230 of PfFabG. The activities of the
mutants were assessed using both an ACP-dependent and an ACP-indepen-
dent assay. The affinities of all the PfFabG mutants for acetoacetyl-ACP
(the physiological substrate) were reduced to different extents as compared
to wild-type PfFabG, but were equally active in biochemical assays with
the substrate analog acetoacetyl-CoA. Kinetic analysis and studies of direct
binding between PfFabG and PfACP confirmed the identification of
Arg187 and Arg230 as critical residues for the PfFabG–PfACP interac-
tions. Our studies thus reveal the significance of the positively
charged ⁄ hydrophobic patch located adjacent to the active site cavities of

PfFabG for interactions with PfACP.
Abbreviations
ACP, acyl carrier protein; FabG, b-ketoacyl acyl carrier protein reductase; FAS, fatty acid synthase; PfACP, Plasmodium falciparum acyl
carrier protein; PfFabG, Plasmodium falciparum b-ketoacyl acyl carrier protein reductase; SPR, surface plasmon resonance.
4756 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS
to a C
n +2
, b-ketoacyl form. The b-ketoacyl-ACP thus
formed is reduced to b-hydroxyacyl-ACP by an
NADPH-dependent b-ketoacyl ACP reductase (FabG).
The b-hydroxyacyl group is then dehydrated to an
enoyl-ACP by a b-hydroxyacyl ACP dehydratase
(FabZ or FabA). Reduction of the enoyl group by an
enoyl ACP reductase (FabI, FabK or FabL) finally
produces C
n +2
acyl-ACP, which can either re-enter
the elongation cycle, or be hydrolyzed to ACO and the
acyl moiety for the synthesis of phospholipids or
sphingolipids, or become diverted for other modifica-
tions [9]. All of the enzymes participating in type II
FAS interact, but not much is known about the resi-
dues involved in interactions.
Plasmodium falciparum ACP (PfACP) is a small pro-
tein, with a flexible conformation, which shuttles the
substrates between the enzymes of the pathway.
PfACP is a nuclear-encoded and plastid-targeted pro-
tein of 137 amino acids that includes leader and transit
sequences. Mature PfACP consists of 79 amino acids
(residues 58–137) with a preponderance of acidic resi-

dues [10]. Two-dimensional NMR [11] has revealed
that PfACP has a defined but flexible tertiary structure
dominated by four a-helices located at residues 4–15
(helix I), 37–51 (helix II), 57–61 (helix II¢) and 66–74
(helix III), all connected by loops with a long struc-
tured turn between helix I and helix II. The unusually
mobile structure of ACP can be best represented as a
dynamic equilibrium between two conformers. Highly
mobile portions of PfACP include the loop regions
and helix II.
FabG is highly conserved across species, and is the
only known isoform that functions as a ketoacyl
reductase in the type II FAS system. Recently, the
crystal structure of P. falciparum FabG (PfFabG) has
been solved [12], and suggests that the interactions of
PfFabG with the 4¢-phosphopantetheine moiety of
PfACP are hydrophobic in nature. Plasmodium FabG
with a lone tryptophan provides an ideal system with
which to study ligand-induced conformational changes
by monitoring the change in its intrinsic fluorescence.
In these studies, we have investigated the interactions
between PfFabG and PfACP using a combination of
computational, biochemical and biophysical methods.
We have been able to identify specific surface features
on PfFabG that are critical for these interactions. In
PfFabG, Arg187 and Arg230 are located in a hydro-
phobic patch adjacent to the active site entrance of
PfFabG, whereas Arg190 is located away from the
active site. Hence, to characterize the role played by
these residues in the interactions of PfFabG and

PfACP, we generated the following mutants: R187E,
R230E, R187A ⁄ R230A, R187E ⁄ R230E, R190E,
R190A and R230K. We also generated R187K, as this
is conservatively substituted throughout the apicom-
plexan group (present as Lys187 in other species of
Plasmodium).
In the type II FAS pathway, the growing acyl inter-
mediates are attached to the terminal sulfhydryl of the
4¢-phosphopentatheine prosthetic group [13], which is
attached via a phosphodiester linkage to the Ser37
located at the beginning of helix II. The primary
gene product is an apoprotein that is converted to
holo-ACP (ACP) by the transfer of the 4¢-phosphopen-
tatheine moiety of CoA to Ser37 by holo-ACP syn-
thase. ACP performs two functions: first, it sequesters
the growing acyl chain from the aqueous environment;
and second, upon binding to one of the type II FAS
proteins, it releases its grip on the fatty acid, which is
inserted into the active site of the enzyme. ACPs from
various natural sources share significant primary
sequence similarity, particularly at the prosthetic group
attachment site, extending to helix II. However, the
individual ACP-binding partners do not share any
common ACP-binding motif.
The molecular details that govern the specific inter-
actions between Plasmodium ACP and type II FAS
enzymes are poorly understood. Here, we report subtle
aspects of the interactions between PfFabG and
PfACP, with emphasis on association constants and
number of binding sites with reference to the cofactor

NADPH. The site-directed mutagenesis studies reveal
that both electrostatic and hydrophobic interactions
play important roles in PfFabG–PfACP complex
formation.
Results
Identification of residues putatively involved in
the PfFabG–PfACP interaction
Multiple sequence alignment indicates that the FabG
sequences from different species, including plants and
bacteria, share a high degree of sequence identity
(Fig. 1A). The crystal structures of FabG enzymes
from Escherichia coli [14], Brassica napus [15] and
P. falciparum [12] are also homologous, and show
FabG to be a tetramer consisting of two homodimers
of monomers arranged in a head-to-tail configuration.
The crystal structure of FabG from E. coli shows that
there is a conserved positively charged patch on its
surface [14,16]. This positively charged patch is posi-
tioned at the entrance of the active site and is involved
in recognition of the highly conserved and negatively
charged a2 helix of ACP. This patch is identical in
E. coli FabG, Plasmodium FabG and counterparts
K. Karmodiya et al. Interactions of PfFabG with PfACP
FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS 4757
from plants. Mutagenesis of E. coli FabG showed that
two arginine residues (Arg29 and Arg172) present in
this patch are central to the binding of ACP [16].
Multiple sequence alignment of FabG sequences from
different species shows that these residues are
conserved in PfFabG too (Arg187 and Arg230,

A
B
C
Fig. 1. Multiple sequence alignment of FabG sequences: (A) P. falciparum (P. falc) (accession number PFI1125c), Bacillus subtilis (B. subt)
(accession number AAC44307), Cuphea lanceolata (C. lanc) (accession number P28643), B. napus (accession number CAC41363), and E. coli
(accession number NP_415611). (B) P. falciparum (P. falc) (accession number PFI1125c), Plasmodium berghei (P. ber) (accession number
PB000052.00.0) and Plasmodium chabaudi (P. cha) (accession number PC000242.01.0). *Residues thought to be involved in the FabG–ACP
interaction. Color scheme: conserved residues blocked in gray, negatively charged residues in red characters, positively charged residues in
blue characters, aliphatic residues blocked in yellow, and aromatic residues in green. (C) Electrostatic potential surface of the Plasmodium
FabG adjacent to the active site entrance. Red indicates negative charge, blue indicates positive charge, and white is hydrophobic. Arg187
and Arg230 are located adjacent to the active site entrance.
Interactions of PfFabG with PfACP K. Karmodiya et al.
4758 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS
respectively) (Fig. 1A). Three residues of PfFabG
selected for the mutagenesis studies, namely Arg187,
Agr190 and Arg230, were highly conserved in all spe-
cies of Plasmodium. Analysis of the Plasmodium FabG
crystal structure shows that the conserved residues
Arg187 and Arg230 are located at the surface, near its
active site entrance (Fig. 1C). We replaced the posi-
tively charged arginines with glutamates to introduce
electrostatic repulsion between PfFabG and PfACP
and to test whether PfACP associates with PfFabG
over the entire predicted surface. We also changed
these positively charged residues to alanines to deter-
mine which of the electrostatic interactions are impor-
tant for promoting the binding to PfACP.
Interestingly, within the Plasmodium genus, Arg187 is
substituted (Fig. 1B) with a lysine.
Expression and purification of PfFabG, PfFabG

mutants and PfACP
The recombinant wild-type PfFabG, its mutants and
PfACP were purified to homogeneity using an Ni
2+
–
nitrilotriacetic acid affinity column as previously
described [17,18]. Figure 2 shows the apparent electro-
phoretic homogeneity of the purified proteins. The
purified proteins on SDS ⁄ PAGE yielded a monomeric
M
r
of 31 000 ± 1000 for PfFabG as well as for
PfFabG mutants.
Gel filtration and CD analyses of the mutants
Changes in the overall shape or the quaternary struc-
ture of the molecule, potentially introduced by muta-
genesis, were first probed using size exclusion
chromatography. Wild-type PfFabG was eluted as a
single peak at a volume of 13.67 mL on a Superdex-
200 gel filtration column [17]. The PfFabG mutants
were also eluted at the retention volume of their
wild-type counterpart. The elution positions of the
wild-type and mutants of PfFabG corresponded to a
relative molecular mass of 110 kDa (± 10 kDa),
indicating that the enzymes are homotetramers and
that the mutations did not alter the overall shape or
the quaternary structure of PfFabG. CD spectroscopy
was used to investigate potential perturbations in the
secondary and tertiary structure of PfFabG mutants.
CD spectra of wild-type PfFabG and the PfFabG

mutants were superimposable (Fig. S1), suggesting that
the relative contents of a-helical and b-sheet secondary
structure in the PfFabG mutants are not changed as a
result of the individual point mutations.
Kinetic analyses of the PfFabG mutants
In order to evaluate the effects of the mutations on the
specific activity of PfFabG, we used an ACP-indepen-
dent spectrophotometric assay, where acetoacetyl-CoA
was used as a substrate in place of acetoacetyl-ACP,
and the disappearance of NADPH was monitored at
340 nm. As can be seen in Table 1, the kinetic con-
stants (K
m
and ACP-independent specific activities) of
the R187A, R187E, R230A, R230E, R187A ⁄ R230A
and R187E ⁄ R230E mutants remained largely
unchanged with respect to wild-type PfFabG. Wild-
type PfFabG shows less activity with acetoacetyl-CoA
than with acetoacetyl-ACP. All the mutants exhibited
very poor activity in the ACP-dependent spectroscopic
assay, but not in the ACP-independent spectroscopic
assay, which shows that PfFabG mutants are selec-
tively compromised for utilization of the acyl-ACP
substrate (acetoacetyl-ACP). The R230E and
R187E ⁄ R230E mutants had higher K
m
values of
123
45
kDa

35
25
45678
Fig. 2. Purification of PfFabG mutants by Ni
2+
–nitrilotriacetic acid
chromatography. SDS ⁄ PAGE of recombinant PfFabG and PfFabG
mutants. Lane 1: purified wild-type PfFabG. Lane 2: protein molecu-
lar weight markers (MBI Fermentas). Lane 3: R187A. Lane 4:
R187E. Lane 5: R230A. Lane 6: R230E. Lane 7: R187A ⁄ R230A.
Lane 8: R187E ⁄ R230E.
Table 1. Specific activity of wild-type PfFabG and the PfACP
binding site mutants in ACP-dependent and ACP-independent
spectroscopic assays. Enzyme activity was monitored spectro-
photometrically at 340 nm as described in Experimental proce-
dures. Values in brace are presented as percentages.
PfFabG
K
m
(mM)
AcAcCoA
Specific activity
ACP-independent
spectroscopic
assay (UÆmg
)1
)
Specific activity
ACP-dependent
spectroscopic

assay (UÆmg
)1
)
Wild-type 0.43 59.8 70.6 (100)
R187A 0.45 57.9 6.9 (9.77)
R187E 0.47 54.5 6.3 (8.92)
R230A 0.44 55.3 6.1 (8.64)
R230E 0.49 54.2 4.0 (5.66)
R187A ⁄ R230A 0.47 45.6 3.0 (4.24)
R187E ⁄ R230E 0.58 42.6 2.0 (2.83)
K. Karmodiya et al. Interactions of PfFabG with PfACP
FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS 4759
0.49 mm and 0.57 mm, respectively, than wild-type
PfFabG for acetoacetyl-CoA (0.43 mm) [17]. More-
over, the point mutation R187K gives similar results
to those for wild-type PfFabG.
The ACP-dependent activity assay clearly showed
the involvement of the two surface arginine residues of
PfFabG in the interaction with PfACP. In order to
determine the ability of PfACP to function as the
inhibitor, we used a spectrophotometric assay, utilizing
acetoacetyl-CoA, with the indicated concentrations of
PfACP. As can be seen in Fig. 3, wild-type PfFabG
showed inhibition with increasing concentrations of
PfACP, whereas no inhibition with the R187E, R230E,
R187A and R230A mutants was observed. The effect
was more deleterious when arginine was changed to
glutamate than when it was changed to the neutral res-
idue alanine (Table 2).
Interaction of PfFabG mutants with wild-type

PfACP monitored by measuring intrinsic PfFabG
fluorescence
The intrinsic fluorescence of PfFabG decreased when
it was titrated with increasing concentrations of
PfACP. As reported earlier [17], binding of PfACP to
PfFabG, as analyzed by quenching of its fluorescence
at 334 nm, gave an association constant of 400 nm
)1
with n = 1. The value of K
a
was determined for
other mutants using nonlinear least squares fit of the
data, using the Adair equation with one to four
equivalent and independent, as well as equivalent and
interdependent, binding sites (n). The K
a
values for
binding of the R187A and R230A mutants were,
respectively, 150 and 82 nm
)1
. Thus, mutation of
Arg187 and Arg230 to alanine decreased the binding
affinities by three-fold and five-fold respectively. The
K
a
values for the binding decreased even more dra-
matically to 93 and 9 nm
)1
, respectively, in the
R187E and R230E mutants. Apparently, mutation of

Arg187 and Arg230 to an acidic residue, glutamate,
diminishes the strength of the PfACP–PfFabG inter-
action in a relatively more significant manner than
their replacement by a neutral alanine residue. The
effect was more drastic when both the residues were
converted to glutamate, there being an 80-fold reduc-
tion in association constant (Table 2). The data
shown here are in close agreement with those from
the ACP-dependent assay.
The affinity of wild-type PfACP increased three-fold
(K
a
= 1.10 lm
)1
) in the presence of NADPH, and the
number of binding sites increased from one to two,
whereas in all PfFabG mutants examined except
R187K, the number of binding sites remained the
same, with decreased binding of the PfFabG mutants
to PfACP in the presence of NADPH (Table 2). The
maximum effect was observed on binding of the
R230E mutant and the double mutant R187E ⁄ R230E,
120
90
60
Percent activity
30
0
04080
ACP conc (µ

M)
120 160
Fig. 3. Inhibition of PfFabG activity by PfACP. The ability of PfACP
to function as an inhibitor of the condensing enzyme reaction was
evaluated using the spectrophotometric assay utilizing acetoacetyl-
CoA as described in Experimental procedures with 1 lg of PfFabG
(d)or5lg of the mutants R230E (.), R187E ⁄ R230E (Ñ) and
R187A ⁄ R230A (s). The activities of the mutant enzymes were not
significantly affected by the addition of PfACP, suggesting that the
mutation reduced PfACP binding to PfFabG.
Table 2. Binding constants (K
a
) for interaction of wild-type PfACP
with mutant PfFabG, in the absence and in the presence of
reduced cofactor NADPH at 20 °C, obtained using the changes in
protein fluorescence intensity at 334 nm and SPR. Experimental
details are provided in Experimental procedures. n, number of bind-
ing sites for the best value of r
2
; K
a
, association constant for the
best value of r
2
, determined using protein fluorescence (334 nm);
ND, not determined.
Serial
no. Sample
Titrated
with:

Fluorescence
nK
a
(nM
)1
)
SPR
analysis
K
a
(nM
)1
)
1 Wild-type PfACP 1 400.0 350
2 R187A PfACP 1 150.2 ND
3 R187E PfACP 1 93.4 ND
4 R230A PfACP 1 82.8 42
5 R230E PfACP 1 9.7 23
6 R187A ⁄ R230A PfACP 1 79.2 ND
7 R187E ⁄ R230E PfACP 1 5.0 ND
8 R187K PfACP 1 390 ND
9 PfFabG-NADPH PfACP 2 1100.0 1100
10 R187A-NADPH PfACP 1 350.1 ND
11 R187E-NADPH PfACP 1 270.6 ND
12 R230A-NADPH PfACP 1 210.0 130
13 R230E-NADPH PfACP 1 110.4 160
14 R187A ⁄ R230A-NADPH PfACP 1 205.3 ND
15 R187E ⁄ R230E-NADPH PfACP 1 92.3 ND
16 R187K-NADPH PfACP 2 992.5 ND
Interactions of PfFabG with PfACP K. Karmodiya et al.

4760 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS
with association constants of 110.4 and 92 nm
)1
,
respectively.
Interaction of the PfFabG mutants with wild-type
PfACP monitored by surface plasmon resonance
(SPR)
The PfFabG–PfACP interaction data were further ver-
ified by direct measurement of binding of these two
proteins by SPR (BIAcore). Wild-type PfFabG and its
mutants were immobilized on the nitrilotriacetic acid
sensor chip surface, following the manufacturer’s pro-
tocol [19]. Approximately 330–350 resonance units
(RU) of His-tagged PfFabG was immobilized in each
channel of the sensor chip. A continuous flow of buf-
fer was maintained on the chip surface until a stable
baseline was reached. All of the binding studies were
conducted at 20 °C.
Various concentrations of His-tag cleaved PfACP
(ranging from 1 to 200 lm) were passed over the sur-
face of the chip. The PfFabG immobilized surface
devoid of the PfACP from a previous reaction could
be regenerated by passing the buffer alone. A typical
sensorgram for the binding of varying concentrations
of PfACP to immobilized PfFabG is shown in Fig. 4A.
The occurrence of an enhancement in the RUs is indic-
ative of the increase in mass on the chip surface, which
indicates binding. As shown in Fig. 4B, wild-type
PfFabG showed strong binding, whereas the R230E

and R187A ⁄ R230A mutants did not show any signifi-
cant binding at similar PfACP concentrations. The
apparent binding and association constants are shown
in Table 2. There was a three-fold enhancement in
PfACP binding to wild-type PfFabG in the presence of
NADPH. PfACP binding to the mutant PfFabGs was
also enhanced in the presence of NADPH. There was
a three-fold to 100-fold decrease in the affinity of bind-
ing of mutant PfFabGs to PfACP as compared to
wild-type PfFabG (Table 2). The apparent binding
constants determined by fluorescence quenching and
SPR experiments for PfFabG and PfACP are essen-
tially similar. Fluorescence quenching corroborated
with SPR experiments shows the involvement of these
residues in the PfFabG–PfACP interaction.
Allosteric binding of NADPH to PfFabG mutants
in the presence of PfACP
The K
a
value for the binding of NADPH to PfFabG,
with n = 4, was found to be 40.90 lm
)1
, and the bind-
ing exhibited negative, homotropic cooperativity (with
a Hill constant of n
H
= 0.8) [17]. The association con-
stant was nearly equal for all of the PfFabG mutants,
suggesting that there was no effect of point mutations
on the NADPH binding (data not shown). Further-

more, the affinity (K
a
) of PfFabG for its cofactor
NADPH determined by fluorescence spectroscopy in
the presence of 20 lm PfACP was found to be
48.36 lm
)1
and n = 2. Thus, whereas the number of
cofactor-binding sites decreased in the presence of
PfACP, the affinity of PfFabG for NADPH increased,
indicating a negative, heterotropic cooperative effect of
PfACP upon binding of NADPH. This effect of
A
B
100
80
60
60
[ACP] µM vs FabG WT
[ACP] µ
M vs R230E
[ACP] µ
M vs R187A R230A
100 140 180
0 50 100 200150
220 260
Response (RU)Relative binding (RU)
Time (s)
[ACP] µ
M

40
20
20
0
60
40
20
Fig. 4. (A) Sensorgram depicting the binding of PfACP to wild-type
PfFabG. Varying concentrations of PfACP (5, 10, 20, 50, 100 and
200 l
M)in10mM Hepes buffer, containing 150 mM NaCl and
50 l
M EDTA (pH 7.4), were passed over the immobilized PfFabG at
a flow rate of 20 lLÆmin
)1
. The dissociation was studied subse-
quently by passing the same buffer at a flow rate of 20 lLÆmin
)1
.
(B) Direct binding between PfFabG and PfACP measured by
BIAcore. The binding results with the fluorescence assay were
confirmed with the SPR approach (BIAcore) described in Experi-
mental procedures. Wild-type PfFabG and mutants were immobi-
lized on a nitrilotriacetic acid chip, and wild-type PfACP protein
solutions were pumped across the surface. Wild-type PfFabG (solid
line and solid circles) exhibited a strong binding signal, whereas nei-
ther the R230E mutant (dashed line and inverted triangle) nor the
R187A ⁄ R230A mutant (dotted line and open circle) exhibited signifi-
cant binding to the PfACP chip at similar protein concentrations.
K. Karmodiya et al. Interactions of PfFabG with PfACP

FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS 4761
PfACP was not observed with different PfFabG
mutants, as there was no increase in the affinity (K
a
)
and also the number of NADPH-binding sites
remained at four, corresponding to the number of
subunits (Table 3).
Discussion
Our site-directed mutagenesis, kinetic analysis and
binding studies provide evidence that Arg187 and
Arg230 of PfFabG are involved in the interaction with
PfACP. These studies demonstrate that the highly
conserved second a-helix of PfACP recognizes electro-
positive residues on the surface of PfFabG. Whereas
the consequences of mutating the corresponding resi-
dues in E. coli FabG involved in the binding of E. coli
ACP have been investigated earlier, the number of
binding sites and ACP-induced cooperativity for the
interactions of the cofactor NADPH for these mutants
have been characterized for the first time in studies
reported here for Plasmodium FabG. We have used
ACP-dependent and ACP-independent spectrophoto-
metric assays and a fluorescence assay to estimate the
association constants and number of binding sites of
different mutants of PfFabG and PfACP, as well as
the interactions in the presence of the cofactor
NADPH.
Studies with E. coli FabG have identified a con-
served, positively charged patch on its surface [14,16].

The surface of FabG surrounding the active site tunnel
is electropositive. This positively charged patch is posi-
tioned at the entrance of the active site and is thought
to be involved in the recognition of the highly con-
served, negatively charged a2 helix of PfACP. PfFabG
would therefore be expected to also contain a posi-
tively charged region at the active site entrance capable
of interacting with the negatively charged region of
PfACP. The sequences of PfFabG were compared for
similarity with FabGs from other organisms, using
clustalw [20]. As shown in Fig. 2A, the sequence of
PfFabG is most similar to those of plants and bacteria,
consistent with its evolutionary linkage to a photosyn-
thetic bacterium and its location in the apicoplast of
the parasite. Plasmodium FabG also has the character-
istic sequence motif formed by a triad of Ser196,
Tyr209 and Lys213 residues that is involved in the cat-
alytic mechanisms of the enzymes of the short-chain
alcohol dehydrogenase ⁄ reductase family. This provides
a strong indication of involvement of Arg187 and
Arg230 of PfFabG in its interaction with PfACP, as
these two residues are located at the entrance of the
active site tunnel (Fig. 1C) [12]. The sequence align-
ment of FabG from different Plasmodium species,
however, shows the presence of Lys187 instead of
Arg187, so the effects of mutations of these residues
on the kinetics and the ACP-induced cooperativity for
the binding of NADPH were also examined.
To eliminate the ability of Arg187, Arg190 and
Arg230 to form ionic bonds with PfACP, we mutated

them to alanine or glutamate and demonstrated their
moderate effect on PfFabG specific activities (Table 1).
In contrast, when Arg187 and Arg230 were changed to
an alanine or glutamate, the PfFabG mutants were
severely impaired in the ACP-dependent spectrophoto-
metric assay, with a decrease in activity of more than
90% (Table 2) and refractoriness to PfACP inhibition
(Fig. 3). To confirm the importance of Arg187 and
Arg230, we used a fluorescence assay to directly monitor
the interactions between PfFabG and PfACP. The
PfFabG mutants show three-fold to 80-fold less binding
when compared to wild-type PfFabG (Table 2). These
data suggest that Arg187 and Arg230 are the most
important electropositive residues involved in PfFabG–
PfACP interactions. Furthermore, the effect was more
deleterious when Arg230 was mutated as than when
Arg187 was mutated for interactions with PfACP as
shown by the direct binding assay (Table 2), which
might be the reason for the greater conservation of
Arg230 during evolution (Fig. 1).
In conclusion, our results provide compelling sup-
port for the hypothesis that the highly conserved a2
helix of PfACP recognizes an electropositive ⁄ hydro-
phobic surface feature adjacent to the active site
entrance on PfFabG. Two surface residues, Arg187
and Arg230, located in a hydrophobic patch at the
active site entrance on the PfFabG make contact with
PfACP. The hydrophobic residues on PfACP also
make contacts with the hydrophobic patches on
PfFabG, and in turn help in the proper positioning of

PfACP near the active site.
Table 3. Binding constants (K
a
) of reduced cofactor NADPH for
PfFabG mutants in the presence of PfACP at 20 °C, using the
changes in protein fluorescence intensity at 334 nm. Experimental
details are provided in Experimental procedures. n, number of bind-
ing sites for the best value of r
2
; K
a
, association constant for the
best value of r
2
, determined using protein fluorescence (334 nm).
Serial
no. Sample
Titrated
with: n
K
a
(lM
)1
)
1 PfFabG-PfACP NADPH 2 48.3
2 R187A-PfACP NADPH 4 42.2
3 R187E-PfACP NADPH 4 41.9
4 R230A-PfACP NADPH 4 40.9
5 R230E-PfACP NADPH 4 38.9
6 R187A ⁄ R230A-PfACP NADPH 4 33.5

7 R187E ⁄ R230E-PfACP NADPH 4 32.1
Interactions of PfFabG with PfACP K. Karmodiya et al.
4762 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS
Experimental procedures
Materials
Acetoacetyl-CoA, b-hydroxybutyryl-CoA, b-NADPH, NADP
+
,
imidazole, kanamycin, chloramphenicol and SDS ⁄ PAGE
reagents were obtained from Sigma Chemicals (St Louis,
MO, USA). Protein molecular weight markers were from
MBI Fermentas GmbH (St Leon-Rot, Germany). His-bind-
ing resin and anti-His-tag horseradish peroxidase conju-
gates were obtained from Novagen (Darmstadt, Germany).
Media components were obtained from Difco (Franklin
Lakes, NJ, USA). Hi-Trap desalting and Superdex 200 col-
umns were from Amersham Biosciences (Uppsala, Sweden).
All other chemicals used were of analytical grade.
Strains and plasmids
FabG was cloned into the pET-28a(+) vector (Novagen),
and BL21 (DE3) codon plus (Novagen) was used for the
expression of PfFabG.
Construction of PfFabG mutants
Mutations were introduced into the PfFabG gene in pET–
FabG [17], using an overlap extension PCR method. All
the PfFabG mutants were prepared using the same two
outer primers: FabG-NcoI-F, 5¢-CATG
CCATGGGAAAA
GTTGCTTTAGTAACAG GTGCAGGA-3¢; and FabG
-XhoI-R, 5¢-CCG

CTCGAGAGGTGATAGTCCACCGTCT
ATTACGAAAACTCG-3¢ (with NcoI and XhoI sites
underlined). The internal primers for all the PfFabG
mutants are listed in Table 4. To construct each mutant,
two PCR reactions with pET–FabG as the template,
consisting of one outer primer and the respective internal
primers, were performed, and the products were then
pooled and used as a template for a second PCR using the
outer primers. The PCR products were purified from a 1%
agarose gel using the QIAquick gel extraction kit (Qiagen,
Hilden, Germany) and cloned in the NcoI and XhoI sites of
the pET-28a(+) vector (Novagen). The clone thus obtained
was confirmed by DNA sequencing. Expression and purifi-
cation was as for the wild-type protein as reported earlier
[17].
CD spectroscopy
The correct folding of the mutant proteins was verified by
analysis of their CD spectrum between 200 and 250 nm
(2 nm bandpass) using a JASCO J-810 (Tokyo, Japan)
spectropolarimeter at 20 °C and a 0.1 cm path length
quartz cuvette. The protein concentration was determined
by measuring the absorbance at 280 nm, immediately prior
to collecting the spectrum. The extinction coefficient of His-
tagged PfFabG was 17 690 cm
)1
Æm
)1
, using the ‘peptide
properties calculator’ ( />biotools/proteincalc.html). The measured ellipticity was
converted to molar values for direct comparison of the

mutants with the wild-type protein.
Expression and purification of PfACP
PfACP was purified as described previously [18].
ACP-dependent spectrophotometric assay of
PfFabG activity
PfFabG activity was tested in an ACP-dependent assay
by analyzing the formation of b-hydroxybutyryl-ACP,
measuring the disappearance of its cofactor NADPH,
spectrophotometrically, at 340 nm. The gel reconstitution
assay reported for E. coli [16] could not be used for mea-
suring ACP-dependent PfFabG activity, as the substrate
acetoacetyl-ACP and product b-hydroxybutyryl-ACP were
not separable by electrophoresis under nonreducing condi-
tions. The reaction mixture contained 25 lm PfACP,
1mm b-mercaptoethanol, 65 lm malonyl-CoA, 45 lm ace-
tyl-CoA, 200 lm NADPH, 1 lg of purified E. coli FabD,
0.5 lg of purified PfFabH in 0.1 m sodium phosphate
buffer (pH 7.0), and 0.2–1 lg of PfFabG (or PfFabG
mutant) in a final volume of 95 lL. The PfACP, b-mer-
captoethanol and buffer were preincubated at 37 °C for
30 min to ensure the complete reduction of PfACP. The
substrate for the PfFabG reaction was generated by using
E. coli FabD to transfer the malonyl group from CoA to
PfACP to produce malonyl-ACP, and subsequently
PfFabH to condense acetyl-CoA and malonyl-ACP to
Table 4. Overview of the oligonucleotide primers used to generate
PfFabG mutants. The mutation sites are underlined.
Primer Sequence (5¢-to3¢)
FabGF CATGCCATGGGAAAAGTTGCTTTAGTAACAGGTGCAGGA
FabGR CCGCTCGAGAGGTGATAGTCCACCGTCTATTACGAAAA

CTCG
R187AF TAATAAT
GCGTATGGTCGAATAATTA
R187AR GACCATA
CGCATTATTAATCATTCTT
R187EF TAATAAT
GAATATGGTCGAATAATTA
R187ER GACCATA
TTCATTATTAATCATTCTT
R187KF AATAATAAATACGGCCGAATAATTA
R187KR TAATTATTCGGCCGTATTTATTATT
R230AF AGCTTCA
GCCAATATAACTGTAAATG
R230AR CATTTACAGTTATATT
GGCTGAAGCT
R230EF AGCTTCA
GAAAATATAACTGTAAATG
R230ER CATTTACAGTTATATT
TTCTGAAGCT
R190AF TATGGTGCCATAATTAATATTTCAAGT
R190AR ACTTGAAATATTAATTATGGCACCATA
R190ER TATGGTGAAATAATTAATATTTCAAGT
R190ER ACTTGAAATATTAATTATTTCACCATA
K. Karmodiya et al. Interactions of PfFabG with PfACP
FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS 4763
form b-ketobutyryl-ACP. The reaction was initiated by
the addition of NADPH. A decrease in the absorbance
at 340 nm was recorded for 2 min. The initial rate was
used to calculate the enzyme activity.
ACP-independent spectrophotometric assay of

PfFabG activity
The activity of PfFabG was assayed at 25 °C by monitor-
ing spectrophotometrically the decrease in absorbance at
340 nm due to the oxidation of NADPH to NADP
+
(Jasco V-530 UV–visible spectrophotometer). The standard
reaction mixture in a final volume of 100 lL contained
50 mm sodium phosphate buffer (pH 6.8) containing
0.25 m NaCl, 200 lm NADPH, 0.5 mm acetoacetyl-CoA
and 0.2–0.8 lg of PfFabG [17]. The assay mixture was
preincubated for 5 min at room temperature before the
reaction was initiated by the addition of substrate or
enzyme. Reactions with appropriate blanks were also per-
formed. The kinetic parameters were determined by non-
linear regression analyses. The data were also evaluated by
double reciprocal plots. The ability of PfACP to inhibit
PfFabG activity in the spectrophotometric assay was tested
by incubation of varying concentrations of PfACP with
PfFabG protein at room temperature for 5 min before the
addition of acetoacetyl-CoA to initiate the reaction.
Fluorescence titration of PfFabG–PfACP binding
Equilibrium binding of various ligands to PfFabG was
measured by fluorescence titration [17] at 20 °C using a
Jobin-Yvon Horiba spectrofluorimeter (bandpass of 3 and
5 nm for the excitation and emission monochromator,
respectively). The fluorescence spectrum of PfFabG was
studied by exciting the samples at 280 nm and recording
the emission spectrum in the range 300–400 nm, with an
emission maximum at 334 nm, due to its lone trypto-
phan. Aliquots of 3 lL of PfACP (from the stock solu-

tions of 100 lm) were added to 0.5 lm PfFabG in 3 mm
Hepes (pH 7.5), 100 mm NaCl, 2 mm b-mercaptoethanol
and 10% glycerol. The solution was mixed after the addi-
tion of each aliquot, and the fluorescence intensity in the
range 300–400 nm was recorded as the average of three
readings. Samples were excited at 280 nm. The effect of
NADPH on PfACP binding to PfFabG was studied by
titration of NADPH (3 lL) into 0.5 lm PfFabG (in
3mm Hepes, pH 7.5, 100 mm NaCl, 2 mm b-mercapto-
ethanol and 10% glycerol), from a stock solution of
5mm, and the corresponding decrease in the fluorescence
intensity of its lone tryptophan was monitored at
334 nm. A double reciprocal plot of the fluorescence
intensities and ligand concentrations, from the data
obtained by titration of a fixed concentration of PfFabG
with ligand, gave the fluorescence intensity at infinite
ligand concentration (F
a
).
Correction for the inner filter effect was performed
according to the equation
F
c
¼ F antilog½ðA
ex
þ A
em
Þ=2
where F
c

and F are the corrected and measured fluorescence
intensities, respectively [21], and A
ex
and A
em
are the solu-
tion absorbance values at the excitation and emission wave-
lengths, respectively.
The fluorescence data were fitted by the Adair equation
[22], with number of sites n = 1 to 4, K being the
association constants: for a single site, Y = K[X] ⁄ (1 +
K[X]); for two equivalent and independent sites, Y =
(K
1
[X] + 2K
1
K
2
[X]
2
) ⁄ (1 + K
1
[X] + K
1
K
2
[X]
2
); for two equi-
valent and interdependent sites, Y =(2K

1
[X] + K
1
K
2
[X]
2
) ⁄
(1 + K
1
[X] + K
1
K
2
[X]
2
); for three equivalent and indepen-
dent sites, Y =(K
1
[X] + 2K
1
K
2
[X]
2
+3K
1
K
2
K

3
[X]
3
) ⁄ (1 +
K
1
[X] + K
1
K
2
[X
2
]+K
1
K
2
K
3
[X]
3
); for three equivalent
and interdependent sites, Y =(3K
1
[X] + 2K
1
K
2
[X]
2
+

K
1
K
2
K
3
[X]
3
) ⁄ (1 + K
1
[X] + K
1
K
2
[X]
2
+ K
1
K
2
K
3
[X]
3
); for
four equivalent and independent sites, (K
1
[X] + 2K
1
K

2
[X]
2
+3K
1
K
2
K
3
[X]
3
+4K
1
K
2
K
3
K
4
[X]
4
) ⁄ (1 + K
1
[X] + K
1
K
2
[X]
2
+ K

1
K
2
K
3
[X]
3
+ K
1
K
2
K
3
K
4
[X]
4
); and four equivalent
and interdependent sites, Y =(4K
1
[X] + 3K
1
K
2
[X]
2
+
2K
1
K

2
K
3
[X]
3
+ K
1
K
2
K
3
K
4
[X]
4
) ⁄ (1 + K
1
[X] + K
1
K
2
[X]
2
+
K
1
K
2
K
3

[X]
3
+ K
1
K
2
K
3
K
4
[X]
4
). All calculations were carried
out with sigma plot 2000 software.
Analysis of PfFabG–PfACP interaction by SPR
Biospecific-interaction analysis was performed using a BIA-
core 2000 biosensor system (Amersham Pharmacia Biotech,
Uppsala, Sweden). The immobilization of PfFabG on the
flow cell of a nitrilotriacetic acid sensor chip involved regen-
eration of the sensor surface with a 1 min flow of 350 mm
EDTA in 10 mm Hepes (pH 8.3), 150 mm NaCl and 10%
glycerol at a rate of 20 lLÆmin
)1
on the sensor surface. This
was followed by injection of 500 lm NiCl
2
in 10 mm Hepes
(pH 7.4), 150 mm NaCl, 10% glycerol and 50 lm EDTA for
5 min at a rate of 5 lLÆmin
)1

on the sensor surface. Nearly
70 RU were coupled. Wild-type and mutant PfFabG were
immobilized on different flow cells of the activated sensor
chip by injecting 125 nm protein in 10 mm Hepes (pH 7.4),
150 mm NaCl, 10% glycerol and 50 lm EDTA at a rate of
2 lLÆmin
)1
. Nearly 350 RU were coupled, where 1 RU cor-
responds to an immobilized protein concentration of approx-
imately 1 pg Æmm
)2
. One flow cell was kept as the reference
surface. The reference surface was treated in the same way as
the ligand surface, except that the PfFabG was not passed
over this surface. This is important to normalize the chemis-
tries between the two flow cells. All measurements were car-
ried out in 10 mm Hepes (pH 7.4), 150 mm NaCl, 10%
glycerol and 50 lm EDTA (HBS buffer) [23]. In order to
determine the association rate constant for the binding of
PfACP to the immobilized PfFabG, PfACP without His-tag
was passed over the surfaces at various concentrations at
Interactions of PfFabG with PfACP K. Karmodiya et al.
4764 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS
20 °C in Hepes buffer at a flow rate of 20 lLÆmin
)1
. For the
determination of dissociation rate constant, the same buffer
was passed at a flow rate of 20 lLÆmin
)1
. To study the effect

of NADPH on the PfFabG–PfACP interaction, PfACP was
diluted to the desired concentrations in HBS buffer contain-
ing 200 lm NADPH, and the binding studies were performed
in the same buffer with NADPH.
Evaluation of kinetic parameters
The association (k
1
) and dissociation (k
)1
) rate constants
are obtained by nonlinear fitting of the primary sensorgram
data using bia evaluation software version 3.0. The sensor-
grams were fitted globally to the 1 : 1 Langmuir dissocia-
tion model (Eqn 1) to obtain the k
)1
values:
R
t
¼ R
t
o
e
Àk
À1
ðtÀt
0
Þ
ð1Þ
where R
t

is the response at time t, and R
t0
is the amplitude
of the initial response. The measured k
)1
values were used
to determine the k
1
values using the 1 : 1 Langmuir associa-
tion model (Eqn 2):
R
t
¼ R
max
½1 À e
Àðk
1
Cþk
À1
ÞðtÀt
0
Þ
ð2Þ
where R
max
is the maximum response, and C is the concen-
tration of the analyte in the solution. The ratio of k
1
and k
)1

yields the value of the association constant K
a
(k
1
⁄ k
)1
). v
2
and residual values were used to evaluate the quality of fit
between the experimental data and the binding model [24].
Acknowledgements
We thank the Department of Biotechnology (DBT),
Government of India for their financial support to
N. Surolia. K. Karmodiya acknowledges the CSIR,
Government of India, for a senior research fellowship.
References
1 Surolia N & Surolia A (2001) Triclosan offers protection
against blood stages of malaria by inhibiting enoyl-ACP
reductase of Plasmodium falciparum. Nat Med 7, 167–173.
2 Waller RF, Keeling PJ, Donald RG, Striepen B, Hand-
man E, Lang-Unnasch N, Cowman AF, Besra GS,
Roos DS & McFadden GI (1998) Nuclear-encoded
proteins target to the plastid in Toxoplasma gondii and
Plasmodium falciparum. Proc Natl Acad Sci USA 95,
12352–12357.
3 Witkowski A, Ghosal A, Joshi AK, Witkowska HE,
Asturias FJ & Smith S (2004) Head-to-head arrange-
ment of the subunits of the animal fatty acid synthase.
Chem Biol 11, 1667–1676.
4 Ramya TNC, Surolia N & Surolia A (2002) Survival

strategies of the malarial parasite Plasmodium falcipa-
rum. Curr Sci 83, 101–108.
5 Surolia A, Ramya TNC, Ramya V & Surolia N (2004)
FAS’t inhibition of malaria. Biochem J 383 , 1–12.
6 Sharma S, Ramya TNC, Surolia A & Surolia N (2003)
Triclosan as a systemic antibacterial agent in a mouse
model of acute bacterial challenge. Antimicrob Agents
Chemother 47, 3859–3866.
7 Ramya TN, Mishra S, Karmodiya K, Surolia N &
Surolia A (2007) Inhibitors of nonhousekeeping
functions of the apicoplast defy delayed death in
Plasmodium falciparum. Antimicrob Agents Chemother
51, 307–316.
8 Ramya TN, Karmodiya K, Surolia A & Surolia N
(2007) 15-Deoxyspergualin primarily targets the traffick-
ing of apicoplast proteins in Plasmodium falciparum.
J Biol Chem 282, 6388–6397.
9 White SW, Zheng J, Zhang YM & Rock CO (2005)
The structural biology of type II fatty acid biosynthesis.
Annu Rev Biochem 74, 791–831.
10 Gardner MJ, Tettelin H, Carucci DJ, Cummings LM,
Aravind L, Koonin EV, Shallom S, Mason T, Yu K,
Fujii C et al. (1998) Chromosome 2 sequence of the
human malaria parasite Plasmodium falciparum. Science
282, 1126–1132.
11 Sharma AK, Sharma SK, Surolia A, Surolia N & Sar-
ma SP (2006) Solution structures of conformationally
equilibrium forms of holo-acyl carrier protein (PfACP)
from Plasmodium falciparum provides insight into the
mechanism of activation of ACPs. Biochemistry 45,

6904–6916.
12 Wickramasinghe SR, Inglis KA, Urch JE, Muller S,
Van Aalten DM & Fairlamb AH (2006) Kinetic, inhibi-
tion and structural studies on 3-oxoacyl-ACP reductase
from Plasmodium falciparum, a key enzyme in fatty acid
biosynthesis. Biochem J 393, 447–457.
13 Rock CO, Cronan JE Jr & Armitage IM (1981) Molec-
ular properties of acyl carrier protein derivatives. J Biol
Chem 256, 2669–2674.
14 Price AC, Zhang YM, Rock CO & Stephen WW (2001)
Structure of beta-ketoacyl-[acyl carrier protein] reduc-
tase from Escherichia coli: negative cooperativity and its
structural basis. Biochemistry 40, 12772–12781.
15 Fisher M, Kroon JTM, Martindale W, Stuitje AR,
Slabas AR & Rafferty JB (2000) The X-ray structure of
Brassica napus beta-keto acyl carrier protein reductase
and its implications for substrate binding and catalysis.
Structure 8, 339–347.
16 Zhang YM, Wu B, Zheng J & Rock CO (2003) Key
residues responsible for acyl carrier protein and beta-
ketoacyl-acyl carrier protein reductase (FabG) interac-
tion. J Biol Chem 278, 52935–52943.
17 Karmodiya K & Surolia N (2006) Analyses of co-opera-
tive transitions in Plasmodium falciparum b-ketoacyl-
ACP reductase upon co-factor and acyl carrier protein
binding. FEBS J 273, 4093–4103.
K. Karmodiya et al. Interactions of PfFabG with PfACP
FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS 4765
18 Sharma SK, Modak R, Sharma S, Sharma AK, Sarma
SP, Surolia A & Surolia N (2005) A novel approach for

over-expression, characterization, and isotopic enrich-
ment of a homogeneous species of acyl carrier protein
from Plasmodium falciparum. Biochim Biophys Res
Commun 330, 1019–1026.
19 Nieba L, Nieba-Axmann SE, Persson A, Hamalainen
M, Edebratt F, Hansson A, Lidholm J, Magnusson K,
Karlsson AF & Pluckthun A (1997) BIACORE analysis
of histidine-tagged proteins using a chelating NTA sen-
sor chip. Anal Biochem 252 , 217–228.
20 Higgins D, Thompson J, Gibson T, Thompson JD,
Higgins DG & Gibson TJ (1993) clustal w: improving
the sensitivity of progressive multiple sequence align-
ment through sequence weighting, position-specific gap
penalties and weight matrix choice. Nucleic Acids Res
22, 4673–4680.
21 Sharma SK, Kapoor M, Ramya TNC, Kumar S,
Kumar G, Modak R, Sharma S, Surolia N & Surolia A
(2003) Identification, characterization, and inhibition of
Plasmodium falciparum b-hydroxyacyl-acyl carrier
protein dehydratase (FabZ). J Biol Chem 278 ,
45661–45671.
22 Campanacci V, Lartigue A, Hallberg BM, Jones TA,
Orticoni MTG, Tegoni M & Cambillau C (2003) Moth
chemosensory protein exhibits drastic conformational
changes and cooperativity on ligand binding. Proc Natl
Acad Sci USA 100, 5069–5074.
23 Kapoor M, Mukhi PL, Surolia N, Suguna K & Surolia
A (2004) Kinetic and structural analysis of the increased
affinity of enoyl-ACP (acyl-carrier protein) reductase
for triclosan in the presence of NAD+. Biochem J 381,

725–733.
24 Schuck P (1997) Use of surface plasmon resonance to
probe the equilibrium and dynamic aspects of interac-
tions between biological macromolecules. Annu Rev Bio-
phys Biomol Struct 26, 541–566.
Supporting information
The following supplementary material is available:
Fig. S1. The far-UV CD spectra of wild-type PfFabG
and PfFabG mutants.
This supplementary material can be found in the
online version of this article.
Please note: Wiley-Blackwell are not responsible for
the content or functionality of any supplementary
material supplied by the authors. Any queries (other
than missing material) should be directed to the
corresponding author for the article.
Interactions of PfFabG with PfACP K. Karmodiya et al.
4766 FEBS Journal 275 (2008) 4756–4766 ª 2008 The Authors Journal compilation ª 2008 FEBS