of Plasmodium falciparum acyl carrier proteinRole of the 4¢-phosphopantetheine group in the stability of the holo form of Plasmodium falciparum acyl carrier protein Rahul Modak1, Sharmis
Trang 1of Plasmodium falciparum acyl carrier protein
Role of the 4¢-phosphopantetheine group in the stability of the holo form of Plasmodium falciparum acyl carrier protein
Rahul Modak1, Sharmistha Sinha2and Namita Surolia1
1 Molecular Biology and Genetics Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur, Bangalore, India
2 Molecular Biophysics Unit, Indian Institute of Science, Bangalore, India
Malaria continues to exact the highest mortality and
morbidity rate after tuberculosis ‘The scourge of the
tropics’, malaria is endemic in 100 countries in the
world Approximately 500 million cases of malaria are
reported every year, and 3000 children die of malaria
every day [1] Our recent demonstration of the
occur-rence of the type II fatty acid synthesis (FAS) pathway
in the malaria parasite, Plasmodium falciparum, and
its inhibition by triclosan, an inhibitor of the
rate-determining enzyme of type II FAS, enoyl-acyl carrier protein (ACP) reductase, proved the pivotal role played by this pathway in the survival of the malarial parasite The essential role of fatty acids and lipids in cell growth and differentiation, and the occurrence of
a different type (type I) of fatty acid biosynthetic pathway in the human host from that of the malaria parasite, make this pathway an attractive target for developing antimalarial agents [2,3]
Keywords
apo-ACP; conformational stability; holo-ACP;
isothermal unfolding; 4¢-phosphopantetheine
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 2208282021
E-mail: surolia@jncasr.ac.in
(Received 29 January 2007, revised 15 April
2007, accepted 1 May 2007)
doi:10.1111/j.1742-4658.2007.05856.x
The unfolding pathways of the two forms of Plasmodium falciparum acyl carrier protein, the apo and holo forms, were determined by guanidine hydrochloride-induced denaturation Both the apo form and the holo form displayed a reversible two-state unfolding mechanism The analysis of iso-thermal denaturation data provides values for the conformational stability
of the two proteins Although both forms have the same amino acid sequence, and they have similar secondary structures, it was found that the – DG of unfolding of the holo form was lower than that of the apo form at all the temperatures at which the experiments were done The higher stabil-ity of the holo form can be attributed to the number of favorable contacts that the 4¢-phosphopantetheine group makes with the surface residues by virtue of a number of hydrogen bonds Furthermore, there are several hydrophobic interactions with 4¢-phosphopantetheine that firmly maintain the structure of the holo form We show here for the first time that the interactions between 4¢-phosphopantetheine and the polypeptide backbone
of acyl carrier protein stabilize the protein As Plasmodium acyl carrier pro-tein has a similar secondary structure to the other acyl carrier propro-teins and acyl carrier protein-like domains, the detailed biophysical characterization
of Plasmodium acyl carrier protein can serve as a prototype for the analysis
of the conformational stability of other acyl carrier proteins
Abbreviations
AAS, acyl-ACP synthase; ACP, acyl carrier protein; AcpS, holo-ACP synthase; apo-ACP, Plasmodium falciparum acyl carrier protein (apo form); FAS, fatty acid synthesis; holo-ACP, Plasmodium falciparum acyl carrier protein (holo form); holo-ACP, acyl carrier protein (holo form); LEM, linear extrapolation model; 4¢-PP, 4¢-phosphopantetheine; PfACP, Plasmodium falciparum acyl carrier protein (both apo and halo forms).
Trang 2The type II FAS pathway, found in most bacteria,
plants and the malaria parasite, consists of distinct
enzymes, each catalyzing individual reactions required
to complete successive cycles of fatty acid elongation,
in contrast to the multifunctional enzyme catalyzing all
the steps of the type I FAS pathway [4,5] ACP is an
essential component of both type I and type II fatty
acid synthesis pathways Whereas in the type I FAS
pathway, it is an integral part of the multifunctional
enzyme, it is a discrete entity shuttling acyl groups to
the successive enzymes in the type II FAS pathway
ACP is a small protein of molecular mass 8–10 kDa
It plays essential roles in a myriad of metabolic
path-ways Assorted functions involve fatty acid and lipid
biosynthesis, lipid A formation, membrane-derived
oligosaccharide biosynthesis, and activation of RTX
(repeats in toxin), toxins of Gram-negative bacteria [6–
13] In particular instances, specialized ACPs operate
in restricted pathways such as rhizobial nodulation
signaling, and polyketide and lipoteichoic acid
synthe-sis [11,12]
ACP plays a pivotal role in fatty acid synthesis as
well as in its utilization It carries the growing acyl
chain from one enzyme of the FAS pathway to the
other in a sequential manner Given its crucial roles in
metabolism, the high degree of conservation of ACP’s primary structure is not surprising The three-dimen-sional structure of Escherichia coli ACP is the proto-type of bacterial and plant ACP structures [14–17] The solution structure of ACPs consists of a three-helix bundle and a short fourth three-helix, all connected by loops, with a long, structured turn between helices I and II ACP in its holo form exists in a dynamic equi-librium between the two conformers [14–22]
ACP is synthesized as an apoprotein (apo-ACP) and undergoes post-translational modification by holo-ACP synthase Holo-ACP synthase transfers the 4¢-phosphopantetheine (4¢-PP) group from CoA to a particular serine residue of apo-ACP (Ser37 in PfACP) The growing fatty acid chain is attached
to the terminal sulfhydryl group of the phospho-pantetheine, the only sulfhydryl group in most ACPs All known ACPs (or ACP-like domains) undergo this modification, and all share sequence similarities around the modified serine [22]
PfACP is a protein of 137 residues, inclusive of signal and transit sequences, required for targeting of the protein to the apicoplast The mature protein com-prises 79 amino acids (residues 58–137) [23] Recently, the solution structure of P falciparum holo-ACP
Fig 1 (A) PfACP expression: SDS ⁄ PAGE (15%) showing the elution profile of PfACP with N-terminal His-tag Lane 1: supernatant of isopro-pyl-b- D -thiogalactopyranoside-induced E coli cultures transformed with pET-28a(+)-ACP Lane 2: protein markers; the protein bands corres-pond to 116 kDa, 66.2 kDa, 45 kDa, 35 kDa, 25 kDa, 18.4 kDa, and 14.4 kDa (from top to bottom) Lanes 3–7: different fractions of PfACP eluted at 50 m M imidazole (B) PfACP expression: native PAGE (12%) showing the ratio of holo-ACP and apo-ACP in the eluted fractions from an Ni–nitrilotriacetic acid agarose column Lanes 1–3: different fractions of PfACP eluted at 50 m M imidazole (C) Size exclusion chro-matography profile of PfACP: holo-ACP dimer has been separated from a mixture of apo-ACP and holo-ACP monomers by size exclusion chromatography using a Superdex 75 column (30 cm) equilibrated and eluted with 20 m M Tris (pH 6.5) and 200 m M NaCl Peak 1: holo-ACP dimer Peak 2: mixture of apo-ACP and holo-ACP monomers (D) Separation profile of holo-ACP dimer and apo-ACP and holo-ACP mono-mers: 12% native PAGE showing the separation of holo-ACP dimer from a mixture of apo-ACP and holo-ACP monomers Lane 1: holo-ACP dimer without dithiothreitol Lane 2: holo-ACP dimer with dithiothreitol Lane 3: mixture of apo-ACP and holo-ACP monomers (E) Removal
of His-tag from recombinant PfACP For the cleavage of His-tag, 1 unit of thrombin was used for 1 mg of PfACP at 25 C for 2 h On 12% native PAGE, ACPs with and without His-tag showed significant differences in mobility Lane 1: holo-ACP with His-tag Lane 2: holo-ACP without His-tag Lane 3: mixture of holo-ACP monomer and apo-ACP with His-tag Lane 4: mixture of holo-ACP monomer and apo-ACP with-out His-tag (F) Separation of apo-ACP and holo-ACP by anion exchange chromatography Elution profile of apo-ACP and holo-ACP on a MonoQ HR 5 ⁄ 5 anion exchange column Peak 1: apo-ACP Peak 2: holo-ACP (G) Separation of apo-ACP and holo-ACP; 12% native PAGE showing the separation of apo-ACP and holo-ACP by anion exchange chromatography Lane 1: mixture of apo-ACP and holo-ACP Lane 2: purified apo-ACP Lane 3: purified holo-ACP (H) Dynamic light-scattering data of PfACP (a) Particle size distribution of apo-ACP The solid lines indicate the accumulation percentages of particles (b) Particle size distribution of holo-ACP The solid lines indicate the accumulation percentages of particles (I) Sucrose density gradient sedimentation analysis Forty micrograms of apo-ACP and holo-ACP were layered on top of a 4 mL continuous 0–10% (w ⁄ v) sucrose density gradient, and this was followed by centrifugation, fractionation and 12% native PAGE, as described in Experimental procedures Protein bands were visualized by silver staining (a) Lane 1: apo-ACP Lane 2: holo-ACP Lanes 3–9: fractions 18–12 of sucrose density gradient for apo-ACP Lanes 10–15: fractions 18–13 of sucrose density gradient for holo-ACP (b) Lanes 1, 2 and 3, respectively, are fractions 16–18 of sucrose density gradient for holo-ACP under oxidizing conditions (c) Apo-ACP (O) and holo-ACP (h) in each fraction was quantified by measuring the intensity of the silver-stained protein bands using QUANTITY ONE software and plotted against the fraction number (AU, arbitrary unit) (d) The apparent molecular masses of apo-ACP and holo-ACP were estimated on the basis of the linear regression of the fraction number of the molecular mass markers cytochrome c (CyC), carbonic anhydrase (CA), and BSA From the calibration curve of the sucrose density gradient, the estimated molecular masses of apo-ACP, ACP monomer and holo-ACP dimer are 16.75 kDa, 21 kDa and 26.5 kDa, respectively.
Trang 3(holo-ACP) has been solved by Sharma et al and it is
found to exist in conformational equilibrium between
the two states [24,25] These two states have been
iden-tified as the major and the minor forms of the
holo-ACP structure, on the basis of their percentage
contri-butions (65% and 35%, respectively) to the overall
structure of the protein The structures of the major
and minor conformations of holo-ACP bear close
resemblance to that of E coli butyryl-ACP, with rmsd
values of 2.24 A˚ and 2.19 A˚, when superimposed on
their backbone atoms
In the present study, we report the detailed
biophysi-cal characterization of both apo-ACP and holo-ACP
to ascertain their conformational stabilities An
inter-esting outcome of the study, reported for the first time,
for this large family of essential proteins, is that the 4¢-PP prosthetic group imparts considerably higher conformational stability (– DG) to holo-ACP as com-pared to its apo-ACP counterpart
Results
Expression and purification of ACP The mature PfACP (without the signal and transit sequence) was expressed in E coli BL21 (DE3) cells with an N-terminal His-tag PfACP was purified by Ni–nitrilotriacetic acid agarose affinity chromatogra-phy to homogeneity, as shown in Fig 1A The purified protein on 15% SDS⁄ PAGE gel has a monomeric
A
E
B
Holo-ACP dimer
1 2 3
1 2 3
Holo-ACP dimer 1
2
Retention time (min) Apo-ACP with his-tag
Apo-ACP Holo-ACP with his-tag
Holo-ACP
A280
Retention time (min)
Holo-ACP
Holo-ACP
Apo-ACP
1
1 2
CyC Apo Holo CA
BSA
2.0 1.8 1.6 1.4 1.2 1.0
18 16 14 12 10 8
Fraction no.
Log (M.W.)
6 4 2 0
1800
1600
1400
1200
1000
800
600
400
200
0
10
3 4 5 6 7
1
14.4
18.4
25
35
45
66.2
116
2 3 4 5 6 7
8 9 10 11 12 13 14 15 1 2 3
2 3
Holo-ACP
A280
Trang 4molecular mass of 9 kDa The ratio of holo-ACP
and apo-ACP was determined to be in the range of
±50% by 12% native PAGE (Fig 1B)
Heterologously expressed PfACP is partly converted
to holo-ACP by E coli holo-ACP synthase Holo-ACP
forms a disulfide-bonded dimer through the thiol
group of phosphopantetheine in a nonreducing
envi-ronment The holo-ACP dimer was separated from the
mixture of apo-ACP and holo-ACP monomers by size
exclusion chromatography under nonreducing
condi-tions (Fig 1C,D) From the calibration curve for the
Superdex 75 column, with standard globular proteins,
the apparent molecular mass of holo-ACP dimer and
the mixture of holo-ACP monomer and apo-ACP were
found to be 33 kDa and 25 kDa, respectively
Purified holo-ACP and the mixture of apo-ACP and
holo-ACP monomer were subjected to thrombin
clea-vage to remove the histidine tag from the protein
Approximately 90% ACP cleavage was achieved, and
uncleaved ACP was removed by passage through an
Ni–nitrilotriacetic acid affinity column (Fig 1E)
Apo-ACP and holo-Apo-ACP monomers from the mixture were
purified by anion exchange chromatography using a
Mono Q HR 5⁄ 5 column (Fig 1F,G) The elution
pro-file (Fig 1F) shows that apo-ACP has weaker affinity
and eluted with 190 mm NaCl, whereas holo-ACP
was eluted with 200 mm NaCl MALDI-TOF MS yielded molecular masses of 9418.845 Da (calculated 9417.65 Da) and 9752.831 Da (calculated 9751.65 Da) for apo-ACP and holo-ACP, respectively [Figs 2Aa,b]
Dynamic light-scattering studies of PfACP Apo-ACP and holo-ACP yielded hydrodynamic radii
of 1.95 ± 0.05 nm and 1.9 ± 0.1 nm, respectively, confirming that they have a single species over the entire experimental concentration range [Fig 1Ha,b]
Sucrose density gradient sedimentation
In sucrose density gradient sedimentation experiments, both apo-ACP and holo-ACP were detected between fractions 12 and 18 [Fig 1Ia–d]; apo-ACP showed a major peak in fraction 15, whereas holo-ACP showed
a peak in fraction 14 [Fig 1Ia–d] From the calibration curve of the sucrose density gradient, the estimated molecular masses of apo-ACP and holo-ACP mono-mers are 16.75 kDa and 21 kDa, respectively The dimeric peak of holo-ACP was found in major amounts in fraction 17 when sucrose density gradient sedimentation for holo-ACP under oxidizing condi-tions was performed
Fig 2 (A) Molecular mass determination of apo-ACP and holo-ACP Molecular masses of apo-ACP and holo-ACP were determined with an Ultra Flex TOF ⁄ MALDI-TOF mass spectrometer (a) Mass spectrum of holo-ACP single major peak (9752.83 Da) [holo-ACP (calculated 9751.65 Da)] (b) Mass spectrum of apo-ACP, showing a single major peak of molecular mass 9418.84 Da [apo-ACP (calculated 9417.65 Da)] (B) Secondary structure of ACP The secondary structures of both apo-ACP (O) and holo-ACP (h) were determined by far-UV
CD spectroscopy CD spectra show the presence of only a-helices as the secondary structure element in both apo-ACP and holo-ACP (C) Guanidine hydrochloride-induced transitions for holo-ACP (.) and apo-ACP (O) at 30 C as monitored by CD at 222 nm The proteins were in buffer containing 5 m M NaCl ⁄ Pi , 100 m M NaCl and 2 m M dithiothreitol, plus the indicated concentration of guanidine hydrochloride The solid lines indicate the best-fit values for each curve (D) Refolding of ACP Far-UV CD spectra of native apo-ACP (d) and holo-ACP (.), and
refold-ed apo-ACP (O) and holo-ACP (,) The CD spectra of native and refoldrefold-ed PfACP overlap, which shows that isothermal denaturation of PfACP
is completely reversible (E) Fluorescence spectra of ACP at 25 C The samples were excited at 280 nm, and emission spectra were recor-ded from 295 nm to 350 nm No change of emission maxima from 305 nm was observed for denatured holo-ACP (,) and apo-ACP (d), and native holo-ACP (.) and apo-ACP (O) For both forms, the only change in fluorescence intensity was observed upon denaturation (F) Refold-ing of ACP Both apo-ACP and holo-ACP denatured in 6 M guanidine hydrochloride were refolded by dialysis (4 · 1000 mL) against 5 m M
Na ⁄ K phosphate (pH 6.5), 100 m M NaCl and 2 m M dithiothreitol Both apo-ACP and holo-ACP are completely refolded after complete dena-turation with 6 M guanidine hydrochloride, as revealed by the equal mobilities of the native and refolded proteins on 12% native PAGE Lane 1: native apo-ACP Lane 2: refolded apo-ACP Lane 3: native holo-ACP Lane 4: refolded holo-ACP There is no degradation of holo-ACP
to apo-ACP during the refolding process (G) AcpS assay Both native and refolded apo-ACPs were used as substrates for the AcpS assay,
to convert them to lauroyl-ACP The reaction mixtures are checked on 12% native PAGE Lane 1: refolded apo-ACP Lane 2: refolded holo-ACP Lane 3: reaction mixture with native apo-ACP Lane 4: reaction mixture with refolded apo-ACP (H) AAS assay Both native and refolded holo-ACPs were used as substrates for AcpS assay, to convert them to lauroyl-ACP The reaction mixtures were checked on 20% conformation-sensitive PAGE with 5 M urea Lane 1: refolded holo-ACP Lane 2: reaction mixture with refolded holo-ACP Lane 3: reaction mixture with native holo-ACP (I) Confirmation of AcpS reaction product The molecular masses of apo-ACP and holo-ACP were determined with an Ultra Flex TOF ⁄ MALDI-TOF mass spectrometer (a) Mass spectrum of refolded apo-ACP, showing a single major peak of molecular mass 9420.639 Da [apo-ACP (calculated 9417.65 Da)] (b) Mass spectrum of reaction mixture with native apo-ACP, showing a single major peak of molecular mass 9943.976 (Da) [lauroyl-ACP (calculated 9935 Da)] (c) Mass spectrum of reaction mixture with refolded apo-PfACP, showing a single major peak of molecular mass 9941.941 (Da) [lauroyl-ACP (calculated 9935 Da)] (d) Mass spectrum of refolded holo-ACP, showing a single major peak (9759.106 Da) [holo-ACP (calculated 9751.65 Da)].
Trang 5C
E
D
190 –1.8e+6
–2.5e+6 200 0
0
1
[GdnHCI] M
Wavelength (nm)
Wavelength (nm)
–2.0e+6 –1.5e+6 –1.0e+6 –5.0e+5 0.0
5.0e+5
Wavelength (nm)
1
holoACP
C12-ACP
a
b
c
9941.941 9943.938 9920.639
9758.108
d
1.0e+6
1.5e+6
2.0e+6
2.5e+6
3.0e+6
3.5e+6
4.0e+6
4.5e+6
–1.6e+6 –1.4e+6 –1.2e+6 –1.0e+6 –8.0e+5 –6.0e+5 –4.0e+5 –2.0e+5 0.0 2.0e+5 4.0e+5
a
b
A
Trang 6Biophysical studies with ACP
The conformations of both apo-ACP and holo-ACP at
pH 6.5 have been determined by far-UV CD
spectros-copy Wavelength scans from 190 nm to 250 nm show
that both forms of ACP have predominantly a-helices,
which is in accordance with known ACP structures
(Fig 2B) [24]
The conformational stability of holo-ACP and
apo-ACP was determined by chaotrope-dependent
unfold-ing at different temperatures The reversibility of the
isothermal denaturation of ACP was shown by the
return of CD and fluorescence signals upon
refold-ing after complete denaturation with 6 m guanidine
hydrochloride (Fig 2D,E) It was also found that the
refolded apo-ACP and holo-ACP have mobilities
com-parable to that of the nondenatured wild-type
counter-parts on 12% native PAGE, which further confirms
the reversibility of the transition (Fig 2F) Unfolding
experiments monitored by the change in mean residue
ellipticity at 222 nm [h]222 demonstrated that both
forms undergo a two-state unfolding transition
(Fig 2C) Unfolding transitions were also monitored
by following the tyrosine fluorescence at 305 nm upon
excitation at 280 nm This was done because PfACP is
devoid of tryptophan but contains one tyrosine
resi-due Fluorescence emission spectra of fully denatured
PfACPs showed no shift in their emission maxima
from 305 nm, but there was a substantial increase in
fluorescence intensity (Fig 2E) The isothermal
dena-turation probed both by the far-UV CD and
fluores-cence coincided well for both apo-ACP and holo-ACP, indicating that their denaturation process is a two-state reaction (Fig 3A,B)
Several representative guanidine hydrochloride-depe-ndent denaturation experiments were done in the range 15) 60 C The results of the analysis of these curves using the linear extrapolation model (LEM) are shown
in Tables 1 and 2 and Fig 4A–C There is a slight temperature dependence of the Cm value (Fig 4A), but the m value is independent of temperature, as per the expectations of LEM (Fig 4C) The mean
m values for the apo form and the holo form are ) 1.64 kcalÆmol)1Æm)1 and ) 1.97 kcalÆmol)1Æm)1, respectively The DGwater values showed strong tem-perature dependence, with maximum stability at 30C (Fig 4B)
Unfolding experiments were also monitored by the change in fluorescence anisotropy of the single tyrosine residue at the C-terminus of ACP The isothermal denaturation probed by fluorescence anisotropy corre-lated well with the far-UV CD and fluorescence quenching studies for both apo-ACP and holo-ACP, further indicating that their denaturation process is
a two-state reaction (Fig 3C) These data further indicate that apo-ACP has lower stability than holo-ACP
Acyl-ACP synthesis assay with apo-ACP
E coli holo-ACP synthase (AcpS) has been cloned and expressed in the laboratory as a His-tagged protein
C
Fluorescence (Fit) 1.0
0.8
0.6
0.4
0.2
0.0
1.0 0.8 0.6 0.4 0.2 0.0
GdnCI concentration [ M ]
GdnCI concentration [ M ]
0 0
1 1
[GdnHCI] M
Fluorescence (Fit) Fluorescence (Expt)
Fluorescence (Expt)
Fig 3 Comparison of guanidine hydrochlo-ride-induced transitions (A) Comparison of guanidine hydrochloride-induced transitions
of apo-ACP at 30 C as monitored by far-UV
CD at 222 nm (d) and tyrosine fluorescence
at 305 nm (O) (B) Comparison of guanidine hydrochloride-induced transitions of holo-ACP at 30 C monitored by CD at
222 nm (m) and tyrosine fluorescence at
305 nm (n) (C) Comparison of guanidine hydrochloride-induced transitions of apo-ACP (O) and holo-ACP (d) at 30 C monitored by fluorescence anisotropy of tyrosine fluorescence at 305 nm.
Trang 7E coliAcpS thus expressed has broad substrate
specif-icity It utilizes apo-ACP and various acyl-CoAs as
substrates to give corresponding acyl-ACPs This
prop-erty of AcpS was utilized to check the extent of the
reversibility of folding of apo-ACP An acyl-ACP
synthesis assay clearly showed that both native and
refolded apo-ACP are equally and quantitatively con-verted to lauroyl-ACP (Fig 2G,I)
Acyl-ACP synthesis assay with holo-ACP
E coli acyl-ACP synthase (AAS) utilizes holo-ACP as
a substrate and converts it to acyl-ACP E coli AAS also utilizes fatty acids with various chain lengths as substrates, producing the corresponding acyl-ACPs This property of AAS was utilized to check the correct refolding of holo-ACP An acyl-ACP synthesis assay clearly showed that both native and refolded holo-ACP are partially converted to lauroyl-holo-ACP (Fig 2H) The band intensity indicates that the extent of conver-sion of refolded holo-ACP is comparable to that of its native counterpart
Table 1 Parameters obtained from the fit of isothermal unfolding data by fitting Eqs (1)–(5) ND, not determined.
Temperature (K)
Cm holo-ACP ( M )
Cm apo-ACP ( M )
m-value holo-ACP (kcalÆmol)1Æ M )1)
m-value apo-ACP (kcalÆmol)1Æ M )1)
DGwater holo-ACP (kcalÆmol)1)
DGwater apo-ACP (kcalÆmol)1)
Table 2 Average Cm and m for apo-ACP and holo-ACP in the
experimental temperature range.
Protein
Average Cm ( M )
Average m (kcalÆmol)1Æ M )1)
C
280 3 4
290 300 Temperature [K]
1 2 3
Temperature [K]
280
0
–1
–2
–3
–4
290 300 Temperature [K]
310 320 330 340
Fig 4 Effects of temperature on the
best-fit Cm(A), m value (B) and DGwater(C) for
the guanidine hydrochloride denaturation
curves at 10 different temperatures for
holo-PfACP (d) and apo-holo-PfACP (O) The solid line
indicates the best-fit values of Cm (A) and
DGwater
Trang 8Overexpression of several ACPs including E coli ACP,
has been reported to be toxic for E coli The active
forms of ACP, the holo-ACPs, are even more difficult
to overexpress in E coli, presumably due to
ineffi-ciency of the E coli holo-ACP synthase in modifying
the ACP in vivo, resulting in the production of mostly
apo-ACP, which has been shown to have inhibitory
effects on E coli growth [26] In our previous studies
[24,27] and also in this study, we have standardized
conditions for overexpression of PfACP in E coli with
high yield (30–35 mgÆL)1 E coli culture) PfACP
appears to be converted to ACP by E coli
holo-ACP synthase These studies also show that Pfholo-ACP is
utilized as a substrate by E coli holo-ACP synthase
and ACP phosphodiesterase Although optimization of
culture conditions yields mostly holo-ACP [27], we
show that both holo-ACP and apo-ACP can be
over-expressed together and purified to homogeneity
The secondary structures of apo-ACP and
holo-ACP, as determined by far-UV CD spectroscopy, have
shown the predominance of a-helices and a very low
percentage of b-pleated sheet in PfACP Analysis of
CD spectra using k2d analysis software (http://www
embl-heidelberg.de/andrade/k2d.html) has shown
that both apo-ACP and holo-ACP contain 56%
a-helix, 10% b-pleated sheet and 34% random coil in
their secondary structure, demonstrating that PfACP
has a similar secondary structure to the other ACPs
and ACP-like domains [14–22] Hence, detailed
bio-physical characterization of PfACP could serve as a
prototype for determining the conformational stability
of other ACPs The NMR structure of PfACP has
been solved recently [24,25,27]; this study augments the
structural data and elucidates the interactions
respon-sible for the conformational stability of PfACP
The size exclusion chromatography profile of PfACP
showed that the apparent molecular mass of PfACP
monomer is 25 kDa, whereas the actual molecular
masses of apo-ACP and holo-ACP are 9.4 kDa and
9.7 kDa, respectively, as is evident from MS studies
The dynamic light-scattering experiments showed
both apo-ACP and holo-ACP exist as single species
[Fig 1Ha,b] The sucrose density gradient
sedimenta-tion showed that the apparent molecular masses of
apo-ACP and holo-ACP are 16.75 kDa and 21 kDa,
respectively The glutaraldehyde crosslinking
experi-ment (Suppleexperi-mentary material) showed that both
apo-ACP and holo-apo-ACP exist as monomers in solution
under reducing conditions, and that holo-ACP
parti-ally forms a dimer by forming a disulfide bridge
invol-ving the SH group of its pantothenyl moiety under
nonreducing conditions only Therefore, the increased apparent molecular masses of monomeric apo-ACP and holo-ACP are not due to oligomerization but are perhaps due to their relatively higher hydrodynamic radii
The chaotrope-induced unfolding was almost fully reversible in both forms of the protein Removal of the perturbation makes the protein regain its native form The unfolding reactions of both forms are simple two-state processes, A´U The transitions monitored by the two probes (far-UV CD and tyrosine fluorescence
at 305 nm) that report the secondary and tertiary structures of the protein were completely superimposa-ble, thus proving it to be a two-state process [38] Both native and refolded PfACP have comparable mobilities
on 12% native PAGE, and both of them are equally utilized as substrates by E coli AcpS and AAS, which further shows the complete refolding of PfACP The fact that it is a small protein with a few hydrophobic residues perhaps explainsd a lack of nonspecific aggre-gation and the ease with which it can be reversibly unfolded by the chaotrope guanidine hydrochloride Detailed analyses of the stability curves obtained
by chemical denaturation are consistent with the LEM The chaotrope-induced equilibrium unfolding
of PfACP, followed by fluorescence, fluorescence anisotropy and far-UV CD, showed no evidence for the existence of stable intermediates, substantiating the assumption of a simple two-state transition (Fig 3A,B) Guanidine hydrochloride-induced dena-turation experiments on PfACP are consistent with the LEM of protein unfolding [29]
It is apparent from the solution denaturation studies that the holo form of the protein has greater stability than the apo form The differences in the unfolding thermodynamic parameters of the two forms are given
in Tables 1–3 In the entire experimental regime, it is seen that the holo form presents better stability than the other form (Fig 4C) The DG of stability is on an average 20% greater in the case of the holoprotein as compared to the apoprotein Similarly, the Cm of the holo form always lies above the apo form at all tem-peratures at which the experiments were conducted The values of Tg, DHg and DCp for the respective
Table 3 Thermodynamic parameters of holo-ACP and apo-ACP analyzed on the basis of stability curves drawn for fitting Eqn (6).
Protein
DHg (kcalÆmol)1)
DCp (kcalÆmol)1ÆK)1) Tg(K) Holo-ACP 53.08 ± 1.09 1.18 ± 0.11 343.16 ± 1.48 Apo-ACP 49.52 ± 1.58 1.02 ± 0.13 337.20 ± 1.82
Trang 9proteins were obtained from the fit of Eqn (6) in
Experimental procedures It is interesting to note that
although the values of DHg and DCp are comparable,
the Tg values of the proteins vary slightly; there is a
difference of almost 6C between the Tg values of the
proteins, the value being higher for the holo form
There have been contrasting reports about the
inter-action of the 4¢-PP group with the polypeptide
back-bone and its effect on the stability of holo-ACP The
average major conformation of the holo-ACP NMR
structure was analyzed for ligand–protein contacts
[30] (http://ligin.weizmann.ac.il/cgi-bin/lpccsu/LpcCsu
cgi) to determine the contacts between the 4¢-PP group
and polypeptide backbone The greater stability of the
holo form may be due to the fact that the 4¢-PP group
(structure shown in Fig 5A) makes a number of
favo-rable contacts with the amino acid residues at the
sur-face of the protein by virtue of the presence of several
hydrogen bond donors and acceptors in it
Further-more, there are several hydrophobic interactions that
hold the structure firmly Closer scrutiny of Fig 5C
reveals that whereas most of the surface of the
holo-ACP is lined by charged residues (shown in blue⁄ red),
the interface between the cofactor and the protein is
predominantly hydrophobic (represented by gray)
Interestingly, a few constructive interactions between
carbon and oxygen atoms were also detected at the
4¢-PP–protein interface These favorable interactions
might result from atypical CH–O hydrogen bonds
According to a report by Jiang et al., these atypical
hydrogen bonds play an especially crucial role in
sta-bilizing the protein–protein interface [31] All of the
interactions reported in Fig 5 in the range between 3
and 6.0 A˚ are identical in the major and minor
frac-tions of the holoprotein in solution In fact, the
archi-tecture of the bound cofactor in holo-PfACP is like an
arch, where the proximal and the distal ends are closer
to the protein and the middle portion is away from it
Consequently, we notice that although there are a
sub-stantial number of attractive interactions between the
4¢-PP group and the protein, they are balanced in a
subtle manner This may be because the free
move-ment of the 4¢-PP group is required for its biological
activity, and hence extensive contacts between the
4¢-PP group and the peptide backbone may not be a
desirable property, which in turn explains the delicate
manner in which the stability of holo-ACP is
regula-ted The differences in stability between the two
pro-teins (as indicated by the values of DGwater, Cm and
Tg) hence arise from the changes in the surface of the
protein because of the interactions of the cofactor with
the protein This is further substantiated when one
overlays the two forms of the protein (Fig 5B) The
rmsd in this case happens to be 0.20 A˚ Again, differ-ences are seen mostly in the loop regions where the 4¢-PP binds the protein
In summary, our studies demonstrate that holo-ACP has higher stability than apo-ACP This work also shows that the 4¢-PP group makes some contacts with the polypeptide that stabilize the holo-ACP structure
Experimental procedures
Chemicals and reagents
Imidazole, kanamycin, dithiothreitol, guanidine hydrochlo-ride, thrombin from bovine plasma, sinapicnic acid, trifluor-acetic acid, sucrose and SDS⁄ PAGE reagents were obtained from Sigma-Aldrich (St Louis, MO) Media components were obtained from Difco (Franklin Lakes, NJ) All other chemicals used were of analytical grade All enzymes were obtained from NEB (Ipswich, MA), MBI Fermentas GmbH (St Leon-Rot, Germany) and Promega (Madison, WI)
Strains and plasmids
E coli DH5a cells (Gibco BRL, Carlsbad, CA) were used for cloning of the gene pET-28a(+) vector (Novagen, Darmstadt, Germany) and E coli BL21(DE3) cells (Nov-agen) were used for the expression of PfACP
Cloning and expression of PfACP in E coli
PfACP was cloned as described previously [27] The plasmid containing PfACP was transformed into E coli BL21(DE3) cells (Novagen) The culture was grown at 37C with vigor-ous shaking (160 r.p.m.) in LB broth (Difco) to a cell den-sity of D600 1 The culture was then induced with 0.2 mm isopropyl-b-d-thiogalactopyranoside, and further incubated
at 37C for 4 h to a D of 2.5 After induction, cells were harvested at 5000 r.p.m for 10 min, and the resultant pellet was stored at) 70 C if not used immediately
Purification of holo-ACP and apo-ACP
All the purification steps were carried out at 4C unless otherwise indicated
The cell pellet was resuspended in lysis buffer containing
20 mm Tris⁄ HCl (pH 8.5), 200 mm NaCl, and 10 mm imi-dazole Lysozyme (2 mg) was added, and the mixture was incubated on ice for 30 min Cells were disrupted using a probe-type ultrasonicator (Vibra-Cell; Sonics and Materials, Newtown, CT, USA) MgCl2and MnCl2were added to the lysate to final concentrations of 10 and 2 mm, respectively, and the mixture was incubated at 35C for 2 h [32] Cell debris was removed by centrifugation at 30 000 g for
30 min using a Sorvall RC5C PLUS (Thermo Fisher
Trang 10Scientific, Waltham, MA, USA) The supernatant obtained
was applied to an Ni–nitrilotriacetic acid metal affinity
col-umn [agarose resin; (Qiagen, Hildon, Germany)]
equili-brated with the lysis buffer The column was initially washed with column buffer (same as lysis buffer) The protein was eluted using a step gradient of 50 mm to 1 m
II
IV
C
A
B
I
I
III IV
Fig 5 Interactions of the 4¢-PP moiety with the holo-ACP protein The average major conformation was used for ligand–protein contact ana-lysis (A) The interacting atoms are labeled The green dotted lines indicate hydrophobic interactions, and the blue lines denote CH–O hydro-gen bonds The 4¢-PP group is linked to the protein by the Ser37 O-c atom Only residues 31–38 of the protein make extensive contacts with the cofactor For the sake of better understanding of the interactions, the entire figure has been divided into four parts (I, II, III and IV):
I, interactions with amino acids 30–33; II, interactions with amino acids 33–34; III, interactions with amino acids 35–37; IV, interactions with amino acids 37–38 (B) The overlay of the apo (green) and holo (orange) forms of the protein (rmsd ¼ 0.20 A˚) (C) Diagram showing the nat-ure of the surface in holo-ACP It should be noted that the protein has a greater number of charged exposed surface (indicated by blue ⁄ red) than hydrophobic ones The red line denotes the area in the protein that makes contact with the cofactor.